Optoelectronics Theory & Practice
http://www.archive.org/details/OptoelectronicsTheoryPractice
Texas Instruments
PDF
Texas-OptoelectronicsTheoryPractice text TEXAS INSTRUMENTS INCORPORATED
TEXAS INSTRUMENTS INCORPORATED
Optoelectronics:
Theory and
Practice
621, 36 OPT
SHORT LOAN fcailf
Texas Instruments Electronics Series
McGraw-Hill Book Company
Optoelectronics
TEXAS INSTRUMENTS ELECTRONICS SERIES
Applications Laboratory Staff of Texas Instruments Incorporated
DIGITAL INTEGRATED CIRCUITS AND OPERATIONAL-AMPLIFIER AND OPTO-
ELECTRONIC CIRCUIT DESIGN
Applications Laboratory Staff of Texas Instruments Incorporated
ELECTRONIC POWER CONTROL AND DIGITAL
TECHNIQUES
Applications Laboratory Staff of Texas Instruments Incorporated
MOS AND SPECIAL-PURPOSE BIPOLAR INTEGRATED CIRCUITS AND R-F POWER TRAN-
SISTOR CIRCUIT DESIGN
Applications Laboratory Staff of Texas Instruments Incorporated
POWER-TRANSISTOR AND TTL
INTEGRATED-CIRCUIT APPLICATIONS
CarrandMize MOS/LSI DESIGN AND APPLICATION
Crawford
MOSFET IN CIRCUIT DESIGN
Delhom
DESIGN AND APPLICATION OF TRANSISTOR SWITCHING CIRCUITS
The Engineering Staff of Texas Instruments Incorporated
CIRCUIT DESIGN FOR AUDIO, AM/FM, AND TV
The Engineering Staff of Texas Instruments Incorporated
SOLID-STATE COMMUNICATIONS
The Engineering Staff of Texas Instruments Incorporated
TRANSISTOR CIRCUIT DESIGN
Hdrtel
OPTOELECTRONICS: THEORY AND PRACTICE
Hibberd Hibberd
INTEGRATED CIRCUITS SOLID-STATE ELECTRONICS
The IC Applications Staff of Texas Instruments Incorporated
DESIGNING WITH TTL INTEGRATED CIRCUITS
Kane and Larrabee
CHARACTERIZATION OF SEMICONDUCTOR MATERIALS
Luecke,Mize,andCarr
SEMICONDUCTOR MEMORY DESIGN AND APPLICATION
Runyan
SEMICONDUCTOR MEASUREMENTS AND INSTRUMENTATION
Runyan
SILICON SEMICONDUCTOR TECHNOLOGY
Sevin
FIELD-EFFECT TRANSISTORS
Optoelectronics
Theory and Practice
Edited by Alan Chappell
Texas Instruments Ltd.
Original German Version: Volkmar Hdrtel
assisted by Eilhard Haseloff, Gerhard Jahn, and Gunther Suhrke Texas Instruments Deutschland GmbH
McGRAW-HILL BOOK COMPANY
New York St. Louis San Francisco Auckland Bogota
Dijsseldorf Johannesburg London Madrid Mexico
Montreal New Delhi Panama Paris Sao Paulo
Singapore Sydney Tokyo Toronto
.
Library of Congress Cataloging in Publication Data Main entry under title:
Optoelectronics.
(Texas Instruments electronics series)
Includes index.
1. Optoelectronics. I. Chappell, Alan.
TA1750.068
>21.38'0414
78-8021
ISBN 0-07-063 7^-5
L-
© Copyright 1978 by Texas Instruments Limited. All
rights reserved. Printed in the United States of
America. No part of this publication may be reproduced,
stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording, or otherwise, without the prior written permission
of Texas Instruments Incorporated.
34567890 HDHD 8543210
y
j»N No.
The circuits, sub-assemblies and procedures described in this book have been tested by Texas InstrumentsLtd. (T.I.L.); but this does not guarantee their reliability in
operation.
jftiii
CLASS No.
$2}
3t=>
2 25
o^t
Neither can T.I.L. give any warranty that these circuits etc. are not covered by patent rights of third parties.
G/i
T.I.L. reserves all rights in this book. Without express permission from T.I.L.
neither the book nor parts of it may be duplicated or distributed in any manner by
photocopies, microfilms or any other process. This applies also to the right of reproduction in public.
< IS* I shoc<-
Texas Instruments reserves the right to make changes at any time in order to improve design and supply the best product possible.
Information contained in this publication is believed to be accurate and reliable. However, responsibility is assumed neither for its use nor for any infringement of patent or rights of others
which may result from its use. No license is granted by implication
or otherwise under any patent or patent right of Texas Instruments
or others.
Design/typography David Muriel Presentation Unit
Foreword
Over the years progress in electronics has given birth to an ever increasing number of circuit components for a great variety of uses. The so called optoelectronic devices in
particular, which make use of the mutual interaction of radiation and the electronic structure of materials, have become widely
used in recent years. This has been brought about chiefly through improvements in the semiconductor manufacturing process which have enabled optoelectronic devices
to be used economically in many and varied
applications.
The purpose of this book is to provide both a theoretical and practical introduction to these optoelectronic components. It is
written primarily for engineers, technicians
and students who need to acquire the
background knowledge and the ability to use these devices. With this in mind particular attention has been paid to providing an abundance of practical hints and suggestions as well as the necessary theoretical background to enable users to
develop their own circuits and applications.
The Editors.
Contents
Page
1
1.1 1.2 1.3 1 .4 2 2.1 2.2 2.3
13
3
13
3.1
14
3.2
16
3.3
19 2 21 2.1
22 2.2 22 2.2.1 23 2.2.2 24 2.2.3 26 2.2.4 26 2.2.5 27 2.3 27 2.3.1 28 2.3.2 29 2.3.3
29 2.3.4
33 3 35 3.1 36 3.2 36 3.2.1 37 3.2.2 38 3.3
40 3.3.1
Introduction
Physics of Optical Radiation Optical radiation and light Basic definitions
The quantum nature of radiation The dualism of waves and particles Wavelength and propagation speed Radiation and luminescence phenomena Atomic and band model Luminescence, Fluorescence, Phosphorescence Luminescence phenomena in semiconductors,
injection luminescence Photoelectric effect External photoeffect Internal photoeffect Junction photoeffect
Principles of calculation in radiation physics and optics Radiant flux - Luminous flux, Radiant energy Light quantity Parameters related to the radiation source Radiant emittance - Luminous emittance Radiant intensity - Luminous intensity
Radiance -- Luminance
Units of luminance Radiant efficiency and luminous efficiency Parameters related to the receiver
Irradiance - Illuminance Irradiation - Light exposure Relationship between irradiance or illuminance and the
reflected radiance or luminance
Spectral radiation - physical units
Laws of radiation
Solid angle Lambert's radiator
Fundamental law of photometry Lambert's cosine law Calculation of radiation with small surface radiators and surface receivers Inverse square law
vu
43 4 45 4.1 45 4.1.1 46 4.1.2 46 4.2 46 4.2.1 47 4.2.2 48 4.2.3 50 4.2.4 51 4.2.5 51 4.2.6 51 4.2.7
57 5 59 5.1 59 5.2 60 5.3 62 5.3.1 64 5.3.2 67 5.3.3
69 5.4
69 5.4.1
69 5.4.2
70 5.4.3
74 5.5
74 5.5.1
76 5.5.2
77 5.6 77 5.6.1
80 5.6.2
83 6 85 6.1 85 6.2
86 6.3 86 6.4 87 6.4.1 88 6.5
Laws of Radiation from a Black Body Black and "Non-black" bodies
Black bodies Non-black bodies
Laws of radiation Planck's law of radiation Stefan-Boltzmann law Wien displacement law
Emittivity KirchhofFs law Radiation isotherms
Reduced law of radiation
General and Photometric Evaluation of Radiation
The human eye
Optical sensitivities Photometric evaluation of radiation
C Determination of the conversion constants C and
Photometric radiation equivalent Photometric radiation equivalent of the photopic sensitivity of the eye
Calculation of the photometric radiation equivalent K
for different radiation sources
K Calculation of the photometric radiation equivalent
of a Planck radiator with the temperature laid down for
the definition of the candela
K Calculation of the photometric radiation equivalent A for standard light Calculation of the photometric radiation equivalent K
of the luminescence radiation from light-emitting diodes Conversion of radiometric units into photometric, photopic units Conversion of radiometric units into photometric units for Planck radiation Conversion of radiometric units into photometric units for the luminescence radiation from light-emitting diode
Actinic value
Actinic value of a radiation for photodetectors and for
the human eye
Actinic value of Planck radiation and luminescence radiation for the eye as a photodetector
Interaction between optical radiation and matter Absorption, transmission and reflection factors Spectral transmission factor and spectral absorption
factor Scatter Reflection of radiation Spectral reflectivity
Basic laws of absorption, attenuation and scatter
Vlll
90 6.6 96 6.7
Absorption and transmission spectra Refraction
101 103 103 104 104 104 104 104 105 105 105 107 107 109 109 109 109 109 110 110 111 113 113 113 114 114
7
7.1 7.1.1 7.1.2 7.1.3 7.2 7.2.1 7.2.2 7.3 7.3.1 7.3.2 7.3.3 7.3.4 7.3.5 7.3.6 7.3.6.1 7.3.6.2 7.3.6.3 7.3.6.4 7.4 7.5 7.6 7.6.1 7.6.2 7.6.3 7.6.4
Radiation sources Natural radiation sources
The Sun The Moon Clouds
Artificial radiation sources
Open fire
Filament lamps Luminescence radiators Hot gases Discharges through gases Discharge lamps Discharge tubes
Xenon lamps
Metal-vapour lamps Sodium-vapour lamps Low-pressure mercury-vapour lamps High-pressure mercury-vapour lamps Fluorescent lamps Mixed-light lamps Flash-discharge tubes in photography Luminescent diodes
Silicon-doped GaAs diodes GaP diodes Zinc-doped GaAsP diodes Considerations on quantum efficiency
117 119 120 120 121
8
8.1 8.2 8.2.1 8.3
Photodetectors Photodetectors with external photoeffect Photodetectors with internal photoeffect Photoconductors or photoresistors Junction photodetectors
123 9
125 126
9.1 9.1.1
130 132 132 134 134
9.2 9.2.1 9.2.2 9.2.3 9.2.4
134 9.2.5
136 9.3
Parameters of IR Detectors and Junction Photodetectors Quantum efficiency of junction photodetectors Quantum efficiency and photocurrent gain of avalanche photodiodes Spectral sensitivity of junction photodetectors Scottky-Barrier PIN photodiodes Planar diffused Si photodiodes Photodiodes by the CDI process Effect of temperature on the spectral sensitivity of photodiodes
Possibilities for shifting the spectral sensitivity of photodiodes Evaluation of radiation by non-amplifying
junction photodetectors for monochromatic radiatk
137 9.4
1 39 9.5
144 9.6
145 9.7
151 153 154
9.8 9.8.1 9.9
Evaluation of radiation by non-amplifying junction photodetectors for chromatic radiation (mixed radiation) Evaluation of radiation by amplifying junction photodetectors Area-dependent sensitivity values for junction photodetectors Actinic values of IR luminescence radiation from
GaAs diodes for silicon junction photodetectors
Dark current of junction photodetectors Dark current of avalanche photodiodes Sensitivity parameters of IR detectors
159 161
167 168
170
173 176
176 181 184 186 191
10
10.1
10.2 10.2.
10.2.2
10.2.3 10.3
10.3.1 10.3.2 10.4 10.5 10.6
Parameters common to emitters and receivers
Evaluation of the radiation and receiving characteristics of optoelectronic components Optical tolerances of optoelectronic components Effect of wafer centering, lens quality, distance from lens to wafer, refractive index of epoxy resin
and shape of dome and case
Effect of internal case reflections and wafer geometry Tolerance levels which occur Coupling characteristics, transfer ratios and contrast current ratios of short optical links Coupling characteristics and transfer ratios
Contrast ratio Half-power and half-value points
Dynamic data Reliability of optoelectronic components
205 207 213 213 215 217
11 11.1 11.2 11.2.1 11.2.2 11.2.3
217 11.2.4
217 11.2.5
218 11.2.6
218 11.2.7
220 11.3 223 11.4 224 11.5 226 11.6 228 11.7
Parameters of Luminescence diodes
Quantum efficiency
Thermal calculations
Basic principles Determination of the thermal resistance
Loss power calculation for luminescence diodes in plastic packages Loss power calculation for luminescence diodes in metal cans with infinitely large heat-sinks Loss power calculation for luminescence diodes in metal cans without heat-sink Loss power calculation for luminescence diodes in metal cans with heat-sink
Calculation of the maximum permissible forward
current
Radiant power Radiant efficiency
Spectral radiant efficiency Electrical parameters Pulse operation
231 233 233
235
237
238
240
240 242
242
245
12 12.1 12.2 12.3 12.4 12.5 12.6
12.6.1 12.6.2 12.6.3 12.7
Radiation measurements General Considerations Measurement of colour temperature of standard
A light
Measurement of radiant power with thermal photodetectors
Measurement of radiant power of standard light A
with the thermopile
Measurement of irradiance of standard light A with
Si photodetectors Measurement of a luminescence diode radiation with Si photodetectors General measurement problems Measurement of relative spectral sensitivity with a
monochromator Measurement of irradiance of a luminescence diode
radiation with Si photodetectors Measurement of the total radiant power of a luminescence diode
247 249 250 252 252 254 259 260
13 13.1 13.2 13.3 13.4 13.5 13.6 13.7
Optoelectronic couplers Direct optoelectronic couplers Reflection optoelectronic couplers Optocouplers with non-stationary source emission Simple examples of optocouplers Optocouplers with lenses Optocouplers with unmodulated optical radiation Optocouplers with modulated optical radiation
263 265 265 269
14 14.1 14.2 14.3
Operation of luminescence diodes with direct current Operation through series resistances Operation from constant-current sources Drive with logic circuits
273 275 276 276 276 277 280
285 288 288 288
288
291 293
15 15.1 15.2 15.2.1 15.2.2 15.2.3 15.2.4
15.2.5 15.3 15.3.1 15.3.1.1
15.3.1.2
15.3.1.3 15.3.2
Photodetector circuits Principle of operation Detector circuits for two-pole junction photodetectors Direct relay control with phototransistors Photo-Darlington circuits Control of thyristors and triacs with phototransistors Driving of transistor- and operational amplifiers with phototransistors, photodiodes and photocells Driving of multivibrators with phototransistors Detector circuits for three-pole phototransistors
Operating modes of three-pole phototransistors Photocell and photodiode operation of the TIL 81
phototransistor
Advantages of the TIL 81 phototransistor in photocell and photodiode operation Phototransistor operation of the TIL 81
Driving of amplifiers with three-pole phototransistors
XI
294 298
302 304
15.3.3 15.4
15.5 15.6
Photo-trigger circuits and photomultivibrators Simple optocouplers with filament lamp and two-pole phototransistors Logic circuits with phototransistors
Photodetector circuits to drive TTL circuits
309 311
312
313 318 320
16 16.1
16.2
16.2.1 16.2.2 16.3
Modulated transmitters with luminescence diodes The simplest modulator circuits for luminescence diodes Sine-wave-modulated transmitter with luminescence diodes Modulation operation with bias voltage sources Modulation operation with constant-current sources Pulse-modulated transmitter with luminescence diodes
325 17 327 17.1 328 17.2
Photodetector circuits for modulated radiation Circuits with phototransistors Circuits with the TIL 81 photodetector as photodiode and as photocell
335 18
337 18.1 338 18.2
Practical measurement of the photocurrent sensitivity of Si phototransistors Theoretical circuits of test equipments Test equipment for measurement of the relative spectral sensitivity of phototransistors
341 19
343 344 345 346
19.1 19:2 19.3 19.4
Light measurement with Si phototransistors in
electronic flash units Principle of an electronic flash unit
Types of exposure control
Circuit of a flash unit Si phototransistors for the automatic flash exposure control
349 351 352 352 354 355 355 357
20
20.1 20.2 20.3 20.4 20.5 20.6 20.7
Circuits with light-emitting diodes Simple indicators Diode tester Logic tester Polarity and voltage tester Large-format seven-segment display unit Analogue indication of digital values
Analogue measuring instruments with LED indication
361 21 363 21.1 365 21.2 368 21.3 371 21.4 374 21.5
Numeric and alphanumeric display units Seven-segment display units Multiplex operation of display units Numerical display units with integrated logic Monolithic display units 5 x 7-point matrix display units
xu
38 1 22 383 22.1 384 22.2 385 22.3 387 22.4
Direction-dependent photocell units Principle of operation Direction-dependent counter Direction-dependent optocouplers
Digital control knob
389 23 391 23.1 394 23.2
Optoelectronic rangefinder Phase measurement as a measuring principle Practical circuit of the rangefinder
399 401 401
401 402 403 403 404 404 406
408 409 410
24
24.1 24.2
24.2.1 24.2.2 24.2.3 24.2.3.1 24.2.3.2 24.3 24.4
24.5 24.6 24.7
Data transmission with optocouplers Interference on transmission links Construction and characteristics of optoelectronic couplers Current transfer ratio Mechanical construction Dynamic performance Photodiode operation Phototransistor operation Simple transmission links Improvement of the switching performance of optocouplers Optocouplers in photodiode operation Duplex operation with optocouplers
Common-mode suppression of optocouplers
413 25 413 25.1 416 25.2 418 25.3
Light exposure switch for photographic enlargers Principle of construction of a light exposure switch The timing system Practical circuit of the light exposure switch
419
421 421 421 423 425 425 426 426 428 429 429 431 431 433
26
Optoelectronic couplers as switches for analogue
signals
26.1 26.2
Semiconductor switches and potential isolation The phototransistor as a switch in the optocoupler
26.2.1 Steady-state performance
26.2.2 Dynamic performance
26.3 26.3.1
Application of optocouplers in a digital voltmeter
Voltage measurement method used
26.3.2
Practical circuit of the digital voltmeter
26.3.2.1 Analogue section
26.3.2.2 Digital section
26.3.2.3 Voltage converter
26.4
D.C. voltage amplifier with chopper
26.5
Line tester
26.5.1
Test principle
26.5.2
Practical circuit of the line tester
435
Index
Xlll
Introduction
The term "Optoelectronics" is today
understood to mean the production,
utilisation and evaluation of electromagnetic
radiation in the optical wavelength range
and its conversion into electrical signals.
Two basic components are needed -- a
radiation source as a transmitter and a
photoelectric converter as a receiver.
Components which emit or are
sensitive to radiation in the UV, IR and
visible range are defined as "optoelectronic
components". Optoelectronics is today
classified as a subsidiary area of
telecommunications, while its origins go
back into the last century. Thus, in 1873,
Smith discovered the change in conductivity
when selenium is irradiated, and in 1887 Hertz discovered the effect named after
him, by which the spark discharge of a
spark gap starts at lower voltages under
UV radiation. In 1888, Hallwachs found the
effect which bears his name, that
negatively-charged metal plates lose their
UV charge under
radiation, but positively-
charged plates do not. Numerous further
examples are named in the literature.
the 1N2175 silicon photodiode. Through
continuous further developments, TI have also provided the largest range of optoelectronic components, as is shown by a range of over 125 standard products, which take account of the most widely-varying needs of the user.
As the following list shows, optoelectronic components are used today in most branches of industry and areas of daily life.
Power engineering, data-processing, machine control, heat engineering, the automobile industry, space systems, sound and video recording, photography, consumer electronics, medicine, telecommunications, environmental protection, domestic
appliances etc.
The fields of application can be divided into
six groups:
1
Measuring, monitoring, evaluation, control and testing of given light sources.
With the introduction of transistors in the fifties, semiconductor optoelectronics
first made use of the radiation-dependence, which was undesirable in normal use, of ordinary diodes and transistors, for
radiation receivers. After further research work, semiconductor radiation sources were also manufactured in the early sixties.
As the largest semiconductor manufacturer in the world, Texas Instruments recognised at an early stage that optoelectronic semiconductor products will find widespread application. TI carried out pioneer work in the research, development and manufacture of optoelectronic components. Thus, as long ago as 1957, Texas Instruments introduced the first solar cells and in 1959
Optoelectronic devices with unmodulated
optical radiation.
Optoelectronic devices with modulated
optical radiation.
Optoelectronic devices for alphanumeric
displays.
Optoelectronic devices for recording or transmission of images.
Optoelectronic devices for image reproduction.
xv
In group 1 , a radiation source is defined as
a natural, existing light source, if its spectral emission distribution lies mainly in the
visible range. Such light sources include, for example, the sun, artificial light or naked flames. Fields of application for optoelectronic components include illumination meters in photometry, exposure meters, illumination switches, flash time controllers and flash release in photography, dust density meters for environmental protection, fog density meters and parking-light switches in transport, flame monitors and combustion monitoring devices in heating engineering, monitoring devices on sanitary installations, twilight switches for monitoring the lighting of streets and shop windows, brightness controls on television receivers, in relation to the ambient lighting, brightness regulators in lighting, inspection and control equipment, machine tool controls for the measurement of lengths, positions and angles, revolution counters for the control of motors and converters to transform radiation and light energy into
electrical energy.
Group 2 comprises optoelectronic devices with unmodulated optical radiation. In these, the non-directional emission from an optical
radiation source is usually converted into a
parallel beam and directed onto a photosensitive receiver. The receiver converts the
incident radiant energy into a DC signal
which is processed further and evaluated in
DC amplifiers. This describes the principle
of the direct light-beam or optoelectronic coupler. If the receiver receives the transmitted radiation through a mirror or
reflector, then we have a reflection coupler.
This group includes couplers for production control, the counting of individual items, as safety devices in machines, as punched card and tape readers and code and text readers, as distance and angle detectors in control devices and for the detection of pointer positions in measuring instruments.
Group 3 is concerned with radiation which is modulated_at the transmitting end. The receiver converts this modulated radiation
AC into an
signal. In other respects, the
principle of Group 2 is retained. The
AC receivers arc designed as
amplifiers and
can therefore be considerably more
sensitive and have greater temperature-
DC stability than
amplifiers.
The applications cover those of Group 2 and additional new areas such as optical sound recording and reproduction, optical telephony, remote controls, garage door
openers, range-finders, alarm systems, signal
transmission with electrical isolation and AC
telegraphy.
Optoelectronic indicating components, indicating units and displays, which radiate in the visible range, are classified in Group 4. In the simplest form, filament lamps, discharge lamps or light-emitting diodes (LEDs) are used for operational and warning indications. In general terms, a display is a device for the representation of numerical and alphanumeric information. These include, among others, mechanical indicating units, cold cathode indicator tubes, seven- or fourteen-segment indicator units, projection displays and alphanumeric displays on image tubes. The segment displays exist with different light sources, e.g., with gallium arsenide phosphide (GaAsP) diodes, incandescent filaments,
discharge lamps, AC luminescence lamps,
fluorescent displays by means of accelerated electrons and translucent or reflecting
liquid-crystal display units.
Group 5 concerns itself with optoelectronic devices for image recording. By means of a lens system, an image is projected either onto a photocathode or on a target. Image converters convert the image appearing on the photocathode, by means of an electron beam, into a visible fluorescent image on the anode or camera tubes scan
the image projected on the target. On the
target load resistance, every point on the image is converted into an electrical signal corresponding to its brightness.
Group 6 includes optoelectronic devices for picture reproduction. This is understood to mean the electronic reproduction of images on a screen. They include the
conventional television picture tubes as well as radar and oscilloscope tubes and the
more recent flat picture displays.
On the optical side, optoelectronics touches
on the fields of geometrical optics,
physiological optics, physical optics,
quantum optics, applied optics and in
future possibly the field of integrated optics. Geometrical optics is concerned with the propagation, refraction and reflection of rays in accordance with geometrical laws, physiological optics is concerned with the process of vision and the evaluation of light by the eye, and physical or wave optics with the propagation of light as waves (diffraction, interference,
polarisation, colour dispersion). Quantum
optics deals with the laws of temperature radiation, the atomic model, excitation conditions and luminescence and emission phenomena, the field of X-rays and the interaction between radiation and matter. Applied optics is concerned with optical instruments such as magnifying- devices, microscopes etc. Integrated optics deals with optical circuits in a manner analoguous to integrated electronic circuits.
On the electronic side, optoelectronics
concerns the production of modulation and the control of electrically operated radiation sources, the drive and deflection circuits of image recording and reproducing tubes, and the further processing and evaluation of the electrical signals produced from a photo-
electric converter.
As these statements show, the field of opto-
electronics, with its applications, is extremely extensive and many-sided; the
present book therefore has the following
limits:
The first part deals with the fundamental principles of optoelectronics, photometric and radiation units, black bodies and Lambert radiators, laws of radiation, radiation, luminescence and photoemission phenomena. In the second, technical part, the emphasis lies on the description, calculation and application of optoelectronic semiconductor components. Finally, the third part is concerned with circuits proven in the laboratory and in practice.
Volkmar Hartel
1
Physics of Optical Radiation
1.1 1.1.1 1.1.2 1.1.3 1.1.4 1.2 1.2.1 1.2.2
1.2.3
1.3 1.3.1 1.3.2 1.3.3
Optical Radiation and Light Basic Definitions
The Quantum Nature of Radiation The Dualism of Waves and Particles
Wavelength and Propagation Speed Radiation and Luminescence Phenomena The atomic and band model Luminescence, Fluorescence, Phosphorescence Luminescence phenomena in semiconductors, injection luminescence
Photoelectric effect External photoeffect Internal photoeffect
The junction photoeffect
r AA
Physics of Optical Radiation
1.1
Optical Radiation and Light
Wavelength range
Designation of Radiation
100 nm- 280 nm
280 nm - 315 nm
315 nm - 380 nm
380 nm - 440 nm
440 nm - 495 nm
495 nm - 558 nm
580 nm - 640 nm
640 nm -- 750 nm
750 nm - 1 400 nm
1-4 JUm -
3//m
3 (Jm- 1 000 jUm
UV-C UV-B UV-
Light - violet Light - blue Light -- green Light - yellow Light - red
IR-
IR-B IR -C
Table 1.1 Subdivision of the optical radiation
spectrum according to DIN 5031
1.1.1 Basic Definitions
Optical radiation is understood to mean
electromagnetic radiation in the range of
wavelengths between 10 nm and 1 mm. This
range is illustrated in Figure 1.1 as part of
the whole electromagnetic spectrum. The
optical radiation band consists of the sub-
UV ranges
(Ultraviolet), visible radiation
(Light) and IR (infra-red). The transitions
between the individual ranges are fluid.
According to DIN 5031, Part 7, the UV range starts at 100 nm. The UV and IR
ranges are divided into sub-groups A, B and
C and the visible range into the relevant
colours, as shown in Table 1.1. The
expression "Light" only relates to the
optical radiation perceived and evaluated
by the human eye.
1.1.2
The quantum nature of radiation
Until the beginning of this century, electromagnetic radiation, including optical radiation, was considered to be continuous trains of waves.
10
200 300 390 455 492
577 597 622
Ext reme
j
Far Near Wiolet Blue (ireen Yellow Orange
1
Ultraviolet
'
Visible light
N
II 770 1,5x10' bx\<? 4x10*
IX 10* (nm)
Red
> Near
t 1 t
Med ium Far Extreme"
Infrared
/
^^
Gamma rays
"**-*
Cosmic rays r
X-Rays
^
^
1
1
i
ill
-- till -- I
1
11 'T
f
r-- t
1
II!
i
ict 10 io"' i<r* ior' icr 6 10" 10~* 10"' ior* 10"' I
mm [
waves
'
1/
1 1
1
1
1
Radio waves
Audio frequencies Long electrical oscillations
A T
1 1
i
i
i
i
10 10' 10' 10* 10s 10* 10' 10* 10* 10'° 10" 10 11 10' 3 10 14 0"n)
Figure 1.1
Electromagnetic Radiation Spectrum
Physical investigations and considerations on the phenomena of the external photo-
electric effect led, however, to the
recognition of the fact that radiation does . not interact with matter continuously, but
in small portions which cannot be further
sub-divided, the so-called quanta. Such a radiation quantum, also called a photon, corresponds to a certain amount of energy, dependent on the frequency of the radiation,
which has a minimum value for any given
frequency. For the relationship between
the energy and the frequency of a quantum
of radiation, the equation
W = h -" Ph
(1.1)
applies,
where h
=
6
62
.
-34
10 '
watt,
2 sec
(W.s ) Planck's constant and V = the
frequency of the radiation in Hz.
In the optical range, the amounts of energy in the individual quanta are so small that the quantum structure of such radiation is beyond the limits of most conventional measurement methods and observations.
All quanta arise from changes in energy in atoms and molecules. The radiations which are of importance for opto-electronics have their origin in the outer electron orbits of the atoms. In the normal state, each electron is in the physically lowest possible level, where it has a certain amount of potential and kinetic energy. By excitation processes, e.g., by the introduction of electrical, thermal or radiant energy, electrons can temporarily leave their basic state and occupy a higher level with a correspondingly higher energy content. This state is not stable; therefore after a very short time the excited electrons fall back to the basic state, while emitting a quantum of radiation, the frequency of which, as shown in formula (1.1), corresponds to the energy difference between the two levels.
1.1.3
The dualism of waves and particles
If the nature of electromagnetic radiation is
investigated, then dependent upon the experimental conditions, sometimes it appears to behave like a wave, but at other
times like a stream of particles (corpuscles).
Therefore, when describing radiation, these two aspects are considered.
Electromagnetic radiation can be demonstrated to exhibit typical wave characteristics such as interference/refraction and polarisation, which can obviously be interpreted by the periodic nature of a
wave-train. In theoretical physics, the propagation of electromagnetic radiation is
derived from a wave process with the aid of Maxwell's equations.
From the corpuscular viewpoint, radiation takes on the character of a stream of particles. Each photon is then considered
as an elementary particle with zero stationary mass, moving at the speed of light. The shorter the wavelength of radiation, the more prominent does the
corpuscular nature become in comparison with the wave character. In the range of gamma rays, therefore, their particle nature becomes the predominant characteristic.
1.1.4
Wavelength and propagation speed
The speed of propagation of light in vacuo,
and also approximately in the atmosphere, is c = 2-998 . 10 8 m/s.
For practical calculations, the rounded value of c = 3 . 1 8 m/s is adequate. The relationship between the three fundamental
values, speed of light c , wavelength X and frequency V is given by
c = X. v
(1.2)
By combining the equations (1.1) and (1.2), the wavelength of a photon can be calculated, if the photon energy Wpft is known.
c -h X=
WPh
(1.3)
The photon energy is usually stated in electron volts (eV). One electron volt corresponds to the kinetic energy received by an electron through acceleration in an electric field with a potential difference of one volt.
For the conversion of eV into the SI unit Wsor J:
1 eV = 1.602 .10~ 19 Watt sees (Ws)
-19
1 eV = 1.602 . 10
Joules (J)
The equation (1.3) can be simplified asa numerical equation, if the numerical values of the natural constants and the conversion
factor for eV into Ws are inserted:
3.
8
10
·
6-62.
10~ 34
X=
-19
Wph · 1-6 · 10
Where X is measured in jum C is measured in ms
and Wph is measured in eV
This results in the numerical equation
1-24
Wph
(1.4)
From these equations it can be seen, that
the greater the energy of the photons, the shorter does the wavelength of the electromagnetic wave becomes.
1.2
Radiation and Luminescence Phenomena
1.2.1
Atomic and Band Model
For the physical explanation of radiation
phenomena and also of the photoemission which is described later, an understanding of the atom and band models is necessary The atomic model describes, in a simplified
representation, the spatial structure of an atom, while the band model gives information on the energy content of electrons,
both in single atoms and also in combinations of several atoms, e.g., in a crystal lattice. In Figure 1.2, the atomic model for germanium is shown as an example.
M V^'7 / I! h
\r^S4^7 \ \ --y
Figure 1.2
Model of germanium atom
Recent work has shown the atom to be extremely complex but for simplicity it can be considered to be composed of negatively charged electrons, orbiting around a charged nucleus. The atomic nucleus itself consists of positively charged protons bound
together with neutral particles called
neutrons. The number of protons determines which chemical element is concerned. In an electrically neutral atom, the number of electrons and protons is equal. The
electrons are located in specific orbits around the nucleus, and these are grouped into so-called shells; in Figure 1.2 these
shells are shown with their normal designa-
M tions K, L, and N.
Nearly all atomic nuclei with more than 82 protons and a few smaller nuclei are unstable and disintegrate, through a process known as
as nuclear fission, spontaneously over varying periods of time. Through this
fission process the basic elements are converted into others. Radium, for example, disintegrates to form the stable element Lead, with the emission of radiation from the nucleus which is characterised into u-, i>, and the short-wave 7-rays. While this radiation originates in the nucleus of the atom, the emission and absorption of optical radiation takes place within the cloud of electrons surrounding the nucleus.
The energy state of the electrons is described by four quantum numbers, The quantum numbers are designated by small letters:
n = Principal quantum number (associated
with a particular shell)
1 = Secondary quantum number (subgroup is the shell and shape of the electron
orbit)
m = Magnetic quantum number (location
in space of the angular momentum vector
of the orbit)
s = Spin quantum number (angular momentum of the electron itself about an
imaginary axis)
All electrons associated with an atom or a molecule differ from one another in at least one quantum number. This principle of exclusion of two identical electron states, the
so-called Pauli exclusion principle,
determines the maximum number of
electrons within a given shell, an intermediate shell or an energy level. If all the electrons of an atom are at their lowest possible energy level, that is, in the innermost shells, then the atom is in its basic
state.
With the single atom, the electrons adopt exactly defined, discrete energy states. In a
crystal, on the other hand, because of the interaction of the electrons belonging to the different atomic nuclei, the previous discrete bands divide into ranges, the energy bands. These ranges are illustrated in the so-called band model, while in semiconductor technology one restricts oneself to the valency band and the conduction band, which are of interest here, with a "forbidden gap or band" between them. Figure 1.3 shows a few examples.
In the cases of metals and insulators, the valency band is occupied by electrons, which are fixed in their places. Thus, no movable electrical charges are possible within the valency band. In the case of insulators, the conduction band is unoccupied, i.e., it is free of charge carriers, while in the case of a metal each atom gives up one or more electrons into this band. These electrons are then very loosely bound to a given atomic nucleus and are therefore free
to move in the crystal lattice. Their number and mobility in the conduction band
determines the conductivity of the substance.
In contrast to the metals, which are known to be good electrical conductors, the
semiconductor, at low temperatures has almost all electrons in the valency band, so that it is then almost an insulator. With rising temperature, more and more covalent bonds 1 break apart, since through the external supply of energy in the form of heat, a certain number of electrons can leave the valency band and move up into the conduction band. This is possible without difficulty with semiconductors, since the width of the forbidden band is very narrow in comparison with that of
A insulators. hole, is then produced in the
valency band and behaves like a positively charged particle which can migrate within the band. The presence of holes in the valency band and electrons in the conduction band causes the conductivity of semiconductor materials to lie somewhere between those of metals and insulators. The width of the forbidden band, which is
1 Chemical bond between the atoms of the crystal lattice
*KES*S
Pure semiconductor
Forbidden band
Conduction band Donor level
p- or n-doped semiconductor
Figure 1.3
Energy bands of a few materials
generally stated in eV, determines the
necessary minimum energy, which must be
supplied to an electron, in order to raise it from the valency band to the conduction band.
In the preparation of semiconductor devices "impurities" are intentionally added in small, defined quantities to the undoped or "intrinsic" semiconductor material. Silicon and germanium are two quadrivalent or group IV semiconductors which are
commonly used for producing useful
electrical devices. However, semiconducting
compound materials such as gallium arsenide can also have characteristics which can be readily utilised. The conductivity of
the basic material is modified as required by adding either trivalent (group III) atoms, so called "acceptors," to give P-material or pentavalent (group V) atoms, so called "donors," to give N-material. Since the donor level is close, in energy, to the
conduction band, a very small amount of
energy is sufficient to raise an electron of the donor substance into the conduction band. Therefore, the pentavalent donors, with one electron which is only loosely
bound to the atom, increase the basic conductivity. The acceptors in turn attract either loosely bound or free electrons, in order to fill the gaps (holes) caused by their addition. The acceptor level is close, in
energy, to the valency band.
The conductivity of the doped semiconductor can be altered by the presence of an
external energy source.
The excitation energy can be supplied in the form of heat, by photons (light energy) or by the application of an external voltage
(electrical energy). If the excitation takes place through photons (irradiation), the term "internal photoeffect" is used. If an excited electron is in the conduction band, this state of excitation is not stable. After a certain time, the electron falls back again and recombines with a hole. Electromagnetic radiation, corresponding to the energy difference liberated, can then be emitted.
Under certain conditions, electrons can be released completely, by energetic excitation, from their parent substance, for example, from alkali metals or certain oxides, and can move freely in space. This process is called emission. If this emission is caused by light quanta, then the term "external
photo-effect" is also used.
On the basis of our knowledge of the atomic model, we know that excited electrons are
located in orbits with higher energy
levels and that on their return to the basic state, electromagnetic radiation can be
emitted. A distinction can be made between
three kinds of radiation:
Radiation, which occurs through the return of the electron to the basic state by direct recombination.
Fluorescence radiation.
:
Phosphorescence radiation.
From the band model, a distinction is made between four kinds of recombination:
crystals, in which electrons have been raised in energy by absorbed light do not return at once to the basic state, with emission of the luminescent light, but are stored in energy levels, which are somewhat below the starting energy needed for luminescence.
Recombination between free electrons of the conduction band and free holes of the valency band.
Bioluminescence: Is part of chemiluminescence. It occurs in nature, e.g., in glow-worms and fireflies.
Recombination between free electrons of the conduction band and bonded holes of the acceptor level.
Recombination between electrons of the donor level and free holes of the valency band.
Recombination between electrons of the donor level and bonded holes of the acceptor
level.
The kind of recombination is determined by the characteristics of the semiconductor and its doping.
1.2.2
Luminescence, Fluorescence, Phosphoresence
For all cases of light emission, which do not have their cause solely in the temperature of the material, E. Wiedemann introduced the term "Luminescence" as long ago as 1889. This is understood to mean radiation phenomena in the visible range. In a broader sense, it is also understood, in the literature, to mean the optical radiation range. Depending on the kind of excitation energy, which causes the luminescence phenomenon, distinctions are made, among others, between:
Chemiluminescence: Occurs through certain chemical reactions, in which energy is liberated and emitted as radiation, e.g., phosphorus glows through oxidation in the air.
Cathodoluminescence: Occurs through accelerated fast electrons, which, on collision with atoms, excite the corresponding valency electrons and cause the emission of radiation or light. Typical examples are television and oscilloscope
tubes.
A.C. Electroluminescence: Obtains the excitation energy through an electric field, e.g., in the dielectric of a
capacitor (Destriau effect). The luminescent
capacitor contains the thin luminescent dielectric and also a transparent electrode.
Photoluminescence:
Is caused by fluorescence. The exciting radiation, for example, UV, is more
energetic than the radiation emitted in the visible range.
g Radioluminescence Obtains the excitation energy through
X-rays or gamma rays.
Thermoluminescence: Excitation by raising the temperature of
Betaluminescence:
The excitation energy is beta radiation.
Crystal luminescence: Is produced by the deformation of certain
crystals.
Triboluminescence: Occurs through the supply of mechanical energy with certain crystals. For example, quartz or zinc sheets glow with a faint light through rubbing, drilling, scratching
etc.
As well as classification by the excitation energy, luminescence phenomena are also classified according to the way in which
they occur:
Fluorescence: With this type of radiation, the excited
electrons fall back, in one or more steps, within about 10~8 seconds, to the basic state and light is emitted. In this process,
the excitation energy generally has a higher
quantum energy than the radiation emitted.
Fluorescent substances act, to a certain extent, as frequency converters. In contrast to phosphorescence, fluorescence only gives an emission, as long as an external supply (e.g., radiation) is maintained.
Phosphorescence:
A radiation, with which the excited
electrons at first remain in a metastable state. This metastable state occurs under the influence of activators (foreign metallic atoms in small concentrations in the basic material), while the electrons fall back into the basic state after a dwell time of varying duration. Phosphorescent materials radiate both during the presence of the excitation energy and also after this excitation energy is switched off, according to the after-glow time.
1.2.3
Luminescence phenomena in
semiconductors, Injection luminescence
In a semiconductor diode operated in the forward direction, the junction region is enriched with electrons and holes. These two kinds of charge carrier recombine with one another, and at every recombination an electron is transferred from the conduction band into the valency band. At the same time it gives up the amount of energy, which corresponds to the difference in energy between the conduction band and the valency band.
Depending on the given conditions, the energy thus liberated can be converted into radiant energy (photons) or into heat (lattice vibrations of the crystal, also called phonons). If a photon conversion takes place in the semiconductor materials known up to the present day, radiation in the range from infrared to the visible range appears. Since they are caused injection of charge carriers into a junction region, radiation phenomena of this kind are called injection luminescence. The probability of photon radiation taking place depends to a great extend on whether the material used is a "direct" or an "indirect" semiconductor.
Both on the basis of the wave-particle dualism and also according to the theory of wave mechanics, a wave function can be ascribed to a particle of matter. In this process, a moving particle, e.g., either an electron or a hole, can be treated mathematically like a wave. From this wave function, a term which is important for semiconductor considerations can be derived by quantum mechanics which is the wave-number vector K. This value, which is also known as the propagation vector, is
proportional to the momentum (p = m.v) of
the moving charge-carrier, as long as the particle can be considered to be "free". In this case it can also be proved that the
W energy of the particle is a quadratic
function of K. In a crystal with its periodic three-dimensional lattice the conditions are more complicated, through the interactions of the lattice components with the moving charge carriers. Here, the
W function = f(K) is no longer a quadratic
function, as before, but the curve which is
now produced can contain several maxima
and minima. Also, the shape of the curve depends on the geometrical crystal direction, in relation to the major crystal axes, in
which the "particle wave" is moving. A
few examples are shown in Figure 1.4. In these, the holes always have an energy
maximum at K = 0, while the curve shapes
differ for electrons.
The probability that an electron will remain is always highest, where its energy becomes a minimum, while the holes endeavour to
reach a level with the maximum possible
energy in the valency band. At the points, where a minimum of electron energy is
directly opposite a maximum of hole
energy, the electron can fill the hole, in a recombination, without a change of the
wave-number vector K or of its momentum
Semiconductors, where this recombination is possible, are called "direct semiconduc-
tors". If the electron energy minimum and the hole energy maximum are not directly
opposite, then a recombination can only take place with a simultaneous change of
K. In this case, we speak of an "indirect
semiconductor".
The physical law of the conservation of
momentum in a self-contained system requires, that when a light quantum is
either absorbed or emitted from a semi-
conductor, the momentum of the light
quantum causes a corresponding change
in momentum in the crystal system. If the momentum values are calculated, both of
a moving charge carrier and of a light quantum in the wavelength range which is
of interest, it is found, that the momentum of the light quantum is negligibly small in
comparison with that of the charge carrier, so that in practice, only the changes of
momentum of the charge carriers need to
be taken into account, even though light quanta are involved in the process.
Electron transitions in direct semiconductors take place without significant change of
momentum, so with recombinations in
these materials the probability of the emission of radiation is high. Things are different with indirect semiconductors. In the case of a recombination, here, as well as the energy given up, a change in
momentum must also be taken into account.
Under these conditions, the production of phonons is again probable, since as well as
the energy, these also take up a momentum
f LE
- LI
t
\
_w
LI
\
\
~1 LE
w eV => 1 -38
Direct transition
DE %. V3 V2 VI
.4? i
-K
K-
a Si (100)
DE
-K
K-
c. GiAiUOO)
Figure 1.4
Band diagrams in the energy-momentum graph. The abbreviations denote LE = Conduction
DE W electron,
= hole, = Energy, K = wave number vector, LI = Conduction band, VI, V2,
V3 =
Valency
bands,
ilOO),
<1 1 1) = Miller's
1
indices
1 Miller's indices are the reciprocal values of the points of intersection of the crystal axes
with cut surface.
10
of the same order of magnitude as the electron momentum. However, a small yield of photons is also possible here with a
W few materials. Figure 1.4 a shows the (K)
diagram of silicon. If can be seen from the
curve shape, that it is an indirect semiconductor. Injection luminescence does
not occur with silicon. The conditions are similar with germanium (Figure 1.4 b). Although there is a minimum of electron
K energy at = here, no radiation emission
occurs, since the conduction electrons are to be found in the deeper energy minima at the sides, from where a direct transition is not possible.
In contrast to the semiconductors just
described, gallium arsenide (GaAs) shows a
deepest electron energy minimum at K = 0, directly opposite the maximum in the
conduction band (Figure 1.4 c). As a direct
semiconductor, GaAs is capable of converting
the energy liberated on recombination into
radiation. From the width of the forbidden
AW band
=1-38 eV, using the equation
(1.4), the wavelength X = 898 nm is
obtained. This radiation lies in the infrared
range and is therefore not visible to the
human eye. It is, however, important,
that silicon photodetectors are particularly
sensitive just in this wavelength range, so
that GaAs light-emitting or luminescent
diodes find many applications in opto-
electronics in combination with Si detectors.
By suitable doping of the intrinsic or undoped semiconductor materials the performance in terms of the radiated wavelength or the efficiency, can be varied. With pure GaAs, the radiation efficiency is of the order of a few percent and increases somewhat with silicon doping, for example
to 1 2% with the infra-red emitting diode TIXL 1 2 (at room temperature), while the
wavelength increases at the same time to 925 nm.
A further important material for light-
emitting diodes is gallium phosphide (GaP), an indirect semiconductor, from which light-
emitting diodes with good quantum yield
can be produced. This is achieved by adding
the so-called isoelectronic centres in
addition to the donors and acceptors. With nitrogen additions, green light is obtained, while zinc-oxygen produces red light. The best published quantum yields lie around
10% for red and around 1% for green.
A partial replacement of the arsenic by
phosphorus in gallium arsenide gives a mixed crystal, which is used as gallium
arsenide phosphide (GaAsP) in many light-emitting diodes. Depending on the
proportion of phosphorus, radiation in the
spectral range red to yellow is achieved. Up
to a phosphorus-arsenic ratio of about 4 : 6,
GaAsP is a direct semiconductor, which emits red light. With increasing phosphorus
content, the material changes to an indirect semiconductor, the radiation wavelength becomes shorter (yellow) and the radiation efficiency falls. Exactly as-GaP, however, the decreasing efficiency with shorterwavelength light is substantially compensated for by the increased sensitivity of the eye. Very often, a mixed crystal with the composition GaAso.56Po.44> which emits
radiation at a wavelength of 650 nm with an
efficiency of 0.1%, is used. The luminance
achieved is still completely adequate, even in brightly lit rooms.
By means of special design measures, it is possible to build semiconductor lasers from GaAs of GaAlAs with a small aluminium
concentration. Optical feedback is achieved through the use of parallel cleavage surfaces
perpendicular to the plane of the PN junc-
tion or alternatively through plane surface grinding. If the threshold current density is exceeded with such devices, they emit coherent beams of radiation. Typical values of the threshold current density lie
around 15 . 10 3 A/cm2 . With special GaAlAs
heterostructures, threshold current
densities under 10 A/ cm are achieved; such injection lasers can then run at room
temperature in continuous pulse operation.
Efforts are still being made, to search for
new materials which will produce practical
11
light-emitting diodes, particularly at either
end of the currently achievable spectrum range. For a blue luminescent diode, useful results are expected from the material gallium nitride (GaN), but various technological problems still have to be solved here.
In future developments in this field, the so-called II-VI compounds should also have a certain part to play. These have long been used as phosphorescent materials (e.g., the well-known fluorescent materials for television and oscilloscope tubes). They
include, among others: Barium sulphide (BaS), Barium selenide (BaSe), Barium Telluride (BaTe), Cadmium sulphide (CdS), Cadmium selenide (CdSe), Zinc sulphide
(ZnS), Zinc selenide (ZnSe). Table 1.2
shows, among other factors, the band energy spacing AW, the wavelength and the
transition type of various semiconductors, which are of importance for optoelectronics.
The highest efficiencies which have so far been achieved for radiation output lie in
the infrared range. Using special phosphorescent materials with lanthanides as doping elements, it is possible, by double excitation of electrons, to convert
the infrared radiation of a GaAs diode
into visible green light. Although this radiation conversion process has only a low efficiency, the luminance of the phosphor achieved is sufficient, because of the high green sensitivity of the eye. Figure 1.5 shows a luminescent diode working on this principle.
A characteristic property of luminescent
diodes is the relatively narrow-band spectral range of the light quanta emitted. This can also be seen from Figure 1. 6, in which the emission power of various semiconductor materials, and also the spectral sensitivity of a silicon photo-detector and that of
the human eye, are plotted against the
wavelength.
It should also be mentioned, that luminescent diodes, depending on their doping and manufacture, can be modulated up to very high frequencies (see later chapter on rise and fall times).
Finally, in Figure 1. 7, the construction principle of a planar gallium arsenide phosphide luminescent diode is shown.
The PN junction must lie as closely as
possible under the surface, so that a good
Material
Band energy
spacing (eV)
Ge
Si SiC
InSb InAs
GaSb
InP
GaAs GaAsP GaP GaP GaN
0-66 1-09 2-5 0-18 0-36 0-7 1-26 1-38 1-90 2-19 1-8
3-1
Wavelength
at 300 K
(Mm)
0-496 6-9 3-45 1-77 0-985 0-898 0-65 0-565 0-69 0-4
Radiation equivalent
(Lm/W)
Radiation range
Transition type
200
70 590
5-5 0-3
Blue
IR - C IR - C IR - B IR - A IR - A
Red
Green Red
Violet
Indirect Indirect Indirect Direct Direct Direct Direct Direct Direct Indirect Indirect Indirect
Table 1.2
Band energy spacing and wavelength of various semiconductors
12
radiation yield is achieved. This
requirement is now satisfied by using
extremely thin films.
Figure 1.5 In this light-emitting diode, the fluorescent screen, consisting of a special phosphorescent material, converts the infrared radiation into green light
5 OS-
" 4Io-3
Figure 1.6
Spectral sensitivity of various semiconductor materials and the human eye
Light emission Oxide mask
P-layer
. Front contac
;
+v
(
e>
_JE:
rSk. ' Jr1
;
N-layer GaAsP
GaAs substrate ^v^v^^^^\vv\vvvvk^Vi\^vvvv^^^^^v^^^v^wv^^^vvvv*.v^y^;
Rear contact
(-ve )
Figure 1. 7
Construction principle of a planar GaAsP
light-emitting diode.
1.3 Photoelectric effect
It has already been described how electrons
can be excited by irradiation energy and can be raised into the conduction band or into an orbit with a higher energy level. If
the electrical conductivity of the substance
is thereby raised, we speak of an "internal photoeffect". In semiconductors with a PN
junction, there is a "junction photoeffect", which, in principle, is also an
internal photoeffect. The escape of "free"
electrons out of an irradiated substance (e.g., from alkali metals) is called an
"external photoeffect". As a general term for these processes, the designation "photo-
electric effect" is used.
The spe£tral photosensitivity of a substance is mainly determined by its absorption capability for photons falling on it. In the wavelength range of the maximum spectral photosensitivity of a substance, photons with the corresponding wavelength are absorbed closely under
the surface of the substance (surface photo-effect). For longer-wave length radiation, the corresponding substance
becomes transparent.
A shorter-wavelength radiation penetrates
deeper into the same substance. At the same time, the photosensitivity decreases. However, in a short-wave spectral range,
the photosensitivity can increase slightly again through the so-called volume photo-
effect.
1.3.1 External photo-effect
If the energy of the incident photons of
electromagnetic radiation - in this case optical radiation - are greater than the
escape energy of an electron of the irradiated substance, then free electrons can escape
from the latter. The minimum energy value,
which corresponds to the escape energy, is defined as the "photoelectric
13
threshold value". From equation (1.4), the wavelength of a radiation depends on the
photon energy Wpjj. If the value of the
WA escape energy
is inserted in this
equation for Wph, the limiting wavelength,
below which the irradiated substance
becomes sensitive to radiation and gives up
free electrons, can be calculated. The
numerical equation for the limiting wave-
length reads:
1-24
XG = WA
(1.5)
In the periodic system, the inert gases, which have fully-occupied electron shells,
are follwed by the alkali metals, which as well as one or more full shells, have one further incomplete shell, occupied by only
A one electron. slight excitation energy is
sufficient for this single electron to escape.
W Table 1.3 Shows the escape energy A and
the limiting wavelength Xq for various
alkali metals. In Figure 1.8, the relative spectral sensitivity of the alkali metals is
illustrated in relation to the radiation
wavelength.
For monochromatic irradiation, the number
of electrons emitted increases in proportion to the irradiance, provided that the photon energy is greater than or equal to the escape
energy. The difference between the photon energy and the escape energy is taken up by the electron as kinetic energy., and the electron moves through space with a
corresponding speed.
WKin = Wphoton-WA
and from equation 1.3
WKin h. c -WA
(1.6)
/\U |320-
280
*240'
$ 200
5 160
1 120
80 '
I I 40. S
.Na
kic
sjlb
Cs
360 400 440 480 520 560 600 640 680 720 nm
Figure 1.8
Relative spectral sensitivity of alkali metals.
The alkali metals are particularly suitable for applications, where the external photoeffect is desired. Typical examples of the external photoeffect are photocathodes in vacuum and gas-filled photocells, photomultipliers, image converters and television camera tubes.
1.3.2 Internal photo-effect
In contrast to the external photoeffect, in
which free electrons escape from the material, with the internal photo-effect the excited electrons remain within the
substance. When the material is irradiated,
electrons are raised from the valency band
Alkali metal
Lithium Sodiun Potassium Rubidium Caesium
W Minimum escape energy A
2-4 eV 2-28 eV 2-25 eV 2-13 eV 1-94 eV (1.36 eV)
Limiting wavelength Xq
517 nm 543 nm 551 nm 582 nm 639 nm (915 nm)
Table 1.3
Minimum escape energy W^ and limiting wavelength Xq of alkali metals
14
into the conduction band and increase the electrical conductivity of the irradiated material. This increase in electrical conductivity is called "photoconduction". In the case of metals, because of their high
basic electrical conductivity, the increase in conductivity through irradiation is
insignificant. The internal photoeffect is put to practical use in both intrinsic and doped
semiconductors.
Intrinsic semiconductors are manufactured to have a very high degree of purity. The intrinsic conductivity is caused by thermally generated electrons and holes, which are
present in equal concentrations. When the
substance is irradiated, additional free
charge carriers are produced by the photon energy and thus the conductivity is improved. Through the deliberate doping of
a semiconductor with either donors, or acceptors an N-type or P-type semiconductor is obtained. With the doped semiconductor, the main point of interest is that through irradiation, the density of a quasi-free charge carrier (electrons or holes) is increased. In order to cause excitation, the photons must have a
minimum energy value, which is determined by the band spacing or the distance from
the band boundary to the doping level.
As with the external photoeffect, the limiting wavelength, below which the doped semiconductor is sensitive, can also be calculated. In the equation (1.5), instead of the escape energy, the band spacing
AW is inserted. Table 1.4 shows examples
of various semiconductors with the band spacing and limiting wavelength. The
Semiconductor
Band spacing AW
in eV (Slight differences are to be met within the literature, depending on whether the band spacing has been
determined through the dielectric constant or the absorption limit).
Limiting wavelength \q
in /urn
Ge
Se
Si
Te AlSb CdS ZnS CdSe CdTe GaAs GaSb InAs InP InSb PbS PbSe PbTe
0-67 2
112
0-33 1-6 2-4 3-6 1-8
1-5 1-35 0-8
0-35 1-25 0-18 0-37 0-27 0-3
1-85 0-62 1-1 3-75 0-775 0-520 0-345 0-69 0-83 0-9 1-35 3-54 0-995 6-9 3-35 4-58 4-13
Table 1.4
Band spacing and limiting wavelength of various semiconductors, calculated for the
internal photo-effect
15
internal photoeffect is utilised in low-cost
photo-resistors, which can work in the UV,
visible and near infrared range and in highquality IR detectors, which are used up to a wavelength of 30 jum (see Figure 1.9).
" '
'1
1
» Limit sei by photon noise
Si
\
/A \
10"
\ InAs
GeffV) \20OK
Ambient temperature
290 K, u · 2ra
y j- io"
/ * V^~\
\|"SbffV)
\ /X^S-S"^ (HgCOTe
1/1 \0^5e^\
///Or\
1
ll I
.
.
300K
.
i
.
1
i
i
. .1
t
20 pm
Figure 1.9
D Specific detectivity + (\ 1000,1) (special
measure of sensitivity for IR detectors) of high-quality IR detectors from Texas
Instruments
1.3.3
The junction photo-effect
The junction photoeffect is an internal photoeffect and is utilised in photodiodes, photo avalanche diodes, phototransistors, photo field-effect transistors and photothyristors. These components are manufactured as doped semiconductors where smaller quantum energies are
necessary to raise the conductivity than for
the intrinsic semiconductor. The construction of a photodiode or a phototransistor is the same in principle as in a "normal" diode or a "normal" transistor. In Figure
1.10, the schematic structure of a semiconductor diode is illustrated.
On the left, we see the P-zone, doped with
acceptors, and on the right the N-zone, doped with donors. Without an ex :ernal voltage source, the free electrons and holes are in continuous random thermal motion. As a result of this motion, some electrons pass through the boundary layer into the P-zone and conversely some holes pass
into the N-zone, where they recombine with the oppositely-charged charge carriers. As a result, a narrow region on both sides of the boundary layer becomes depleted of mobile charge carriers, and the positively charged donor ions in the N-region and the negatively-charged acceptor ions in the P-region, which are fixed in their places in the crystal lattice, remain within this depletion region. These ions build up a space charge in the junction layer, which is depleted of mobile charge carriers, and through its electric field this space charge impedes the diffusion of the charge carriers through the junction layer more and more, so that finally an equilibrium condition is established in which the diffusion current and the current occurring through the presence of electric field are balanced. In this equilibrium condition no measurable voltage appears at the external connections. The internal potential difference in the semiconductor is exactly compensated by the semiconductor external connection contact potentials.
If a PN junction is externally irradiated then
additional free charge carriers are
produced and start to move according to the direction of the electric field set up by the fixed space charges. The newly produced electrons then arrive in the N-region and the holes in the P-region. The result of this
is, that the potential difference in the semiconductor changes in comparison
with the non-irradiated state and we can measure a photo-voltage Vp at the outer connections of the irradiated diode. Thus we have obtained a photocell, similar to those used in solar cells. The no-load
voltage of a photocell rises at first with increasing irradiance but very soon approaches a saturation value, which cannot be exceeded. If the diode connections are short-circuited then a short-circuit current will flow which is almost proportional to the irradiance over a wide range.
PN junctions are not only operated as photocells. By applying an external voltage
in the reverse direction, they can also be
16
Junction region +
<).©(>·© (3
©0Q-0tD QC3Q0-6
0'
1
Interna] potential,
+, .
related to the anode
Dark condition Illuminated condiiiur
Figure 1.10
Schematic construction of a semiconductor photo-diode
used as radiation deteqtors. Without any irradiation, only a reverse current flows which is generally negligibly small, this is the so-called dark current. With increasing irradiation of the junction, the number of additional free charge carriers thus produced increases and raises the reverse current quite considerably. The selection of the operating
mode - photocell or operation with reverse voltage - mainly depends on the application
and on the matchir"] conditions to the subsequent circuit.
A further component in this category is the
phototransistor. It is several times more sensitive than the photodiode, but its upper limiting frequency is lower. As with
the diode, at first a photocurrent is produced by irradiation of the system, but this is considerably amplified by the
17
transistor effect. Photo transistors with the base connection brought out offer a number of additional possibilities for
varying the mode of operation, e.g.,
adjustment of the working point, increasing the cut-off frequency with a simultaneous reduction of the sensitivity, photo-diode operation by not using either the emitter or collector connection.
As a result of the geometrical dimensions,
the collector-base diode generally has a higher photosensitivity than the
emitter-base diode. As with every normal
silicon-transistor, the reverse emitter-base
breakdown voltage is only about 6 V, because of the emitter doping which is higher than that of the base and the collector. Furthermore, both junctions can also be used as photocells.
Finally, as a very sensitive photoelectric component, let us also mention the Darlington phototransistor, which, as well as the actual phototransistor, also in one case contains an additional emitter follower.
18
2 Principles of
calculation in radiation physics
and optics
2.1
2.2 2.2.1
2.2.2 2.2.3 2.2.4 2.2.5 2.3 2.3.1 2.3.2 2.3.3
2.3.4
Radiant flux - Luminous flux Radiant energy - Luminous energy
Units related to the radiation source
Radiant emittance
Luminous emittance
Radiant intensity - Luminous intensity Radiance - Luminance
Units of luminance
Radiant efficiency - Luminous efficiency
Units related to the receiver
Irradiance - Illuminance Irradiation - Light exposure
Relationship between irradiance or illuminance and the reflected radiance or
luminance
Spectral radiation - distribution units
19
Principles of calculation in radiation physics and optics
The definitions of the physical and optical values have been compiled strictly on the basis of the DIN Standards 1301, 5031, 5033 and 44020, as revised from 1964 to 1972. For the practical worker in formulae listed
from the DIN standards have also been shown in simplified form under certain
stated conditions.
For convenience the optical spectrum can be divided into the two invisible parts (IR
and UV) and the visible part (see Section
1.1). Historically and for practical purposes the visible part was of chief importance. The parameters defining this part of the spectrum are defined as "photometric"
units. The definition of an optical (photo-
metric) unit includes the weighting in accordance with the ocular sensitivity V(\)
laid down by the CIE (International
In this chapter we also define radiometric units for the optical spectrum, which are by
analogy also applicable, to the photo-
metric units. For example: The radiant flux is termed photometrically as luminous flux and is defined as follows: The luminous flux is the radiant flux emitted, which, when falling on the retina of the eye, causes an
impression of light, weighted in accordance with the spectral sensitivity V(\) of the eye.
The symbols for the radiometric units are given the index "e" (energetic), those for the photometric values the index "v"
(visible).
2.1
Radiant flux - Luminous flux, Radiant energy -- Luminous energy
The radiant flux or radiant power e is the
quotient of the radiant energy or quantity
dW of radiation
e , transmitted by radiation,
divided by the time dt. The total radiant
energy emitted in all direction by a radiation source in one second is called the
total radiant flux.
"^ ^C
q^\e = d\Vedt
(2.1)
W The radiant energy e is obtained by
integration of the radiant flux 4>e over the
time t.
We = J'4Vdt (2 .2)
W With constant radiant energy e per unit
time, the simplified form:
a we e t
(2.3)
is obtained from equation (2.1).
If the radiant flux 4>e is uniform over the time t, then the simplified form:
We = $e · t
(2.4)
is obtained from equation (2.2).
The unit of radiant flux 4>e is the Watt.
{%] =W
(2.5)
W The unit of radiant energy e is the watt-
second.
[We ] = Ws = J
(2.6)
The spectral portion of the radiant flux, to which the human eye is sensitive, is called the luminous power or luminous flux 4>v - The unit of luminous power is the Lumen.
[*v] = lm
(2.7)
21
W The unit of luminous energy v is the
Lumen-second.
[Wv ] = 1ms
(2.8)
Table 2.1, below, shows a summary, in which the various definitions are listed.
2.2 Units related to the radiation source
By analogy with electronics, the radiation sources can be defined as "transmitters" and radiation detectors as "receivers". In
the following sections, the units at the
"transmitting end" will be dealt with.
2.2.1
Radiant emittance - Luminous emittance
M The radiant emittance e is the quotient of the radiant flux d$e emitted from the
surface element dAs, divided by the area of the surface element dAs- Thus it is the
density of the radiant flux per unit area.
A plane surface element can only radiate in
a hemisphere or a semi-infinite space. The
following formula applies:
*dAS
M d<J>e , e dAS
(2.9)
The radiant flux 4>e is obtained by integra-
M tion of the radiant emittance e over all
surface elements dA<j.
4>e = /M e .dAS
(2.10)
$ If the radiant flux e leaving the surface
As is uniform over all surface elements dA$,
then (2.9) simplifies to:
M<,=-
AS
(2.11)
M If the radiant emittance e from every
surface element dAs is equal then the
equation becomes, from (2.10):
% M A = e . S
(2:12)
The unit of radiant emittance ML is the
m2
watt per
--W
m [Me ] =
(2.13)
M The unit of luminous emittance v is the
Lumen perm2
=4 [Mv] m
(2.14)
These various definitions are summarised in Table 2.
Parameter
Symbol
Dimension
Unit
radiometric
radiant flux
*e
power
W
radiant power
(Po)
photometric
radiometric photometric
luminous power luminous flux
radiant energy
quantity of light (luminous energy)
(Pv)
We
Wv
power
lm
power x time Ws power x time 1ms
Table 2.1
Summary of radiant and luminous flux and radiant and luminous energy parameters.
22
radiometric photometric
Parameter radiant exitance (radiant emittance)
luminous exitance (Luminous emittance)
Symbol
Me
Mv
Dimension
power area of
active region
power area of
active region
Unit
Wm~ 2 (Wcm-2 )
1mm-
Table 2.2
Summary of radiant emittance and luminous emittance paramters
2.2.2
Radiant intensity Luminous intensity
n = <t>e
Ie ·
2.18)
The radiant intensity Ie is the quotient
dO of the radiant flux e leaving the
radiation source in a given direction, divided by the solid-angle element d£2
covered by the radiation. Radiant intensity can only be associated with a radiation source which is considered as a point (see
Section 3.3).
-A
u = d<J>e dO
I
(2.15)
The radiant power 4>e obtained by
integration of the radiant intensity I e over the the solid angle elements d£2.
<t>e = /l e .dn
SI
(2.16)
If the radiant power 3>e is uniform in all solid-angle element d£2, the following simplified form applies:
*R
U=-n
(2.17)
This is obtained from the equation (2.15).
The solid angle fi is stated in steradians,
or sterads, abbreviated as sr (see Sections 3.1 and 10.1):
_
27Tsr
m>]=^-=i«=»
(2.19)
The unit of radiant intensity is the watt per
steradian.
[Ie sr
(2.20)
The unit of luminous intensity is the Lumen per steradian or candela, abbreviated
to cd:
lm
[Iv = =cd
sr
(2.21)
A "black-body radiator" (see the chapter
on radiation laws of the black body), with a temperature of 2042°K (since 1969, 2045°K, the solidification point of
platinum, applies) has a radiant intensity
per cm of surface in the normal
direction of approx; 31-3 W/sr. Photometrically, this corresponds to a luminous intensity of 60 cd.
If the radiant intensity I e is the same in every solid-angle element dfZ, then the
simplified form applies:
The various definitions are summaried in
Table 2.3.
23
Parameter
Symbol
radiometric
radiant
le
intensity
photometric
luminous
Iv
intensity
Table 2.5
Summary of radiant and luminous intensity parameters
Dimension
power
solid angle
power
solid angle
Unit
Wsr 1
lmsr ; cd
2.2.3
Radiance -- Luminance
The radiance Le is the quotient of the
radiant intensity dl e of radiation which leaves or passes through a surface element
dAs in a given direction, divided by the
projection dAg.cos i^of the surface element, if ip is the angle between the direction of radiation and the normal to the surface.
dL dAg · cos <p
(2.22)
2
d 4>P
LPe = dAg · cos i/> . dft
(2.23)
For the same given direction, the radiant
intensity I e is obtained by integration of the radiance L e over the surface element
dAS :
Ie = / L e · dAs · cos (£
AS
(2.24)
M The radiant emittance e is obtained by
integration of the radiance L e over the
solid angle elements dft:
Me = / Le · dft
ft
(2.25)
The radiant power <l>e is obtained by integration of the radiance Le over the
surface elements dAs and the solid angle
elements dft:
% = f / Le . dAs . cos i/? . dft
AS ft
< 2 - 26 >
The limits of the integral of the radiance Le
over the solid angle ft in equation (2.26) or of the radiant intensity I e over the solid angle in equation (2.16) depend on the radiation angle of the radiation source (see radiant intensity distribution curve). With plane surface radiators, the integration is carried out over the semi-infinite space from to 2;rsr. In the case of sources radiating in all spatial directions, the integration is carried out over the infinite space from to 47Tsr. (See Sections 3.1 and
10.1).
If the radiant flux 4>e leaving the surface
As is uniform over all surface elements dA s and in all solid angle elements dft,
then the simplified forms of equation (2.22)
are obtained:
IP
Le =
As cos .
1/7
equation (2.23):
(2.27)
<k.
L,, =
As · cos i/J . ft
(2.28)
If the radiance from all surface elements
dAs i s tne same in every solid angle element
dft then the simplified forms are obtained from equation (2.24):
24
radiometric
Parameter radiance
Symbol
photometric
luminance
Table 2.4
Summary of radiance and luminance parameters
Dimension
power area of active
region . solid angle
power area of active
region . solid angle
Unit
Watt
m . sr
Watt
")
cm ( 2 . sr
ftLa
Candela
m
As I e = Le ·
cos <f>
(2.29)
equation (2.25)
Me = L e . S2
(2.30)
and equation (2.26):
$ As e = Le ·
· cos if) . £2
(2.31)
The unit of radiance is the Watt per (m'
steradian):
W
[L e
m2 . sr
(2.32)
The unit of luminance is the Lumen per
m (m2 steradian) or Candela per
:
lm
cd
m m [Ly] = 2
=
2
. sr
(2.33)
All important parameters, are summarised in Table 2.4
m cd .
asb
cd . cm , sb
L
9 29 . 10
2919
6-45 . 10
1 nstead of cd . also Nil
2-957 . 10
00929
2054 . 10
2919
6-452
1 in - 2-54 cm 1 ft = 0-3048 n
1 L (Lambert) =
cd . cm
. JT
10-764
1 IL (l-ootlambert)
2
'
cd . ft
. ff
3 . 426
1550
33-82
1-076. 10 J
10-764
3-426. 10 "
4869
155
2-957. 1 2
929
2 054
3-382 10 .
I
1-0764.10 :
1
452-4
2-211 10". I
IT - 929 cm'
2 I!
=
00929
m2
2
lm
-
10-764
2 It
I
2 in
-6-4516 cm 2
1
2 II
=9
2903 dm 2
I
. cd
1 cd
Table 2.5
Conversion factors for the various common units of luminance
25
--5 r
r
2.2.4
Units of luminance
In practice, the need foi mathematical determination of the luminance arose at an early date. Therefore consequently there are still several commonly-used units of luminance, which originated in past history.
The most common will be explained below
and the conversion factors will be stated in Table 2.5.
The SI Unit
is now specified by law and it is calculated
lm
cd
m m in 2
or t~-
sr
In the literature and data-sheets, however, the following time-honoured measurement units are often to be found.
The Stilb (sbj
--m^ is obtained from the unit jby means of
the conversion
factor
4 10
and
relates
to
the
unit of area 1 cm .
-- --m lsb
1
cd
2
cm
=
a4 10
cd
f
TheApostilb (asb) takes direct account, with the factor it, of the radiation conditions from a plane surface element of the Lambertian radiator. The luminance of reflecting
surfaces, among others, was previously measured in "asb".
-- -- 1 cd
1 asb =
it
m
The Lambert fL) also takes account, with the factor it, of the radiation conditions of the Lambert radiator. It is greater than 1 asb by the
4 factor 10 .
L=-- -- 1
4 cd
1
10
it
nC
2 cd/ft
belongs to the same group and the Stilb,
while
1
2 cd/ft
relates
to
1
2 ft
=
929
cm2
and the Stilb relates to 1 cm2 of radiator
surface.
cd
cd
1--5
ft
=
10-764--
m
The Foot-Lambert (fL)
also relates to
1
2 ft
and
also, like the
Apostilb, has the solid angle 12 - 7Tsr
included.
--m 1
fL =
,, 3-4 26
cd j-
2 cd/in
belongs to the same group as the Stilb,
while
it relates
to
1
2
inch
=
6-4516
cm
of
radiator area.
1c--d; 2 in
=
cd
1550--
m2
2.2.5
Radiant efficiency and luminous
efficiency
% The radiant efficiency of a radiation
source is the quotient of the total radiant power <Pe produced, divided by the input power P.
% The luminous efficiency of a light source
is the quotient of the total luminous power $y produced (total luminous flux) divided by the input power P.
*,,
*te=-
(2.34)
(2.35)
For the input power P and the radiant power 4>e , the watt is the unit used. The luminous power is measured in Lumens. The radiant efficiency is usually stated as a percentage. The luminous efficiency is quoted in the unit Lumens per watt.
26
Lm
=
W [T?vl
(2.36)
In the case of a black-body radiator, the luminous efficiency corresponds to the photometric radiant equivalent of the black body (see conversion of radiometric into photometric values).
The radiant efficiency TJe of a radiation source in the wavelength range from Xj
to X2 is the ratio of the radiant flux emitted
in this range to the power needed for its
generation (for spectral values, see Section
2.3.4).
X2
/ <&e,X.dX
Tfe=-
(2.37)
2.3 Units related to the receiver
Radiation detectors acting as receivers of radiation will be termed "receivers" for
brevity. The units related to them will
be explained below.
2.3.1
Irradiance - Illuminance
The irradiance Ee is the quotient of the radiant flux d 4>e divided by the irradiated
surface element dAE
^
d<I>e
Ee =dAE
(2.38)
The radiant power d>e falling on the
A receiver area E is obtained by integration
of the irradiance E e over all surface elements dAE .
d> e
=
/Ee
.dA E
AE
(2.39)
If the incident radiant power 3>e is equal on
all surface elements dAE , then the simplified
form:
AE
(2.40)
applies for equation (2.38).
If the irradiance E e is equal on all surface elements dAE> then the simplified form:
A E E = <**>
e.
(2.41)
applies for equation (2.39).
m The unit of irradiance is the watt per
:
=--W
m [Ee ]
(2.42)
The irradiance Ee must not be confused with
M the radiant emittance e , since both values
have the same unit.
Unit 1 lx (Lux) 1 lm.cnT
lx
= 10"
lm. cm
-4 10
1
fc
Remarks
0-0929
1
2 ft
=
0-0929
nT
4 0-0929 . 10
Previously alsp Phot (ph)
instead of lm . cm
1 fc
=
(footcandle)
10-764
H
10-764 . 10
1
m 1
2
=
10-764
2 ft
Table 2.6
Conversion factors for the various common irradiance units
27
radiometric
Parameter irradiance
Symbol Ee
photometric
illuminance
Ev
Table 2.
Summary of irradiance and illuminance parameters
Dimension
power
area
power
area
Unit
Wm~ 2 (WcnT
foot candela
lm . m~
The unit of illuminance is the Lumen per
m or Lux:
lm [Ev ] =" -=lx
(2.43)
In Table 2.6, the conversion factors are
listed for the various common units, while
in Table 2. 7, the most important terms are again summarised.
2.3.2 Irradiation
Light exposure
The irradiation H e is the integral of the
irradiance E e over the time t.
\ He = J'Ee .dt
^AE
(2.44)
If the irradiance Ee is constant during time t,
then the simplified form:
HP Ee .t
(2.45)
applies for equation (2.44).
The measurement unit for irradiation is the
m watt-second per :
=W--s
[He!
m
(2.46)
The measurement unit for light exposure
m is the Lumen-second per or the Lux-
second.
lms
m [H vj = 2~ : lxs
(2.47)
This data is summarised in Table 2.8.
radiometric photometric
Parameter radiant exposure
(irradiation)
light exposure
Symbol
He
Hv
Table 2.8
Summary of irradiation and light exposure parameters
28
Dimension
energy area
luminous energy
area
Unit
Wsm-2
lm . s
m,2 ; lx . s
2.3.3
Relationshio betwen irradiance or illuminance and the reflected radiance or luminance
Originally, the reflected luminance LV)r
A from a diffusely reflecting plane surface
(see Section 6.3, Reflection of radiation)
could be determined very simply from the incident illuminance E v , since the product
of the illuminance Ev in Lux and the reflectivity p gave the reflected luminance Lv r in Apostilbs. Hence the reflected
luminance or reflected radiance calculation
will be carried out with new and old measurement units, for greater ease of understanding. The reflected luminance or radiance value will be needed typically for
calculations involving reflection optocouplers. Here, only a uniformly diffusely
A relecting plane surface r will be considered
as the reflecting light or radiation source.
The reflected radiance Le>r is firstly dependent on the irradiance E e , or the reflected luminance Lv r on the illuminance Ev and the reflectivity p. The irradiance or illuminance is calculated from the equation (2.40). The reflectivity (see
Chapter 6) is the ratio of the reflected radiant flux <J^>eP,rr to the incident radiant flux <!>,'e,o
^ or of the reflected luminous flux v ,r to tne $ incident luminous flux v,o-
$r
(2.48)
$r
L1r
=
A
.fi
r
(2.50)
Because Lambert's law applies, the solid
angle fi is equal to n . O^ (see Sections 2.24, 3.2, 6.3 or DIN 5469, Item 1.8). If fi
is replaced by IT . flo, then equation (2.50)
fcr
Lr =
A r . IT . STo
(2.51)
If
the
reflected
radiant
flux
4> ei r
or
the
reflected luminous flux <J>Vjr is replaced by
the equations (2.40) and (2.48), then the
final simplified form becomes:
A p E .
.
r
A n r .
. fio
p- E n . Q)
(2.52)
In the obsolescent measurement unit, the Apostilb, the factor 7Tsr is taken into account, so that for the previouslymentioned calculation of the reflected illuminance LV)I in Apostilbs, the fpllowing formula applies:
Lv,r asb
-v,r
(2.53)
For the calculation of the reflected
luminance Lvr , the equation (2.52) reads:
P-Ey
l-vj
IT. fto
(2.54)
The reflected radiance Le>r or the reflected luminance Lv>r is calculated in accordance
with equation (2.28):
For the calculation of the reflected
radiance Le>r , the equation (2.52) becomes:
$r L=
A r r . cos if . 12
(2.49)
^e,r
P -Ee
IT. fto
(2.55)
$ If the radiant flux e ,o falling on the
A reflecting surface r , or the luminous
flux <&v , falls perpendicularly on the
A surface r , the factor cos if = 1 and
can therefore be omitted:
2.3.4 Spectral radiation-distribution units
The definitions stated so far for physical or radiometric units such at radiant flux
29
d> e,
radiant
intensity
Ie,
radiant
emittance
Me , radiance Le and irradiance E e do not
relate to the spectral distribution of the
radiation, "the spectral radiation distribution
is taken into account by the "spectral
density of the radiation-distribution
units". These units are related either to a
differential range of the wavelength dX or
of the frequency &V i.e. they are functions of the wavelength X or of the frequency V.
Example:
The spectral radiance Le \ is the proportion of the radiance dLe related to
a small wavelength interval dX.
dLg Le ,X :
dX
(2.56)
Through. integration of the spectral radiance
Le> X over all wavelength intervals dX, the radiance Le is obtained:
Le = / Le,X-dX
o
(2.57)
By identification with the index X for the
wavelength or v for the frequency, the symbols for the radiation-distribution units are converted into those for the spectral radiation distribution units {Table 2.9).
Provided there is no risk of confusion with
other spectral units, the designation "spectral density of a radiation-distribution unit"
can be shortened to "spectral unit". For example, the "spectral density of the radiant flux" can be abbreviated to "spectral
radiant flux".
Parameter Spectral irradiation Spectral radiance Spectral radiant power Spectral radiant power Spectral radiant intensity Spectral radiant emittance Spectral irradiance Table 2.9
Summary of spectral units
Formula
H e ,X
_
dH e
dX
dLe
Le ,X dX
*e,X
d4>e
dX
_ d<J>e *e,v
dv
Ie,X
_ dl e
" dX
Me ,X
_ dMe dX
dEe
Ee,X
=
dX
Practical unit
WsnT 2 (nm" 1 )
(2.58)
Wm-2 -1
-1
sr (nm )
(2.56)
W(nm_1 )
(2.59)
W(Hz-1 )
-1
-1
Wsr (nm )
Wm-2
-1
(nm )
(2.60) (2.61) (2.62)
Wm-2
-1
(nm )
(2.63)
30
Figure 2.
Relative spectral radiance distribution of a
Planck radiator as a function of\
Figure 2. 1 shows the relative spectral
radiance distribution of a Planck's radiator
as a function of the wavelength X. The relative spectral radiance Le X,rel 's tne quotient of the spectral radiance Le ,X at the
wavelength A, divided by the maximum spectral radiance Le)X,max at the maximum wavelength Xmax :
Le,X,rel
Le,X Le.Xmax
(2.64)
The proportional radiance q is the quotient
of the integral of the spectral radiance
Le \ over the wavelength range between X and X2 divided by the integral of the spectral radiance Le>X over the whole
wavelength range:
X2
/Le ,X-dX
A l
</Le,X- dX
(2.65)
This equation is most easily solved either graphically, with a special slide rule, or
scientific calculator.
The proportional radiance q is a relative
value. In practice, it is often sufficient to use relative calculations and models of the spectral distribution of the radiation. In these it is immaterial, which radiation unit is taken into account.
For radiation such as that from a temperature
radiator with a known distribution temperature Tv (see Chapter 4), the relative spectral radiation distribution SX
has been introduced as a relative unit for
each wavelength interval dx . SX will also
be called a radiation function. If the spectral radiation distribution units are represented generally by XX, then the
radiation function SX is the quotient of the
value XX at the wavelength X, divided by
^ the maximum value XXmax at e
wavelength Xmax -
XX sx=
XXma
(2.66)
The radiation leaving any radiation source
is exactly described in its spectral distribution
by the radiation function SX- As an example, let us mention the standard light source (as defined in DIN specs) which is very important in optoelectronics. The spectral radiation distribution S(A)X of the
standard light A has been stated in the wavelength range from 320 nm to 780 nm
in a table in DIN 5033, Part 7.
Since, for standard light A, the value for
S(A)560 nm at tne wavelength X = 560 nm
is made equal to 100, the equation (2.65)
must be completed by a constant factor.
X (A)X
S(A)X :
const-A
X (A)Xmax
(2.67)
A In general, the standard light is produced
by a temperature radiator with distribution
temperature Tv = 2856°K. The values for SX
of this temperature radiation can be
determined, apart from the calculation
stated in Chapter 4, with the "Radiation
Calculator" available from GE. At Tv = 2856°K and X = 560 nm, the value for SX*0-35 is read off. The constant
A which applies for the standard light can
be calculated with the formula (2.68).
a const, - S <A) 560 nm _ 100 _ 2g6
S560nm
°-35
(2.68)
31
With the aid of this constant, the value for S(A)X can be calculated from S\.
S(A)X = S\ . a const,
(2.69)
At a wavelength X = 660 nm, the value for S\ ** 0-6 is read off. The value for
S(A)660 nm amounts to:
S(A)660nm = S660nm const. A =
0-6.286=171.6
(2.70)
This value agrees approximately with the
m DIN standard value for S(A)66Q n =
171.96.
Since the spectral distribution of Planck radiators is described unambiguously by
the distribution temperature Tv (see Chapter 4), the radiation function S\ for
these radiators is often to be found as a
function of Tv and A in books of tables.
32
Laws of radiation
3.1 3.2 3.2.1 3.2.2 3.3
3.3.1
Solid angle Lambert's radiator
Fundamental law of photometry Lambert's cosine law Calculation of radiation with small surface radiators and surface receivers Inverse square law
33
1
Laws of radiation
3.1 Solid angle
The solid angle is a sterometric value. Figure 3.1 shows a spherical sector of a sphere.
Q The value of the solid angle of a
spherical sector (Figure 3.2) is determined
by the ratio of the section of the spherical
A surface to the square of the radial distance
r. From this:
A^
n=-r. ^o
(3.1)
equation (3.1), area per area. In radiation calculations, the possibilities arise, that either the unit sr is not converted with
m m respect to 2 /m2 , or 2 /m2 is not
converted into sr. The correction factor flo = 1 sr, which is inserted, prevents such
calculation errors (see equation (2.19)
and section 10.1).
The greatest possible solid angle is formed by the sphere. If the radial distance is related to r = 1, e.g., r = 1 m, then the solid
angle is:
4 7rm
Tm fi =
"
7
no = 47r.fio = 47rsr
(3.2)
Semi-infinite space, or a hemisphere with unit radius r = 1 m, has the solid angle:
lm £2 =
-.n = 2 7r.fi = 2 7rsr.sr
52
(3.3)
Figure 3.
Sector of a sphere
Figure 3.2
Calculation of the solid angle Slofa
spherical sector
The unit of the solid angle is the steradian, abbreviated to sr. The general dimension of
the solid angle is, in accordance with
Figure 3.3
Two-dimensional representation of a
spherical sector. A = Area cut out of the
spherical surface by the spherical sector (spherical cap}, h = height of the section,
R = radius of the circumscribing circle, y= half aperture angle, r = radial distance
(from the centre to the surface of the
sphere) = 1,1 = r-h.
35
Figure 3.3 shows a two-dimensional representation of a spherical sector. Thus the solid angle Q, can be calculated with the half aperture angle <p
--A _ 2r.7T.h ^
fi = ;fto=
5
.fl,,
r
r
2r . 7T(l-cosy?)
U>
2 7T(l-COS<£) . fio
(3.4)
With the values R and 1, the solid angle
can also be calculated:
A
_ 2 7T r (r-1)
r
r
27T(1-
)fio
V^ +
z
l
= 27T(1-
"o
/£ + 1
(3.5)
To clarify this, Figure 3.4 shows the relationship between the solid angle £2 and
the half aperture angle For example: <J>.
lsr = (£= 32-72°
i 2irSI
f 5-
n = f(*>)
/
1 )
1
1»-
3
2
5»-
1
-
10
20
30
40
50
60
70
80 90°
Figure 3.4
Dependence of the solid angle £lon the half aperture angle <pofa spherical sector
3.2
Lambert radiator
The lambert radiator is an imaginary, ideal, exactly diffusely radiating black-body radiator (see Laws of Radiation). These conditions are satisfied by a uniformly
heated black spherical radiator.
The Lambert radiator has a uniform
radiance in all directions (Figure 3.5). Therefore, with this radiation source, the calculations are very simple
Figure 3.5
Radiation diagram of the Lambert radiator
3.2.1
Fundamental law of photometry
The radiances of the surface elements of a Lambert radiator are equal, even if they are observed at different angles. The areas
Asi and As 2 are simplified plane surfaces. From Figure 3.6:
36
6
Figure 3.
Section of the radiator surface of a Lambert radiator
-- Lei
=
Le 2=-Ie,o
Asi
_
h,o
AS 2
_
l e,y
Aps2
The fundamental law of photometry can
only be expressed exactly with higher mathematics. It reads:
_ I e,i/J
AS2 · cos ip s
(3.6)
Ie Ie yj
= Radiant intensity in the normal
direction = Radiant intensity in the direction
of angle </?
dAs2 · cos ips · dA E · cos ipg
d2
4>.
=
L (
· fio
(3.9)
In this form, in addition, it is restricted
to vacuum only. The distance r should always
be greater than the photometric limiting distance (see Section 3.3).
(Suffixes: S= Transmitter, E = Receiver,
P = Projected).
If the power radiated at the angle </$ frorn As2 falls perpendicularly at a great distance r on the small receiver surface Ap£, then the incident radiant power is:
% = Le : cos^s .As2.fi
Le . cos^s · As2 · ApE .Oo
(3-7)
If the receiver surface element Apg is a
projection of A£, the incident radiant power is:
% ^ A n A S 2 - cos ^S ·
E
cos
·
i^E
^ = Le
·
(3.8)
3.2.2
Lambert's Cosine Law
The proportionality of the radiated and of the incident radiant power to cos </> can be determined by experiment. This is called
Lambert's cosine law (Figure 3. 7).
In this diagram, dA$ is a surface element of a Lambert radiator. The radiant flux per
element of solid angle, the radiant intensity,
shows a maximum in the normal direction.
The radiant flux per unit of solid angle decreases symmetrically on all sides, in
proportion with cos <p:
e,ip
e,o
cos i/J
.
(3.10)
The total radiant power i£e of the surface
element dA§ of a Lambert radiator is the
37
integration over semi-infinite space, for the
M radiant emittance e :
M L L e = e J* cos </? . df2 = e . Tt . JT^>
(3.12)
Figure 3. 7 Lambert's cosine law
product of I eo and the integral over all
solid angle elements d£2in the semi-
infinite space.
*e = !e,o / cos ^ . dfi = Ieo .tt.Qq
(3.11)
The equation (3.6) stated, that the radiance
Le of a small plane surface As (strictly
speaking surface element dAg) of a Lambert radiator is equal for all directions. Thus, from equation (2.25), one obtains, by
Exact Formulae
Me = d$e
dAS
(2.9)
3.3
Calculation of radiation with small surface radiators and surface receivers
A small radiator surface can be regarded,
in simplified form, as a point radiator, if the distance r to the receiver is very large in
comparison with the maximum dimension
of the radiator surface (at least a factor of
10). This minimum distance is called the photometric limiting distance. The surface
radiator has its whole radiant flux in semiinfinite space. Therefore it is advantageous
M to work with the radiant emittance e ,
i.e., the total radiant flux of a surface element in semi-infinite space is related to the small finite area of the surface element dAg.
The following relationships are taken as
the basis for the radiation calculations:
Simplified Formulae
In the case of homogeneous radiation as with the Lambert radiator:
MP
AS
(2.11)
Figure 3.8
Projection of the transmitter area
For the apparent area of the surface element dAps:
dAps = dA§ . cos ifi
(3.13)
For the apparent area Apg:
Aps = As . cos (p
38
(3.14)
Vox the radiant intensity in the direction SE:
I e ,v? = h,o · cos*P
(3.10)
For the radiance of the point S in the
direction SE (see quation (2.22):
w dle,<£ dAps
dl,e,<£
dAs . cosip
d 2 <£e
dAs . cos^J . dii
2
d
4>e
=
Le
dQ, . dAs · C0SV
(see equation 2.26)
(3.15)
(2.23) (3.16)
=I Ie,i/J e,o -cosv>
(3.10)
With uniform radiance (see Section 2.2.3 and 3.2.1):
e,i£ _ Ie,i^
Aps A§ . cos^p
<*V
Le = As . cos/? . £1
d> e
=
Le
.
£2 .
As
.
cos£
(3.19)
(2.28) (2-31)
In order that the rays meet the receiver surface at about the same angle, the distance r must be great in comparison with the size
of the area. This minimum distance is the
photometric limiting distance.
Figure 3.
Projection of the transmitter and receiver
areas.
For the solid angle from the point S to
dAE :
jri diZs
=
d APE -~Y~
.
iAj
=
dAE
cosyfr j
.
iZo
(3.17)
For the solid angle from point E to dAps".
^ -- ^ dii£
=
dAps j--
.
iZo
=
dAscosv>S
j
·
_,
(3.18)
APE "S=-T~-
^o
=
AE
n cos>PE
i .^o
r
r
(3.20)
-- i^ A A =
PS j-.
i^o
=
S . cosfls
5
· iio
(3.21)
39
If dfi is replaced in equation (3.16) by the equation (3.17), then the radiant flux <&e falling on dAg, and thus the fundamental photometric law, is obtained (see equation
3.9):
$ n d,22^e = rLe dAE · cos
E dA ·
S · c0SlfiS
j
"
· "o
r
(3.22)
The symmetrical construction of
equation (3.22) allows dAE t0 be considered as the radiator and dAg as the receiver, if the radiance L e is the same in
both cases.
dAs . cosies
The expression
=
can be replaced
r
by dfig:
dA 2
d <l>e
=
Le
. dftE
.
E . cosi/^
(3.23)
A ,, E As . cosflE .
· cosifls
*e = Le
=
· S2o
(3.26)
A E L <£e = e . fl . E . cosine
(3.27)
In equation (3.22), Le and dAs · cosies can
be replaced by dl e ^:
z
d
q>e
=
dl ee,i<0p ^
z2
SZ-q«
(324)
^ In equation (3.24), d 2 <&e , dA E and dl e
can be replaced by E e and I ei /> (see equation 2.38). The photometric inverse
square law is obtained:
(3.25)
A E . coscpg
&e = htfT.
2
-^o
r
Ie^.cosfc
^e
7
· ^K)
(3.28) (3.25)
3.3.1
Inverse square law
This states, that the irradiances at two surfaces, which are struck by the rays from a point source of radiation at equal
angles, are inversely proportional to the
square of the distance of these areas from the source of radiation {Figure 3. 1 0)
Pninl-tnnrrr or ratnauon
^^
-prr-
J^^^^r-~~---1__ 1
^|
^'
2--__
3 __
Figure 3.10 Decrease in irradiance with distance (inverse square law)
40
The mathematical expression for this reads:
Ee ,l Ee> 2
^o I e ,^ . cosi/?e ·
ii
dp cos^e e I ,i^> ·
.
*2
(3.29)
Ee \ x\ Ee,2 r?
(3.30)
41
4
Laws of Radiation from a Black Body
4.1 4.1.1 4.1.2 4.2 4.2.1 4.2.2 4.2.3 4.2.4 4.2.5 4.2.6 4.2.7
Black and "Non-black" Bodies Black bodies Non-black bodies Laws of radiation Planck's law of radiation
Stefan-Boltzmann Law Wien Displacement Law
Emittivity
Kirchhoffs Law
Radiation isotherms Reduced law of radiation
43
'
1
Laws of Radiation from a black body
4.1
Black and 'Non-black" Bodies
4.1.1 Black bodies
Every body which has a higher temperature than Kelvin emits radiation due to its own temperature. These radiation sources
are called temperature radiators.
The laws of radiation are derived from a
theoretical ideal radiator, the black-body or black radiator. This is understood to
mean a body which completely absorbs radiation of any wavelength falling upon it at any angle. Furthermore, the black body emits, in every direction and at any wavelength, the maximum possible radiant energy, as compared with other temperature radiators of the same temperature, geometrical shape and dimensions.
In practice, black -body radiators can only be produced for limited temperature ranges. The best-known model of a black body is the cavity radiator {Figure 4.1): an enclosed cavity, the walls of which are impervious to heat and are at the same temperature, is
filled with "black radiation" (see DIN
5496). In this case, the cavity radiation
depends only on the wall temperature, so that it is almost equivalent to the radiation of a black body of equal temperature. The cavity emerges through an aperture, while its character as black radiation does
not change.
In order to obtain substantially ideal black radiation, first the aperture is kept small in relation to the surface of the cavity and secondly the reflectivity of the cavity walls is reduced with suitable blackening agents.
Radiating surfaces differ more or less in their radiation performance from that of a black body, depending on their quality.
Figure 4. Cavity radiator
1 or the identification of temperature radiators, various temperature terms and data are needed: The unit of absolute
temperature T is the Kelvin (K).
[T] =K
(4.1)
The other commonly used unit of
temperature t is the degree Celsius ( C).
[t]= C
(4.2)
The Kelvin is defined as the 27 3- 16th part of the thermodynamic temperature of the triple point of water (see DIN 1301).
1
IK =
. 273-16 K
273-16
(4.3)
The temperature t in degrees Celsius is the difference between any given thermo-
dynamic temperature T in K as compared with the freezing point of water i.e. T =
273-16 K.
t = T-T = T- 273-16 K
(4.4)
Absolute zero, T = K, is obtained from
equation (4.4) with
45
OK- t =
27316 K = -273-16 C (4.5)
If the thermodynamic temperature is
represented as the sum of T and a Celsius
temperature, the following equation applies:
T=T +t
(4.6)
The black body, at any temperature Ts
and for any wavelength X, has a spectral radiance which is determined by Planck's law of radiation. For every wavelength X, any temperature radiator is allocated
that temperature T s of the black body, at
which the temperature radiator has the same spectral radiance as the black body.
For the temperature radiator under
consideration, at wavelength X, this
temperature is called the spectral radiation
temperature
T s
or
black-body
temperature.
For non-black temperature radiators, it is
always lower than the true temperature.
The colour temperature Tf of a radiator is that temperature of the black body, at which the black body gives the same impression of colour (has approximately the same spectral radiation distribution in the visible range) as the radiator under
consideration. A colour temperature can be
allocated to temperature radiators, but not in all cases to luminescence radiators.
The distribution temperature Tv of a
radiator is that temperature of the black body, at which the radiation function of the radiator under consideration, in a wavelength range to be stated, between Xi and X2, is strictly or approximately proportional to the radiation function of the black body. If the stated spectral range includes the visible range, the djstfibwtion temperature corresponds to the colour temperature.
The ratio temperature Tr is that
temperature of the black body, at which the ratio of the spectral radiances for two different wavelengths \\ and X2 is equal to that of the radiator under consideration.
4.1.2 Non-black bodies
Grey emitters are described as non-black radiators and their spectral distribution of the radiance is strictly or approximately proportional to the distribution of the spectral radiance of the black body. The emittivity in semi-infinite space (see Section 4.2.4) is equal or constant for all wavelengths of the spectral range under consideration. The colour temperature Tf, the distribution
temperature Tv , the ratio temperature Tr and the true temperature T of a grey emitter have the same value in each case (see DIN
5496).
Tf =Tv
T =T r
(4.7)
In practice, non-black temperature radiators are only grey emitters for a given spectral range.
Selective radiators are non-black radiators. Their spectral distribution of radiance does not correspond to the spectral distribution of radiance of the
black body. They have spectral emission bands or lines. Selective radiators are mostly
luminescence emitters.
Mixed radiators are also non-black bodies.
Their spectral distribution of radiance corresponds to that of a grey emitter, but with additional superimposed emission bands or lines. Mixed radiators emit temperature and luminescence radiation simultaneously.
4.2
Laws of radiation
4.2.1
Planck's law of radiation
Planck's law of radiation permits spectral radiation calculations for the black body.
To simplify the radiation conditions, a plane radiating surface As of a black body
46
is assumed. The radiance of every surface
element dAg is constant, since the blackbody is also a Lambert radiator. The total
radiant flux from the plane surface As,
M the radiant emittance e s, is emitted in
semi-infinite space only. The spectral
M radiant emittance X e>s? will be calculated
as follows:
lit. Co.h
Me,s,X=Co
X.k .T
Xs . (e
-1)
(4.8)
Co = Velocity of light k = Boltzmann's constant h = Planck's constant
A$ 2 h c .
.
.
· coap
Ie,s,X :
X.K.T
^.Xs .(e
-1)
(4.10)
4.2.2
Stefan-Boltzmann Law
The total radiation in semi-infinite space is given by integration of radiation law (4.8) over the whole wavelength range
Me,s = /Me ,s,X-dX
(4-11)
M = O.T4
(4.12)
The factor 2 applies for unpolarised radiation, while it is omitted for linearly polarised radiation. With linearly polarised radiation, the vector of the electrical field strength of an electromagnetic wave only
oscillates in a certain direction, perpendicular to the direction of propagation. With unpolarised radiation, the direction of oscillation is subject to a continuous irregular change. This radiation can be visualised as the statistical superimposition
of two field-strength vectors, perpendicular
to each other, Ex and E y , each with half
the radiant power as compared with the unpolarised radiation. Since Lambert's law applies exactly for black-body radiation, the spectral radiance can be calculated from equations (3.12) and (4.8).
l«aX =
2v . 4> h Cp .h
X.k.T
J2o.Tr. Xs . (e
-1)
2.<£.k
c h -
X.k.T «o.X5 .(e
(4.9)
The spectral radiant intensity can be determined in accordance with equations (2.29) and (4.9), if the size of the emitting area Ag is known:
15 h 3 Co a =5-67 . 10 8° Wni or
a =5-67 . 10~ 12 Won 2 K 4
(4.13)
The Stefan-Boltzmann law, corresponding
to equation (4.1 2), states that the radiant
emittance of the black body is proportional to the fourth power of its temperature. In this, a is the Stefan-Boltzmann constant.
In optoelectronics, the radiant
M emittance e>s of a black body often
relates to the temperature 1\ = 2042 K
(previous temperature for the definition of
a candela; now 2045 K) and 1 2 = 285 ° K
previous colour temperature of a tungsten
lamp for the -measurement of the photo-
current sensitivity of Si photodetectors;
now 2856 K).
ForTi:
Me s ,l
=
5-67
.
10" 12 Wcnf 2 K~ 4
44 .2045 .K
,
(4.14)
M Wcm e ,s,l
= 5 " 67 ·
10_I2
·
175 ·
10 12
-2
Me s,l = 99 Won-2 ,
47
.
The radiance Le>Sj i in any direction in semiinfinite space can be calculated with equation (3.12). For the temperature Ti = 2045 K:
_Me> l
Le,s,l
(4.15)
Le,s,l
99 W
7T. lsr cTM,m 2
(4.16)
Le,s,l
=
31-55
Wcm~
2
-l sr"
\Y
TT(K) 2000 \
1000
2
4
6
8
X GTM0
For T2 = 2856 K, the following is obtained
for the radiant emittance:
Me,s,2 = 5-67 . 10
12 Wcm
2K
4
4 .2856
K4
(4.17)
Wcm Me)S)2 = 5-67
.
10" 12 ·
6-65 .
13
10
Me ,s,2 = 377Wcnf 2
According to equation (4.15), the radiance
Le s 2 in any direction amounts to:
Le,s,2
377 W
cm Tt . 1 sr .
(4.18)
Le,s,2 = 120Wcm
2 .sr
'
Figure 4.2
The Wien displacement law
\nax.T·= 2080 yon K
In the literature, the following dimensions
are often found:
W
Me ,s,Xmax=
1-309.
,-is
10
cm2 (Atm)K s
M As in Section 4.2.2, the values of e SjXmax
for the temperatures T^ = 2045 K are also
of interest here, as is the temperature of
the sun, with T3 = 6000 K, for comparison.
Me,s,l,Xmax
1-309.
10~ 1S
W.
2045 s
Ks
.
K cm2
5
. /xm .
-2
-1
= 46W. cm (Mm )
(4.20)
4.2.3
Wien Displacement Law
It, is deduced from the radiation law, that a
black body emits its maximum radiant
power at a given wavelength Xniax> (^max is temperature-dependent). The spectral radiant emittance at Xmax increases very steeply in proportion with the fifth power
of the Kelvin temperature {Figure 4.2).
W
Me,s,Xmax = 1-309 . 10" 18 cm2 (nm)K s
(4.19)
M e,s,2,X,max
W 1-309 .10 ls .2856 s . K 5
K cm2 . /xm . 5
= 248 W. cm
2 <JXm
1 )
Me s 3 Xr,
1-309. 10 ls W. 6000 s -K s cm2 . Mm . K 5
10178 W. cm 2
Mm
1 )
(4.21) (4.22)
48
As a further example, Me^^max wil1 be calculated for the temperature T = 2000 K
Me,s,Amax
1-309. 10" 15 W. 2000 5 .K5
cm K 2
s
. (/Urn) .
41-8 Wcm
2 (jUm
')
(4.23)
This value agrees with that of 4.18 · 10
Wm~ 2
l
()Um~ )
in
Figure
4. 4.
From equations (4.15) and (4.23), the maximum spectral radiance Le,s,Xmax at
the emission temperature T = 2000 K
amounts to:
By inversion of the formula (4.27), if the temperature is known, the maximum
wavelength Xmax at maximum emission of
the black body can be calculated.
2880 Urn K
'Vnax
(4.28)
The following examples indicate the
significance of the law (4.28):
An IR vidicon is to take a thermal radiation picture of the human body. The IR vidicon
target must therefore have high sensitivity to
the maximum energy radiated from the human body at the maximum wavelength Xmax . In simplified form, the human
body is assumed to be a black-body radiator with a temperature T = 300 K. From
equation (4.28), its maximum wavelength is
Me ,s,Amax
Le,s,Amax =
(4.24)
2880 Mm K
Xmax =
: = 9-6 fJm
300 K
(4.29)
Le,s,X,max
W 41-8
n. lsr. cm (/im)
(4.25)
1
1
Le.s.Atta^l^WcnfV (Mm" )
This value agrees with that of 1-36 . 10 s W.m 2
i
·sr" ' (Mm ) in Figure 4.5.
At this wavelength, the vidicon target must
have its maximum sensitivity. On the other
hand, a black -body radiator, at a
temperature of 2045°K (definition temperature of the candela) has a maximum
wavelength:
2880 Mm K XmaxX = 2045 K = 1-4 Mm
(4.30)
It is specified in the radiation law, that a
black body shifts its maximum wavelength Xmax wrtn the maximum emission to
shorter wavelengths with increasing
temperature. The equation (4.26) states,
that the product of the maximum wavelength Xmax and the emission temperature T remains constant:
vmax T = constant
(4.26)
Furthermore, at a temperature of 2856 K
(colour temperature of the scientific tungsten lamp for measurement of the photocurrent sensitivity of Si photodetectors) the black-body radiator has a
maximum wavelength of
2880 Mm K
X,max
= 1-04 Mm 2856 K
(4.31)
\nax · T
h .c
K . 4-965
2880. 10 6 mK
(4.27)
K and finally, at a temperature of 6000
(temperature of the sun), it has a
maximum wavelength of
Xmax · T = 2880 Mm K (Wien constant)
2880 Mm K
vmax
6000 K
480 nm
(4.32)
49
By conversion of the formula (4.27), if the maximum wavelength Xmax is known, the temperature of the black body can be
calculated.
2880 jum K T=
ax
(4.33)
If the maximum wavelength Xn^ of the
black body corresponds with the wave-
length of the maximum photopic
sensitivity of the eye (daylight vision)
Vji
= 555 nm, then its temperature is
2880 urn K
T=
=5189K
0-555 /Xm
(4.34)
The Wien displacement law is illustrated
in a graph in Figure 4.2. In this, the results of equations (4.29) to 4.31) and (4.34) can be read.
4.2.4 Emittivity
The radiation from all bodies in a wavelength range under consideration is dependent firstly on the temperature and
secondly, in the case of non-black bodies, on the material composition and surface condition.
The ratio of the emitted radiation of any
given temperature radiator to the emitted
radiation of a black body of the same temperature is known as the emittivity 6.
The ratio of the spectral radiance Lg \ of a
temperature radiator in the normal
direction to the surface (optical axis) to the
spectral radiance Le jSjXof the black body
at equal temperature is defined as the spectral emittivity normal to the surface <*X) n
Le,X,n
e(X)n :
Le,s,X
(4.35)
The ratio of the spectral radiance Le>X,<£ of
a temperature radiator in a given direction to
the spectral radiance Le>s \ f the black-body
of equal temperature is called the directional
spectral emittivity e (\ip):
e(K<p)
s,\ip
Le,s,X
(4.36)
All non-black bodies have a directional
< spectral emittivity e (X,<$ of 1 ; in many
cases only a few percent. Integration of the
directional spectral emittivity over the wave-
length range between Xi and X2 will be defined as the directional band emittivity e (<$b and that over the whole wavelength
range as the directional total emittivity
e(<0 = J Le,X,<i · d X /Le)S>\. dX
(4.37)
M The ratio of the radiant emittance e of a
temperature radiator to the radiant
M emittance e>s of the black body of
equal temperature is called the emittivity in
semi-infinite space .-
Ue en =
M,e,s
(4.38)
As is shown by Figure 4.3, the spectral
M radiant emittance e \ of a non-black
body is always less than the spectral
M radiant emittance e s \ of a black body at
the same radiator temperature and wavelength.
Figure 4.3
Spectral radiant emittance of a black and non-
black body as functions of the wavelength X
50
4.2.5
Kirchhoffs Law
In the case of a temperature, radiator, the directional spectral emittivity, for every temperature and every wavelength, is equal to the spectral absorption for incident radiation from the same direction (see Section 6.1).
e(X,<0 = a(X^»)
(4.39)
M The spectral radiant emittance \ 6)St is
represented as a function of the wavelength
X. The calculation in equation (4.23) can be verified in this graph.
Figure 4.5, on the other hand, shows radiation isotherms on a log-log scale. The
spectral radiance Le>s>X is represented
as a function of the wavelength X. The
calculation in equaiton (4.25) can be verified with this graph.
4.2.6 Radiation isotherms
The spectral radiation units of the black body can be represented firstly, as a
function of temperature w^th the wavelength as a variable, as isochromatic lines and secondly as a function of the wavelength with the temperature as a variable as radiation isotherms.
Figure 4.4 shows radiation isotherms of the black body on a linear scale.
10'
H / 1
I1
/y
i i ~r~
i
1
»/i
'
I
10 77 T1
.
// *%oc
't
[ft
/ #" /i
J/
t
W1 r */
/'
fr I; ,
r/
t |
If / 1 / / /
'"o-l »J 04M04I 2 4 » »»
^
s.
\i
\ 's
s
» 20 40 00 100
>dmO
.
.
1
Figure 4.5
Radiation isotherms L e>s< \ = ffh) on log-log
scale
Figure 4.4
M Radiation isotherms \ e>St = ffN on a
linear scale.
4.2.7
Reduced law of radiation
In practice, both absolute values of the spectral density of a radiation unit at a wavelength which is of interest, X, and also relative units of a proportion of the radiation (see equation 2.65) are needed from a Planck radiator with a given temperature T. If the spectral density of the radiation units is represented generally by XX, then the spectral density of a radiation unit from a Planck radiator
with constant temperature T and at a given wavelength X can be calculated from
the following general equation:
XX
X\-(-
-)-
XXm;ax
XXm;ax
(4.40)
51
The ratio of X>A to X\Amax in the bracket
in equation (4.40) is the radiation function
SX- From the equations (2.66) and (4.40),
one obtains:
X\=S\.x\max
(4.41)
The spectral radiant emittance is usually selected as a spectral radiation unit for XXThe spatial distribution of the radiance or radiant intensity from the Planck radiator under consideration will not then be
needed.
The values of the spectral radiant emittance for the black body are inserted in
equation (4.41):
M M e,s,X = SX
e,s,Xn
(4.42)
The radiation function SX can be read off,
with the product of the desired wavelength
X and the temperature T of the black body (X.T)/cm. K in the Table 4.1 of the graph in Figure 4.6. From this graph, the radiation
function SX can also be seen from the quotient of the desired wavelength X and the maximum wavelength Xmax of the
black body.
X.T
cm . K
0-0800 0-0825 0-0850 0-0875 0-0900 0-0925 0-0950 0-00975 0-10 0-11 0-12 0-13 0-14 0-15 0-16 0-17 0-18 0-19 0-20 0-21 0-22 0-23 0-24 0-25 0-26 0-27 0-28 0-29 0-30
sx
0-0015 0-0020 0-0030 0-0042 0-0060 0-0083 0-011 0-014 0-017 0-038 0-074 0-128 0-191 0-265 0-349 0-430 0-521 0-613 0-685 0-735 0-790 0-844 0-895 0-931 0-972 0-982 0-996 1-000 0-997
Table 4.1
Radiation function S\ = f(X . T)
X.T
cm . K
0-31 0-32 0-33 0-34 0-35 0-36 0-37 0-38 0-39 0-40 0-41 0-42 0-43 0-44 0-45 0-46 0-47 0-48 0-49 0-50 0-52 0-54 0-56 0-58 0-60 0-62 0-64 0-66 0-68
SX
0-984 0-965 0-945 0-921 0-898 0-875 0-851 0-829 0-806 0-783 0-759 0-734 0-711 0-686 0-664 0-641 0-618 0-595 0-572 0-548 0-507 0-473 0-435 0-400 0-367 0-339 0-313 0-289 0-267
52
X.T cm . K
0-70 0-74 0-78 0-82 0-86 0-90 0-94 0-98 1-0
1-1 1-2 1-3 1-4 1-5 1-6 1-7 1-8 1-9 2-0 2-2 2-4 2-6 2-8 3-0 3-2 3-4 3-6
SX
0-249 0-216 0-187 0-161 0-139 0-121 0-106 0-093 0-089 0-064 0-049 0-038 0-028 0-023 0-019 0-015 0-0125 0-010 0-0085 0-0060 0-0044 0-0032 0-0024 0-0019 0-0015 0-0012 0-0010
6
7
M The spectral radiant emittance X e)S; of a
black body with a temperature of
T = 2045 K, at the wavelength X = 0-93 jum, is to be determined. At its maximum wavelength Xmax = 1-4 jLlm (see equation
4.30), the spectral raidant emittance
Me ,s,Xmax = 46 Wcnf '(/mf 1 ) (see
equation 4.20). The product (A · T) gives:
X. T = 0-93 /xm . 2045 K = 0-19 cm K
(4.43)
The quotient of the desired wavelength X and the maximum wavelength Xmax is
Xmax
0-93 /urn = 0-664
1-4 Aim
(4.44)
The radiation function S\ can be read off
from Table 4.1 or from Figure 4.6 with the value from equation (4.43), or from Figure
4.6 only with the value from equation (4.44). One obtains:
SX= 0-613
(4.45)
M The spectral radiant emittance e)S)\,
from equation (4.42), is:
M Wcm = e s ,X ,
0613
.
46
2 (jum
')
=
28-2Wcnf 2
-1
(/Lftn )
(4.46)
If, from a Planck radiator at a given temperature T, the ratio is formed, from equivalent radiation units, of the shorter-wavelength radiation up to a
selected wavelength
SX= f (r )andSX=f(X.T) Xmax
X, to the total radiation, this ratio is called
the proportional radiation q (see also equation 2.65):
/Xx e ,x.dX
J'X e ,X-dX
(4.47)
If the spectral radiant emittance of a black-
body radiator is inserted in the equation (4.47), then the following formula is
obtained:
0-2 0-3
0-5 0-7 1
Figure 4. Radiation function
Figure 4. Proportional radiation q = f(~h)
53
J*Me,i,X- dX
/M e ,s,X-dX
(4.48)
Figure 4. 7 shows the proportional radiation q as a function of the selected wavelength
X, with various temperatures T of the black
body:
With any given temperature T of the black
body, and with the product (X . T)/cm. K, which is formed from the selected wavelength
X and the temperature T of the black
body, the proprotional radiation q is read off from the Table 4.2 or from the graph in Figure 4.8.
The shorter-wavelength proportional
radiation q^ from a black body with
temperature T = 2856 K, up to the
wavelength \\ = 615 nm, is to be
determined. The product
(Xi
T) .
is:
h.T 0-615 . 10~4 cm . 2856 K
01756 cmK
(4.49)
X.T
cm . K
0050
0-052 0-054
0056
0-058
0060
0-062 0-064 0-066
0068
0-070 0-072 0-074 0-076 0-078 0-080
0082
0-084 0-086
0088
0-090 0-092
0094 0096
0-098
0100 0110 0120
0-130
Q
1-3652. 10~ 9 3-6788 . 10" 9 9- 1749 . 10~ 9 2-1358.10" 8 4-6745 . 10~ 8 9-6798 . 10" 8 1-9069 . 10 7 3-5907 . 10" 7 6-4902 . 10" 7
11 302 . 10~ 6
1-9025 . 10~ 6
31045 . 10" 6
4-9236 . 10~ 6 7-6070 . 10" 6 1-1473 . 10" 5 1-6923 . 10~ 5 2-4453 . 10" s 3-4668 . 10" s 4-8287 . 10" 5 6-6159. 10~ 5 8-9269 . 10" s 1-1874. 10~ 4 1-5586 . 10~ 4 2-0204 . 10" 4 2-5885 10~4
.
3-2804 . 10~ 4 9-2957 . 10~ 4 2-1727. 10~ 3 4-3866 . 10~ 3
X.T
cm . K
0-140 0-150 0-160 0-170
0180
0-190 0-200 0-210 0-220 0-230 0-240 0-250 0-260 0-270 0-280 0-290 0-300 0-310 0-320 0-330 0-340 0-350 0-360 0-370 0-380 0-390 0-400 0-420 0-440
Table 4.2
Proportional radiation q=f(\- T)
Q
7-9053 · 10" 3 1-3023 · 10~ 2 1-9962 . 10"" 2 2-8858 . 10~ 2 3-9754 . 10" 2 5-2613 · 10" 2
2
6-7331 . 10 8-3750 . 10~ 2 1-0168 . 10" 1 1-2091 . 10 1 1-4122. 10" 1 1-6239 . 10" 1 1-8423 . 10" 1
_1 2-0653 . 10 2-2911 . 10" 1 2-5183. 10" 1 2-7454 . 10" 1 2-9712 . 10" 1
31947 . 10" 1
3-4150 . 10" 1 3-6314 . 10""' 3-8432 . 10" l
_1
40502 . 10
_1 4-2518. 10 4-4479 . 10"
_1 4-6382. 10 4-8227 . 10" 1
51738.10" 1
5-5012 . 10" 1
54
X.T
cm . K
0-460 0-480 0-500 0-520 0-540 0-560 0-580 0-600 0-620 0-660 0-700 0-740 0-780 0-820 0-860 0-900 0-940 0-980 1-00
110
1-20 1-30 1-40 1-50 1-60 1-70 1-80 1-90
200
q
5-8057 . 10" 1 -1
60880 . 10
6-3494 . 10 ' 6-5912.10 1
1
6-8146 . 10 _1
70209 . 10
7-2116. 10 * 7-3877 . 10 ' 7-5505 . 10 ' 7-8402 . 10" J
80885 . 10"
8-3020 . 10" 1 8-4861 . 10 ' 8-6455 . 10" 1 8-7840 . 10 1 8-9048 . 10" 1 9-0105. 10" * 9-1033. 10" 1 9-1455 . 10" 1 9-3217. 10" 1 9-4532. 10" 1 9-5331 . 10" 1 9-6304 . 10" 1 9-6909 . 10" 1 9 7390. 10" 1 9-7777 . 10" 1 9-8091 . 10" 1 9-8349 . 10 ' 9-8563 . 10" 1
the radiation in the wavelength range between Xi and X2 in relation to the total radiation, then this ratio is also called the "proportional radiation q":
*2 /Xe,X.dX
q-^
J'X e ,X.dX
(4.53)
If, for easier calculation, the spectral
radiant emittance of a black body is again inserted in equation (4.53), the new formula:
X2 JMe,s,X.dX
Xl
·lMe )S ,XdX
(4.54)
Figure 4.8
V Proportional radiation q = f (X-
With the value from equation (4.49), the proportional radiation qi is determined from Figure 4.8 or by interpolation from Table 4.2, or better still from a substantially extended table.
-2
qi '3-45. 10
3-45%
(4.50)
From the same black body, the shorterwavelength proportional radiation q2, up
to the wavelength X2 = 510 nm is to be
determined. The product (X2 . T) is:
\2 .T = 0-51 . 10" 4 cm. 2856 K
= 01456 cmK
(4.51)
The proportional radiation q2 is determined with the value from equation (4.51), as
before. We obtain:
-2
q2«105 . 10
1-05%
(4.52)
If, as in section 2.3.4 and similarly to equation (2.65), the ratio is formed for a
Planck radiator with a given temperature T
and with equivalent radiation values, from
is obtained.
From the previous formula, the proportional radiation qi up to the wavelength Xi and the proportional radiation q2 up to the
wavelength X2 can be determined. If the longer wavelength is selected for Xi and
the shorter wavelength for X2, then the
difference between the proportional
radiation qi and the propertional radiation
q2 gives the desired proportional radiation
q=qi-q2
(4.55)
The proportional radiation q from a
black body with temperature T = 2856 K
is to be determined in the wavelength
range between Xj =615 nm and X =510
nm. The value for qi = 3-45% has been determined in equation (4.50) and the
value for q2 = 1-05% in equation (4.52). From equation (4.55), the proportional
radiation q is:
q = 3-45%- 1-05% = 2-4%
<4.S6)
Note:
The temperature slide-rule "Temperaturstab Nr.922", from Aristo, is suitable for the
calculations of equations (4.16, 4.18, 4.28 to 4.34, 4.47, 4.48, 4.50 and 4.52).
55
The calculations of equations (4.12, 4.14 to
4.23, 4.28 to 4.34, 4.38, 4.45, 4.47, 4.48, 4.50 and 4.52) can be carried out
with the "Radiation Calculator" or the Radiation Calculation Slide Rule from General Electric.
56
5
General and
Photometric Evaluation of Radiation
5.1 5.2 5.3 5.3.1 5.3.2 5.3.3 5.4
5.4.1
5.4.2 5.4.3
5.5 5.5.1 5.5.2
5.6 5.6.1 5.6.2
The human eye as a light receiver
Optical sensitivities Photometric evaluation of radiation spectra Determination of the conversion constants
C and C
Photometric radiation equivalent Photometric radiation equivalent of the photopic sensitivity of the eye Calculation of the photometric radiation
K equivalent for different radiation
spectra Calculation of the photometric radiation
equivalent K of a Planck radiator with the
temperature laid down for the definition
of the candela Calculation of the photometric radiation
K equivalent for standard light source A
Calculation of the photometric radiation
equivalent K of the luminescence
radiation from light-emitting diodes (LED's) Conversion of radiometric units into photometric, photopic units Conversion of radiometric units into photometric units for Planck radiation Conversion of radiometric units into photometric units for luminescence radiation from light-emitting diodes
Actinic value Actinic value of radiation spectrum for
photodetectors and for the human eye
Actinic value of Planck radiation and luminescence radiation with the eye as a photodetector
57
5.1
The human eye as a light receiver
The luminous flux from an observed object passes through the optical system of the eye to the retina, so that a true image of the object is produced there. The optical system of the eye consists of the transparent cornea, the aqueous humour, the iris with the pupil, the lens and the vitreous humour. Since the curvature of the lens of the eye can be varied (variation of the total
refraction), the effective focal length of this optical projection system is variable. This capability of the eye to adapt itself to produce focussed images of objects at different distances, is called accommodation.
The pupil of the iris adapts the area of
its aperture within certain limits to the
effective luminance in the field of vision.
With high luminances in the field of vision,
the pupil has a small aperture area and
conversely with low luminances it has a
large aperture area. The aperture area of
the pupil acts like a variable stop, so that
the solid angle for the incident luminous
flux on the individual points of the retina
is variable in the ratio of about 1:16. The retina of the human eye consists of
about 1 rod-shaped visual cells and
6 7.10
cone-shaped
visual
cells.
After
their
·sensitivity threshold is exceeded, the cones
respond to differences in brightness and
the wavelength of the incident light. They
pass their stimuli, due to the incoming
light, in the form of electrical pulses, each
through an optic nerve fibre to the cerebral
cortex. Above a luminance in the field of
-
vision of about 10 cd.m
predominantly
,
only the cones are excited. The cones are
therefore always excited in daylight, so
that cone vision is also called "daylight
vision".
The cones are located, in a very dense
General and photometric
evaluation of radiation
concentration, mainly in the central part of the retina (yellow spot). Here, the eye has its greatest resolution capability of about one minute of arc. Through continuous
oscillating movements, the eye can make
use of this high resolution capability over a relatively wide angle of the field of view.
The rods only respond to differences in
brightness. They are mainly excited at
a luminance, in the field of vision, below
about
10~ 3
-2
cd.m . The
cones
can
no
longer process such low luminance
values in the field of vision. Therefore,
at night or in darkness, only the rods are
still sensitive, so that rod vision is also
designated "night vision". The rods pass
their excitation, due to light stimuli, in the
form of electrical pulses, in inter-
connected groups, each through an optic
nerve fibre to be cerebral cortex. In the
central part of the retina there are hardly
any rods, but they are distributed, in relatively lower concentrations as compared
with the cones, over the remainder of the
surface of the retina. Through, the
interconnection of the individual rods into
groups, each group works on one optic
nerve fibre, so that the effective receiving
area is increased. This characteristic
causes a considerable multiplication of the
sensitivity in the case of rod vision.
This capability of the eye to adapt itself to different brightness sensitivities with cone and rod vision and the capability of varying the pupil area according to the luminance in the field of view, is called adaptation.
5.2 Optical sensitivities
From considerations of radiation physics,
the excitation of the visual cells of the
human eye occurs through absorption
59
of the incident visible radiation. The photosensitive material in the rods and the
three different photosensitive materials in
the cones absorb the incident radiation with different spectral sensitivity distributions. Although there are three types of cone, in the case of cone vision, the resultant spectral absorption of all cones is assessed together for the sake of simplicity. With daylight vision, the eye only works in the spectral brightness sensitivity range of the
cones. The spectral brightness sensitivity
of the eye, adapted for brightness, with -
a luminance of 10 cd.m in the field of view, is defined as the photopic sensitivity V(X) of the eye. Here, the eye with normal colour vision has its highest
sensitivity V(X) max at the maximum wavelength Xmax = 555 nm. The limits of
the photopic sensitivity range of the eye
lie around \\ = 380 nm and X2 = 780 nm.
The ratio of the spectral radiance
Le,555nm ° r Le.Xmax at the maximum wavelength Xmax = 555 nm to the spectral radiance Le X at another wavelength X in the photopic sensitivity range of the
eye gives, with equal excitation of the
cones of the eye, adapted to brightness,
the photopic sensitivity V(X) for this
wavelength:
V(A) = Le 'Xmax
Le,X
(5.1)
The sensitivity v(X) of the eye is a relative value. In Table 5. , the V(X) values are
listed, in coarse steps, as a function of the wavelength X. Figure 5. 1 shows the V(X) curve plotted with these V(X) values.
In the case of night-vision, the eye only works in the spectral brightness sensitivity
range of the rods. The spectral bright-
ness sensitivity of the eye, adapted to darkness, with a luminance of 10
-2 cd.m n the field of view, is defined as the scotopic sensitivity V'(X) of the eye. Here, the eye with normal vision has its
highest sensitivity V'(X) max at a maximum
wavelength Xmax = 507 nm. The limits
of the scotopic sensitivity range of the
eye lie around \y = 330 nm and X2 =
730 nm. The ratio of the spectral
radiance L' e ,507nm or L' e ,Xmax at the
maximum wavelength Xmax = 507 nm, to the spectral radiance L' e \ at another wavelength X in the scotopic sensitivity
range of the eye, which is necessary for
equal excitation of the rods of the eye,
when adapted for darkness, gives the
scotopic sensitivity V'(X) of the eye.
V(X) = L'e.X,max
a
(5.2)
The scotopic sensitivity V'(X) of the eye is
a relative value. In Figure 5.1, the scotopic
sensitivity V'(X) is shown by a broken line,
as a function of the wavelength X. All the V(X) and V'(X) values represent the
measured values ascertained by the CIE
through extensive series of measurements. Because of the change in definition of the candela, in which the temperature of the
black body has been redefined from 2042 K
to 2045-5 K, the latest V(X) and V'(X)
vlaues can be looked up in DIN 5031,
Part 3.
5.3 Photometric evaluation of radiation
An absolute evaluation of a radiation from
the brightness perceived by the eye is not possible. Sensations of brightness can give different results under different
conditions of observation.
Monochromatic radiation in the visible range, entering the eye, causes an excitation of the receiver cells, which is proportional to V(X)Le ,X- &K. If several different monochromatic radiations strike the retina simultaneously, then the total excitation of the visual cells concerned is equal, to a sufficient degree of accuracy, to the sum of the individual excitations. Even with a large number of monochromatic radiations,
60
. io V(X) v<X)
1
0-05
002
0-01 0-005 0-002
0001 0-0005 0-0001
S00
/
V(X)
Scotopic
/
--< t /
^= s- ^ N
\
v.
^
V
V
s
/
/t
\ \
\
s
Pholopic
/
I
V 3=
--/ / /
n
-- -^
/
r
/
\ \
-
/
i
1
--t
/
\
1
1
--I /
f
f
--
H
f--
/
1
/
T1-
/
\ -v-
i
\
\
\
V
\
\
1
1
1
--
--11 --f --h
1 1 1
I
i
<00
500
A--
-v-
-- \
i
-\ \
\
i
VX
\ \
1
\
i [_ \
600
TOO
Figure 5.
Sensitivities VCty and V'f'h) of the eye as functions of the wavelength \
or in the extreme case with a continuous spectrum, this law, which is called the theorem of additivity, is still valid. In this case, the excitation is proportional to jV(X)-Le ,X-dX. Additivity is the prerequisite for a physiological evaluation of
the radiation by the human eye. If two different monochromatic or mixed
radiations give rise to the same impression of brightness under the same observation conditions, they are given the same
evaluation factor.
A physical radiant unit, evaluated
physiologically by the eye, is defined as a photometric unit. Photometric parameters have their own system of measurement units (see Chapter 2). The scotopic and photopic units have the same numerical values, if the black-body radiation
to be evaluated has the solidification point of platinum as its base temperature. The scotopic units are identified, for distinction from the photopic units, by
the stroke on the corresponding symbols.
61
Every photometric unit is basically derived
from the equivalent physical radiant
unit. The value of a particular photometric
X unit v \ is calculated for a mono-
chromatic radiation at the wavelength X^,
by multiplying the equivalent spectral
X radiation unit e Xi either by the
evaluation factor of the photopic sensitivity
V(X) i
of
the
eye
and
the
conversion
constant C or by the evaluation factor of
the scotopic sensitivity V'(X) of the eye
and its conversion factor C. In general, the
following relationship applies for
monochromatic radiation:
XV)X= Xe,X-V(X).C
and
X'v,X=X e>X-V , (X).C
(5.3) (5.4)
If the value of a photometric quantity Xv
is to be determined for radiation made up
of various wavelengths, the total radiation
must be evaluated wavelength by wavelength with the value of the corresponding sensitivities V(X or V'(X) of the eye. The spectral values of the physical radiant
X unit e , multiplied by the corresponding
sensitivity V(X) or V'(X) are integrated over the relevant wavelength range. The product of the integral and the corresponding
C conversion factors C or give the value
of the required equivalent photometric
parameter.
780 nm
Xv C/Xe,X.V(X).dX 380 nm
(5.5)
730 nm
XV C'/Xe,X.V'(X) dX
330 nm
(5.6)
The photometric evaluation of radiation is carried out, in accordance with previous definitions, with the radiance and luminance in the field of vision. These values are inserted in the equations (5.3) to (5.6). For
the luminance, the following relationship applies in the case of monochromatic radiation:
W,X=Le,X-V(X).C
(5.7)
L'v^=Lc^.V'(X)C
(5.8)
For wide band radiation, the following
applies:
780 nm
Lv = CjLe,X.V(X).dX
380 nm
(5.9)
730 nm
L,v = C'/Le,X.V'(X).dX 330 nm
(5.10)
5.3.1
Determination of the conversion constants
CandC
Since the sensitivity V(X) or V'(X) of the eye, as a relative value, has no units, the
C conversion constants or C* each contain
a constant numerical factor and corresponding units, according to the different definitions of the physical radiation and photometric units. The unit of the conversion constant
C is obtained by inversion of the equation
(5.3):
[XV ,X1
[C]
[Xe ,M . [V(X)]
(5.11)
The luminance LV>X and the spectral radiance Le X are inserted in equation (5.11):
[Lv,Xl [C]
[Le,Xl . [V(X)J
(5.12)
After insertion of the units in equation
(5.12), the unit of the conversion constant C
is obtained.
m m lm . sr ' .
2 (jum * ) i
W m W [C] =;
-~2 i
. sr
.
(pun ')
(5.13)
C The unit of the conversion constant
is obtained by inverting the equation (5.4):
62
[C]
[X'y,\]
[Xe ,Xl · [V<*)1
(5.14)
[C]
[L'v.X]
V X [Le ,Xl - [ '< )1
(5.15)
m - 1 lm. sr' .
2 (Mm ') lm
W m W lC ~ '
sT 1
.
.
-2
-1 )
.(Mm ~
(5.16)
The numerical value of the conversion
constant C must be calculated in accordance
with the definition of the light unit in
DIN 5031 , by inversion of the formula (5.9) for wide band radiation:
Lv
C= 780 nm
/Le,s,X-V(X).dX
380 nm
(5.17)
The candela, abbreviated cd, applies as the light unit for photopic vision. The
luminous intensity 1 cd is defined as
m 1/60 cm2 or 1/600 000 2 of the surface
of a black-body radiator at the temperature
of platinum solidifying under a pressure
of 101325 N/m2 (T = 2045 K) in the
normal direction. According to equation
(2.27), with a radiator surface As 1 cm the luminance Lv is:
lv
Lv A§. cosifi
60 cd
cd
j-- =60
cm 1
.1
cm2
lm
= 60
cm sr
(5.18)
The integral in the denominator of the equation (5.17) is solved, by multiplying
the absolute value of the total radiation
described above by the relative photopically evaluated proportional radiation q (see equation 2.65 and 4.47).
780 nm v Le,s,2045 K- <1 = /Le,s,X · (*> · dX
380 nm
(5.19)
The absolute value of Le>s>2045 K was
calculated in Section 4.2.2. From equation
(4.16) it is:
Wcm Le,s,2045 K = 31 " 55
~2 sr_1
The proportional radiation q to be
evaluated photopically is determined graphically with the aid of Figures 5.2 and
5.3. Here, firstly the radiation function
S\,2045 K of a black body at the temperature 2045 K and secondly the
photopic sensitivity V(X) are plotted as functions of the wavelength X. The
calculation of the radiation function SX is stated in Section 4.2.7. The product of the ordinates V(X) and SX is given by the curve V(X)-SX- The ratio of the area Ai measured by planimetry, under the curve
V(X)-SX to be measured area A 2 under the
curve S\ gives the proportional radiation q.
~M
q 2
(5.20)
For the better measurement of the area
\l under the curve V(X)-SX, this section is shown enlarged in Figure 5.3.
The values inserted in equation (5.20) give
the proportional radiation
18-2 mm'
mm q =
6440
- = 2-82 . 10"
(5.21)
If the values from equation (4.16) and (5.21) are inserted in equation (5.19), then
the photopically evaluated radiance is obtained:
780 nm
J'Le>s ,X-V(X).d\
380 nm
Wcm 31-55
.-2 -11
sr
.
2-82.
10"
89.10
3 Wcm
2
sr
'
(5.22)
The required numerical value of the conver-
sion constant C is obtained by insertion
63
3
1
VWSx
0-8
1 v(X)
^V^-.-Sx^CMSK
0-6
A2
0-4
0-2
45K
fk
1
2
Figure 5.2
Photopic evaluation of the black-body radiation ofT= 2045 K
! \/ \
Figure 5.
Photopic evaluation of the black-body
radiation of T = 2045 K. Enlarged
section from Figure 5.2, for planimetric
measurement of the area A j
of the values from the equations (5.18) and (5.22) in the equation (5.17):
60 lm cm sr 89. 10" 3 Wcm' 2 sr_1 <674 1mW
(5.23)
By the same calculation method, the ratio of the luminance (see equation 5.18) to the radiance of the black body radiator at the
solidification temperature of platinum,
T = 2045 K, evaluated in accordance with
the scotopic sensitivity V'(X), gives the conversion factor
C = 1725 1mWT !
(5.24)
5.3.2 Photometric radiation equivalent
To simplify the conversion of physical
radiation units into photometric units, the product of the corresponding conversion
64
C constants C or with the relevant
sensitivity V(X) or V'(X) is formed. The
product C-V(X) is defined as the photometric radiation equivalent K(X), and C'V'(X) as the photometric radiation equivalent K'(X).
The simplified equation for the calculation of the photometric radiation
equivalent K'(X) reads, correspondingly:
fr'v,X K'(X) =
*e,X
(5.30)
K(X) = C . V(X)
C K'(X) = . V'(X)
(5.25) (5.26)
The photometric radiation equivalent K(X)
or K'(X) represents the relevant conversion factor between radiant and photometric
units with monochromatic radiation. From
the equations (5.3) arid (5.25), the general simplified formula for the calculation of the photometric radiation equivalent K(X) for monochromatic radiation of wavelength
Xis obtained:
K(X) = XV ,X
*e,X
(5.27)
From equations (5.4) and (5.26), the
simplified formula is obtained for the calculation of the photometric radiation equivalent K'(X :
K'(X) = X'V ,X *e,X
(5.28)
If equivalent photometric and radiometric
values are inserted in equations (5.27)
and (5.28), then, as in equation (5.13), by
reciprocal cancellation of their solidangle and area units, for example, the ratio
of the luminous flux <JV X t0 the radiant flux 5
<I>e X is obtained. The photometric
radiation equivalent for monochromatic
radiation K(X) is therefore defined as
"the ratio of the luminous flux 3\, X to trie 5
radiant flux ^^foi the monochromatic
radiation of wavelength X". The simplified
equation reads:
K(X)=*--v'X-
$e,X
(5.29)
If the maximum value for V(X) at the maximum wavelength Xmax = 555 nm, of V(X) max = 1, is inserted in equation (5.25), the conversion constant C represents the maximum value of the photometric radiation
equivalent K(X) max .
K(X) max
(5.31)
If the conversion constant C is replaced
by the equation (5.17), the basic equation is obtained for the calculation of the
maximum value of the photometric radiation equivalent K(X) max :
K(X) max
Lv
780 nm
/Le,s,X-V(X).dX
380 nm
(5.32)
If the corresponding values from equation
(5.23) are inserted in the equation (5.32),
the maximum value of the photometric
radiation equivalent is defined as "the
ratio of the luminance, fixed at Lv =
60
lm
cm~
2
_1
sr
of the black body
at
the
solidification temperature of platinum
T = 2045 K, to the radiance of the black
body at the same temperature, evaluated in
accordance with the photopic sensitivity
V(X) of the eye".
K(X) max
60 lm . cm
2
.sr
1
89.10~
3
Wcnf
2
-1
sr
;6741mW
(5.33)
According to DIN 5031, 1970 Edition, this value has been laid down at 673 lm/W.
K(X) max = 673 lmW
(5.34)
The reciprocal of the maximum value of
65
the photometric radiation equivalent
K^max or K(^)555 nm is defined in
the literature as the "mechanical light
equivalent M".
M= 1
K(X) max
(5.35)
From equation (5.33), the value for
K^)max is inserted in equation (5.35).
The previous value of the mechanical
M light equivalent is:
-- -- M =
1W
-
= l-47mWlm_1
680 lm ============
(5.36)
The maximum value of the photometric
radiation equivalent K(X) max permits
the photometric radiation equivalent K(X) to be calculated at any given wavelength
X For this, the conversion factor C from
equation (5.25) is replaced by K(X) max
in accordance with equation (5.31):
K(X) = K(X) max .V(X)
(5.37)
Wavelength
Spectral brightness sensitivity of the eye
Photometric radiation equivalent
_X_ V(X);
nm
(Ly=
lOcd.m
2
)
K(X)
lm
W
380 400 410 420 430 440 450 460 470 480 490 500 510 520 530 540 550
000004
0-0004
00012
0-0040 0-0116 0-023 0-038 0-060
0091 0139
0-208 0-323 0-503 0-710 0-862 0-954 0-995
0-0272 0-272 0-816 2-72 7-89 15-64 25-84 40-8 61-88 94-52 141-44 219-64 342-04 482-8
58616
648-72 676-6
For K(X)max , the value 673 lmW-1 is
inserted:
K(X) = V(X) . 673 ImWr-l
(5.38)
In Table 5.1, values of the photometric radiation equivalent K(X) are listed
according to the wavelength X (prior to 1970: K(X) rnax = 680 ImW-1 ).
By the same calculation method, a value of the photometric radiation equivalent K'(X) can be determined at the wavelength X. The maximum value of the photometric
radiation equivalent K'(X) max is:
K'(X) max 1725 ImW-l
(5.39)
In Figure 5.4, the photometric radiation equivalents K(X) and K'(X) are shown as functions of the wavelength X.
555
1
560 570 580 590 600 610 620 630 640 650 660 670 680 690 700 710 720 730 740 750 760
0-995
0-952
0-870
0-757
0-631
0-503
0-381
0-265
0175 0107 0061 0032 0017 00082 00041 00021 000105 000052 000025
0-00012
000006
680
676-6 647-36 591-6 514-76
42908 34204 25908
180-2 119
72-76 41-48 21-76 11-56
5-576 2-788 1-428 0-714 0-3536 0-170 0-0816 0-0408
The following calculations now only relate
to the photometric radiation equivalent K(X) of the photopic sensitivity V(X) of the eye.
Table 5.1
The photopic sensitivity of the eye, Vfty, and the photometric radiation equivalent K()j as functions of the wavelength X
66
-- 1
-- ----
1000
100
10 s
1
01
IT
Scot
-->
J /
i
t
/
t
1
/
---f
1
1
1 1
-f--
1
1
--1
J
1
4
1
1 1 1
·
1
1 1
-f-
/ / /
--
f<"-
1
I
j-
1
/
/
/
t
.
\ \
\
y
\ \
--\
KM
Pholavfc
*
\P
\
I
\
\
\\
c i
*V
-- \
\
\
»L
k--
\
\
-- \ \ -1 \
L
1
\
V-
_\_
\
\
\
)
\
\
Figure 5.4
Photometric radiantion equivalent K(~\) and K'fa as functions of the wavelenth A
5.3.3 Photometric radiation equivalent of the photopic sensitivity of the eye
With the aid of the equation (5.5), for a photopically evaluated mixed radiation, the
K radiation equivalent of the total
radiation will be formed, by putting the
photopically evaluated radiation in relation to the total radiation. For this, the
conversion constant C is replaced, in
accordance with equation (5.31), by the
maximum value of the photometric radiation equivalent K(X) max :
K(X) r
K:
780 nm
.jXe;X.V(X). dX 380 nm
c/ Xe,X dX
(5.40)
If the same radiometric values, e.g., the
X radiant flux, are inserted for e X in
equation (5.40), then after calculation, the
K radiation equivalent is obtained as the
ratio of the photopically evaluated radiant
flux to the total radiant flux:
67
780 nm
K(X) max ./<J>e ,A.V(X).dX
380 nm K=
J*e,X-dX
(5.41)
The ratio of radiation evaluated by the
eye to the total radiation is the visual
efficiency V
780 nm
/$e,X · V(X) . dX
380 nm V=-
J'$e,X- dX
(5.42)
Equation (5.42), inserted in (5.41), gives the following relationship for K:
K = K(X) max .V
(5.43)
The photopically evaluated radiant power
in the numerator of equation (5.41) corres-
ponds, as shown in equation (5.5), to the
photometric luminous power 4>v. The denominator corresponds to the total
radiant
flux
e <£> .
The
photometric
radiation equivalent K of a mixed radiation
is therefore defined as the quotient of the
luminous power <J>y and the total radiant
flux <£».
For any given radiometric parameters, in
general:
Xv = Xe . V . K(X) max
(5.47)
As shown by equations (2.34) and (2.35),
the ratio 4>v to <J>e also corresponds to the ratio of the luminous efficiency T?v of a radiation source to the radiant efficiency
Tfe of the same radiation source
T?v
K
T?e
(5.48)
In addition to this, Figure 5.5 shows the
variation of the radiation equivalent K and
the luminous efficiency Tfy of the black body
as a function of the absolute temperature.
K 2l
(5.44)
Corresponding to the equations (5.27) and (5.40), the general equation must
read:
K
X,
(5.45)
When the visual efficiency V is known,
the photometrically evaluated radiant flux is obtained from the equations (5.43) and (5.44):
% V = «J>e
. K(X) max
(5.46)
Figure 5.5
Radiation equivalent K and luminous
efficiency 7]v of the black body as a
function of the absolute temperature T
68
5.4
Calculation of the photometric radiation
K equivalent for different radiations
5.4.1
Calculation of the photometric radiation
K equivalent for a Planck radiator at the
temperature specified on the definition of the candela
K L e ,2856 K
(5.49)
From equation (4.18), the value of the radiance Le> 2856 K of a black body at
the temperature T = 2856 K or of a
Planck radiator with distribution
temperature Tv = 2856 K is:
Le ,2856 K^OWsr^cnT 2
The photometric radiation equivalent K
for the radiation of a black body at the temperature T = 2045 K, which applies for the definition of luminous intensity, will be calculated in accordance with equation
(5.45). In this equation, the values of the luminance and radiance for the radiation source described can be inserted from the equations (4.16) and (5.18).
We obtain:
Lv
K=
Le
60 lm . cm
-1 sr
K=
-2
31-55 W. cm
1
sr
(5.49)
1-91 lmW
(5.50)
The result can be verified by means of
Figure 5.5 with K = f(T).
5.4.2
Calculation of the photometric radiation
equivalent K for the standard light A
The standard light A has already been
mentioned in Section 2.34. It is used for the measurement of the photocurrent sensitivity of germanium and silicon radiation detectors. If the user has a tungsten lamp calibrated to the distribution
temperature of Tv = 2856 K, the
illuminance, measured with relatively wide tolerances by a Luxmeter, can be converted by means of the radiation equivalent into the irradiance. The photometric radiation
equivalent of the standard light A can be
calculated from equation (5.49):
The luminance Lv is calculated by inversion
of equation (5.32); this gives:
780 nm
Lv
=
K(X) max
. J'Le>s ,X
V(X)
·
.
dX
380 nm
(5.51)
The integral is solved in accordance with
equation (5.19):
780 nm
M Le ,2856 K q = /Le,s,A V
d^
·
380 nm
(5 52)
The proportional radiation q, to be
evaluated photopically, will be determined graphically with the aid of Figures 5. 6 and
5. 7. The product of the ordinates of the photopic sensitivity V(A) of the eye and the
radiation function SX,2856 K 8 ives the
curve V(X) · Sx,2856 K. The ratio of the measured area Ai under the curve V(X) ·
SX,2856 K to the measured area A 2 under the curve S\,2856 K 8 ives tne proportional
radiation q, as was stated in equation (5.20):
For a better planimetric measurement, the
area A} under the curve V(X). S\,2856 K
is shown enlarged in Figure 5.7.
376 mm
= 0-025
15036 mm' ====
(5.53)
If the values from equations (4.18) and
69
V(X)Sx 80
/ 1
1,
/
/
*2
1
1 1-- V(X) SX.2856K
1/ <\\ 1
--Sx,2«S«K
XOim)
Figure 5.
K Photopic evaluation of the black-body radiation at T= 2856
(5.53) are inserted in equation (5.52), the photopically evaluated radiance is obtained:
= 3 Won
2 sr
'
(5.54)
From equation (5.51), the luminance is:
Lv = 6801mW
l
. 3Wcm
2 sr
l
20401m cm
2 -1
sr
(5.55)
The required photopic radiation
equivalent K of the standard light A can
be calculated by inserted of the values equations (4.18) and (5.55) in the formula
(5.49).
2040 1m cm
2 sr
'
120Wcm_2
_1
sr
= 17 lmW"
(5.56)
The result can be verified with Figure 5,5,
which shows the curve K = f(T).
5.4.3 Calculation of the photometric radiation
equivalent K of the luminescence radiation
from light-emitting diodes (L.E.D.S)
The photometric measuring instruments
which have been common up to the present
are not very suitable for the measurement of the selective radiation from light-emitting diodes. The spectral sensitivity of their photodetectors is corrected with a special filter to the overall photopic sensitivity of the eye. This correction shows relatively large tolerances in a few wavelength ranges. The selective radiation from lightemitting diodes can therefore only be measured with a photodetector which is
exactly photopically corrected to their
relatively small wavelength range. The
light-emitting diodes of various colours must be measured in each case with the appropriately corrected photodetector or sensor.
70
4
l00 T
Vft) ^s.
*%
Mmnl"
Figure 5.
Photopic evaluation of the black-body radiation atT= 2856 K. Enlarged section from Figure 5. 6 for planimetric measurement of the area A \ (on left of diagram)
Photopic evaluation of the radiation from a red-emitting GaAsP diode, e.g., for the
diodes TIL 209A and TIL 220 (on right of diagram)
At the start of the development of lightemitting diodes, there were still no suitable photometric measuring instruments available. Therefore, the measured radiant power from a light-emitting diode was previously stated in data sheets.
With the following calculation of the photometric radiation equivalent for luminescence diodes with selective light emission, the user can determine the photometric luminous power with an uncorrected radiant power meter, provided
that the radiation function SX.LED of an
LED luminescence radiation is determined
exactly with a monochromator. If, on the other hand, the relative spectral radiation
distribution, stated in the data sheets, is guaranteed within close tolerances by the semiconductor manufacturer, the luminescent
radiant power <J>e can be converted directly
K from its radiation equivalent into the
photometric luminous power q\. With LED's giving a green light output, the spectral distribution can be assessed superficially by subjective colour comparison. For LED's with a red output, however, the subjective colour
71
comparison can give totally false assessments of the photometric luminous power. As a result of poor manufacture, LED's intended to have a red light output can also radiate
in the near infra-red range. Such defects can easily be detected with an infra-red converter. The radiant power determined with an unweighted radiant power meter may only be used, if the infra-red part of this radiation is also taken into account. Texas Instruments guarantee the minimum,
typical or maximum wavelength stated in their data sheets with the maximum
emission of the corresponding light-emitting diodes or display units. In addition, the typical relative spectral radiation distribution, the radiation function S\: is stated as a function of the wavelength in graphs.
As an example, the typical radiation
function S^leD of the red-emitting
diodes TIL 209 and TIL 220 is plotted in Figure 5. 7, with the photopic sensitivity of the eye V(A), as a function of the wavelength A. The product of the
ordinates V(X) and S^LED gives the
< curve V(X) · S^LED- The ratio of the area
A3 under the curve V(X) · S^led to the area A4 under the curve S^LED Sives the
effective visible proportional radiation q (see equation 5.20).
A3 A4
160 mm 2
mm q = 1350
= 0-1185
2
(5.57) (5.58)
The photopically evaluated radiant power is calculated similarly to equation (5.19):
780 nm
*e,LED · q = J'$e,X
V(X) dX .
380 nm
(5.59)
For example, the minimum guaranteed radiant power <£e ,min,LED is taken from
the data sheet of the luminescent diode TIL 220 under "Radiant Power Output"
po,min- It is 25 fJW. The value from
equation (5.58) and the data-sheet value are inserted in equation (5.59). The photopically evaluated radiant power is
then:
780 nm
W J*e,X- V(X)-dX= 25. 10~ 6 . 0-1185
380 nm
= 2-96 /M
(5.60)
The required typical radiation equivalent Ktyp of the luminescence radiation from the red-emitting GaAsP diodes TIL 209 and TIL 220 can be calculated with equation (5.41) or the inverted equation (5.46) as
follows:
Ktyp
780 nm K(^)max /#e,X-V(X).dX
380 nm
00
i'*e,X-dX
(5.41)
-- 680 lmW
Ktyp =
l
.
2,96 /iW
:
=
80-5
lmW-1
25 /iW
(5.61)
A simplified calculation of the photometric K radiation equivalent typ for the selective
luminescence radiation from light-emitting diodes is possible, by reading the K(A) value from Table 5. 1 for the typical
wavelength with the maximum emission of
the corresponding diode. The typical
wavelength Xp with maximum emission of
the light-emitting diodes TIL 209 and TIL 220 is:
Xp = 650 nm
(5.62)
For this wavelength, the value of the photometric radiation equivalent is read from Table 5.1
K(X)'typ = 72-76 lmW" 1
(5.63)
This value does not agree exactly with that from equation 5.61, since on the one hand
72
1
a graphical integration is always subject to a certain inaccuracy and also the function V(A) is curved.
A further very simple calculation with
reasonable accuracy is as follows:
The arithmetic mean is formed from the
radiation equivalents of the shorter-
wavelength half-power point Hi and the longer-wavelength half-power point H2
(see Section 10.4):
K(X) H l + K(X) H 2 K:
typ
(5.64)
The red luminescence radiation from the GaAsP diodes TIL 209 and TIL 220 has
its typical wavelength Xp for maximum
emission at Xp = 650 nm, as is already shown in equation (5.62). Its typical bandwidth between the half-power points Hj and H2 is 20 nm. The wavelengths of the
typical half-power points therefore lie
around
Xhi = 640 nm
(5.65)
and
Xf{2 = 660 nm
(5.66)
The corresponding values of the photo-
metric radiation equivalent for these
wavelengths are read off from Table 5. and are inserted in equation (5.64):
K(X) H 1
=
119
-1
lmW
(5.67)
K(X)H2 = 41 - 481mW
(5.68)
1191mW 1 + 41-48 lmW 1 Ktyp =
= 80-24 lmW"
(5.69)
This result is near to the graphical solution using equation (5.61).
A somewhat more accurate method is to
calculate with "Kepler's rule". For this,
the radiation equivalents are needed at
V ^Hl and ^H2 :
K(X)Hl+4.K(X)p + K(X) H2 Ktyp-
(5.70)
Applying this to the light-emitting diodes TIL 209 and TIL 220, with the aid of Table 5.1:, one obtains:
Ktyp
119
lmW -1
+
4.
72-76
_1 lm\V
+
41-48
lmW" 1
= 75-25 lmW-1
(5.71)
This value is more accurate than that
previously determined.
The green luminescence radiation from the
GaP diodes TIL 21 1 and TIL 222 has its
typical
wavelength
X
p
for
maximum
emission at
Xp = 565 nm
(5.72)
Their typical bandwidth between the half-
power points H} and H2 is 35 nm. The
wavelengths of the typical half-power points
thus lie at
Xt-ji = 547-5 nm
(5.73)
and
Xt-[2 = 582-5 nm
(5.74)
The values of the photopic sensitivity V(X) of the eye can be taken from DIN 5031, Part
3. We obtain:
V(X) 565 nm = 0-9786
(5.75)
V(X) 547 nm = 0-9874
(5.76)
V(X) 582 nm = 0-8494
(5.77)
From equation (5.37), the radiation
equivalents K(X) for these wavelengths are:
73
K(X) = K(X) max .V(X) K(X) p = 680 lmW-1 .0-9786 = 665-6 lmW" 1
K(X hi = 680 lmW 1 . 0-9874
_1
= 6721mW
-1
K(X) H 2 = 680 lmW · 0-8494 = 577 lmW-1
(5.37) 5.78) (5.79) (5.80)
Because of the sharper curvature of the V(X) function in the green spectral range, Kepler's rule, as in equation (5.71), is advisable for the calculation:
Ktyp _ 672 lmW_1 + 4. 665-5 lmW_1 + 577 lmW-1
= 6521mW
(5.81)
With the examples shown, the user can himself determine the corresponding radiation equivalents for other selective luminescence radiations from light-emitting
diodes.
5.5 Conversion of radiometric units into photometric, photopic units
5.5.1
Conversion of radiometric units into photometric units for Planck radiation
The calculation of the total radiant power
of a Planck radiator for the distributor temperatures which are of interest has been described in Chapter 4. Planck radiation is used mainly for the measurement of the photocurrent sensitivity of
germanium and silicon photodetectors. The
measured or calculated photocurrent of a photodetector does not, however, depend only on the incident irradiance, but also
on the weighting of the radiation given by
the spectral sensitivity of the photodetectoir. Irradiance measurements of Planck radiation are carried out exactly with wideband radiant power meters. But very often only photometric and selective
radiant power meters are available. The evaluation of radiation by selective radiant power meters and by silicon photodetectors requires knowledge of the spectral sensitivity values, which have not yet been dealt with
(see Chapter 9).
The photometric evaluation of radiation is carried out with the radiation equivalent, which is described in detail in Sections 5.3 and 5.4. The photometric
K radiation equivalent permits the simple
conversion of radiometric units into
photometric units. The general conversion formula for total radiation is obtained by inversion of equation (5.45):
Au -- Jv . Ap
(5.82)
X If the irradiance E e is inserted for e in
equation (5.82), the illuminance Ev is
obtained for Xv.
Ey - K . Ee
(5.83)
The irradiance values used for the measurement of the photocurrent sensitivity of germanium and silicon photodetectors
lie between 1 mW/cm and 20 mW/cm2 .
Very sensitive phototransistors are
measured with the irradiance values 1 mW/ cm , 2 mW/cm and 5 mW/cm , while
less sensitive or older phototransistors and photodiodes are measured with the
irradiance values 9 mW/cm2 and 20 mW/ cm . From equation (5.50), the
photometric radiation equivalent at the temperature for the definition of the
specified light unit of Planck radiation is:
K = 1-91 lmW" 1
For this Planck radiation, the following illuminance values are obtained in accordance with equation (5.83):
74
forEe = 1 mWcm
x
:
mW Ey = 1-91 1
' . 1 mWcm 2 = 191 lx
(5.84)
for Ee = 2 mWcrrT : Ey = 1-91 lmW l . 2 mW,cm 2l _= 38-2 lx
(5.85)
for Ee = 5 mWcm Ey = 1-91 lmW l . 5 mWcm
= 95-5 lx
(5.86)
forEe = 9 mWcm
2
:
Ev
=
1-91
-1
lmW
·
9mWcm-2
=
171-9
lx
(5.87)
forEe = 20mWcm
2
:
Ey = 1-91 lmW" 1 . 20 mWcnT 2 = 382 lx
(5.88)
The photometric radiation equivalent of the
A standard light which is specified for the
measurement of the photocurrent sensitivity for germanium and silicon photodetectors, from equation (5.56), is:
K=17 1mW_I
For Planck radiation related to the standard light A, the following illuminance values are obtained in accordance with equation (5.83):
for Ee = 1 mWcirf 2 :
Ev = 17 lmW" 1 . 1 mWcm-2 = 170 lx
(5.89)
E»(lx)
100
200
500
1000
Figure 5.8
Conversion of the irradiance into illuminance; with the distribution temperatures of
Planck radiators as variables
75
for E P = 2 mWcm' -2.
Ev = 17 lmW ' . 2 mWcm 2 = 340 lx
(5.90)
for E e = 5 mWcm : Ev = 17 lmW" 1 . 5 mWcm" 2 = 850 lx
(5.91)
for Ee = 9 mWcm-~2.
Ev = 17 lmW-1 . 9 mWcrrf 2 = 1530 lx
(5.92)
for E e = 20 mWcm-2.
Ev = 17 lmW" 1 . 20 mWcrrf 2 = 3400 lx
(5.93)
Comparison of the examples (5.84) to (5.88) with the corresponding examples (5.89) to (5.93) will clearly show, to the practical worker, that a deviation from the specified distributor temperature for the standard
A light causes large measurement errors.
The conversion of the irradiance to the illuminance of Planck radiation can be seen from the graph in Figure 5.8. The distribution temperature is the variable. Corresponding to the above conversion, the irradiance can be determined more simply by measuring the illuminance with an accurately calibrated Luxmeter. In doing this, care is to be taken, that no reflected or ambient radiation falls on the photodetector of the Luxmeter.
5.5.2 Conversion of radiometric units into photometric units for the luminescence radiation of light-emitting diodes
The radiometric units of the luminescence radiation of a light-emitting diode will be converted in accordance with equation (5.82) into the equivalent photometric
units:
X K X : v
t yp . P
(5.94)
The calculation of typical radiation
equivalent Ktyp for luminescence radiation was described in Section 5.4.3. From
equation (5.71), the typical radiation
equivalent
K
ty p
for
the
luminescence
radiation of the red-emitting GaAsP diode
TIL 220 is:
Ktyp 72-25 lmW
The minimum guaranteed radiant power
mm <t»e
has a value of 25 /iW (see Section
5.4.3 or data sheet of the TIL 220). If this
radiant power and the radiation equivalent
from equation (5.71) are inserted in
equation (5.94), the minimum guaranteed
luminous power <tV is obtained for the
typical spectral radiation distribution of
the TIL 220:
<*V,min = $fe,min K typ
< 5 - 95 >
4v min = 25 juW . 72-25 lmW-1 = 1-88 min ;
(5.96)
This minimum luminous power still gives no. information on the photometric data
of a semiconductor radiation source. These depend mainly on the construction and also on the radiation distribution pattern of the element. In the early days of light-emitting diodes, the radiance or luminance perpendicular to the wafer surface, i.e., in the normal direction, was stated in the data sheets. For four different reasons, however, this practice has been abandoned and a number of well-known American firms agreed, for future light-emitting diodes and display units, to state the luminous intensity or radiant intensity in the normal
direction, with the units cd or mcd and Wsr~ ' or mWsr~ 1 . The reasons for this
measure were as follows:
1
The statements of luminance in the initial period had been erroneously related to the whole wafer area. As a result of the
76
geometrical shape of the element and the arrangement of bonds pads the whole area does not light up.
Similarly, in the case of light-emitting diodes with either clear or coloured lenses or plastic cases the corresponding, considerably larger effective luminous area had not been taken into account, but here too the data was related to the total wafer
area.
In addition, the different units of luminance confused the users (see Section 2.2.4)
The result was that some users adopted
the practice of assessing the light-emitting
diodes by subjective visual observation.
For the calculation and measurement of the luminous and radiant intensity of lightemitting and radiation-emitting diodes, reference is made to Chapter 10, Section
10.1.
GaP diode, for an equal impression of
brightness on the eye under equal
conditions of observation, are very
different, assuming that both light-emitting
diodes have the same luminous intensity
distribution. To compare the green-emitting
GaP diode with the red-emitting GaAsP
diode, the luminous power of the red-
emitting GaAsP diode from equation (5.96)
and the typical radiation equivalent for the
green luminescence radiation of the GaP
diode will be inserted in the rearranged
-3
equation (5.95). With 4\ = 1-88 . 10 lm
-1
and Ktyp = 652 lmW
we thus obtain:
,
3y
*e = Ktyp
1-88. 10 3 lm
<**6 652 lmW l
= 2-88 jUW
(5.99) (5.100)
Under the assumptions described above for equal brightness excitation of the eye, from equation (5.100), only 2-88 fJW is needed for the green-emitting GaP diode, while the red-emitting GaAsP diode has to deliver
25 AAV.
The light-emitting diode Type TIL 210 has,
to a first approximation, a radiant intensity distribution in accordance with Lambert's
law. By rearrangement of the equation (3.11),
the luminous intensity can be calculated in simplified form.
*v,nmuin v,min
7T.no
(5.97)
If the value from equaton (5.96) is inserted in equation (5.97), the value of the
minimum luminous intensity in the normal
direction for this diode is obtained:
v,mm
1-88 mlm
n. fio
0-6 mlm -= 0-6 mcd
(5.98)
The necessary radiant powers of a redemitting GaAsP diode and a green-emitting
5.6 Actinic value
5.6.1
The actinic value of radiation spectrum for photodetectors and for the human eye
The spectral sensitivity of the eye has been described in Sections 5.3 and 5.4. There,
it has also been shown, how various
radiation spectra can be evaluated with the aid of the curves V(X) and K(X). Just like the eye, every semiconductor photodetector has a sensitivity curve, which is characteristic of the material and its doping and with which similar calculations can be carried out. Further details on the spectral sensitivity of photodetectors are contained in Part II.
77
The absolute sensitivity of a photodetector is designated by s. The general definition of the sensitivity of the eye or of an electronic photodetector is described by the following equation.
X
(5.101)
The general value Y denotes the output value
of the photodetector, while the general
X value represents an input value to the
photodetector, evaluated by radiometric, photometric or other means.
In relation to the eye:
1
The relative sensitivity srei corresponds to the visual efficiency V, see equation (5.42):
780 nm /4>e>X · V(X) . dX 380 nm Srel - V = ~55
J*fe,X- dX
(5.102)
The absolute spectral sensitivity s(X) corresponds to the photometric radiation
equivalent K(X):
S(X)=K(X) = K(X) max .V(X)
(5.103)
The relative spectral sensitivity s(X) rei corresponds to the photopic sensitivity V(X) of the eye:
s(X)
s(X) rei =
s(X) max
: V(X)
(5.104)
The absolute sensitivity s corresponds to the photometric radiation equivalent K:
K = K(X) max .V
(5.105)
The values of these sensitivity quantities depend on the wavelength X of a mono-
chromatic radiation or on the total spectral distribution of the radiation and on the spectral sensitivity distribution of the eye. Similar, corresponding relationships apply for any given photodetector.
The sensitivity for a radiation spectrum Z will be denoted by s(Z) and that for a
radiation spectrum N by s(N). Here, the
radiation Z is any radiation to be investigated in more detail, while the
radiation N represents a selected reference
radiation. The corresponding input values are denoted by X(Z) or X(N) and the corresponding output values by Y(Z) or Y(N). The sensitivity with radiation Z or
N radiation is determined by
Y(Z) s(Z) =
X(Z)
(5.106)
Y(N)
s(N) =
X(N)
(5.107)
An unweighted radiometric input value Xe
is obtained by integration of the
corresponding spectral values Xe X over {h e
wavelength range which is of interest. The
radiometric input values X(Z)e and X(N)e are given by
X(Z)e = /X(Z)e>X-dX
(5.108)
X(N)e = J'X(N)e,X-dX
(5.109)
The input values X(Z) e and X(N)e can be
represented as relative units, related to a
maximum value, if, in accordance with
equations (5.108), (5.109) and (2.66), the integration is carried out over the wavelength range which is of interest with the radiation function S(Z)\ or S(N)\ of the particular
radiation Z or N:
X(Z)rel = J'S(Z)X . dX X(N)rel = /S(N)X.dX
(5.110) (5.111)
The output values Y(Z) and Y(N) are determined by integration of the
78
corresponding spectral input values, in this
case X(Z)e>X and X(N)C)X, and the spectral
sensitivity s(X) of the photodetector over the wavelength range which is of interest:
Y(Z) = fX(Z) t>\. s(X) . dX
(5.1 12)
Y(N) = /X(N) e ,\ . s(X) . dX
(5.1 1 3)
The output units Y(Z) and Y(N) can also
be stated in relative terms, if, in accordance with equations (5.1 1 2), (5 .1 1 3) and (2.66), the integration is carried out over the wavelength range which is of interest with the relevant radiation function S(Z)\ or S(N)X and the relative spectral sensitivity s(X)iej of the photodetector:
Y(Z) rei = J"S(Z)X · s(X) rel . dX Y(N)rel = JS(N)\ . s(X) rei - dX
(5.1 14) (5-1 15)
In optoelectronics, the actinic value a(Z)
of a radiation spectrum Z is the ratio of its effect on a photodetector, related to
the effect of a reference radiation spectrum N. Here, the effect of the radiation corresponds to the sensitivity s(Z) or s(N) of the photodetector.
s(Z)
a(Z) = -
s(N)
(5.116)
The equations (5.112 to 5.116) presuppose, that the output parameters Y(Z) and Y(N) are proportional to the corresponding input parameters X(Z) and X(N). If there is no proportional relationship between the input and the corresponding output parameter then in many cases the actinic value can be calculated by equating the output
parameters:
Y(Z) = Y(N)
(5.117)
For this case, the extended form of the equation (5.116) is obtained by insertion of equations (5.106) and (5.107) in equation (5.116):
s(Z) X(N)
a(Z) =
=
s(N) X(Z)
(5.118)
The general extended form of equation
(5.116) reads:
s(Z) Y(Z).X(N)
a(Z) =
=
s(N) X(Z) .Y(N)
(5.119)
The actinic value, which is calculated for two radiation spectra with radiometric input
parameters for a given photodetector, is given the definition ae(Z).
To calculate the actinic value of a
radiation spectrum for a given linear photodetector the absolute values can be inserted in equation (5.1 19) in accordance with equations (5.108), (5.109), (5.1 12) and (5.113), or the relative vaiues in accordance with equations (5.1 10), (5.1 1 1), (5.1 14) and (5.115):
ae(Z)
JX(Z)e,X. s(X) .dX./X(N)e ,X.dX /X(Z) e>X. dX . jX(N) e>X · s(X) . dX
(5.120)
a e(Z)
JS(Z)X s(X) rei . dX . |S(N)X · dX ~ /S(Z)X . dX . J"S(N)X · s(X) rei · dX
(5.121)
The actinic value, which is calculated for two monochromatic radiation spectra, from the radiometric input values for a given photodetector, is designated ae(Z)X- For monochromatic radiations, the equation (5.120) can be simplified, so that the following formula is obtained from equation
(5.119) with spectral sensitivity values:
s(Z)X
ae(Z)X : s(N)X
(5.122)
The equations (5.120) and (5.121) can be
simplified correspondingly to equations
(5.103) and (5.104). The actinic value of a
79
radiation spectrum of the eye is therefore calculated by the ratio of the radiation equivalents for the corresponding
radiations sources.
For two radiation spectra Z and N of any
given spectral composition, the actinic
value ae(Z) of the radiation Z for the eye can be determined with the following
formula:
K(Z)
ae(Z)
K(N)
(5.123)
The actinic value ae(Z)\ of a monochromatic radiation Z for the eye can also
be calculated with the radiation equivalent K(X) for the corresponding monochromatic
radiations:
K(Z)X
ae(Z)X
K(N)X
(5.124)
5.6.2
Actinic value of Plank radiation and luminescence radiation with the eye as a photodetector
The actinic value of a radiation spectrum for the eye can be calculated from the
radiations spectra defined in Sections 5.3 to 5.5 with the equations (5.123) and
(5.124). To distinguish the various
individual radiation spectra,
1
Planck radiation with the temperature
Tv = 2045 K, which defines the unit of
luminous intensity, the candela, is designated by "PL";
Planck radiation with the temperature
Tv = 2856 K, which defines the standard
light A, is designated by "PA";
The actinic value, which is calculated from photometric input values for an electronic
photodetector, is given the definition
av(Z)\ for two monochromatic radiations and the definition av(Z) for two chromatic
radiation spectra.
If the input values X(Z) and X(N) are related to the absolute photopic sensitivity K(X) of the eye, then the modified form of equation (5.120) reads:
ay(Z)
/X(Z) e ,X · s(X) . dX . /X(N) e ,\ . K(X) . dX
~
J'X(Z) e
\
.K(X) .
dX
.
J'X(N) e ,X
·
«(*)
·
<*X
(5.125)
If the input values X(Z) and X(N) are related to the relative spectral sensitivity V(X) of the eye, then the modified form of equation
(5.121) reads:
ay(Z)
JS(Z)X
·
s(X) rei
.
dX .
/S(N)X
·
V(X)
.
dX
" /S(Z)X · V(X) . dX . /S(N)X . s(X) rel . dX
(5.126)
The luminescence radiation of the redemitting GaAsP diode TIL 220 is designated
by "LR" and
The luminescence radiation of the greenemitting GaP diode TIL 21 1 is designated by "LG".
The radiation equivalents of these radiations
are:
K(PL) = 1-91 lmW K(PA) = 17 lmW-l
(5.50) (5.56)
K(LR) = 72-25 lmW K(LG) = 6521mW_1
(5.71) (5.81)
If the radiation PL serves as the reference
radiation, then, from equation (5.123),
the actinic value of the designated radiation
on the eye is:
_1
K(PA) 17 1mW
ae(PA) = ^^77=
K(PL)
1; -n9i1
,lm,,W,_ i
= 8-9 (5.127)
80
:
K(LR) 75-25 lmW
ae (LR) K(PL)
1 91 lmW
= 39-398 (5.128)
K(LG) 652 lmW l
^ ae(LG)
= 7^77
K(PL)
=
,
1-91
,lm,,W,_!
=
341-36
(5.129)
If the radiation PA serves as the reference
radiation, then the actinic value of the designated radiation on the eye is:
-- -- ae(PL)
=
-K(LPL)
K(PA)
=
1-91 lmW-1 17 lmW"
= 0-1124
(5.130)
K(LR) 75.25 lmW
ae(LR) =
K(PA)
=
17
:
1mW"
= 4-4265 (5.131)
K(LG) ae(LG) =
K(PA)
-- 652 lmW = 38-353
17 1mW
(5.132)
If the radiation LR serves as the reference
radiation, then the actinic value of the designated radiation on the eye is:
ae (PL)
K(PL) K(LR)
^ 1-91 lmW,-i = 0-0254
75-25 lmW
(5.133)
ae(PA)
K(PA) K(LR)
17 1mW-l
--j^ = 0-226
72-25 lmW
(5-134)
K(LG) 6 52 lmW
ae(LG)
=
K(LR)
=
75-25
-r
lmW -l
=
8-6644
(5.135)
LG If the radiation
serves as the reference
radiation, then the actinic value of the designated radiation on the eye is:
ae(PL)
K(PL) K(LG)
1-91 lmW-l
= 0-00293
652 lmW"
(5.136)
K(PA) 171mW '
ae (PA)
=
K(LG)
=
652
lmW
r = 0-02607
(5.137)
-- K(LR) 75-25 lmW l
ae6(LR) = K(LG) = 652 lmW 1 = 0-1 154
(5.138)
If the actinic value is described for the
example (5.135), then the luminescence
radiation of a green-emitting GaP diode is actinically more effective on the eye by
the factor ae(LG) = 8-6644, as compared with the luminescence radiation of a redemitting GaAsP diode TIL 220, which is
selected as the reference radiation.
If the actinic value is described for the example (5.138), then the luminescence
radiation of a red-emitting GaAsP diode TIL 220 is actinically less effective on the eye by the factor ae (LR) = 0-1 154, as compred with the luminescence radiation of a green-emitting GaP diode, which
is selected as the reference radiation.
For the examples (5.135) and (5.138), the following relationshios apply
ae(LG)
1
a e (LR)
1
= 8-6644 0-1154
(5.139)
ae(LR)
1
ae(LG)
1
= 0-1154 8-6644
(5.140)
81
6 Interaction between
optical radiation
and matter
6.1
6.2
6.3 6.4 6.4.1 6.5
6.6 6.7
Absorption, transmission and reflection
factors
Spectral transmission factor T(X) and th spectral absorption factor a(X) Scatter Reflection of radiation Spectral reflectivity p(K) Basic laws of absorption, attenuation and scatter Absorption and transmission spectra Refraction
83
6.1
Absorption, transmission and reflection
factors
The conversion of radiation into either
a different spectral distribution or into
another form of energy, e.g., heat, is called absorption. The transparency of a medium to a radiation is the transmission. The
throwing back of radiation in all possible
directions, that is, also to the radiations
source, is called reflection. The absorption, transmission and reflection of a medium
are relative to the incident radiant flux.
The absorption factor a is the ratio of
the
absorbed
radiant
flux
d> a
to
the
incident radiant flux 4> . The transmission
factor r is the ratio of the radiant flux
passed
through
d> tr
to
the
incident
radiant
flux $-.
For this, the following relationships result:
tr
T =-
P=
(6.1) (6.2) (6.3)
The following relationship exists between
the absorption factor, transmission factor and reflectivity:
tt+ t+ p= 1
(6.4)
These values -depend on the wavelength of
the incident radiation.
Interaction between optical radiation and matter
6.2 Spectral transmission factor T(X) and spectral absorption factor 0(A)
In either the transmission factor T or the
absorption factor a is considered for a
monochromatic radiation, one speaks of
the spectral transmission factor t{\) or the spectral absorption factor a(X):
fr^tr T(X) =
*X,o
(6.5)
$X,a Q(X)
$\,o
(6.6)
The spectral transmission and absorption
factors are functions of the wavelength (or
the frequency). The adjective "spectral" has a different meaning here, as compared
with a spectral radiometric parameter, since the latter is described, precisely, as the spectral density of a particular
radiometric parameter. The spectral transmission and absorption factors have the
bracketed suffix (X) attached to their
symbols. The following relationships exist
between the values T, a and the spectral
values T(X), a(X
^tr_ /d^ .T(X .dX
$o
J'*X,o · dX
(6.7)
% a = <Sfe_ /<3X, .tt(X).d\ J<*%o dX
(6.8)
The limits of the integration need to be stated in each case.
A vacuum is completely transparent to
radiation. Any solid, liquid or gaseous matter, on the other hand, attenuates
radiations. Substances which are transparent to light can also show similar transparency
85
in the near IR range. Metals are only transparent to incident radiations up to a thickness of the order of magnitude of the wavelength. The transmission spectrum can be measured with a spectro-photometer. With a calibrated radiation source, firstly the direct radiant flux and secondly the radiant flux through the medium to be
investigated are ascertained at every wavelength. With different film thicknesses
of the medium, and also with inhomogeneous
substances, the spectral transmission or
absorption distribution varies. Zones with high absorption (low transmission) are called spectral bands. With spectro-photometers of high resolution, they can be
broken down into spectral lines.
The particles are very much larger than the wavelength. The scatter is not selective.
IR radiation is therefore transmitted better in certain media (e.g., mist, clouds) than light would be, since its wavelength is longer. The IR transmission of thin pulverised layers depends on both the particle size, and on the wavelength. Lamp-black completely absorbs visible
sunlight, while medium to long-wavelength IR radiation is partially transmitted.
6.4
Reflection of radiation
6.3 Scatter
The deflection of rays from the original direction of propagation in the same medium is called scatter. It occurs through reflection of an incident beam of radiation, while the
irradiated bulk elements radiate, as secondary radiation sources, in all possible directions. Part of the stray radiation is also thrown back to the radiation source. Pure scatter, that is, without absorption, can exist in transparent gases. Scatter generally increases the absorption, since
the path through the medium is lengthened. On the other hand, scatter always reduces
the transmitted radiant flux.
In the reflection of radiation, a distinction
is made between:
1
Directional reflection,
Diffuse reflection and
Mixed directional and diffuse reflection.
In Figure 6.1,* directional reflection, from a plane surface which obeys the optical laws, is illustrated. Figure 6.2 shows an ideally diffuse reflection. Diffuse reflection occurs at rough reflecting surfaces. Finally, in Figure 6.3, a mixture of directional and diffuse reflection is shown.
Depending on the size of the scattering particles, a distinction is made between three kinds of scatter:
1
The particle size is very small in proportion to the wavelength (molecular scatter). The scatter is very selec ive.
Tne particle size is hardly any smaller than the wavelength. The scatter is still
selective.
Diffuse reflections are difficult to measure. If the measurement is to be carried out
in one direction, the procedure of the IBK measurement conditions will be followed: the angle of incidence is 45 . The reflected beam is measured in the normal direction.
The law of reflection described the path of the rays at a plane mirror with "Angle of incidence 0% = angle of reflection 0^." In
Figure 6.4, the plane mirror is rotated
through the angle d> the reflected beam is deflected by twice this angle. In practice, a
86
1
Figure 6. Directional (mirror) reflection
Figure 6.3
Mixed reflection
Figure 6.2 Ideal diffuse reflection
beam, or bundle of rays, is always involved, necessitating careful adjustment of the plane mirror. The plane mirror must have constant reflectivity over the cross-section of the beam and a completely flat mirror
surface.
The flatness of optical mirrors is stated in fractions of the wavelength of the yellow sodium or the green mercury line. Metal mirrors are used for uniform reflections over a large wavelength range. The losses are dependent on the wavelength. With new
Figure 6.4
The law of reflection
mirrors they amount to a few percent and with older mirrors up to 50%. Hence for monochromatic radiation sources, interference reflectors (p = 99%) are generally
used.
6.4.1 Spectral reflectivity p(X)
If the reflectivity p s considered for a monochromatic radiation, it is designated
as the spectral reflectivity p(X).
87
P(X)
(6.9)
The spectral reflectivity is a function of the wavelength (or the frequency). The meaning
of the adjective "spectral" is explained in
Section 6.2. The following relationship exists between the reflectivity p and the
spectral reflectivity p(X):
_ f$\o · P(X> . dX P
J<*V · dX
(6.10)
The Unfits of the integration are to be stated in each case.
Similarly to equation (6.4), the following relationship exists between the spectral values a(X), r(X), and p(X):
a(X) + T(X) + p(X) = 1
(6.11)
Selective reflections are represented in spectral reflection curves. With metals, the reflectivity increases with the wavelength.
For IR radiation, both aluminium and aluminium paint have a very high
reflectivity (see Figure 6.5).
'
1
1
08
0-6
04
0-2
0-5
1
3
5
10
3
Figure 6.5
Selective reflection from aluminium
Non-metals show reflection bands . They
are characteristic of each substance. A
spectral band with high reflectivity is
described as a metallic reflection. Quartz, for example, shows metallic reflections in several bands {Figure 6.6). Transparent
Figure 6.6
Selective reflection of quartz
substances show glass-like reflections. When
the incident radiation is perpendicular or at an acute angle to the normal, the reflectivity is low, but with a larger angle to the normal (grazing the surface) the reflectivity obtained is very high.
In the IR range (X> l^lm), a few
substances with very low reflectivities can be classified as black, e.g., asbestos, cotton, rubber, wood, silk.
The reflection capability of thin granulated layers on a solid background depends to a great extent on the base material.
6.5
Basic laws of absorption, attenuation and
scatter
The attenuation of radiation is mainly caused by absorption and scatter. If a monochromatic, parallel radiation with the
spectral density of the radiant flux <&\ falls
perpendicularly on a bulk element dV of
an attenuating medium, then the attenuation of the radiation, if no reflection occurs, is
proportional to the film thickness dx, as
long as d$\ is sufficiently small in
comparison with $\. The following applies:
*X,x - ^X,(x+dx) _ d$\,x
®Kx
$\,x
-#X)-dx
(6.12)
88
In this:
^VXx+dx) = Transmitted spectral radiant
flux,
dj^x
= Incident spectral radiant flux,
d$\ x
j3(A)
= Loss of spectral radiant flux
= Spectral attenuation coefficient
The proportionality factor or attenuation coefficient (3(A) depends on the properties of the media (or medium) present in the
bulk element dV and on the wavelength of
the radiation.
If the bulk elements within a particular material do not cause any interactions between one another (e.g. change in aggregation) and if the absorption and the scatter are independent of one another, the attenuation is expressed by
d$\,x
*fc,x
~[a(A) + 0(A)] . dx
(6.13)
where a(A) = spectral absorption coefficient
m in
at a wavelength A,
and 0(A)
= spectral scatter coefficient in
m at a wavelength A.
The absorption coefficient (only if the reduction in intensity is due to true absorption and not to scatter) is the reciprocal value of the depth of penetration
d£. It defines the distance covered in a
medium, in which the original incident radiation is attenuated to 1/e (37%). Under the above assumption, the absorption of radiation by an absorbent particle is independent of the particle concentration. At a given wavelength A the absorption coefficient is proportional to the number of absorbent particles or the particle concentration na (Beer's law). This reads as
follows:
a(A) = a'(A) . n a
(6.14)
where a'(A) = spectral absorption coefficient,
related to the concentration.
Under the same conditions, the following
applies for the spectral scatter coefficient 0(A):
0(A) = (f(\) . n s
(6.15)
where n s = Concentration of scattering
particles
and O^A) = spectral scatter coefficient, related to the concentration.
The coefficients a'(A) and cf(A) have the same units. Thus, for the attenuation in a layer thickness dx, one can insert:
^^ = -[a'(A).na +0'(A).ns ] dx
*A,x
(6.16)
With a finite layer thickness S, the integration produces the Lambert-Beer Law, in which
&\ denotes the incident and ^^tr tne
transmitted spectral radiant flux.
$fc,tr
d^x
x=S
Jf -ZT^ =: -[a'(A) . na + (f(K) . n s ] /dx
#A,x
x=0
(6.17)
$Vr
X-- ,
ln
=-
[a'(A)
.
na +
C(A)
.
n s]
S .
^A,o
(6.18)
^o.e- <^tr =
+0 '
[a (X)
-
na
n ' (X '
S s ] .
(6.19)
The spectral transmission factor r(A is:
*A,o
(6.20)
'
7UAf}A>-6e-l 2 (X)
-
na
+tf (X>-
n s]
S -
(6.21)
89
If the radiation loss is solely due to absorption without scatter, then after integration as above, Lambert's law is
obtained:
'
a (X
«Vr = *\,o.e-'
- nal- S
^ 22)
The spectral transmission factor affected by
absorption only, 7tX) a , is:
7(X)a = e-[ a'(X)na].S
(6.23)
With constant exponents, the transmission of a material depends only on the quantity of material traversed by the radiation. The assumptions previously made, for example that a' is independent
of na and (f is independent of N s , are not
always valid. If the absorbent particles interact with one another, the total absorption will be determined not only
by the number of absorbent particles, but also by the change in concentration. Under
certain circumstances, such changes in concentration can convert the type of absorbent molecule into another. In the case of changes in concentration, Beer's law is no longer applicable.
The spectral transmission factor and the
spectral absorption factor are often stated as the pure spectral transmission factor t(X)r and the pure spectral absorption factor 0<X)r. In these, the reflection losses are not taken into
account. The pure spectral transmission factor t(X)r is defined as the ratio of the
transmitted radiant flux *J\tr to tne
incoming radiant flux Q\\:
*A)R "4*4
(6.24)
The pure spectral absorption factor ob(X)r is defined as the ratio of he absorbed radiant flux <J>X,a to the incoming radiant flux ^^j:
g* c*A) R =
<*X,i
(6.25)
6.6
Absorption and transmission spectra
The spectral parameters which depend on the media and substances can be determined for a given layer thickness. The manufacturers of optical components use graphs on their data sheets, to define the characteristics parameter for a given layer thickness as a function of wavelength.
Usually, absoroption or transmission spectra are used. Assuming that no reflection occurs, the absorption scale runs in the opposite direction to the transmission scale. In each case they add
up to 1 00%, since t + a = 1
n
*X,0
Vji J *Xji d*
i
*X.tr
X -0
x =S
Figure 6.
The derivation of the absorption law for a differential layer dx
The layer thickness affects the absorption and transmission characteristics. The
statement concerning the layer-thickness is omitted if the spectrum or the absorption coefficient is available. Very complicated conditions can arise in attenuation calculations for example in the case of the earth's atmosphere. Here, the absorption, the scatter and the
refractive index depend, not only on the wavelength, but also on further variables such as temperature, water vapour content and atmospheric pollution. In most cases, the attenuation or the transmission in the earth's atmosphere is determined for a given small wavelength range, from graphs as a function of
distance (Figure 6.8). In this graph,
90
(Km)
Figure 6.8
Transmission T in the atmosphere as a function of the distance r, with various attenuation
coefficients (5 as variables 91
300
400
500 600 700 800
1000
X (nm)
Figure 6.
Spectral transmission of light-guides
.i~
,
(%) 80 · 604020-
\ Soda-lime glass
Flint glass
2
4
6
XOim)
^
Figure 6.12
Spectral transmission of thin glass films
Figure 6.10
Spectral transmission of PVC film
"\ r
80-
IWR.60
40
1
20
I]JW\ipAr-\ n (
'
V
\
V
1
\l
2
4
6
8
10
12
14
XOim)
_
Figure 6.11
Spectral transmission of polystyrene
Figure 6.13
Spectral transmission of thin films of water
different attenuation coefficients are used as
variables. The attenuation coefficients which occur must be known in the
particular application
Figure 6.9 to 6.15 show a number of
transmission spectra, as examples. In
practice, numerous spectra are needed for
92
1 100-
V
80-
f«R
<*) 60-
Methanol
0,03 mm
40-
20-
W
0-
D
2
4
6
8
10
12
4
XGmi)
*
Figure 6.14
Spectral transmission of methanol
80-
Mr
( %) 60-
40-
20-
\ OT
MIK A
y \
U
Acetone
03 mm
2
4
6
8
10
12
14
Figure 6.15 Spectral transmission of acetone
_.
t" 0-8 o(\)
0-6
J0-4
v
n
1
05 mm
A
>--\
X
\
0-01 mm
f
11
3-'h 0-2
IAA,
2
4
6
8
10
2
> fnm*
m
Figure 6.16 Spectral absorption for various water films
-- (M
'| ><
0-3
0-2
01
/
/
/
B
10
12
14
16
the widely varying possible applications. Several physical institutes, in various countries, have extensive documentation
of transmission spectra for a large number of
substances.
Following these examples, Figure 6.16 shows the spectral absorption for various water films and Figure 6.17 the absorption coefficient a = f(X) for the
filter No. 1173 formerly produced by Texas Instruments. Furthermore, Figures 6.18 and 6.79 give the spectral
transmission of this filter.
In addition, Figures 6.20 and 6.21 show the spectral transmission of various interference filters from the Schott
Figure 6.17
Absorption coefficient a = f(~K) for the
filter No. 1173 formerly from TI
Company, Mainz, West Germany; also in Figures 6.22 and 6.23 the curves are given
for various IR-transmitting black glasses, also from Schott. With these, the pure transmission factor t(X)r is plotted as the ordinate, the reflection losses are not taken into account.
Emergent
T(A)r =
radiant flux
Incoming
In such cases, the reflectivity is usually stated separately.
93
mm mm Transmission curve for glasses 1
and 1 1
(hick
100
( *·>
60 40
·i= /
v~»>-~
1 mm
11 nun
20
I
\\
C
2
4
6
»
10
2
14
16
It
Figure 6.18
Spectral transmission of the filter No. 1173 formerly from TI
Transmission curve for glasses coated with PbF2- Composition Ge2gSb|2Se60 100
<*> «0 40 2(
2
4
6
>
10
2
4
16
1S
Figure 6. 19 Spectral transmission of the filter No. 1173 with the composition stated above
94
1-
TOO 50
(%) <0 30
20
10
s
1
0.1 0.01
500
/^
/1 /I
\
/ \\
/
\
600
700
MO
X(nm)
p
Figure 6.20 Spectral tranmission of a double-band filter (Schott)
1
50
« (*>
30
f
10
I
S
1 0.)
0.01
^ //
\V
\
/
0:001
.6
0.8
1.0
U
1.4
1*
!.»
2
>Omi)
»
Figure 6.21
Spectral transmission of an IR-band filter
(Schott)
r(»* 040
S"
/ «GN» "*
0-50
1
/
/ /
"1
1
f/ '"'
041
1 0001
OOOOI
A|
1
700
«00
900
1000
X(nm)
»·
L_
Figure 6.23
Spectral transmission of the IR-transmitting
glass RGN9 from Schott
Figure 6% 22 Spectral transmission of various IRetransmitting black glasses from Schott
L
Figure 6.24
RG Spectral transmission of the filter
830
from Schott
The characteristics of another very interesting filter from Schott, is shown by Figure 6.24, this has a good transmission
for the 900 nm radiation from GaAs diodes.
The spectral sensitivity of silicon photodetectors can thus be matched optimally to the spectral emission of GaAs diodes.
Interfering stray short-wave radiation is thus
substantially suppressed.
95
6.7 Refraction
If a light ray falls at an angle on the boundary surface of the transparent media air and water, it is deflected from its original direction at the boundary surface. The refracted ray forms the angle of refraction with the perpendicular to the surface. {Figure 6.25). The wider the angle with which the incident ray strikes
Perpendicular
j
Air
/
technical calculations, a value of n = 1 can also be used for air.
With various materials, the refractive index n is dependent on the wavelength. As a result, the angle of refraction is a function of the wavelength. The separation of a mixed radiation which thus occurs is
called dispersion.
In practice, refractive indices are normally shown for optical components graphically as a function of X. Figure 6.26 shows, as
Refraction plane
/
[
A
i\
j\
-- r*% i
/ |
/
1
/
!
/ !
Water
Figure 6.25
Refraction of a ray passing from air into
water.
the refracting plane, the greater is its
deflection. If it strikes the boundary surface perpendicularly, there is no deflection. The refractive index, the ratio of the sine of the angle of incidence, sinOg, to the sine of the angle of refraction, sinOg, is constant. Thus, with
Index e = incident Index g = refracted
sinQg sinOg
: const. (6.25)
The reflective index of a transparent substance with respect to vacuum is called the "absolute refractive index" and is denoted by the index n. Here, a refractive index of 1 is allocated to vacuum. In
n(X) 2-7000 800 600 400 200
2
4
6
""-^
8
10
1
L
Figure 6.26
The refractive index n = f(~K) for the IR filter No. 1173
an example, the refractive index n as a
function of the wavelength A for the IR
filter No. 1173 formerly from Texas
Instruments.
The law of refraction describes the path of a ray when passing from one medium into
another with a different refractive index n. At the same time as the change in direction, a change in the propagation speed of the ray, proportional to the reflective index, also takes place:
smOg
sini°g
H
(6.26)
ce - Propagation speed in medium e
Cg = Propagation speed in medium g
96
By rearrangement of this equation,
constant ratios are obtained:
^ !^ in
!
=
= constant
ce
cg
(6.27)
If these ratios are multiplied by the speed of light in vacuum, c /c gives, in each case, the refractive index for the
corresponding medium in relation to vacuum as the medium for the incident ray (c =
speed of light in vacuum).
ne . sinafe = ng . sinO^ = constant
(6.28)
The product of the refractive index and the
since of the corresponding is constant in
refraction (Law of refraction or Snell's Law). The law of refraction gives the ratio of
the refractive indices:
Perpendicular
Figure 6.27 Refraction from a less dense to an optically
denser medium
"eg' ne
(6.29)
Ray paths are always interchangeable. Therefore it is immaterial, whether the ray starts from the medium e or from the medium g. However the ratio of the refractive indices must be selected in the
right direction.
The refraction of a ray from one medium
to another medium which is optically denser
<n (n e
g), see Fig. 6.27, takes place towards
the normal to the surface. The refraction
of a ray from an optically dense medium
>n into a less dense medium (ne
g), see
Fig. 6.28, takes place away from the
normal. In case of reflection at the denser
medium, the reflected component of a
wave is given a phase shift of A/2, but not in
case of reflection at the less dense medium.
The transition from an optically denser
> medium into a less dense medium (ne ng)
shows strong reflections with increasing angle of incidence. Above a limiting angle (critical angle), total reflection occurs. The
limiting angle Oq is obtained as:
Perpendicular
Figure 6. 28 Refraction from an optically denser medium
to a less dense medium
sinO(j
(6.30)
Example: The critical angle for the transition from
glass to air is:
97
n Air sinac :
"Glass
1
= 0-667 =>Og = 42
l' 5
(6.31)
For the transition from GaAs to air, the
Oq critical angle
is:
n Air sinOQ :
nGaAs
0-278 =>Og= 16-1
3-6
(6.32)
As is shown by the last example, the small exit angle, above which total reflection occurs, has a very unfavourable effect on the radiant efficiency of GaAs diodes. Total reflection however can be utilised to useful effect for example in light guides.
In general, a reflection occurs whenever a ray is refracted. If the absorption of the
second medium is very slight or non-
existent, then for Figures 6.27 and 6.28:
% % = - $ r
(6.33)
% 4> = g
- p <J> .
(6.34)
* =* g
.(l-p)
6.35)
Here:
3> = incident radiation power,
4>g = refracted radiant power,
<t>j = reflected radiant power.
The reflectivity depends on the refractive indices of the two media. If the absorption of the second medium is very slight or nonexistent, then one obtains, for the
reflectivity with perpendicular incident radiation:
np -- nP ^
P^-1 -)
n g
+
ne
(6.36)
From equation (6.3), the reflected radiant power is:
<% = %.(-
n
g
+
ne
(6.37)
In the case of the passage of a ray through air-glass-air, the reflectivity is:
P=(-- (-- no -- ni i) 2 .2 =
1-5
--
1 -2
r)
.2 =
n2 + ni
1-5 + 1
(6.38)
In factor 2 occurs because part of the radiation
radiation is reflected twice. In total, 8% of
the incident radiant power is reflected. In the IR range, the refractive indices lie between 1-2 and about 4, so that reflectivities between 1 and 36% occur.
The reflection losses and with them the unwanted reflections can be effectively suppressed by application of an antireflection layer ("bloom") to both sides of optical glasses. The anti-reflection effect takes place through interference. Of course, dut to the wavelength-dependence of the interference phenomena, it can only be made particularly effective (T>99%) for a limited wavelength range. Also it is dependent on the angle of incidence.
In practice, the radiant power distribution in flat parallel plates is often of interest.
When using filters, this power distribution
through refraction and reflection must also be taken into account. With a completely absorption-free substance, the radiant power is divided by multiple reflection (see Figure 6.29).
On entry into the medium of the plate,
the radiant power <t> is attenuated by the
factor (1 -- p) and again, on entering the
original medium, by the factor (1 - p), that
is, in total, by (1
-
2 p) .
Figure
6.29
gives
information on the power distribution of the
$ incident radiation .
In practice, for the refracted radiant power
*<g
*g = <fr .(l -p) 2
(6.39)
98
*g3 = *o ·(' ff P *rl s *o P>
·o-tl-»)*-p""*J6
*o-(l-P?V**g4 *o-(l !>}'«&
Figure 6.29
Reflection and refraction at a plane, parallel-faced plate.
99
Radiation sources
7.1
Natural Radiation Sources
7.1.1 The Sun
7.1.2 The Moon
7.1.3 Clouds
7.2
Artificial Temperature Radiators
7.2.1 Open Fire
7.2.2 Filament Lamps
7.3
Luminescence Radiators
7.3.1 Hot Gases
7.3.2 Discharge through Gases
7.3.3 Discharge lamps
7.3.4 Discharge Tubes
7.3.5 7.3.6
Xenon Lamps Metal Vapour Lamps
7.3.6.1 Sodium Vapour Lamps 7.3.6.2 Low-pressure Mercury-vapour Lamps
7.3.6.3 High-pressure and Very-high-pressure
Mercury-vapour Lamps
7.3.6.4 Fluorescent Lamps
7.4
Mixed-light Lamps
7.5
Flash-discharge Tubes in Photography
7.6
Luminescent Diodes
7.6.1 Silicon-doped GaAs Diodes
7.6.2 7.6.3 7.6.4
GaP Diodes Zinc-doped GaAs and GaAsP Diodes Consideration of Quantum Yield
101
Radiation Sources
In semiconductor optoelectronics, three kinds of radiation source are mainly of interest: filament lamps, semiconductor light-emitting diodes and daylight (natural temperature radiators). Other radiation sources, such as discharge lamps, are seldom used in electronic systems except for specific military applications. In special industrial applications, the radiation from incandescent metals or burning gases is analysed.
The ambient and/or background radiation from natural and artificial radiation sources
can severely affect an electro-optical system, if the spectral radiation distribution of the interfering radiation source falls within the spectral sensitivity range of the photodetector.
Radiation sources are classified into groups according to the type of radiation or
radiation phenomenon utilised and
taking into account the many basic types which have been developed, as is shown in
Figure 7.1. Here the most important
radiation sources are systematically summarised.
7.1 Natural Radiation Sources
7.1.1
The Sun
The sun has approximately the spectral radiation distribution of a black body with
the distribution temperature Ty = 6000 K.
The solar radiation undergoes absorption in the atmosphere in some wavelength
ranges.
On the surface of the earth, the solar spectrum contains characteristic H2O and
Radiation sources
Thermal radiation sources
Luminescence radiators
c
1 fogfcM TM*.pe»W« ndfa.w[
MM*l T »tottoft»wv«» | |^«», tomHW^xiv.x- ndtoon.) fcjrfkbl turnip ace radioes]
J
pk-jvci.lv h.HlK--
[lm.--jmlf-n.tni MihmanceT]
P
]
Vacuum filament lamp Cms-I riled litament lamp
liH.-vmlew.vnl rttctiiU Silicon carbide rod*
Ounpositc lamp* |
I
[
lifameni with mercury
vu|i»ur lamp connected
I- wellies Polar lielmlAuror
KelkcitiH! heavenly bodice
Luminescent %url'ace*
lli>i paws, vtmc will temperature radial in
Rjdkilumtnvwnt illuminated
s ipn> Phololuminetveni illuminated
Incandescvnt pa*- nun lie Bun*cn flame t-xhauM flame*
(j»> dKcharpe Limp*
(,lim discharge lamp Dhcliarec lube Plasma display Xenon lump Xenon flash tube
Metal vapour lumps
Mercury -vapour lamp I luorescent lamp Sodium vapour lamp Tellurium-vapour lamp
Colour TV lube
Lfcrtl-emittin? diode
Black-and-white TV lube Gar" diode*. GaAsP diodes
Oscillograph tube
Semieonduciof display
IR-emmine tu mine wem diode
UaAs diode. InAs diode
Figure 7.1
Systematic summary of the most important radiation sources.
103
CO2 absorption bands (Figure 7.2).
7.2.1
Open Fire
In heating installations, fires are often monitored by optoelectronic means. In
this case the 1 2 Hz flicker frequency of the flames is mostly used. The spectral radiation distribution depends on the
burning substance and its temperature. In some wavelength ranges, the flames radiate like "grey emitters", since they show the spectral radiation distribution of a Planck radiator. Figure 7.3 shows,
l»
EW
Figure 7.2 The solar spectrum with absorption bands
The broken line shows the solar spectrum without absorption by the atmosphere. In summer, around midday, with solar radiation, an illuminance up to 1 30 000 lm/m2 can be measured on the surface of the earth. The illuminance of daylight in winter is about 10 000 lm/m .
0-6
N
0-5
3
1
2
3
X(nm)-
7.1.2
The Moon
The light radiated from the moon to the earth has an illuminance of about 0.1 lm/m
7.1.3
Clouds
Clouds reflect the radiation from the sun, and also from the earth, very effectively. Although, their radiance is small, there are normally large areas of clouds and the total radiant power is relatively large.
7.2
Artificial Temperature Radiators
Figure 7.3
K Spectral emissivity of coal at 2000
as an example, the spectral emissivity
of coal at a temperature T = 2000 K.
For the visible range, the spectral emissivity is uniformly E(X) = 0-7.
7.2.2
Filament Lamps
Filament lamps consist of a glass bulb, which is either evacuated or filled with gas. The radiation is normally produced by an incandescent tungsten filament. Filament lamps are temperature radiators which means they have a continuous spectral radiation distribution (see Chapter 4). The
104
working temperatures lie between
2200 K and 3000 K, the melting point
of tungsten being about 3600 K.
The radiant power, luminous power and also the life are dependent, to a great extent, on the voltage. Figure 7.4 shows the variation in the life time and the
t
1 500
7.3.1
Hot gases
Hot gases are generally selective radiators. They have a few emission bands. Through
the combustion process, CO, CO2 and H2O
emission bands also occur (CO = 4-8 jlm; CO2 = 2-7 /An, 4-4 jum and 15 /Urn;
H2O = 2-8 jltm and 6-7 jUm). Carbon
components in the flame cause a continuous
spectral emission distribution similar to that of the black body. Figure 7.6 shows, as an example, the spectral radiation distribution of a low-pressure Bunsen flame.
The continuous spectrum very clearly has emission bands superimposed on it.
110
120
130
Relative working voltage (%)
Figure 7.4
Variation in the luminous power and the life of a filament lamp with deviation from
the rated voltage
luminous power of a filament lamp with deviation from the rated voltage.
Filament lamps have a positive temperature coefficient and can therefore be used, for
example, in Wien bridge circuits, as PTC resistors. The limiting frequency of small filament lamps lies between 15 Hz and 100 Hz. As an example, the frequency response of a scale lamp is illustrated in
Figure 7.5.
7.3
Luminescence radiators
7.3.2 Discharges through gases
Under normal conditions, gases are good insulators, as long as any applied electrical field strength remains efficiently small. This changes, however, if the field strength exceeds a certain level, depending on the type of gas and the gas pressure. Gas discharges, associated with luminescence phenomena, then take place.
Because of the effect of short-wave and radioactive radiations, which occur practically everywhere, and of thermal excitation, a tiny fraction of the atoms or molecules in every gas is always ionised. These ionised particles (free electrons, positively and negatively charged ions)
move under the influence of an electrical field and produce a very weak current, which can only be detected with the most sensitive
measuring instruments. Since no luminescence phenomena are vet t0 " e observed at this stage, the term "dark discharge" is used.
The glow or arc discharges which occur
with increased field strength can appear
in the most widely varying forms. A
distinction is made, for example, between low-pressure and high-pressure discharges.
105
10"
V(mV)
10s
Memuement circuit
+ 1SV -
^W 2N3244 /!><
| lOknM
$p--
-* Oscillograph
fru")
«1
TI
fB»22y F s
[
^
\
10»
~r
mu Uniitinf frequency of nhoMsfg^- 3db v
mA 6 V/300
telephone lamp = 1 8 Hz subsequently
6 V/ 40 mA telephone lamp « 40 Hz 6 dB per ocUve
12 V/ 30 mA Mini-lamp
= 53 Hz
101
10"
Figure 7.5
Frequency response of a 6 V/0-3 A filament lamp
10*
«Hz)-
Figure 7.6
Spectral radiation distribution of a lowpressure Bunsen flame
In principle, a glow discharge can be produced at any pressure, but the luminescence phenomena occurring below a pressure of about 100 mbar are the most
striking.
In a suitably constructed discharge tube,
before a visible flow discharge commences through an increase in the field strength,
the current first rises, because the original
free electrons, being accelerated more and more, ionise neutral particles by giving up their kinetic energy and thus multiply the number of available charge carriers. If the
primary cause of the spontaneously
106
produced charge carriers were now to be
suppressed, the whole flow of current would come to a standstill, since now, too, no more additional charge carriers are formed. This transitional range is called the "Townsend discharge" and, like the dark discharge, is not self-supporting, since an external means of ionisation is necessary to maintain the current.
With further increase in the field strength,
the positive ions are given so much energy of motion, that finally, on collision with the
cathode, they liberate secondary electrons and thus greatly increase the current. Luminescence phenomena, such as are illustrated in Figure 7. 7, then commence. In this, the negative glow and the positive column are of particular technical interest. Figure 7.8 shows the potential, fieldstrength and space-charge distribution of a
glow discharge in schematic form. The gas, which is ionised to a certain extent during
discharges, is called a plasma.
Glow and arc discharges can be maintained,
even without external ionisation; they are therefore called self-supporting discharges.
With further increase in the applied voltage, an arc discharge is finally obtained with very heavy currents, while the voltage on the electrodes collapses to a fairly low value. The cause of the high current lies in the fact that, through the intensive ion bombardment on the cathode, the latter is heated up, so that a plentiful thermal emission takes place.
With a variation of the arc discharge,
which occurs in tubes with a liquid mercury
cathode, it is not a matter of thermal
emission but of field emission. Here, the
arc constricts itself very tightly just in
front of the cathode, while field
strengths up
to
8
10
V/m
are
produced.
Figure 7. 9 shows a typical I-V discharge characteristic. The glow discharge is a cold discharge, since generally no significant heat is produced. The spectrum of the
luminescence radiation, occurring through recombination, is substantially dependent on the pressure and the type of gas. At low pressures, discrete line spectra are obtained, and at higher pressures, through the increasing effect of the particles on one another, these merge into more or less continuous band
spectra. The spatial distribution of the
luminescent layers is also pressure-
dependent. The lower the pressure, the more the areas at the cathode end spread at the expense of those at the anode end. Changes in the electrode spacing only affect the length of the positive column. If the anode is moved into the cathode drop area, the glow discharge is
extinguished.
Arc discharges take place both at low and higher pressures. The spectra, which depend on the type of gas and the pressure, are often accompanied by a noticeable temperature radiation. At very high pressures, continuous spectra are obtained over a wide wavelength range.
7.3.3 Discharge lamps
The discharge lamp has a cathode in the form of a metal cylinder or in the shape
of a beehive. A rod-shaped anode is located
inside the cathode cylinder. Discharge lamps are filled either with neon or helium gas or a corresponding gas mixture. Only the negative cathode glow light is utilised.
The colour of the light is reddish or orange. The current is limited by a series resistance. The limiting frequency lies between 10 kHz and 100 kHz, depending on the type.
7.3.4 Discharge tubes
Discharge tubes are gas-filled glass tubes.
The unheated electrodes are fitted at each
A end of the tube. stray-field transformer
107
Anode Anode glow Anode dark space
Positive column
Faraday dark space Negative glow Discharge fringe Crookes' (Hittorfs) dark space Cathode layer Aston's dark space Cathode
Figure 7.7
The individual regions of a low-pressure glow discharge 108
Potential
V
r
I
Cathode drop
/o
Field strength
V
Anode drop
7.3.5
Xenon lamps
Xenon lamps consist of quartz bulbs with a xenon gas filling. They are fired with high voltage. In the visible range, xenon
lamps have a continuous radiation distribution with a light superimposition of spectral bands. It is therefore possible to allocate a colour temperature to the visible spectrum of the xenon lamp. It is about Tf = 6000 K. Figure 7.10 shows the spectral radiation distribution of a xenon lamp. In the near IR range, two pronounced
spectral bands can be seen. Xenon lamps can therefore also be used in IR transmission
systems.
Anode
Figure 7.8
Potential distribution, field strength and space charge in a low-pressure glow
discharge
supplies the necessary high voltage for ignition and limits the working current through the action of the stray field. With
the exception of the xenon tube, discharge tubes produce a discrete line spectrum. Table 7. 1 shows the light colour of discharge tubes with different gas fillings.
7.3.6
Metal Vapour Lamps
Metal vapour lamps are constructed as glass tubes with a metal vapour filling. Ballast units are necessary for ignition and for current limitation in the operating condition. Metal vapour lamps produce a discrete line spectrum.
7.3.6.1
Sodium Vapour Lamps
These produce almost monochromatic yellow light with a wavelength X = 589 nm.
Ligh colour
Gas filling
Red Yellow
Blue White Green
Neon Helium Neon and mercury Carbon dioxide Neon and mercury with brown coloured
glasses
Table 7.1
Light colour of discharge tubes with
difference gas fillings
7.3.6.2
Low-pressure Mercury-vapour Lamps
These lamps radiate mainly in the UV
range.
7.3.6.3
High-pressure and Very-high-pressure Mercury-vapour Lamps
The spectral radiation distribution of high-
pressure and very-high-pressure mercury
UV vapour lamps lies in the
and, with the
109
1(A) \tr*
Arc discharge
Anomalous glow discharge
Normal glow discharge
Transition zon-
Townsend discharge
icr*
--I
100
ISO
200
Figure 7.9
Typical I- V characteristic of a gas discharge
colours blue and green, in the visible range (see Figure 7.11).
7.3.6.4
Fluorescent lamps
These are low-pressure mercury-vapour
lamps, in which the four characteristic
Hg-vapour spectral lines are initially used.
UV The invisible
radiation excites the
coating of flurescent material, which is
wash-coated onto the interior wall of the
V(V).
_l_
Dark discharge
A glass tube, to fluorescent radiation.
continuous spectrum in the visible range is thus produced. All colours can be achieved by suitable selection of the fluorescent materials. Figures 7.12 and 7.13 show the spectral radiation
distribution S\ = f(X) of a white-light and a warm-white fluorescent tube.
7.4
Mixed light lamps
Mixed-light lamps are composite lamps.
110
500
600
Figure 7.10
Spectral radiation distribution of a high-pressure xenon lamp
They contain, in one glass tube, an incandescent filament and a high-pressure mercury-vapour lamp. The reddish filament lamp light and the greenish-blue mercury vapour light are combined in one lamp. These combined luminescence and temperature radiators are also obtainable with a fluorescent coating on the interior of
the glass tube.
7.5
Flash discharge tubes in photography
In amateur flash units, xenon flash lamps or tubes are used as a flash radiation source. Figure 7.14 shows the construction and the technical data of a standard flash discharge tube. The colour temperature of the xenon-filled flash discharge tubes lies
Figure 7.11
Spectral radiation distribution of a very-high-pressure mercury-vapour lamp
111
t sx
H 1.1
wo
500
600
700
X (nm)
*
Figure 7.12
Spectral radiation distribution of a white fluorescent lamp
ft~ Flexible ignition lead
Arc length Length of fliu lube Overall length Tube Diameter Raw material Construction tiled rode material Ignition electrode Cat filling Load capacity Load capacity Hash entigy Peak cunent Anode voltage Ignition voltage Ignition energy Colour temperature Radiation ipttirum Luminoui efficiency Life llaih repetition rate Klaih lime
20 - 40 mm JO - SO mm 35 - 55 mm mm 3.5 - 5
Hard giaii/Quartz Rod form Nickel/Tung iten
Wire/Metal foil/TraiMparent, conductive layer
Xenon Hard glau 10-20 Wi/cm1 internal lurface Quarti 20-40 Wi/cm' intenuJ surface 20- 100 Wi 100 - 500 A ISO - 400 V 7 - 10 IcV 1 - 5 mW. 5000 - 7000 K At daylight 30 - 50 LnVW 2500 tlaihei
Every 10
l(1000i
Figure 7.14 Standard flash discharge tube for photographic applications
produce higher colour temperatures. With
UV hard-glass tubes, the
component, up
to about 350 nm, is absorbed by the glass.
The spectral radiation distribution
approximately corresponds to the spectral
distribution of daylight with a superimposed
line spectrum. The spectral radiation
distribution of the xenon flash discharge
tube and that of daylight are illustrated in
Figure 7.15.
Figure 7.13
Spectral radiation distribution of a warmwhite fluorescent lamp
between 5000 and 7000 K. Short discharge times with high peak currents
Figure 7.15
Spectral radiation distribution of the xenon flash discharge tube and of daylight
112
7.6
Luminescence Diodes
The radiation of a luminescence diode (light-emitting diode) is produced by the recombination of the charge carriers
injected in the junction region. Injection luminescence has been dealt with in Section 1.2.3. Wafer geometries will be described in Section 11.1.
7.6.1
Silicon-doped GaAs Diodes
In the mass production of wafers for use as luminescence diodes, two technologies have become principally established. The first method is liquid epitaxy, the principle
of which is illustrated, for Si-diped GaAs diodes, in Figure 7.16. Here an-N-type GaAs single-crystal slice, doped with silicon, is
sunk into a gallium melt, saturated with
GaAs at about 900°C and doped with
silicon. While the melt cools down, a deposit forms on the substrate slice.
As a doping substance, silicon behaves here in an amphoteric manner, that is, as a quadrivalent element it can be fitted into the GaAs lattice in two possible ways. Silicon atoms either replace gallium atoms and thus form donors or they replace arsenic atoms and act as acceptors.
Above a temperature of 900°C, silicon occupies more donor places than acceptor levels. Below a temperature of 900 C, on the other hand, it occupies more acceptor levels than donor levels. Through a continuous growth process with an initial temperature of about 950 C and
N-type GaAs
single-crystal slice
Solution epitaxy and
production of the PN junction
Metallisation for application of contacts
Melt (Ga + GaAs + Si) Temperature 950°C-850°C
Figure 7.16
Principle of liquid epitaxy (solution growth) in the vertical reactor for Si-doped GaAs diodes
Separation into individual wafers
Figure 7.17 Simplified production scheme for Si-doped
GaAs wafers
subsequent reduction of the temperature to
about 850°C, a PN junction, formed. Figure
7.17 shows a gre.atly simplified production scheme for silicon-doped GaAs wafers.
7.6.2
GaP Diodes
GaP diodes are produced in a double
113
epitaxy process. An N-type GaP single
crystal, drawn by the Czochralski process and doped with tellurium, serves as
the substrate. In the first liquid epitaxy process to produce red-emitting wafers, a
tellurium doped-n-type GaP layer is
grown. For clarification, Figure 7. 18 shows the principle of liquid epitaxy for
red-emitting GaP wafers.) In the second
stage of the liquid epitaxy process, the
PN junction is formed by growing a p-layer
doped with zinc and oxygen, on top of the tellurium doped n-layer.
proportion of arsenic and phosphorus to produce a useful diode and this problem is overcome in the vapour phase epitaxy process. Here a single-crystal layer of GaAsP or GaAs is allowed to grow on single-crystal GaAs slices. Figure 7.19 shows the principle of vapour phase epitaxy used for the production of GaAsP wafers. The
PN junction is produced in the subsequent
planar process. Finally, in Figure 7.20, a
Temperature IIOO°C~500 C
Figure 7.18
Principle of liquid epitaxy in the horizontal reactor for the production of red-emitting GaP diodes
7.6.3
Zinc-doped GaAs and GaAsP Diodes
This second process has been successfully adopted for the mass production of zinc-
doped GaAs and GaAsP wafers. In this case, vapour phase epitaxy is combined with planar technology GaAsP cannot be drawn from a melt with the correct
Figure 7.19
Principle ofgas epitaxy for GaAsP wafers
simplified production scheme for zincdoped GaAs and GaAsP wafers is shown.
7.6.4
Consideration of quantum yield
The highest quantum yields are at present
obtained with the luminescence diodes produced by the liquid epitaxy process. In the liquid epitaxy process, as compared with vapour phase epitaxy, the slice surface which is composed of disordered diffusion atoms is first dissolved, so that an epitaxial layer of high perfection can be built up out of the saturated gallium melt.
114
N-type GaAs
single-crystal slice
Growth of a GaAs or GaAsP s single-crystal layer by vapour phase
epitaxy
Deposition of a masking layer, usually of Si3N4 + Si02
The masking layer is partially removed again by photolithography, to produce windows
Diffusion process produces the PN junction in the exposed window
areas
Application of a protective layer to improve the long-term stability and to reduce reflection losses
Metallisation for application of contacts
Separation into individual wafers Figure 7.20
Simplified production scheme for zincdoped GaAs and GaAsP wafers
115
8
Photodetectors
8.1 8.2 8.2.1 8.3
Photodetectors with external photoeffect Photodetectors with internal photoeffect Photoconductors or photoresistors Junction photodetectors
117
»
Photodetectors
Photodetectors are classified into various groups according to the photoeffect utilised. Tht> most important kinds are summarised
in Figure 8. 1. The thermal and pneumatic photodetectors, which are mainly used for radiation measurements, are. described as far as necessary in Chapter 1 2 (radiation measurements). The photoelectronic
detectors are distinguished according to whether the external or internal photoeffect is used. Figure 8.2 shows the spectral distribution of the most important semiconductor radiation sources and photodetectors.
8.1
Photodetectors with external photoeffect
The external photoeffect has already been described in Section 1.3.1. Photodetectors
with external photoeffect are divided into non-amplifying and amplifying photocells. In principle, they consist of a photo-
cathode and an anode in a vacuum or gas-
filled tube.
The photosensitive cathode surface is made from alkali metals or their compounds,
such as silicon-antimony, potassium, rubidium, caesium, caesium-antomony or caesium oxide or multi-alkali compounds.
The relative spectral sensitivity distribution of alkali metals is shown in Figure 1.8.
There is a linear relationship between the photocurrent I,, of a vacuum photocell and the incident irradiance E e .
In gas-filled photocells, the photoelectrons, which are accelerated as a result of the
electric field between the cathode and the anode, produce secondary electrons on collision with gas atoms. These increase the electron current to the anode, while the positive gas ions diffuse towards the cathode. The ion bombardment on the photocathode causes a greater decrease in sensitivity with time then with vacuum photocells. Argon is mostly used as the filling gas. The operating voltage lies
around 60 to 75% of the breakover voltage. The limiting frequency of the gas
Tnermel photodetectors
Bolometers Thermocouple Thermopilf
Pyrc-electric, l*o'o<l«'«>ors
Photodetecton with extern*, photoeffect
Photoelectric detectors
Pneumatic photoconductor Photodetecton with internal photoeffect
Photoconductor
Non-amplifying photodetectort
Vacuum photocell
Gaa-nUed photocell
(gju amplification
3-8 apptox.
timet)
Amplifying photodetectort
PhotomuttipUBn Image converter*
FhotoetectromagnetJc detectori
Doped photoconducton
1R. detectof (N-type or P-typ*)
Intrinsic photoconducto rt
Photoretitton IR detectori
Non-amplifying photodetectort
Photocell Pfcotodiode Photoduodiode
Amplifying photodetecton
Phototiamntoii
Phototnyrutori Avalanche photodiodet
Figure 8.1
Systematic summary of the most important photodetectors
119
Semiconductor radiation detectors HgCdTe
uv
-- >'fflflM---»
GaP GaAs
I
GaAIAs
GeCu Semiconductor radiation emitters
HJ
Figure 8.
Spectral distribution of the most important semiconductor radiation sources and
photodetector.
photocell is low in comparison with the vacuum photocell and is usually 5 to 10 kHz
Because of the formation of secondary electrons, gas photocells have a small
M gain of = 3 to 8. The characterisitc Ip =
f(E e ) is not linear and, in the higher voltage range, can be compared with the output characteristics of triode valves. Gas photo-
cells are being increasingly superseded by semiconductor photodetectors.
The combindation of a vacuum photocell
with a secondary electron multiplier
produces a photomultiplier. On impact
with a dynode (auxiliary anode), a photoelectron produces several secondary electrons. In this process, no har nful free ions are produced. The amplification of the photocurrent takes place in several stages. It amounts to gains of several orders of magnitude, depending on construction and type.
8.2 Photodetectors with internal photoeffect
The internal photoeffect has been described in Sections 1.3.2 and 1.3.3. The
photodetectors with internal photoeffect include photoelectromagnetic detectors, photoconductors and barrier-layer
photodetectors. The photoelectromagnetic detectors have so far achieved hardly any
practical significance.
8.2.1
Photoconductors or photoresistors
The photoconductor utilises the change in conductivity of a semiconductor under an incident radiation. Photoconductors are either pure intrinsic semiconductors or N-type or P-type doped semiconductors.
A distinction is therefore made between
intrinsic photoconductors and doped photo-semiconductors.
120
;
Intrinsic photoconductors are produced both for the visible and for the very near IR range from cadmium chalcogenides such as cadmium sulphide (CdS), cadmium selenide (CdSe) and cadmium telluride (CdTe). Intrinsic photoconductors for the near IR range consist of lead salts such as lead sulphide (PbS), lead selenide (PbSe) and lead telluride. For the middle and far IR range, intrinsic photoconductors consisting of indium arsenide (In As), indium antimonide (InSb), tellurium (Te) and mercury cadmium telluride (HgCdTe) are used as IR detectors.
The doped photo-semiconductors are doped with appropriate substances, in order
to achieve a desired spectral sensitivity in
the middle and far IR range up to about 40 /im. So far they are mainly made from germanium or germanium-silicon compounds with appropriate doping. The most important doped photosemiconductors are: Gold-doped germanium (Ge:Au); Gold-, antimony-doped germanium (Ge:Au,Sb); Zinc-doped germanium (Ge:Zn); Zinc-, antomony-doped germanium (Ge:Zn,Sb); Copper-doped germanium (Ge:Cu) Cadmium-doped germanium (Ge:Cd); Gold-doped germanium-silicon alloy
(Ge-Si:Au);
Zinc-, antimony-doped germanium-
silicon alloy (Ge-Si:Zn, Sb).
Generally, the sensitivity of photoresistors are dependent on their "history", so that
they show undesired fatigue phenomena, as compared with junction photodetectors. Because of the long life of the charge carriers, the frequency response of
sensitive photoresistors is relatively low.
Depending on their design and spectral sensitivity, IR detectors are operated at low temperatures in Dewar flasks or cryostats. The most important sensitivity values of IR detectors are dealt with in more detail in Section 9.9.
8.3 Junction photodetectors
These photodetectors are classified into non-amplifying and amplifying photodetectors. The non-amplifying devices include photodiodes, photoduodiodes, Schottky barrier photodiodes and photocells. The amplifying photodetectors include phototransistors, photo-field-effecttransistors, photothyristors and avalanche photodiodes. The parameters of nonamplifying and amplifying junction photodetectors are dealt with in detail in Chapter 9.
121
9
Parameters of IR
Detectors and
Junction Photodetectors
9.1 9. 1 . 1 9.2 9.2.1 9.2.2 9.2.3 9.2.4 9.2.5 9.3
9.4
9.5 9.6 9.7
9.8 9.8.1 9.9
Quantum yield, Q, of junction photo-
detectors
Quantum yield and photocurrent
gain of avalanche photodiodes Spectral sensitivity of junction photodetectors Schottky Barrier PIN Photo-
diodes Planar diffused Si photodiodes
Photodiodes by the CDI process Effect of temperature on the spectral
sensitivity of photodiodes Possibilities for shifting the spectral
sensitivity
Evaluation of radiation by non-
amplifying junction photodetectors for monochromatic radiation sources
Evaluation of radiation by non-
amplifying junction photodetectors for chromatic radiation (mixed radiation)
Evaluation of radiation by amplifying
junction photodetectors Area-dependent sensitivity parameters for junction photodetectors. Actinic value of IR luminescence radiation
from GaAs diodes for silicon junction
photodetectors Dark current of junction photodetectors Dark current of avalanche photodiodes
Sensitivity parameters of IR detectors
123
Parameters of IR Detectors and Semiconductor Junction Photodetectors
9.1
Quantum yield, Q, of junction
photodetectors
For the measurement and calculation of
quantum yield Q of junction radiation
detectors, non-amplifying photodetectors are advantageously operated as photocells. Photodiodes and photocells are physically
identical electrical components. They only differ in the electrical mode of operation.
Photodiodes are operated with an external voltage source in the reverse direction, while photocells (solar cells) themselves generate a voltage, the photo-voltage.
Here, the number of photons per unit time for monochromatic radiation falling on the photodetector is the ratio of the incident radiant power (see equation 2.4) to the energy of one photon (see equation
1-1)
nPh _%Xl1
h.v
(9.2)
The number of electrons per unit time ng is
the ratio of the product of the electrical
photocurrent Ip and the time t to the
electronic
charge
e
=
1-6
.
_,1°9 10
As.
of one
electron.
Junction photodetectors with internal amplification such as avalanche photodiodes or phototransistors, with an external base connection, can also be operated as photocells. Phototransistors obtain their base control current from the photocurrent generated in the collector-base photodiode (see Chapter 1). Their current amplification
M factor depends on the magnitude of the
photocurrent. Therefore, the quantum
Q yield of a phototransistor is conveniently
determined from the collector-base junction, operated as a photocell. Similar considerations apply for field effect phototransisitors and photothyristors, since these also come into the category of junction photodetectors with internal amplification.
The photocurrent Ip of a (non-amplifying)
junction photocell is proportional to the
incident radiant power <J>e . The quantum
yield of a junction photocell can therefore
be defined as the ratio of the number of electrons n£ flowing in the external circuit per unit of time to the number of photons nph falling on the photodetector.
Thus:
Ip.t nE =
(9.3)
If the equations (9.2), (9.3), (1.2) and (1.3) are inserted in equation (9.1), the quantum yield for monochromatic radiation Q(X) is obtained:
Q(X)
=
Ip,X-t-h
.^_
I X-h..c
P;
x e
.
3> e)
-1
X <J>e ,A .
e .
(9.4)
The constants h and c /e (see Chapter 1)
can be replaced by a common constant
factor:
h.c
"Ws 6-62 . 10
2 . 3 . 10 8 ms J
1-6 . 10 19 As
WmA = 1-24 . 10 6
*
(9.5)
The quotient cf tiie photocurrent Ip \ and the incident mono chromatic radiant power dig ^ can be replaced by the absolute
spectral photosensitivity for photocells
(see Sections 9.2 and 9.3):
"E
Q:
t
"Ph _ "E t nPh
(9.1)
J P,X $e,X
s(X)
(9.6)
125
If the equations (9.5) and (9.6) are inserted in equation (9.4), then, with the absolute spectral sensitivity s(X) for monochromatic radiation of wavelength X, the quantum yield Q(X) for the same monochromatic
radiation of wavelength X can be calculated:
-- Q(X) =
.1-24.10 6 WmA
A
(9.7)
If the quantum yield for monochromatic
radiation Q(X) of a junction photocell is equal to unity, then by conversion of equation (9.7), the theoretical absolute spectral sensitivity s(\)th can be determined (see Section 9.2):
WmA s(X>th = 1-24
10 6
l
(9.8)
temperature, in the longer-wavelength limiting sensitivity range. Figure 9. 1 shows
90
|
X'0,9tan 70
60
SO 40
30 X s l,06(«i^
20
10
- 60 -40 -20
20
40
60
80
100
If the absolute spectral sensitivity s(X) of a
junction photocell is known, the quantum yield for monochromatic radiation Q(X) can also be calculated by forming the
quotient of the absolute spectral sensitivity s(X) at the radiation wavelength
X divided by the theoretical absolute
spectral sensitivity s(A)th at the same radiation wavelength X:
s(X)
Q(X) =
s(X) th
(9.9)
The quantum yield for monochromatic
radiation Q(X) is also stated as a relative value. The relative quantum yield for monochromatic radiation Q(X) rei is the
ratio of the maximum absolute spectral quantum yield Q(X) max for monochromatic radiation of the maximum wavelength Xmax to the absolute spectral quantum yield for monochromatic radiation
Q(X) at any given radiation wavelength X:
Q(X) = Q(X) max
Q(X)
(9.10)
The spectral quantum yield for silicon photodiodes is dependent mainly on the
Figure 9.1
Spectral quantum yield Q(\j as a function of the case temperature tg for the Si photodiode TIXL 80
the spectral quantum yield Q(X) as a function of the case temperature Iq for the largearea Si photodiode TIXL 80. The variable in each case is the wavelength X of the incident monochromatic radiation.
9.1.1
Quantum yield and photocurrent
amplification of avalanche photodiodes
For junction photodetectors with internal
amplification, the quantum yield Q(X) for monochromatic radiation can be determined
M if the internal gain is known. The
quantum yield for monochromatic radiation, Q(X), is measured for avalanche
photodiodes, with the amplification factor
M set by the applied reverse voltage, for
the given wavelength of the laser radiation used. In the data sheet, a typical absolute value for Q(X) and also the function Q(X) re l = f(X) are usually stated. Figure 9.2
126
M-100 tc-25»C
80
f<100 dfc\
/
/
//
\ \\
f>10MlfcS. 10
Figure 9.2
Relative quantum yield Q(ty re[ as a function of the radiation wavelength Xfor the silicon avalanche photodiode TIXL 56.
shows the relative quantum yield Q(X) as a function of the radiation wavelength X for the silicon avalanche photodiode TIXL 56. Here, the case temperature (iq) is 25°C
M and the photocurrent amplification = 100
The variable is the modulation frequency
of the incident radiation.
The maximum quantum yield for avalanche
photodiodes is reached with the highest
practicable amplification M. To achieve
this and to make full use of the avalanche
effect, the working point is generally set
a few tenths of a volt below the breakdown
voltage VgR. If the working point is in the breakdown region, the useful signal power is
considerably less than the total noise power
produced through the increased current
Vr noise. The applied reverse voltage
must
therefore be adjusted exactly, in order to
compensate for the tolerances stated in
the data sheet for the breakdown voltage
VBR> typically between 140 and 200 V for silicon types and between 30 and 60 V or 85 and 150 V for germanium. Among
the individual devices within a production
batch, however, a distribution of only a
few volts is normally seen for breakdown
voltage.
The temperature coefficient TKyon of the
breakdown voltage VgR is calculated in
accordance with the following equations:
_ V BR(125°C) - VBR(-55°C)
i.v VBR
C
125°C-- (-55 C)
TKvBR ,rel _ VBR(125°C) ~VBR(--55°C)
VBR(25°C)
100% 180°C
The temperature limits are to be taken from the data sheet in each case. The
typical relative temperature coefficient
TKv BR) rel
for Si avalanche photodiodes is about
0-11%/ C and for Ge devices about
0-15%/°C.
M The photocurrent amplification for
avalanche photodiodes is measured with
a fixed avalanche noise response
Mj threshold. Here, the amplification
is
to be determined for the applied
reverse voltage Vr, at which the noise
deviates from the theoretical characteristic
Figure 9.3 shows the incident radiant
power 4>e and the noise power P^ as
t
/ Signal pow
1
f NoiMpcxn
/i J
1
^/
!«T
1
M lof
>
Figure 9.3
M 3>e and P^ as functions of for the
avalanche photodiode TIXL 56.
127
7"
/lltuWlcm'
0-6 0-8
I
~
I
E* =
I
1 lSmW/an'
1 |
5-9 mW/cm1
ll-8mW/cm3
14
L\
1-6
/
//
22
iz / /
24
/
/
26
2-8
3-2 /
/
_i
J/
^f
i
4-2
13
look n
cvR)-. -4 (£) -
TlXLsJ
N^oov
4.4
1
4-6
50
i
60
SO
Vr(V)
Figure 9.4
Jp = f( vr) characteristics for the silicon avalanche photodiode TIXL 56 as a function the E irradiance e.
128
functions of the photocurrent
amplification M. The incident radiant
$ power e is determined for the measure-
ment of the photocurrent amplification
M, by measuring, with an applied reverse
Vr voltage
= 100 V for Si avalanche
photodiodes, a photocurrent Ip = 1 nA.
Figure 9.4 shows, with the irradiance
Ee as a variable, the increasingly rounded
form of the function I p = f(Vj^) with increasing irradiance E e . As the slope of the function Ip = f(Vj^) decreases, so too
does the photocurrent amplification M.
The typical current amplification for
> silicon avalanche photodiodes is Mtyp
100 and for germanium avalanche photodiodes Mtyp >40. The photo-
M current amplification can be seen as a
function of the applied reverse voltage
Vr for the silicon avalanche photodiodes
TIXL 56 and TIXL 59 in Figure 9.5 and
for the germanium avalanche photodiodes TIXL 57 and TIXL 68 in Figure 9.6.
Figure 9.5
M = f(Vj^) characteristic for the silicon
avalanche photodiodes TIXL 56 and TIXL 59. (The case temperature is tQ =
25°C)
The product of amplification and
i 101
I
bandwidth (M . Af) for a photodetector
1
70
I
measures the modulation frequency of
the incident radiation, for which the
40
photodetector still has an amplification
ofM= l.Thus:
20
fi = (M . Af)
10
(9.11) 7
The (M . Af) product of a photodetector is
4
stated for the working point, at which
it has its maximum amplification M.
2
For avalanche photodiodes, the amplifi-
M cation is measured with a modulated
laser radiation and the value of the amplification-bandwidth product (M . Af) is calculated in accordance with equation (9.1 1). The modulation frequency of the laser radiation is normally about 1 GHz. The wavelength of the laser radiation has
been selected at X= 632-8 nm (HeNe
laser) for the measurement of Si avalanche photodiodes and at X = 1-15 jUm for the measurement of Ge avalanche photodiodes. The product of amplification and bandwidth
75
100
Vp Reverse voltage
-
Figure 9.6
M = f( Vji) characteristic for the germanium
avalanche photodiodes TIXL 5 7 and
TIXL 68.
(M . Af) is typically 80 GHz for Si
avalanche photodiodes and typically
50 GHz for Ge avalanche photodiodes.
129
9.2 Spectral sensitivity of junction
photode tec tors
The spectral sensitivity of a junction photodetector depends on the spectral absorption of radiation falling on the junction. The absorbed radiation causes electrons to leave the valency band and climb to the conduction band. The fact,
that the electrical field of the space-charge zone separates the pairs of charge carriers produced, was described in Section 1.3.3.
The electrons diffuse into the N-region, the holes into the P-region. At the external
connections of a junction photocell, in
no-load operation the maximum
photovoltage and, in short-circuited
operation the maximum photocurrent,
can be measured. If an incident photon
has just the minimum energy to raise an electron from the valency band into the
conduction band, this kind of absorption is called "basic lattice" or "intrinsic"
absorption. Also, this minimum energy
value determines the limiting wavelength, on the long-wave side, for the absorbed radiation. For shorter radiation wavelengths, the absorption generally increases rapidly
in semiconductors. The steep transition between relatively low and relatively high
absorption is usually defined as the absorption edge, or for semiconductors as the basic lattice absorption edge. With photosemiconductors, there follows,
beyond the short-wave side of the absorption edge, the relatively narrow
wavelength range of maximum absorption.
If the limiting wavelength on the longwave side of the absorbed radiation, or the
wavelength associated with the absorption edge, is defined as that wavelength with
the half-value of the maximum spectral
photosensitivity, then the width of the "forbidden band" can be determined in simplified form in accordance with equations (1.3) and (1.5):
*G
(9.12)
In contrast to pure semiconductors, doped semiconductors also absorb a somewhat longer wavelength radiation. To raise an electron from the valency band to the energy level of the foreign atoms, a photon needs less excitation energy, as compared with the intrinsic absorption mode.
Corresponding to the smaller energy difference between the valency band and the dopant level, this absorption follows the basic lattice absorption, on the longwave side. This kind of absorption is called divergent or "extrinsic" absorption.
Beyond the long-wave side of the
absorption edge, in semiconductors, there follows an increasing transmissivity. This is associated with a decrease in absorption and in radiation sensitivity. Figure 9. shows, as an example, the spectra]
100 90
80
(%> 70 60 so 40 30 20 10
.
S" -- -- --
*~
1 1
1 1
1 1
1
1
1\
123 4 5 6
Germinium
\
\
%
i [\
l I V,
\
* ·">
j ~f\
N
10 11 12 13 14 IS 16
XGiro)
..
Figure 9. 7
Spectral transmission t(^), as a function
of the radiation wavelength \ of silicon
and germanium; the layer thickness (S) is
mm 10
in each case.
transmission of silicon and germanium. These materials can also be used as very effective filters against visible and very near
infra-red radiation.
For non-amplifying junction photodetectors.,
the maximum attainable absolute spectral
sensitivity for every wavelength is described by the relationship calculated from the theoretical spectral sensitivity s(X)jh- The
130
theoretical spectral sensitivity s(X) th corresponds to the absolute spectral sensitivity s(X) for a semiconductor photodetector of quantum yield Q(X) = 1. For this, equation (9.8), already stated in Section
9.1, applies:
WA m s(X)th = 1-24
10 6
1
According to this equation, the theoretical
spectral sensitivity s(X) tn is a linearly
increasing function, dependent on the
radiation wavelength. At the imaginary shortest wavelength Xq = 0, the theoretical spectral sensitivity s(X)tft = 0. From a certain longer radiation wavelength,
characteristic of the particular semi-
conductor material, Xj-, the semiconductor
becomes transparent to this radiation.
At this wavelength Xj-, the theoretical
^^ spectral sensitivity s(X)tn ,T is
= 0-
The absorption of the longest wavelength
radiation which is still possible, i.e., before the photosemiconductor shows
transmissivity to radiation, determines
the theoretical maximum attainable
m spectral sensitivity s(X) th,max
accordance with equation (9.8). Here,
the wavelength is defined as Xfj^maxFor non-amplifying semiconductor
photodetectors, all spectral sensitivity
curves consequently lie in the triangle
defined by:
s00th,o = at Xq =
V s(X)th,T =0at
s(X>th,max = max. at Xth >max-
The longest wavelength which is still possible for absorbed radiation, Xth,max
is approximately:
Xtn max = 1*12 /im for silicon and Xth max = 1"8 JUm for germanium.
With this data and the equation (9.8), the theoretical spectral sensitivity triangle can be constructed as shown
Figure 9.8 Theoretical spectral sensitivity triangle as envelope curve for all spectral sensitivity curves, a) for silicon photo-
detectors, b) for germanium photodetectors
in Figure 9.8 as an envelope enclosing all spectral sensitivity curves for nonamplifying silicon or germanium photo-
detectors.
Germanium photodetectors are nowadays
only used for specialised applications.
The lasers radiating on a wavelength in the
range between 1 /xm and 1 -6 lixn, for
YAG example
lasers on 1-06 /Lttn, are
spectrally well-matched to photodetectors germanium avalanche photodiode
photodetectors. Figure 9. 9 shows the
relative spectral sensitivity s(X) rei as a
function of the radiation wavelength X for the new obsolete germanium avalanche photodiode TIXL 68. The
modulation frequency of the incident
radiation is the variable. This photodiode
had an antireflection coating for X =
l;54/lm.
Germanium photodetectors are being superseded more and more by silicon
photodetectors.
The deliberate adjustment of the absolute
spectral sensitivity s(X) of silicon junction photodetectors is of fundamental
importance. The requirments for spectral
131
. 100
\>
s(X)rel<%)
i
60
\
1
\
f> 10 MHz 1
\
\
\
20
\\
f=100kHz\
/
\
0-4
0-6
0-8
10
12
1-4
1-6
1-8
I
X iiim)
·-
Figure 9. Relative spectral sensitivity s(~h) re i as a
function of the radiation wavelength X for the germanium avalanche photodiode TIXL 68, with the modulation frequency
of the incident radiation as a variable.
performance can differ according to the application. For selective radiometric measurements, a linearised, flat spectral sensitivity range is necessary, for photometric applications the spectral photopic sensitivity of the eye, V(X), has to be simulated and for general applications the theoretical
spectral sensitivity limits are desirable.
9.2.1
Schottky Barrier PIN Photodiodes
Within the spectral sensitivity range of nonamplifying junction photodetectors, the Schottky barrier PIN photodiodes have the most linear sensitivity curve between 600 and 900 nm. These photodiodes have a metal-semiconductor junction which is formed by a very thin gold film (of the
order of 15 nm) being evaporated onto N-doped silicon. If a voltage is applied to the Schottky barrier PIN photodiode, with the negative potential to the gold film and the positive to the N-doped silicon, a spacecharge-free region is formed over the whole junction. The charge carriers produced by absorption can diffuse at a relatively high speed through this intrinsic
layer. This is also accounts for the short rise and fall times of these components. Within the depletion region of about 1 jUm, the Schottky barrier PIN photodiodes have a considerably higher absorption coefficient than planar diffused silicon junction photodetectors. Since shorterwavelength radiation penetrates less deeply into silicon than longer-wavelength radiation, Schottky-barrier PIN photodiodes are considerably more sensitive than planar diffused silicon junction photodetectors in the shorter-wavelength absorption range. In the longer-wave absorption range from about 800 nm, the spectral sensitivity s(X) of the Schottky
junction PIN photodiodes is somewhat
less than that of the planar diffused silicon junction photodetectors. Here, the reflection
> of the gold film has an effect (p 30%).
Figure 9.10 shows the spectral sensitivity of Schottky barrier PIN photodiodes without an antireflection coating.
1 0-25
0-20
015
Schot ky barr
\
\ \
010
}
(
1
1
3-20 0'30 040 050 060
0-25 0-35
70 0-80 0'90 100 110
X (ym)
-
Figure 9.10 Absolute spectral sensitivity s(~h) as a
function of the radiation wavelength X for Schottky-barrier PIN photodiodes
without antireflection coating.
9.2.2 Planar diffused Si-photodiodes
Planar diffused silicon photodiodes have
very high quantum yields of about 70% to
132
90% in the range from 700 nm to 900 nm.
The absolute spectral sensitivity s(X) of
these silicon junction photodetectors in this range is already very close to the theoretical sensitivity limit. At shorter
radiation wavelengths the quantum yield Q(X) and the blue-violet sensitivity falls off. So-called "dead zones" on the device surface, which are caused by surface
recombination of diffusion atoms, prevent
a higher quantum yield and higher sensitivities. To raise the blue-violet
sensitivity, firstly the diffusion-induced
surface defects can be reduced by improved technological processes,
secondly the PN junction can be located
very near the device surface and thirdly antireflection coatings can be evaporated on the surface Figure 9.1 1 illustrates the spectral sensitivity s(X) of a normal silicon photodiode and that of a silicon photodiode
Figure 9.11 Absolute spectral sensitivity s(\j as a
function of the wavelength X Curve a:
Normal silicon photodiode. Curve b: Silicon photodiode with increased blue-
violet sensitivity
with increased blue-violet sensitivity.
s(XXA/w)
-0-4 0-3
0:2 0-1
--
--[-- -
a^ ^
^*-
·
X^
··
\
,* *
/ //
7 ti
7
/ T!
7)
/ >' t--
t/
t* f--
/
'/
/7 >--
f--
-s
'
^
^
\V b
wk\ 65 oc 25oc\
\\ \
v
S N
^v -
·,
S V io 5°t N
\
S<JC*>
-V \
\-- \
*
\
n ·fi
·7
C8
c9
0
--
^*-
^
1
Figure 9.12
Absolute spectral sensitivity s(\} as a function of radiation wavelength A for silicon photodiodes produced by the CDI process. Curve a: for a deep function. Curve b: for a shallow junction. The effect of temperature is also shown for both types.
133
9.2.3
Photodiodes by the CDI process
In the so-called CDI process (Collector
Diffusion Isolation), the junction is
even shallower (about 1-1-5 (Jm), as
compared with planar diffused photodiodes. The spectral sensitivity of these silicon junction photodiodes is thus displaced more into the shorter-wavelength region, so that these components are better matched to the sensitivity of the eye. As is shown by Figure 9.12, there are limits to this matching of sensitivity to the V(X) curve, since at the same time the absolute spectral sensitivity s(X) is reduced.
9.2.4
Effect of temperature on the spectral sensitivity of photodiodes
The spectral sensitivity s(X) can vary with increasing temperature. The sum of the energy of a photon and a phonon can give the energy difference needed for the transfer of an electron from the valency
band into the conduction band. The spectral absorption and the spectral sensitivity range can therefore shift toward
longer radiation wavelengths. Figure 9.12 shows the change in the spectral sensitivity of silicon junction photodetectors, caused by the effect of temperature.
9.2.5 Possibilities for shifting the spectral sensitivity
Because of the dependence of their
M amplification factor on the photo-
current, junction photodetectors with internal amplification have a somewhat different absolute spectral sensitivity s(X) from non-amplifying junction photodetectors. In Figure 9.13, the typical relative spectral sensitivity
s(X) rei is shown as a function of the
radiation wavelength X for TIL 63-67
silicon phototransistors.
For certain applications, e.g., for the evaluation of the radiation from a
1-2
Wwi
0-8 0-6 0-4 0-2
3
0-4
0-5
0-6
0*1
0-8
0-9
t-0
11
12
Figure 9.13
Typical relative spectral sensitivity as a function of the radiation wavelength X for TIL 63-
67 silicon phototransistors.
134
GaAs luminescence diode in the IR range, it is necessary, to shift the maximum of
the sensitivity curve towards longer wavelengths. This is done firstly by
100 *(X>rel
SO
M tchedlR
ntlthrily
N \
NA
/
H UnimiiEtttdIR
senatW <y
~"
t - IRndittkmof CiAidiodn
/it
400
500
600
700
tOO
900
1000
Figure 9.14
Relative spectral sensitivity s(X) rei as a
function of the wavelength for Si phototransistors. For the evaluation of the IR radiation from Si-doped GaAs luminescence diodes. Si phototransistors with matched IR sensitivity must be used
suitable adjustment of the junction depth and also by application of an antireflection
« coating for X 900 nm to the component.
Figure 9.14 shows the effect of these measures
Apart from the above-mentioned technological possibilities, the spectral sensitivity of a photodetector can also be modified by externally attached optical filters. The makers of radiation measuring instruments either linearise the spectral sensitivity or they simulate the V(X) curve, by placing a very expensive optical difference filter in front of a Schottky barrier PIN photodiode (see Section 6.5). Semiconductor manufacturers have also produced very lowpriced silicon photodiodes which are reasonably well matched to the V(X) curve, by fitting a green filter in the case of the silicon photodiode. Here too, the absolute spectral sensitivity s(X) is reduced through the low transmission of IR by the filter. As an example of this, Figure 9.15 shows
the relative spectral sensitivity s(X) rei as a
function of the radiation wavelength X
CM
ii
\i
XOim)
»-
Figure 9.15
Relative spectral sensitivity s(~K) rei as a function of the radiation wavelength X of the silicon photodiode Type TIL 77, produced with an attached green filter, in comparison with the photopic sensitivity curve of the eye, V(\)
135
for the now obsolete silicon photodiode TIL 77, made with an attached green filter,
in comparison with the photopic sensitivity of the eye V(X).
The possibilities described of affecting the
absolute spectral sensitivity s(X) of a silicon junction photodetector within the theoretical sensitivity triangle, can be summarised as follows:
1
Variation of the absorption performance by
the junction depth.
Reduction of surface recombination by technological methods.
Improvement of the absorption performance by antireflection coatings on the component surface.
Fitting of external optical filters.
M Adjustment of the amplification factor
of junction photodetectors with internal amplification.
The quantum yield for monochromatic radiation Q(X) is a direct measure of the absolute spectral sensitivity s(X). The
absolute spectral sensitivity s(X) can
therefore be derived from the quantum yield for monochromatic radiation Q(X)
in accordance with equation (9.4):
Q(X)
!p,X · h · co
*e X.X.e ;
(9.4)
If the value of the constants from equation
(9.5) and the wavelength X of the
monochromatic radiation used are inserted in this equation, there then remains, to
define the absolute spectral sensitivity s(X) of a non-amplifying junction detector which is the ratio of the
photocurrent produced I p X to the incident
$ monochromatic radiant power e \. The
relationship which applies for this
_ ^X
s(X)
$e,X
(9.6)
has already been stated as equation (9.6).
For the calculation of the.absolute spectral sensitivity s(X) with the aid of the quantum yield for monochromatic radiation, the equations (9.4) and (9.5) have to be combined with equation (9.6).
We obtain:
X.< s(A) = Q(X) .
h. c
1
= Q(X).X.
1-24 W.Mm
(9.13)
9.3
Evaluation of radiation by non-amplifying junction photodetectors for monochromatic radiation sources
A non-amplifying junction' photo-
detector evaluates an incident mono-
chromatic radiation in accordance with its absolute spectral sensitivity. This is defined, for non-amplifying junction photodetectors, by the equation (9.6), which has already been defined.
The evaluation of radiation by the eye, described in Chapter 5 , is equivalent
in principle to that with non-amplifying junction photodetectors, if the relevant spectral sensitivity is taken into account. In order to understand the relationships which follow, the fundamentals from Chapter 5 will first be mentioned once more.
According to equation (5.103), the absolute spectral sensitivity s(X) corresponds to the equivalent absolute photopic sensitivity of the eye K(X):
s(X) = K(X) = K(X) max . V(X)
(5.103)
136
The absolute sensitivity of the eye, K(X)
is defined by equation (5.27) as the
quotient of the output value X V)X X divided by the input value e>X:
K(X) = *e,X
(5.27)
Here, the output Xv \ is a corresponding
photometric value for monochromatic
X light and the input, e>X is a spectral
radiometric value equivalent to the input
value.
Similarly to equation (5.27), the absolute spectral sensitivity s(X) of a nonamplifying junction photodetector corresponds to the ratio of the output
value Y\to the input value Xe X:
YX
s(X) =
Xe ,X
(9.14)
This relationship is generally valid for nonamplifying junction photodetectors, since
their output value Y\ is proportional to the input value X e \ . In accordance with equation (9.6), the output value Y\
corresponds to the photo-short-circuit
current I P) \ of the junction photodetector,
operated as a photocell, for the incident monochromatic radiation of wavelength
X. The sensitivity for junction photo-
detectors is also generally defined as the
photocurrent sensitivity. To avoid
misunderstandings, it should be mentioned, that an ideal non-amplifying junction photodetector shows the same sensitivity in both the photocell and photodiode modes.
The input value Xe>X in the denominator
of equation (9.14) is either the
monochromatic radiant power 3>e X
falling on the photodetector or the
monochromatic radiant power ^X/A
related to a defined area A. In accordance with equation (2.40), the
monochromatic irradiance E CjX is tnus
obtained:
$e,X E e ,X
(2.40)
To calculate the absolute spectral sensitivity
$ s(X), the values I P)\, e> X or E e ,X are t0
be inserted in the general equation (9.14).
Either the known equation (9.6) or
the following relationship is obtained:
s(X)
Ip,X
E e ,X
(9.15)
The relative spectral sensitivity s(X) rei is defined by the ratio of the absolute spectral
sensitivity s(X) at the wavelength X to the
absolute spectral sensitivity s(X) at the
reference wavelength Xq :
s(X)
s(X) re l =
s(X)
(9.16)
^ tne maximum wavelength Xmax with
maximum spectral sensitivity s(X) max is
inserted for Xq in this expression, the equation (5.104), stated in Section 5.6, is
obtained:
s(X)
s(X) re l =
s(X) r
(5.104)
The absolute spectral sensitivity s(X) can be calculated by rearrangement of the equations (9.16) and (5.104):
s(X) = s(X) rei . s(X)
(9.17)
s(X) = s(X) rel s(X) max
(9.18)
The equation (9.18) corresponds to equation
(5.103) for the analogous absolute spectral sensitivity of the eye, K(X).
9.4
Evaluation of radiation by non-amplifying junction photodetectors for chromatic radiation (mixed radiation)
A non-amplifying junction photodetector
137
evaluates incident radiation in accordance with its absolute sensitivity s. In the case of a non-amplifying junction photodetector, this corresponds, in analogy to equation (5.105), to the photometric
K radiation equivalent of radiation source:
s = K = K(X) max .V
(5.105)
The photometric radiation equivalent K for
chromatic radiation is determined by the equation (5.40):
X2
K(X) max .JXe,X.V(X).dX
Xl
K
J'Xe ,X<iX
(5.40)
In this equation, the sensitivity values for the
human eye can be replaced by the sensitivity
values for non-amplifying junction photodetectors. Thus:
X2 //Xe,X-V(X).dX
--Xl
V Srei = =
/Xe ,X-dX
o
(5.102)
s(X) = K(X)
(5.103)
v s(*)rel = s=K
(5.104) (5.105)
sMmax = K(^)max
(9.19)
If the expressions from equations (5.104), (5.105) and (9.19) are inserted in equation (5.40), then the basic equation, equivalent to the latter, for the calculation of the sensitivity s of a non-amplifying junction photodetector is obtained:
X2
sMmax.fXe,X-s(*)rel-<lX Xl
s= oo
JXe.X · d^
(9.20)
If the expressions from equations (5.105), (5.103) and (5.37) are inserted in equation (5.40), then one obtains, for the sensitivity
X2
/Xe,X · s(X) . dX
Xl
/xe>x.dX
(9.21)
Furthermore, by insertion of equations (5.102), (5.105) and (9.19) in equation (5.40), one obtains the relationship
s - s(X) max . sre i
(9.22)
This expression is the analogous form of the equations (5.43) and (5.105) for the calculation of the photometric radiation equivalent K.
In the equations (5.40), (9.20) and (9.21),
the value Xe \ occurs both in the numerator
and the demoninator. It can therefore be replaced by its relative value, which
characterises the radiation function S\ of the radiation falling on the photodetector (see equation 2.66). The equation (9.20)
then reads:
X2 s(X)max-/SX.s(X) re i.dX
Xl
J'SX-dX
(9.23)
At the same time, equation (9.21) becomes:
X2
J'SX · d(X) . dX
Xl
J'SX-dX
(9.24)
In accordance with this expression, the sensitivity s of a non-amplifying junction photodetector is defined as the rate of
138
.
the output value, weighted both with the absolute spectral sensitivity distribution s(X) = f(X) of the photodetector and with
the radiation function S\ of the incident
radiation:
X2
Y = /S\ · s(X) . dX
Xl
(9.25)
to the incident total radiation as the input value
X = J'Sx · dX
(9.26)
The general equation for the calculation of the sensitivity s is thus given by (see
Section 5.6):
X
(5-101)
If the spectral radiant flux X <t>e) is inserted
X in equation (9.21) for the value e \, then
integration over the wavelength range
which is of interest gives the photocurrent
Ip in the numerator and the incident radiant power 4>e in the denominator:
*e
(9.27)
If the spectral irradiance Ee>\ is inserted
X in equation (9.21) for the value e X,
then integration over the wavelength
range which is of interest gives the
photocurrent I p in the numerator and the
irradiance Ee in the denominator:
Ee
(9.28)
In contrast to equation (9.27), where the
sensitivity s is related to the radiant
$ power e , here it is related to the irradiance
As an example, Figure 9.16 shows the
V(-g = to 50 V lE-0
"tU-25«C lp<mA)
0*
05
M
0-3
0-2
01
02030405060 70 80
^(W/cm 2 )
-
L
Figure 9.16
The function Ip = f(Ee) for the collectorbase Junction of the phototransistor TIL 81.
typical linear characteristic of the function Ip = f(Ee) for silicon photodiodes or
photocells. The characteristic is applicable
to the collector-base junction of the
phototransistor TIL 81. In phtoodiode operation, the applied reverse voltage may have values up to 50 V. In photocell
operation, the photocurrent Ip represents
the short-circuit current.
9.5
Evaluation of radiation by amplifying
junction photodetectois
An amplifying junction photo-detector
evaluates incident radiation in accordance with its increased (amplified) absolute sensitivity for this incident radiation, s\f
The absolute sensitivity sm of an amplifying photodetector is given by the product
of the non-amplified absolute basic sensitivity s for the incident radiation and the amplification factor M:
sm = s .M
(9.29)
For phototransistors, the amplification
M factor corresponds to the DC current
139
gain B. Further, it can be assumed as an approximation, that the non-amplified absolute basic sensitivity s^g corresponds to the absolute sensitivity s^g of the collector-base junction acting as a
photocell:
SM = SCB · B
(9.30)
The absolute sensitivity sm of amplifying
photodetectors is generally defined,
corresponding to equation (5.101), as the
ratio of the output value Y to the input
X value P .
SM
XP
(9.31)
is given by the differential coefficient
of the function Y = f(X e):
dY
M
dX P
(9.33)
But if the output value Y is proportional X to the input value e , then the absolute differential sensitivity sj .can be made
equal to the absolute sensitivity sj^. Thus
in the linear case, the radiation evaluation of amplifying junction photodetectors can be carried out in accordance with Sections 9.3 and 9.4 for non-amplifying junction photodetectors, i.e.:
M Sd = s = s
(9.34)
For phototransistors, the output value Y
is the collector current 1^. Other definitions, such as light current or photocuirent, for example, are incorrect and lead to
X confusion. As the input value e , a
radiometric value should be selected, since phototransistors without spectral correction are sensitive both in the visible
and the IR range. Corresponding to
Section 9.6, the irradiance E e is to be inserted
X as the radiometric input value e :
ic
SM="
(9.32)
For amplifying junction photodetectors,
M the amplification factor depends on the
X input value e . In phototransistors, in
DC particular, the
current gain B is a
function of the photocurrent Iq-q of the
collector-base photocell. Here, photo-
current is defined as the control current
occurring without the amplification
effect, due only to charge-carrier
generation, which would correspond, in
the normal transistor, to the base current.
The absolute sensitivity s^ for
amplifying junction photodetectors is
therefore also dependent on the input
X value e . For non-linear amplifying
junction photodetectors, the absolute
differential sensitivity sj is stated. This
Furthermore, with a non-linear relation-
ship between the output value Y and the
X input value e , the absolute sensitivity
sm of amplifying photodetectors can also
be stated by a finite interval between
Xe2
-Xe
iandY 2
-Y .The 1
)
corresponding differential quotient then
replaces the differential quotient in the
equation (9.33). This statement of
sensitivity is defined as the interval sensitivity
Y2-Y1 AY
X X - e ,2
e> i
"axT
(9.35)
For phototransistors, the function
Y = f(X e) is shown in a graph as the
characteristic curve 1^ = f(E e) for ar applied collector-emitter voltage
VqE = 5 V and a case temperature of
tQ = 25°C. Figure 9.17 shows the typical
function 1q = f(Ee) for the well-known phototransistor TIL 81.
Generally, the function 1^ = f(Ee) is stated
> from an irradiance Ee 1 mW/cm for a
given radiation function S\.
< In practice, irradiances Ee 1 mW/cm
are also used. Here the user has to know
either
the
typical
function
BIy p
=
f(E g)
for the radiation function S\ in question,
140
r l B "0 5 0f
'C itA
VCE = 5V
25
20
/
/
10
/ ·*
1,
5
r
.V V = 1
10
12
14
16
18 20
Ee (mW/cm 3 )
..
were measured. From equation (9.36), the
DC amplification is
9 mA
B = - -=300 40 fiA.
If the DC current gain B is determined
for different collector currents 1^, the
function B = f(Ic) can be plotted in a
graph as a characteristic curve. However,
DC it is better to determine the relative
current gain B rei as a function of the
collector current 1^, since this characteristic
curve can be utilised for the whole corresponding family of phototransistors.
We have:
Figure 9.17
Collector current Iq as a function of the
E irradiance e for the photo-transistor TIL
81. The collector-emitter voltage Vqj? is the variable. The irradiance is generated with
A the standard light radiation.
or Btyn = f(Ic) without taking account of a radiation function SX-
The absolute sensitivity sjyj of the phototransistor can be calculated in accordance with the equations (9.29) to (9.32). For a point on the characteristic B = f(Ic)> the current gain B of a phototransistor with an external base connection is obtained by forming the ratio of the collector current 1^, measured with a given
irradiance in phototransistor operation, to the photo current I^g, measured with
the same irradiance for the collector-base photocell working in the short-circuited mode.
B= ICB
(9.36)
As an example, the current gain B of the
phototransistor TIL 81 will be determined.
With the irradiance Ee = 2-5 mW/cm2 for
standard light A, a collector current 1^; =
mA 9
and a photocurrent Iqq = 30 /MA of
the short-circuited collector-base photocell
B
Brel =
Br
(9.37)
Figure 9.18 shows the relative DC
current gain B rel as a function of the
collector current 1^ for the photo-
transistor TIL 81. The maximum DC
current gain B rel,max nas been standardised
to
the
value
B rel,'max
=
100% -
For
°Pt0 "
DC couplers, the relative
current gain B rei
of the phototransistor is measured as a
function of the collector current Ic(on)
when the GaAs light-emitting diode is
switched on. A defined direct current Ip
flows through the GaAs diode. Optocouplers
are measured at an ambient temperature
of tu = 25°C. In Figure 9.19, the
standardised DC current gain B re l
is shown as a function of the collector current Ic(on) for tne optocoupler TIL 112.
The measurement of the photocurrent
sensitivity of phototransistors is
usually carried out at an ambient
temperature tij = 25°C for plastic
q
transistors and a case temperature tQ= 25 C
for metal-encapsulated transistors. The
Planck radiation used for this purpose,
standard light A, causes heating of the wafer through the longer-wavelength IR
raidation. This effect mainly occurs with
phototransistors with high thermal
resistances. Therefore the measurement
141
B,el(«) 100-
75-
V «-'rr,
S K^
>*
-
»«i - tac)
VCE » 5V;TC -2S°C
»nL 81 -
47 A
i 5
mwJ/?cmy
f°'
s'»" d »' d
'W«
A
25-
0"
10>
1
10*
101 *
'CO-A)-
F#ure9.7S
DC Relative
current gain B re} = flltf of the silicon phototransistor TIL 81
time is kept as short as possible, so that the measured result is as accurate as possible Figure 9.20 shows the dependence of the collector current Iq on the ambient temperature ty for the phtotransistor LS 400.
For photo transistors with external base
connection, the DC current gain B can
be measured, instead of with the photocurrent I^jj of the collector-base diode, alternatively with the base current lg from an external current source. During this measurement the radiationsensitive surface of the phototransistor is covered. With this method, the dependence of the collector current 1q on the applied
collector-emitter voltage V£E> .that is, the function Iq = f(VcE)> can a' s0 ^ e measured on the characteristic recorder.
In this case, the base current I]j is the
variable. If the sensitivity s of the collector-
base photodiode is known for the
particular base current, (see Figure 9.16),
the parameter Ig can be replaced by the irradiance Ee. As a specific example of this, Figure 9.21 shows the typical function Iq = f(Vc£) of the silicon phototransistors TIL 602, TIL 610 and TIL 614 with the irradiance E e as parameter.
In Section 9.2, it was explained that the
absolute spectral sensitivity of an
amplifying junction photodetector is also
dependent on its particular amplification
factor M. If, with monochromatic radiation,
the output value Y\ of an amplifying
photodetector behaves proportionally to
X the spectral input value e \ and if the
M\ amplification factor
or the DC current
gain B\ were independent of the output
value Y\and of the input value Xe>X
142
--I T 1
VCE -5V
I"
lF-0
tU"
1-2
10
0-8 ^"
0-6
0-4
0-2
·1
2 0-4
1
2
4
Standarc tscd tu Oat
m IF* 1 A.
1
10 20 40
100
'CConXmA)
-
Figure 9.19
DC Standardised
current gain Bre[ as
a Junction of the collector current Itfon)
with the luminescent diode conducting,
for the optocoupler TIL 112. The relative
DC current gain has been standardised to mA the value Bre\ = 1 for a current Ip= 1
flowing through the GaAs diode.
(see equation 9.14), then the relative spectral sensitivity s(X)M,rel of amplifying junction photodetectors corresponds to the basic relative spectral sensitivity s(X) rel ( see equation 5.104):
sMrel = s(X)
s(X>M,rel
s(X) max
s(X)m s(X)M,n
(9.38)
If equation (9.14) is inserted in equation
(9.38), we obtain:
YX YX,max
sMM,rel
Xe,X xe,X,max
YX · Xe,X,max
Xe,X · YX,max
(9.39)
If, on the other hand, the amplification
factor M\ or theDC current gain B\ depends on the input value Xe>X, then the
relative spectral sensitivity s(X)M,rel can be
100
t mA
>C(mA)
10
5s
max.
^*-l lyi>. f*
"e:~
1
0-5
~^ '*-- --min.-
>
01
5
50
-25
--
«25
50
75 100°C 1 25 <U(°C)
_
Figure 9.20
Function Iq = f(ttjj for the phototransistor
LS 400. The applied collector-emitter voltage is Vq£ = 5 V.
Figure 9.21
M Typical function Iq = f( Vqe) °f t^le s con
phototransistors TIL 602, 606, 610, 614;
E (The variable is the irradiance e)
143
measured by relation to equal output values
of YX and Y^max-
Y\ = Y X) max
(9.40)
The relative spectral sensitivity s(X)M )lei of amplifying junction photodetecto.rs can be
defined for equal output values YX and
YX,max as tne ratl ° °f trie spectral
X radiometric input value e X,max at tne
radiation wavelength Xmax wltn tne maximum absolute spectral sensitivity
s(A)m max to the spectral radiometric
X input value e> X at the radiation wavelength
X with the absolute spectral sensitivity
s(X)m- By insertion of equation (9.40) in
equation (9.39), one obtains:
*e,X, max
s(^)M,rel :
Xe ,X
(9.41)
Its dimension is:
A mW LlA ' W mW /Mf
(9.42)
Similarly, the absolute spectral photosensitivity s(X) is determined by the equation (9.6):
lp,X s(X) =
*e,X
(9.6)
Its dimension is:
[s(X)]
A
mA
W(jum
l )
mW(/im
! )
MA
L
/iW(/im )
(9.43)
9.6
Area-dependent sensitivity values for junction photodetectors
The sensitivity units s, s(X) and s\[ are defined as the ratio of the output current I to the incident radiant power 4>e or to the irradiance E e . Here, a distinction is made between large-area and small area
photodetectors.
Large-area junction photodetectors are mostly non-amplifying radiation detectors such as silicon solar cells and Schottky
barrier PIN photodiodes. The
manufacturer of such radiation detectors can give exact data on their photosensitive area Ajr. If the receiving area is known, the incident radiant power can be calculated by the equation already defined (2.41).
*e = Ee - AE
(2.41)
For large-area non-amplifying photodetectors, the absolute photo-sensitivity is determined by the equation (9.27):
(9.27)
These sensitivities are defined accordingly as the specific photocurrent sensitivity s or the specific spectral photocurrent sensitivity
s(X).
Small-area junction photodetectors can be amplifying radiation detectors, for example, such as silicon photodiodes, avalanche photodiodes and phototransistors. For these radiation detectors, the photosensitive area of their wafer, or their magnified photosensitive wafer area, due to built-in lenses or a plastic case which transmits radiation, is not exactly known.
The radiant power falling on the photodetector can be calculated only with an excessive error by using equation (2.41).
Therefore, the radiation falling
perpendicularly on to a specified measurement plane is stated. The photodetector device under test (e.g., a silicon photodiode) is located at the centre of the measurement plane in the normal
direction.
Generally, the area of the measurement
A plane is specified at meas = 1 cm .
144
The incident radiant power is related
to this area, so that instead of the radiant
power e <J> , the irradiance E e is to be stated in the unit W/cm . For small-area
non-amplifying junction photodetectors, the absolute photosensitivity s is determined by equation (9.28):
-- s=
Ee
(9.28)
Its dimension is:
A
mA
[s] = Wcm 2 mWcm 2
jUA
/uWcm 2
(9.44)
Similarly, the absolute spectral photosensitivity s(A) is given by equation (9.15):
Ip,X s(A) =
Ee,A
(9.15)
Its dimension is:
[s(A)J
A
Wcm
2 (pm
l )
mA
mWcm 2 (jLtm
*)
MA
2
J
/LlWcm (jUm )
(9.45)
For small-area smplifying junction photo-
detectors, the absolute sensitivity sm is determined by the equation (9.32):
IM SM="
(9.32)
Its dimension corresponds to that of equation (9.42). Similarly, the absolute differential sensitivity sj is obtained from equation (9.33):
dY
M:
dXe
dIM
dE fi
(9.46)
Its dimensions corresponds to those of equation (9.44).
The interval sensitivity sj is given by the
equation (9.35):
= Y2 ~ Yl
= *M,2 ~ iM.l
s-
xe,2 - xe,l E e,2 - E e,l
(9.47)
The equation (9.44) also applies here for
its dimensions:
The relative spectral sensitivity s(A)m rei is determined by the equation (9.41):
_ xe,A,max _ Ee,A,max
s(^)M,rel
*e,A
Ee,A
(9.48)
9.7
Actinic value of IR luminescence radiation from GaAs diodes for silicon junction photodiodes
The basic principles of actinic value were explained in Section 5.6. The sensitivity values needed for silicon photodetectors have been described in this chapter. The actinic value a(Z) of any given radiation spectrum Z for silicon junction
photodetectors is the ratio of the
sensitivity s(Z) when acted upon by this radiation to the sensitivity s(PA) when acted upon by the Planck standard light
A A radiation. The standard light is always
used as the reference radiation, since this
form of radiation has become generally
established for the sensitivity evaluation of silicon junction photodetectors.
For non-linear amplifying photodetectors, it will be assumed, for simplicity, that the actinic value will either be measured within a linear range or with equal output values, or will be calculated in accordance with equation (5.1 17) with Y(Z) = Y(N).
The actinic value of IR luminescence radiation from GaAs diodes for silicon junction photodetectors can be determined from equation (5.121):
145
s(Z)
a(Z) =
s(N)
_ J'S(Z)X . s(X)rei dX . /S(N)\ . dX /S(Z)X.dX./S(N)X.s(X) re i.dX
(5.121)
The IR luminescence radiation from GaAs diodes is denoted by LIR. Its radiation
function is therefore stated as follows:
S(Z)X =- S(LIR>X
(9.49)
The reference radiation, the Planck radiation '.standard light A", is denoted by PA, as in Section 5.6.2. Its radiation function is identified either by its distribution
temperature Tv = 2856 K or by PA.
S(N)X =* SX.2856 K = S(PA)X
(9.50)
The relevant sensitivities of a silicon junction photodetector are characterised in accordance with these radiation functions:
s(Z)
s(LIR)
(9.51)
~ s(Z)X
s(LIR)X
s(Z)X,rel =* s(LIR)X,rel
s(N) =* s(PA)
s(N)X =* s(PA)X
(9.52) (9.53) (9-54) (9.55)
s(N)X,rel =* s(PA)X,rel
(9.56)
The actinic value of IR luminescence radiation from GaAs diodes for silicon junction photodetectors will be calculated in accordance with equation (5.121) with
relative radiometric input values. Instead of a(Z), it is defined as ae(LIR):
a(Z)
ae(LIR)
(9.57)
Equations (9.49) to (9.57), inserted in equation (5.121), give a more convenient representation:
ae(LIR)
s(LIR) rei s(PA) rei
'"
·Wiel S ^0-8-
0-5-
(M-
0-30-2-
01-
3
d -* |s 4
\
X>)re IKX max
·^ <^^ ^1 £1
s$
iI
'/ /
/
i
h S(LIR
}
/
/
* n
I
('
i
/
f
\
//
^' 1,
^^ ,<fk \}A
-- -- .
r- --
<
I
I
0-5
6
0-7
1k'
\
\
Rhj
I
8
0-9
S\,28 S6K" S(PA> .
ii
12
Figure 9.22
Radiation functions S(PA)\ofthe standard light A, S(LIRj\ofIR luminescence radiation from the GaAs diode TIL 31 and the relative spectral sensitivity s(ty rei, s(PA)\ rei and s(LIR)\ rei of a silicon photocell as functions of the
wavelength X
146
--
/S(LIR)\ . s(X) rei · dX . /S(PA)\ dX ~ J'S(LIR)\ . dX . J'S(PA)\ . s(X) rei . dX
(9.58)
Integration of this expression is carried out graphically. For this, Figure 9.22 shows the radiation functions S(PA)\ of the standard
light A and S(LIR)\of the IR luminescence
radiation for the GaAs diode TIL 31, and
also the relative spectral sensitivity
s(X) rei of a silicon photocell. The product of the ordinates S(PA)\and s(X)rel
gives relative spectral sensitivity s(PA)X,rel
for the standard light A. The product of the ordiantes S(LIR)\and s(X) rei gives the
relative spectral sensitivity s(LIR)X,rel
for the IR luminescence radiation of the GaAs diode TIL 31. The ratio of the measured area under the curve s(PA)X rel t0 tne measured area under the surve s(PA)X gives the relative sensitivity s(PA) rei. To give a clearer abscissa scale, the radiation function S(PA)\ is only shown
up to a wavelength X= 1-2 fJm. The plani-
metric measurement of the area under the curve S(PA)\ was carried out up to a
wavelength X = 4 /Urn. This approximation is permissible, since up to X = 4 /£n, covers about 94% of the total radiation of standard
A light and also since, in practical measure-
ments, longer-wavelength IR radiation is naturally absorbed by the glass bulb of
the radiation source.
The ratio of the measured area under the curve s(LIR)\,rel t0 tne measured area under the curve S(LIR)X gives the relative
sensitivity s(LIR) rel-
The actinic value ae(LIR) of luminescence
radiation from the GaAs diode TIL 31 for
a silicon photocell is, in accordance with equation (9.58), the ratio of the graphically determined relative sensitivity s(LIR) rei to the graphically determined relative sensitivity s(PA) rei. Evaluation of the curves in Figure 9.22 gives the following
values:
-- -- As(PA)Xrel 88 cm
s(PA) rei = AS(PA)X = ,3,35g c__mi
= 0-263
(9.59)
--As(LIR)X,rel _ 16 "5 cm
s(LIR) rel = AS(LIR)X
i1a8 cm2
= 0-917
(9.60)
-- L aeV(LIR)
= s(LIR) rel s(PA) rel
=
0-917
~
0-263
=
4_9_ (9 .6 i)
The actinic value of IR luminescence radiation from the GaAs diode TIXL 16 for
the photocell with a relative spectral
sensitivity as shown in Figure 9.22 was also determined by measurement. The irradiance falling on the photocell was
in one case 5 mW/cm and in another case 20 mW/cm2 for both of the two radiations PA and LIR. The photocurrent,
Ip, measured with the photocell shortcircuited, is shown in Table 9.1.
The actinic value ae(LIR) is:
for 5 mW/cm2
Type of radiation Irradiance E e
Photocurrent Ip
PA
5 mW/cnT
20 mW/cm2
3-5 mA
13-7 mA
LIR
5 mW/cmz
20 mW/cm2
9-27 mA
37-08 mA
Table 9.1 Measured
photocurrents
at
various
irrcdiances
and
with
different
types
,
_,.
.
of radiation
, for
t,.he
calculation of actinic value a e (LIR)
147
9-27 mA
mA ae(LIR) = 3-5
= 2-65
(9.62)
for 20 mW/cm2 :
37-08 mA
mA ae(LIR) = 13-7
= 2-7
(9.63)
The different actinic values a e (LIR) arise through measurement errors, which are primarily due to the deviation of the radiation function S\ of the filament lamp used from that of an ideal Planck radiator in the IR range, despite having the same colour temperature. The actinic value of IR luminescence radiation from GaAs diodes varies for Si phototransitors. The spectral sensitivity of a phototransistor also depends on the
DC current gain B (see Section 9.5).
Figure 9.23 again shows, as for Figure 9.22, the radiation functions S(PA)\
and S(LIR)Xfor the GaAs diode TIXL
26 and also the relative spectral sensitivities s(A) re i and s(PA)\ rei. In contrast to Figure 9.22, almost the
whole radiation function S(PA)X is shown. The other functions are compressed into the first third of the graph. The function s(LIR)X rel nas therefore not been shown. It corresponds approximately with the function s(LIR)\ rel ln Figure 9.22.
The actinic value of IR luminescence
radiation from GaAs diodes being
detected by phototransistors can be
determined, for example, if the output
variables have equal values. Measurement
in this way is not always possible, so that
in these cases the output variable is to be
measured in each case with equal values of
the input variable. The relative DC current
gain B rei of phototransistors is given for
the collector currents ]q (p\) for the
radiation PA and Iq (JJR) f° r tne
radiation LIR, measured with equal
irradiance values in each case, by the
characteristic curve B rei = f(Ic) (see
Figure 9.18). The relative DC current gain
for the incident radiation LIR is designated
B(LIR)rel anc* f° r tne incident radiation PA
with B(pa) rel- The quotient Kg of the
DC relative
current gain values for B(PA)rel
1
CM
0-9 -
kX
0-7--
"Ti
/
n
s
0-6 -
/, .
0-5 0-4--
/ A/
03-
/V
1\
0-2-
// 0-1-
y A 1
A)X,«1
0) 4's *)rel ~>Q
)ma>
|
!
«
!
i
ft
\1
/
SK28S6K = S(PAh
0* 0-3 0-4 0-5
0-7 0-8 0-9 ]
11 12
1-5
2
2-5
i
Figure 9.23
Radiation function S(PAj\of standard light A, S(LIRJ\ofIR luminescence radiation from the GaAs diode TIXL 26 and the relative spectral sensitivity s(\ re\ and s(PAj\ rei of a silicon phototransistor as functions of the wavelength X
148
and B(LiR) rei is the correction factor for the collector current Ic(LIR) :
kb B(PA)rel
B(LIR)rel
(9.64)
The product of the correction factor Kg and
the collector current Ic(LIR) § ives tne corrected collector current Ic(LIR)Kor«
which would appear directly with the DC
current gain B(PA), if this value were
constant for all types of radiations.
IC(LIR)Kor = Kb · Ic(LIR)
(9.65)
With this method, the non-linearity of
the DC current gain B of a phototransistor
can be eliminated from the actinic value calculation in many practical cases. But here, the conditions mentioned must be satisfied, i.e., the function B re i = f(Ic)
must be known and furthermore the measured irradiances E e (PA) for the PA radiation and E e(LIR) for the LIR
radiation must have exactly the same
values without errors:
Ee(PA) = Ee(LIR)
(9.66)
Thus the actinic value of IR luminescence radiation from GaAs diodes detected by silicon phototransistors is determined by the ratio of the corrected collector current
kXLIR)C or R t0 the collector current
IC(PA).
The sensitivities s(PA) and s(LIR) of the LS600 phototransistors were measured
for both PA and LIR radiation with an rrradiance Ee = 5 mW/cm .
In accordance with the statement on the
device data sheet, the applied collector-
emitter voltage must be Vce = 5 V. In
Table 9.2, the measured collector
currents
Ic(PA)
and
kXLIR)
in
each
case >
with the associated values of the relative
DC current gain B rei, are listed. The values of the relative DC current gain
B(PA)rel and R B( Li )rel can be seen from Figure 9.18. The correction factor Kb was
calculated from equation (9.64), the corrected
corrected collector current Ic(LIR)Corr from equation (9.65) and the actinic value
ae(LIR) from equation (9.67). The mean actinic value a e(LIR) of IR luminescence radiation from the GaAs diodes listed for
the four LS 600 phototransistors is:
2-2 + 2-1 + 2-16 + 2-47 ae (LIR) t y P :
= 2-23
(9.68)
The typical correction factor Kb for the
four phototransistors is calculated as follows:
0-869 + 0-874 + 0-8715 + 0-854
KB,typ "
0-867
(9.69)
s(LIR)corr ae(LIR) =
s(PA)
^(LIRX^orr kXPA)
(9.67)
The actinic value of selective IR luminescence
radiation with a maximum wavelength
(Xmax) of 930 nm was measured for four
LS600 silicon phototransistors. The IR
luminescence radiation LIR with the
maximum wavelength Xmax = 930 nm
relates to the following GaAs diodes TIL 23, TIL 24, TIXL 12, TIXL 13, TIXL 14, TIXL 15, TIXL 16 and TIXL 26.
If no values are stated by the manufacturer for ae(LIR) ty p and Kg^yp, then the values obtained from equations (9.68) and
(9.69) can be used for rough calculations when using IR luminescence radiation from Si-doped GaAs diodes with silicon
transistors. When doing this, a rather large tolerance for the Kg value must be taken into
account. With the typical value for a
ae(LIR) ty p and the typical correction
factor Kg ;typ! the photo current
sensitivity' s(PA) for the standard light A, stated in the date sheet, can be converted
149
Phototransistor LS 600
I£H£1 = s(PA) . 5 mW/cm2 mA
B(PA)rel
0-26 63
0-24 62-5
0-21 61
0-152 58-5
-- mA ]C(LIR) t=
s(LIR)
.
5
mW/cm ,2
0-66
0-58
0-52
0-44
B(LIR)rel
%
kb
72-5 0-869
71-5 0-874
70
68-5
0-8715
0-854
! C(LIR)Corr
mA
0-573
0-506
0-453
0-376
Ae(LIR)
2-2
2-1
2-16
2-47
Table 9.2
Calculation of the actinic value for silicon phototransistors
to the sensitivity s(LIR) for the IR radiation from Si-doped GaAs diodes.
The equations needed are obtained by rearrangement of the formulae (9.67) and (9.65):
IC(LIR) = ae(LIR) . Ic(PA)
KB
(9.70)
ae(LIR) . s(PA) s(LIR) =
KB
(971)
Example calculation:
The collector current Ic(PA) °f tne LS 613 phototransistor was measured
for an incident radiation of standard
A light with irradiance
Ee( AP)
=
5
rnW/cm2 ,
as
mA 1<X?A) = °-47
-
Its photocurrent sensitivity s(PA) is thus:
s(PA)
IC(PA)
Ee(PA)
O*47 mA
5 mW/cm
(9.72)
From equation (9.70) and the average values obtained from equations (9.68)
and (9.69), the collector current Ic(LIR) is calculated for an incident IR luminescence
radiation from the GaAs diode SL 1183 (TIXL 16 with reflector) with the same
irradiance.
2-23. 0-47 mA
IC(LIR) =
1-209 mA
0-867
(9.73)
A collector current of
mA IC(LIR) = I"4
was measured.
Mainly because of the assumed correction
factor Kg = 0-867, the calculation error
still amounts to 1 3-6%. In practice, even
150
greater calculation errors must usually be
taken into account. In general, simplified actinic value calculations are subject to
relatively large tolerances. The calculation of actinic value by graphical integration becomes inaccurate through plotting errors on the individual functions, with their scatter and their planimetric measurement. Determination of the actinic value by direct measurement is only exact, if the radiation functions of the radiation sources used can be measured exactly and if no errors are made in the sensitivity measurements.
9.8
Dark current of junction photodetectors
The reverse current of a semiconductor diode consists essentially of two components: the reverse current caused by the presence of minority carriers and a leakage current with an approximately resistive (Ohmic)
behaviour. In Si devices, the latter is very
small and can usually be neglected. The reverse current due to minority carriers
increases for small reverse voltages, in accordance with an exponential function of the reverse voltage.
Ir = Isat (1 - e
Vr 26 mV )
(9.74)
After an initial rise in the so-called starting range, the reverse current tends asymptotically towards a constant value (the reverse saturation current I^t) so that
V above a reverse voltage of approx. 0.1
the variable part of the exponential
function can be neglected. Up to the
breakdown of the junction, this range is sometimes called the "true" reverse range.
For semiconductor devices, the typical value of the reverse current at a temperature of 25°C is generally stated in the data sheet. Also, the reverse current also has an exponential relationship with the ambient
temperature. Usually, Ir = f(t) is shown on
log/linear paper. The temperature is plotted linearly on the abscissa and the reverse
current (or dark current) logarithmically on the ordinate. The relationship
& between Ir t is as follows:
°
iR.t = lR(25)
K(t e
-
25
C)
(9.75)
where:
lR,t =
lR(25) =
K
=
Reverse current at the specified temperature t,
Reverse current at 25 C (from
data sheet) Temperature coefficient of 06
to 01, average: 0082
In this coordinate system, the function
Ir t appears as a straight line, the slope being given by the temperature coefficient
K and the parallel displacement of the
lines by Ir(25> In junction photodetectors, the reverse current is defined as the "dark current". This is subject to the proviso, that the ambient irradiance must be zero. In Figure 9.24, in principle, the dark current Irj = f(t) can be read off for all nominal photodiodes and phototransistors. The temperature
K coefficient serves as the variable. The
value of the dark current Id(25) determines the scale of the ordinate. The Id(25) values (usually a second value is also stated at a higher temperature, e.g. Id(80)) can ^e taken from the data sheet and transferred to the grpah. If these points are joined by a straight line, the function Itj = f(t) is obtained for the corresponding device.
The temperature coefficient K is only
needed, if only one dark-current value
is known. If K is not known, a mean value, K e.g., = 0-82 can be used. This calculation
of the dark current becomes unnecessary, if an appropriate graph is contained in the
data sheet.
The examples given show Itj = f(t) for the phototransistor TIL 78 in Figure 9.25 and for the photodiode TIL 77 in Figure 9.26. It is also common, to indicate
151
'D (nA.)
10'-. 5-
\0'-
2-
10^
5-
2-
S-
2-
5-
25-
25-
2-
5-
////
/ v 2-
1*
20 25 OC
KKS
A AA
/^
/^
/s&
/ //
//
100
120
140
160
180
t(°C)-
Figure 9.24
The dark current Iq as a function of temperature t for junction photodetectors.
152
:
VCE >30 V
Ee =
/
40
60
80
1(°C)
»
Figure 9.25
The dark current Id as a function of
temperature (t) for the phototransistor TIL 78
3V--=
t: :b.-c
'D(nA)
1
0'4
0-1
004
^-- /
001
f
0-004
0-001
20 -10
10 20 30 40 50 60 70 8 ,(°C)
Figure 9.26 The dark current Ij) as a function of temperature ft) for the photodiode TIL 77.
Figure 9.27 The dark current ratio l£}/lfj(25) as a function of temperature ft) for the
photodiode 1N21 75
the normalised relationship Id/!D(25) as
shown in Figure 9.27 for the photodiode 1N2175. In a typical emitter-sensor system, the dark current at the maximum
case, or ambient, temperature of the photodetector should be less than the
minimum useful photociirrent by at least a
factor of 10.
9.8.1
Dark current of avalanche photodiodes
For avalanche photodiodes, the dark current 1^(25) is normally stated separately in the data sheet as the leakage current in the device, the so-called bulk leakage current I D u(25)> anci as tne leakage current on the device surface, the so-called surface leakage current I Su(25)>
Vr for a given applied reverse voltage
and an incident irradiance of E e = 0. The dark current, Itj(25), is the sum of the
surface leakage current, Isu(25)> anc^ tne product of the photocurrent amplification
153
M factor multiplied by the bulk leakage
current Ibu(25) :
lD(25) = Isu(25) + (M - Ibu(25))
(9.76)
The values for Ibu(25) and I su(25) are determined by a relatively difficult
measurement and a simple calculation:
The photosensitive surface of the wafer
of an avalanche photodiode is arranged
exactly at the focus of an optical system.
The adjustment and the quality of the
optical system must ensure, that the
radiation at the focus does not fall on
the whole wafer surface, but only on the
central part of the active area of the wafer.
The measurement is carried out with
monochromatic radiation of wavelength
X = 900 nm. This radiation is modulated
« at 400 Hz. A radiant power of <&e
nW 10
is to fall on the active area of the
wafer.
Vr First, a reverse voltage
of about
quarter of the typical breakdown voltage,
VgRtypg, is applied to the avalanche
photodiode to be measured, that is
VR = y«VB R,ty P
For the TIXL 56 silicon avalanche photo-
V diode this reverse voltage is about 40
and the photocurrent amplification has a
M value of = 1. Then the signal current
IS is measured. As the next step, the
Vr reverse voltage
is increased, so that the
signal current I§ increases to a higher, but
freely selectable value, corresponding to the
amplification factor M\ (see Figure 9.28).
The value of Mi must, however, remain
less than Mj/2 (with regard to Mj, see
Section 9.1.1), otherwise the dark current
and the noise superimposed on the signal
current 1$ would assume unusably high
values. The incident radiant flux 4>e is now interrupted and the value of the
dark current Ipi, which now flows, is
recorded.
Vr Following this, the reverse voltage
is
raised further, until the signal current 1$,
and thus the amplification factor, have
increased to twice the value; i.e.:
M2 = 2 . Mi
(9.77)
The incident radiant flux 4>e is again
interrupted, so that the value of the dark
current which now occurs, Ip2» can *> e
recorded. Figure 9.28 shows the principle of this measurement. The surface leakage current I Su(25) anc* tne bulk leakage current Ibu(25) are tne calculated as
follows:
!su(25) = !d1 - dD2 - iDl) = 2I D1 - ^D2
(9.78)
!bu(25) = *D2 - *Dl ID2 - !d1
M2 -- Mi
Ml
(9.79)
Figure 9.28
Dark current Id(25) as a function of the photocurrent amplification M. The
measurement points Ij)], Mj and Ij)2 He
in the linear part of the characteristic
M lD(25)=f( >
9.9
Sensitivity values of IR detectors
IR detectors, also called IR signal detectors, are understood to mean infra-redsensitive receivers with comparable sensitivity
154
and noise power for different spectral sensitivities. These detectors include, for example, very sensitive avalanche photodiodes, PIN photodiodes, intrinsic photoconductors and photo multipliers. For the noise voltage, which is of interest, it can be said in general, that as with most noise mechanisms it is roughly proportional to the square root of the bandwidth to be
transmitted.
The signal-to-noise ratio, abbreviated as S/N ratio, is either the ratio of the rms signal current I«j to the rms noise current
IN or of the rms signal voltage Vs to the rms noise voltage V^:
N iN vn
(9.80)
S is the rms value of the output signal of a
detector or detector-amplification system.
The detector is usually exposed to radiation with square-wave modulation. The pulse repetition frequency is either 800 Hz or 1000 Hz with a mark-space ratio of 1:1.
The signal is measured selectively at imod
and a defined narrow bandwidth, (e.g., Af = 5 Hz) and is usually converted to a bandwidth of Af = 1 Hz.
N is the rms value of the detector noise
or the detector-amplifier noise. As a rule,
300 K background radiation falls on the
detector. The noise is also measured
mo selectively at f d and the defined
bandwidth, (e.g., Af = 5 Hz) and is usually converted to a bandwidth of Af = 1 Hz.
The following data is indispensible for the measurement of S/N ratio:
The temperature of the black-body radiation or the wavelength of a monochromatic radiation, with which the most favourable S/N ratio occurs,
the electrical reference bandwidth,
the midland frequency,
the detector reference temperature or
the working temperature for cooled
detectors,
the detector surface area Aj> the incident irradiance Ee , where applicable the angle of view and
the temperature of the aperture.
In order to compare different kinds of detectors, the term noise equivalent power (NEP) has been introduced. It is defined as the power of sinusoidally modulated radiation, which would produce the same rms output voltage in an imaginary ideal noise-free detector, as the real detector, without an
input signal, delivers as noise voltage.
The noise equivalent power, NEP, is the
lower useful threshold of the detector, with a given detector surface area Ajj, where the useful and noise signals are
equal.
The S/N ratio is measured with the aid of a known, weak but still easily detectable
incident radiant power, and the NEP is
obtained approximately from the
following formula:
NEP = Ee . AD . 1
N
.(---)
VaI
N
1
= 4>e
(9.81)
In practice, the measurement unit for NEP
is often stated in watts only, instead of
in W/ \/faz, when related to a bandwidth of Af= 1 Hz.
If the irradiance is chosen for the calculation instead of the radiant power, the term noise equivalent input power, abbreviated NEI, is used.
For this:
NEP
NEI
AD
1
N
VaT- I( }
(9.82)
155
The reciprocal value of the NEP is the
detectivity (detection capability),
abbreviated D. The detectivity is a measure of the least detectable radiant power:
1
D
VAf
S
-·(--
NEP Ee . AD N
(9.83)
The measurement unit for detectivity is
sheet and the noise voltage is measured, then, using the formulae (9.80), (9.81), (9.84), (9.85), the sensitivity can be calculated
as follows:
VN
D VN
NEP . V^f V^D · Af
(9.86)
The measurement conditions for the
parameters NEP, NEI, D and D* are
identified by figures in brackets, e.g.:
The NEP of a detector system is proportional
to the square root of its surface area Atj. The specific detectivity, abbreviated D*, takes account of this factor
1
D*
NEP NEI . \/A^
(9.84)
The reference bandwidth Af of the specific
detectivity is always 1 Hz. In practice, the
measurement unit is often stated in cm/W instead of in cm . VHz/W.
D* black-body -
D* spectral
-
D* spectral max. -
D*(500, 1000, 1) D* ( 6, 1000, 1) D* (X,max, 1000, 1)
where 500 denotes T = 500 K of the black
body as a radiation source.
1000 denotes f = 1000 Hz, pulse repetition frequency of the modulated radiation.
6 identifies the wavelength (X = 6 jllm) of
the monochromatic radiation, with which D *
D* was measured.
As well as the noise power and the
detectivity, the sensitivity s of a detector
is of interest. The sensitivity s is the ratio
V of the rms value of the signal voltage
to
s
the rms value of a known radiant power
falling on the detector:
V,,
$e Ee . AD
(9.85)
From this formula, the dimension of sensitivity is V/W. For the determination of sensitivity, the noise power, bandwidth and the modulation frequency when measuring with intermittent light (if fmo d is far enough below the limiting frequency
of the detector and the measuring apparatus) are not needed. However, in sensitivity measurements, the distribution temperature of the radiation used, the detector temperature and the detector area Aj) are to be defined. If the
NEP or D* are known from the data
Figure 9.29
D* Specific detectivity
as a function
of the wavelength for silicon photo-
detectors with various detector surface
areas: A
0-02
cm
2
,
B
=
0-2
cm 2 ,
C= 1 cm'
156
-- 1
1 denotes Af = 1 Hz as the converted bandwidth of the receiving amplifier, with which D* was measured at fmod-
If the parameters NEP, NEI, D and D* are unknown for a Si photodetector, the NEP
can be calculated approximately with special purpose calculators. Here, the current noise with the dark current Irj and the resistance noise with a load
resistance Rl are taken into account for
the corresponding Si photodetector.
Figure 9.29 shows the specific spectral
detectivity D* of various silicon
detectors as a function of wavelength X. The variable is the detector surface area
with A = 0-02 cm2 , B = 0-2 cm 2 and C = 1 cm 2 . D* (2800, f,l) was converted to D*(Xmax) using the empirical value 5. In Figure 9.30, the specific detectivity D*
(X,450,l) in InAs detectors is shown as a function of the wavelength X. The variable
is the temperature T in degrees Kelvin.
With sensitive detectors, the specific detectivity is mainly dependent on the background radiation (Blip); it can be improved by providing a cooled aperture in front of the detector. The restricted aperture angle is called the viewing field angle ©. Figure 9.31 shows the decrease
2
1
V Theoretically ( sin 6/2)
1
1
9
1 1
8
1
1
7
1
6
t \
S
V
\ 4
\ \
3 ^>x
2 /
1Measure i
»
-T
1
1
^00 K background radiation
V*~
\
""*j? Cooled / aperture
VV [A t /fl SmsitivE -
II
area
1
v
(angle of viewing field) -
20 40 60 80 100 120 140 160 181
Viewing Held angle 6 in degrees
»-
c
+
10'°
77 K -£-
^
[
i
\
\
\
\
\
N
10*
'
^3( OK
1
l-S
1
2-5
3
3-5
4
Figure 9.30
Specific detectivity D* as a function of
the wavelength X for InAs photodetectors
Figure 9.31
D* = f(@) for InAs photodetectors
in the relative specific detectivity with greater viewing field angles for InAs detectors.
K In Figure 9.32, the sensitivity S500
is shown for Ge:Hg detectors as a function of the viewing field angle at 5 K. The variable is the detector surface area At>
Radiation detectors have inertia (see rise and fall times) and their speed performance is determined by the response time or
time constant (time response). The time
constant is inversely proportional to the
limiting frequency. It is determined when the modulated radiation falling on a
detector surface causes a drop in the detector sensitivity of 3 dB as compared with the same radiation, but unmodulated.
157
10'
\ j
GeHgi 300e Kbacground
radiatio*
Aq iir* cm1 i
-
10 s
ad
1
icr
cm1
\
~
10*
\ -
101 )
*D" tr*a**:
20
40
60
80
100
120
S Viewing fieto aneje in degrees
*
1
1
CO 277f
(9.87)
By means of the time constant of the detector, the maximum usable modulation frequency is determined, since above this both the sensitivity s and the quantum yield decrease. The theoretical
measuring set-up for the determination of the time constant of silicon detectors,
for example, is shown in Figure 9. S3.
Figure 9.32
s = f(@D for mercury-doped germanium (Ge.HgJ photodetectors
Krom thermocouple to
r therm ocompensa tor
Chopper frequency setting control
Black-body radiation,
e.g., tungsten lampp wiitthhj J
3F- colour temperatun
28S6 fC
Preamplifier
Temperature control
Power unit for detector bias voltage
Calibration circuit
Figure 9.33 Circuit for measurement of the time constant
Tektronix 122
amplifier
HP-302A spectrum
analyser
Oscillograph
158
10 Parameters
common to emitters
and receivers
10.1 10.2
Evaluation of the Radiation and Receiving Characteristics of Optoelectronic Components Optical Tolerance of Optoelectronic Components
10.2.1 10.2.2
Effect of wafer centering, lens quality, distance from lens to wafer, refractive index of epoxy resin and shape of
dome and case
Effect of internal case reflections and wafer Geometry
10.2.3 10.3
10.3.1
Tolerance levels Coupling Characteristics, Transmission Ratios and Contrast Current Ratios of Very Short Optical Links Coupling Characteristics and Transmission Ratios
10.3.2 10.4 10.5 10.6
Contrast Ratio Half-power and Half-value points
Dynamic Data
Reliability of Optoelectronic
Semiconductor Components
159
10.1
Evaluation of the Radiation and Receiving Characteristics of Optoelectronic Components
Both the radiation intensity I e of a radiation source and also the radiation sensitivity s of a detector are greatly dependent on the angle of emergence or of incidence of the radiation. This dependence is plotted in suitable graphs as the Radiant Intensity Distribution Curve Ie = f(tp) or as the sensitivity distribution curve s = f(i£). For this, polar coordinates are mostly used, but occasionally also
A Cartesian coordinate systems. well-
known example of this is the luminous
intensity distribution curve of a filament lamp, illustrated in Figure 10.1, where
the luminous intensity, which depends on the angle to the axis of summetry, is stated
in cd.
10
Parameters Common to Emitters
and Receivers
On the same principle, Figure 10.2 shows
the emission conditions of a fluorescent tube. Here, the luminous intensity
distribution curve is shown for a standardised luminous flux of 1000 lm for the tube.
From antenna and radar engineering in
particular, it is generally well-known, that the efficiency of a transmitter and also of a receiver can be considerably improved by focussing the radiation into
a beam. The distribution characteristics then form more or less narrow peaks.
In the field of optoelectronics, the functions I e = f(^) and s = f($ are of great importance for evaluating the performance of these components. These are characteristics of approximately spherical shape and also narrow patterns.
The shape of the emission or reception characteristics is fixed by the mechanical construction of the components.
Figure 10.1
Luminous intensity distribution curve, with axial symmetry, of a filament lamp
for 100 lm
Figure 10.2
Luminous intensity distribution curve of a fluorescent lamp for 1000 lm. Plane I perpendicular to the axis of the lamp, plane II through the axis of the lamp
161
The case and its material play a decisive part here. The structural shape of the wafer
A also has an effect. distinction is made
here between flat and hemispherical dome wafers. Many wafers are embedded in an epoxy medium, through the shape and
refractive index of which the radiation characteristic can be influenced in a planned
manner. Lenses may be built to produce
particularly narrow focussing of the maxima. It should be mentioned at this stage, that in the case of components with an asymetric characteristic, more or less pronounced secondary peaks can occur as well as the main peak, and must be taken into account in calculations and in the construction of optoelectronic
systems.
hemispherical surface is broken down, as shown in Figure 10.3, into annular zones. If the zones are sufficiently narrow, then the area of such a zone is:
dA = 2tt . r . sin<£ . d<£ . r
(10.1)
By integration of this expression from to 90 , and taking account of the elementary radiation laws (2.18) and (3.1), the
relationship
-T^) <t>e = Ie .^=Ie
(10.2)
is obtained and, with equation (3.10), the following integral for the radiant power:
For all components with a spherical radiation characteristic, Lambert's cosine law can be applied for the calculation. It is known from Chapter 3, that the intensity I e of a Lambert radiator decreases, according to a cosine function, with increasing angle <p from the normal.
Ie,i/J= h,o<x>sP
(3-10)
A hemispherical space around the Lambert
radiator is therefore not uniformly
A irradiated. maximum radiant intensity is
measured in the normal direction, and this decreases to zero, in proportion to cos# with increasing angle if.
Now, the Lambert radiator can have
allocated to it a given solid angle 12, which is so great, that a spherical cap with this solid angle, and assuming the radiant intensity I e to be constant in space,
receives the same total radiant power as the hemisphere defined by the cosine law. This means, the Lambert radiator is
replaced, in the imagination, by a radiator, which emits with its full intensity I max within a given solid angle £2 and emits no
radiation outside this solid angle.
The solid angle can be determined mathematically as follows: The original
Figure 10.3
Derivation of the solid angle in the case of the Lambert radiator
162
Oo 90
a>p
/ 27T . r. sini/J . I e ,o · cos£ dip . r
90"
^o V = lit . I e ·
· J sin</? · co *P - d
,
sm2,^0"
p^ [-- 27T . Ie>0 . fto .
]=77.Ie, .no
(10.3)
Division
of
the
power
4>
e>
which
is
assumed to be constant, by the maximum
radiant intensity Iei0 gives a solid angle
---- d>e
J2 =
=7T.fio = 7r
sr
I'oe.o
for the Lambert radiator.
(10.4)
This result has already been pointed out in
(3.1 1). The expression J2o or 1 sr has no
unit in itself, but is treated mathematically like a constant in the formulae.
By means of Figure 1 0.4, it will now be shown geometrically that a hollow
sphere corresponding to the Lambert radiation characteristic is irradiated with equal intensity at all points. For every surface element of this hollow sphere, conditions, such as are shown in Figure 3. 9, apply, where both the angle </> and
also the distance a are variables:
4fe,dA£i = f(<Aa)
(10.5)
Since the normal to the emitter and the normal to the receiver always intersect at the centre of the sphere, they form the two equal sides of an isosceles triangle, the base of which is the ray under consideration. For reasons of symmetry, the angles ifls and i0g must therefore also be equal:
iflS=VE = ^
(10.6)
Normal to surface of AEI
e,0°
Normal to surface 1
ofAsi
-|
d Standardised semi-infinite space
Standardised radiant intensity distribution sphere of Lambert radiator
Standardised radiant intensity distribution sphere of a very small plane area A S1 of the
Lambert radiator, radiating on one side in the standardised semi-infinite space with unit
m radius r = 1
163
With the Lambert radiator, the distance 'a' is clearly dependent on the angle <p, namely:
costp
(10.7)
If the results of equations (10.6) and
(10.7) are inserted in (3.22), we
obtain:
d2 <k
dAg dAs . cos<£> .
. cosip
2 (r . cosi£)
(10.8)
Since the only variable in this equation, cos(,0, can be cancelled out, pure constants are obtained and the differential notation can therefore be dispensed with:
$P
Ae- a s ·«o
(10.9)
In this equation, <J>e is no longer dependent on the angle and is therefore a constant value for the hollow sphere under
consideration.
Now let us again consider the hemisphere
with radius r. According to the cosine law, at an angle <p = 60 , the radiant power obtained is:
$ $ $ e,60° = e?0 . cos60° = 0-5 . e ,o
(10.10)
Since, for semi-infinite space, only half the radiant power is emitted here, in comparison with the power in the normal direction, the corresponding points on the distribution curve (a complete sphere with radius r/2) are designated as the half-power points, abbreviated HP.
Next, the solid angle £2 is examined. It is obtained by consideration of the spherical cap within the half-power points with 2i/> = 1 20°. According to equation (3.4):
Q= 27T(1 -cos 60°) .f2o = 7r. sr
(10.11)
If this result is compared with (10.4), then it is seen, that the boundary of the solid angle, in the case of the Lambert radiator, goes exactly through the half-power points of its distribution curve.
It has already been pointed out, that many optoelectronic components have
asymmetric directional characteristics.
An exact calculation of the .solid angle
12j£ which applies here is only possible, if the asymmetric characteristic can be described unambiguously by an equation of the form
h,y = %>) · !e,o
(10.12)
In practice, an attempt will therefore be made, to find a function approximately the same as the characteristic, which can be calculated with a minimal expenditure of mathematical effort. Functions of the form
Ie,i£= cos(n. v?) e.o
(10.13)
are suitable for this purpose, n being a number depending on the degree of narrowness of the distribution.
As an example, Figure 10.5 shows, in addition to the Lambert radiator with n = 1, two maxima with n = 3 and n = 10. Calculation of the solid angle is carried out on the same principle as for the Lambert radiator, and the following
integral is obtained:
90! n Oj£ = 2tt . fio . Jsin<p . cos(nip) . dip
n
cos(l+n)ip
27T.no. [--
2(1 +n)
cos(l-n)i£ "^ -] n
2(1 -n) o
(10.14)
Calculation of this integral for a few selected values of n gives the results stated in Table 10.1.
164
1-5 2 2-5 3
4
5 6 8
10
K ft /sr
1-50 0-867 0-562 0-393 0-222 0-143 0-0993 0-0559 0-0358
VQj ° <Alp/
40-4 30-5 24-4 20-4 15-3 12-2 10-2
7-65 6-12
40 30 24 20
15 12 10
7-5
6
Table 10.1
Results obtained from equation (10.14) for a few selected values of n
Q& Also, from the solid angle
according to
equation (3.4), the angles <pQ, which limit
the cone for Q^, have been
determined. At the same time, the angle
for the half-power points, at which the
radiated power has fallen to half, has
also been stated. As can be seen, with these
peaks, i/J£2 and ^HP are n0 l° ng er exactly
equal, as was the case with the Lambert
radiator. The small differences are not
due to inaccuracies in calculation, but
actually exist. Of course, they are very small
(of the order of 1 to 2%), so that in practice,
in the case of asymmetric radiation maxima which are similar to the functions discussed here, the solid angle can be calculated directly with the aid of the half-power points.
Solid angle calculations are mainly of interest for components which emit radiation. In the data sheets, statements on the radiated power 3>e are generally found. In the case of asymmetric characteristics, the total power, including
any secondary peaks which may be present,
is often stated, but often also only the power of the main area alone.
Nowadays, statements of the radiant intensity I e>0 in the normal direction are most often quoted and it is to be expected, that this parameter will be stated
in future in all data sheets.
The radiation characteristics, which define
the radiant intensity as a function of the
angle i/> in suitable graphs, generally show
relative values, related to the value in the principal direction of emission. In evaluating the graphs, it must not be forgotten, that as a result of various tolerances, which are unavoidable in
0' 9
,,« 01 0-6
05
04 03
02
1
0-1 0-2 0-3 04 0-5 0-6 07 08 0-9
1
Figure 10.5
Functions of the form a = cos nip in polar coordinates
165
mass production, the final results can sometimes differ from the values obtained theoretically. In particular, even small tolerances of the radiation distribution curve can cause a considerable reduction of the
irradiance Ee falling on a photodetector. Further details on tolerances will follow in
Section 10.2.
The problems mentioned will be illustrated by two practical examples: Figure 10.6 shows the radiation emission characteristic
of the GaAs diode TIXL 27. Here, we are
concerned with an approximately spherical pattern, with which the calculation of the solid angle fij^ can De carried out with sufficient accuracy from the half-power points. As can be seen from the graph, the rays through the half-power points have an angle i£ = 67 to the major axis. With equation (3.4), one therefore obtains:
% «27T(l-cos67°) . fio = 3-826 sr
(10.15)
In the data sheets of the TIXL 27, a
minimum radiant power <I>e mm =15 mW
is stated. This gives a minimum radiant
intensity I e ,o,min in tne direction of the major axis of:
le,o,nmuin
*e,min
15 mW
3-826 sr
= 3-92 mW/sr
(10.16)
As a further example, the results of a practical measurement on the GaAs diode TIL 31 will be discussed.
The emitter-receiver distance was selected
at r = 20 cm and a solar cell, with an
Ag active area
= 3-6 cm 2 and a sensitivity
s = 0-51 A/W, was used as the detector.
The radiation of the TIL 31 is first observed
on a projection screen with an IR viewer.
Here, the projection of the wafer is very
clearly visible as a square with a length of
side of 1 =4-3 cm. At a somewhat
·
U,rel
,10°
-np,-
j$y-
*k
^
0°
noa
100%
"~-^?°
^%b
Typical radiant intensity 'distribution cutve
80 %_
Vfc,
«0%-l
C
\ *r.
\\
.
x^X^^
Semi-infinite sP*ce curve
40%__
JVcV'*
4J-
21
1
1100%
i
r
i
50*
20% / %
U
50%
I
|t
lOOif*
Figure 10.6
Calculation of the solid angle £1% in the case of the GaAs diode TIXL 27. This diode has a modified T05 can. The projecting cylindrical cover is removed. The flat wafer is embedded in an epoxy dome
166
greater distance around this square, a number of rings, which arise from the secondary peaks, can also be seen quite faintly. At first, these were not taken into account.
The solid angle of the main maximum can
be calculated with sufficient accuracy on
the basis of the water projection found, from (3.1):
4-32 cm2
^=^-^0 =
sr = 0-0462 sr
202 cm2
(10.17)
From the converted formula (3.4), the
angle of the rays through the half-power points of the idealised main peak to the normal is obtained:
--&K
00462
ip= arc cos (1
--) = arc cos (1
)
lit
27T
= 7^
(10.18)
The measuring instrument connected to the fully irradiated solar cell showed a shortcircuit current Ip = 160 jUA. From this the radiant power falling on the solar cell was calculated with equation (9.27):
IP 160 MA
4k'e,E
s
313-7 JUW
0-51 W/ A
(10.19)
Further, (2.40) gives:
$e (E _ 313-7 MW
AE
3-6 cm'
/iW 87-14
cm
(10.20)
Since the angle of incidence of the rays at the edge onto the solar cell is still very small, the cos ipg in equation (3.25) can
be made equal to 1 without any great error, and the radiant intensity of the main area of the distribution can be
calculated as follows:
-- mW MW Ie
=
Ee
.
l i
=
87-14
.
20 ,2
= 34-86
i^
sr
sr
(10.21)
The radiant power of the main peak is obtained from (2.18) as
% mW $>e = Ie .
= 34-86
. 0-0462 sr
sr
= 1-61 mW
(10.22)
It has been mentioned above, that secondary maxima are present as well as the main peak. In order to include their radiant power as
well, the solar cell was now arranged directly in front of the TIL 31. In this
case, evaluation of this greatly simplified measurement gave a total radiant power
4>e tot 7 mW. From this it can be seen, that only about 23% of the total power
is concentrated in the main peak, while
the remainder is scattered in a fan-like pattern by the secondary maxima.
The calculation of radiation characteristics which differ considerably from the previous idealised curves is rather more difficult. For this, it is convenient to apply graphical methods. In the Rousseau
diagram, the radiation pattern is divided into sufficiently small angular areas and
A integrated graphically. second method
describes a breakdown into constant solid-angle zones, which can also be
integrated graphically.
10.2 Optical Tolerances of optoelectronic
components
In the previous section 10.1, it has already been pointed out, that
optoelectronic components can be
subject to relatively large tolerances.
Distinctions are made here between:
Tolerances on the electrical parameters.
167
Tolerances of the total radiated power $£ of semiconductor radiation sources.
Tolerances of the basic photosensitivity of semiconductor junction photodetectors. Here, all external or built-in optical parts such as lenses, mirrors or reflectors are to be assumed to be dismantled from the photodetector.
Installation tolerances of optoelectronic components on the part of the user.
5a Tolerances of the emission peak of semiconductor radiation sources.
5b Tolerances of the receiving peak of junction photodetectors.
The electrical tolerances and the tolerances on radiant power and photosensitivity of optoelectronic components should be overcome by an optimised circuit design. The radiant power tolerances can be eliminated in many cases by adjusting the flux current Ip through the radiating
diode. In simple applications, the photosensitivity tolerances can be
compensated for by gain adjustments on the following amplifier. The user should examine all measures of circuit design and also keep the mechanical installation tolerances of the optoelectronic components as small as possible, in order to be able to use lower priced, massproduced standard types of optoelectronic component instead of the more expensive
selected types.
The tolerances of the radiation peak of
semiconductor radiation sources and the tolerances of the receiving peak of junction photodetectors must be dealt with in great detail, since these tolerances can have a very drastic effect on the output signal from the detector.
10.2.1 Effect of wafer centering, lens quality,
distance from lens to wafer, refractive
index of epoxy resin and shape of dome
and case
The highest-quality optoelectronic components, which are mostly used in commercial electronic systems, have metal cases with flat windows. Metal cases are preferred, since they have better thermal resistance Rt^ jq and are easy to mount on heat sinks. Flat windows are installed intentionally, so that the user can design the focussing of the radiant power optimally for his needs, with simple
lenses or high-quality reflector, glass fibre or lens combinations.
Epoxy resin domes are not used, since otherwise these components cannot be used for temperatures above 100 C. The GaAs power diodes are constructed with a dome wafer for a better radiation yield.
Components with flat windows often have emission or reception maxima with wide aperture angles. The deviation of the
principal direction of emission or reception from the normal is mainly
determined only by the wafer centering. Here, the tolerances are not great, since
the spatial distribution alone gives a
drop in power of a few percent in the normal direction. In addition, the aperture angle of such maxima is usually so
large, that despite the tolerances, the aperture angles of attached lenses or lens systems are covered almost without a decrease in power.
As an example, Figure 10.7 shows a distribution curve of an optoelectronic component with a tolerance deviation. The aperture angle between the halfpower points is about 135 . The drop in power in the normal direction as compared with the main emission or reception direction is aobut 4%.
"Low-cost" GaAs power diodes have wafers with flat geometry and are
installed in a metal case. For a better
168
--
Figure 10.7 Radiant intensity or sensitivity distribution of an optoelectronic component affected by
tolerance. The aperture angle is about 135 , the drop in power in the normal direction about 4%
radiation yield (see Section 6.6 and
Chapter 1 1), an epoxy-resin dome or
lens is fitted on the wafer. Here, somewhat greater radiant intensity
distribution tolerances are to be expected. They are mainly determined by the wafer
centering, the refractive index of the epoxy resin and the shape of the epoxy dome. Figure 10.8 shows a radiant-intensity distribution curve, subject to tolerance,
for a "Low cost" GaAs power diode. The
aperture angle betwen the half-power
OO
10°
100
& ^r- "\
^'"T
\
,
T~~-
/~\
«
J -. ~^tL ^yfr
x^ *k / V-;
4y
1 ^^<
§//
J fr
/ / />
Jitl
ii I
I
I
1
20
\y V
\ A^v ^ ' V
\ /
"
^
'' s
f\'\~~~
Wy\ / y K r\- ~h
;</ Vx \
n//*S
1
J
|
1
|
10
x5<vfex-rT\\
§
3pr\ M \
\
\
!
1
1
|
s
'e.rel
Figure 10.8
Radiant intensity distribution of a "Low-cost" GaAs Power diode, subject to tolerance
169
points is about 110 , while the drop in radiant intensity in the normal direction, as compared with the main direction of radiation, is around 15%. Despite the
larger radiant intensity distribution
tolerances of such GaAs diodes, the
aperture angles of attached lenses are covered without significant losses of radiant power.
Low-power GaAs diodes also have, in
general, wafers with a flat geometry. They are installed in metal cases with flat windows, if close radiant intensity distribution tolerances and high operating temperatures are called for.
Epoxy resin droplets are applied to flat wafer surfaces, if somewhat higher radiation yields and radiant powers are desired. The radiant intensity
distribution tolerances then increase. For
focussing the emission peak of GaAs
diodes or the reception peak of junction photodetectors, i.e., if high radiant intensities or high sensitivities are required, glass lenses are fixed on the metal case. The optical quality of these very small lenses is low as compared with the usual single lenses. The distance from the wafer to the lens is subject to the assembly tolerances. Therefore the distribution tolerances increase once
more. The wafers of "Low cost" lowpower GaAs diodes or junction photodetectors are completely embedded in
epoxy-resin cases. Here the distribution tolerances are increased yet again.
10.2.2
Effects of internal case reflections and wafer geometry
The emission or reception peaks of optoelectronic components are also affected by internal case reflections and the wafer geometry. Flat wafers have either a cathode bond (GaAs diodes) or a base and an emitter bond (junction photodetectors)
on their surface. A bond can sometimes
be located in the centre of the wafer. The
output power distribution can then show dips in the normal direction. This effect can also occur through the shape of epoxy resin cases. Figure 10.9 shows a distribution curve with a dip in the principal
direction of emission. Discontinuities in radiant power in the radiant intensity distribution profile can be detected visually with an IR viewing device, by observing the
radiation from a GaAs diode falling on a
projection screen. Corresponding to the radiation distribution shown in Figure 10.9, a more weakly reflecting circle will be projected in the normal direction, with a more strongly reflecting annular area surrounding it.
Optoelectronic components in metal transistor cases with fitted lenses can show additional undesired secondary maxima due to internal reflections from the case, as well as a closely fo cussed peak in the main
direction. In Figure 10.10, the radiant intensity distribution of a low-power
GaAs diode with unwanted secondary peaks is shown. If this radiation in the normal direction falls perpendicularly on a
projection screen, then a strongly reflecting circular area will be projected in the main direction of radiation and two surrounding less strongly reflecting annular areas at intervals from it.
Furthermore, the radiant intensity
distribution of a GaAs diode can be assessed visually with the IR viewer by
a second method. For this, transparent paper is stretched out flat in the normal
direction to the GaAs diode. The radiant intensity distribution on the
transparent plane can be seen very impressively and clearly, since it corresponds to the radiant intensity distribution in the polar coordinate system.
With epoxy resin and metal cases for lowpower GaAs diodes, the radiation also emerges through the base of the package through internal reflections in the case. For epoxy resin cases, this is clearly understandable, since the base of the case is
170
Figure 10.9
Radiant intensity or sensitivity distribution of an optoelectronic component affected by tolerance. The distribution profile shows a dip in radiant intensity or sensitivity in the
normal direction
transparent to 1R radiation. With metal cases, the radiation emerges through the plastic insulation of the connecting wires,
which, although opaque to light, sometimes transmits radiation even in the very near IR range. This effect is mentioned, since
Figure 10.10
Radiant intensity distribution of a GaAs diode with secondary maxima
171
in extreme cases it could cause cross-talk interference with adjacent radiation links.
GaAs diodes which are overloaded from the point of view of loss power can also
change their radiant intensity distribution. Dead areas can then form in the wafer. With an IR viewer, it is easy to ascertain, whether only part of the wafer is still emitting radiation'. Figure 10.1 1 shows
the wafer of a GaAs diode overloaded in this way. Its emission decreases more and more from the centre to the left-hand
side. It is equally possible that only a small corner area will emit radiation.
N
HP' x
mHHmK^
Rp '
HHHk-
R
Hi / /
Figure 10.11
Emitting wafer of an overloaded GaAs diode with dead zones
For the enlarged observation of the wafer
of an IR-emitting GaAs diode, an IR viewer
with an additional lens or with an external
lens will be needed. For GaAs diodes
with built-in, strongly refracting lenses, it is simpler to observe the wafer. Here,
the radiation from these GaAs diodes is
already focussed so that the wafer can be shown as an image on a flat projection screen and observed with the IR viewer. Wafer defects, bonds and bonding wires are then clearly visible. If the radiance of
a GaAs diode is considered over all
Figure 10.12 Sensitivity distribution over the active wafer area of a junction photodetectqr
active parts of the wafer area, it cannot be regarded as constant. In the same way, the sensitivity of a junction photodetector is not constant over all the active'parts of the wafer area. Figure 10.12 shows, in simplified form, the sensitivity distribution over the active wafer area of a photodiode. Here the sensitivity increases towards the outer zones. Various people have investigated the sensitivity distribution of Si photodetectors over the photo-
sensitive area. From what has been said so
far, the reasons for the distribution
tolerances of low-power GaAs diodes
with closely focussed emission peaks or of junction photodetectors with closely focussed reception peaks can be
summarised. The factors which are
responsible are the wafer centering, the quality of built-in lenses, the distance from lens to wafer, the refractive index of the epoxy resin, the shape of the epoxy
dome or case, bonds on the wafer and
bonding wires, internal case reflections and last but not least the radiance distribution
over the active wafer area of GaAs diodes
or the sensitivity distribution over the active wafer area of junction photo-
detectors. The stated causes for the tolerances clearly show, that to achieve a uniform mode of representation, the radiant intensity of GaAs diodes and the
172
Figure 10.13 Radiant intensity distribution curve or sensitivity distribution curve of a "cross-eyed' optoelectronic component. The principal direction deviates from the normal by 9-5
sensitivity of junction photodetectors must always be measured in the normal direction instead of the main direction of emission or reception. Figure 10.13 reproduces the distribution curve of a "cross-eyed" optoelectronic
component. The principal direction deviates from the normal by 9-5 . Here,
the sensitivity of a detector or the radiant
intensity of a GaAs diode is 15% higher
in the main direction than in the normal
direction.
10.2.3 Tolerance levels
Generally, data on the minimum, typical or maximum sensitivity for junction photodetectors only relates to the normal direction of the radiant power data of GaAs diodes only relates to the corrected emission peak in the normal direction. Thus the only case which can occur is that
a junction photodetector shows a higher sensitivity in the principal reception direction or a GaAs diode shows a higher radiant power or radiant intensity peak in the principal emission direction. The sensitivities for junction photodetectors or the radiant power values for GaAs diodes stated in data sheets mainly relate to the distribution curves illustrated. Closely focussed distribution maxima are sometimes represented with idealised
distribution curves.
Users often set very strict tolerance requirements, which are not attainable in
mass production, especially on optoelectronic components with built-in lenses or with epoxy-resin cases. It should be strictly noted that due to the inherant tolerances of these components comapred to devices without epoxy encapsulations and lenses they will have less optical precision. The built-in lenses only represent a poor substitute for external and optimally
selected individual lenses.
173
In electro-optical precision instruments or systems, optoelectronic components with
flat windows are mainly used in conjunction
with attached external lenses or objectives. For reasons of space, in optoelectronic reading heads, or for reasons of cost, e.g., for consumer applications, external
individual lenses cannot be used in many
cases. Here, at least, taking account of the tolerances, one can have resources to selected optoelectronic components with
lenses.
The selection types needed, e.g., for optoelectronic components with lenses, can be determined in a rough tolerance
calculation in conjunction with radiant intensity and sensitivity measurements for the components in question.
For roughly-selected GaAs diodes with
lenses:
Radiant power tolerances
*e,max ^.
<*Ve.min
(10.24)
Estimated emission peak tolerances in the measurement direction
«K,rmax ^C.min
(10.25)
The radiant intensity tolerance is then:
^e.max ^K.max ·3.2 = 6
^e.min ^jC.min
(10.26)
For preselected Si photodetectors with
lenses:
The difference in the photocurrent sensitivity amounts to
smax
Smir
(10.27)
Installation tolerances:
The deviation of the opposing optical axes in each case for the receiver and emitter, which are caused by assembly and by temperature-dependent stresses in the installation support material, depend on the; precision of the mechanical structure. The
overall installation tolerances are estimated by the inclusion of a factor 2.
d Transmission medium:
In many cases, the tolerances of the transmission medium are not taken into account.
However, in an atmosphere containing water, dirt and oil, e.g., for machine controls, these tolerances are by no means neelieible..
f
Overall tolerance factor:
The product of the factors from a to d gives the overall tolerance factor of
a.b. c.d«6 . 3.2. 1 36
(10.28)
An overall tolerance factor of 36 corres-
ponds, for example, to a photo-current in the detector of
mA mA !p,min = 0-1
and I p>rnax = 3-6
or as a second example of
^.min = 5 mA and Ip>max = 180 mA
(see also Figure 10.15).
Depending on the application and the selection of components, the overall tolerance factor can be even greater or also less. For very short-range radiation links, the overall tolerance factor can be reduced by the amount of the installation
tolerances, since the emitter irradiates the whole detector.
GaAs diodes usually show the greatest
tolerances, since, as already discussed, the semiconductor manufacturer only guarantees
the minimum radiant power. The radiant intensity tolerances of GaAs diodes can be
reduced by about a factor of 2 by adjusting the diode forward current Ip. Radiant
174
1.
Quantity (%) 20
IS
10
5
30 100 2
350 450 5 50 650 800 1100 1400
25 Quantity (%)
20
15
10
5
» )
30 1
250 350 450 550 650 800 11 00 1400
Figure 10.14a Percentage numbers of phototransistors as a function of the relative photocurrent sensitivity for a photo-transistor type of
batch A
Figure 10.14b Percentage numbers of phototransistors as a function of the relative photocurrent sensitivity for a phototransistor type of
batch B
intensity tolerances can also be taken into account by the photodetector amplifier, by varying the gain. With high tolerance requirements, it is convenient, if the user, during his goods-inward inspection, at
least selects the GaAs diodes supplied for the relative minimum, typical and maximum
radiant intensity.
GaAs diodes with low radiant intensities are
grouped for Si photodetectors with high radiant intensity for Si photodetectors with low sensitivity.
For mass production, it should be taken into account, that the distribution of the
radiant intensity values for a GaAs diode
type or the distribution of the sensitivity values for a Si photodetector type can vary from batch to batch within the data sheet tolerances. For one phototransistor type, the selection of which only related to an uppe photocurrent sensitivity limit, the relative photocurrent sensitivity sre i is illustrated in Figure 10.14a for
Batch A and in Figure 10.14b for Batch B.
A further reduction of the overall
tolerance factor is achieved, by choosing closely selected photodetectors with sensitivity differences of
smax = 1-5-2
smin
Here, a large number of selected Si-photodetectors, developed by Texas Instruments, are available to the user. To achieve the
smallest overall tolerance factors, the
GaAs diodes and Si photodetectors
recorded or selected in the customer's goods inward inspection are combined as matching pairs in each case.
Competent manufacturers of electronic systems have already used this method for years. In general, in the many simple visible light or applications, which only call for a Yes-No decision, relatively large
tolerance factors are acceptable with appropriate circuit design. Here, a signal
is only initiated when a fixed irradiance
is exceeded or not exceeded.
175
10.3 Coupling characteristics, Transmission Ratios and Contrast Current Ratios of
Very Small Photocell Junctions.
mA [(. = SO
'((Onl
1
10.3.1
Coupling Characteristics and Transmission Ratios
Very small opto isolators are, as the name implies, light or IR sensitive devices with very short distances between source and detector. They also include the optoelectronic reading heads and optocouplers. The maximum source-detector distance
corresponds to the photometric limiting distance. For this, an opto isolator can be designed with a theoretical accuracy of 1%. For shorter distances, practical calculation of a very small opto isolator performance is hardly possible. For this reason, the semiconductor manufacturer states the coupling characteristic between
GaAs diodes and Si photodetectors. Here,
the emitter and the detector face one another in the same optical axis. The coupling characteristic is represented in a graph as a function of the photodetector output signal in relation to the sourcedetector distance. Figure 10.15 shows, as an example, the theoretical coupling characteristic for very small optoisolators or optoelectronic reading heads. The possible
i "On" collector current Ic(On) 's illustrated
as a function of the source-detector distance
for typical pair combinations with the GaAs diodes TIL 24 and the phototransistors LS 600.
The expression "On" relates to the switched-
on or conducting state of the GaAs diode.
The Ic(On) tolerances correspond
approximately to the tolerance calculation
state in Section 10.2. For a distance of
r = 0-25 inch, a typical "On" collector
mA current of Ic(On),typ = 2
is obtained.
This output current is high enough to
drive either subsequent amplifiers or
interface circuits. The following Figures
10.16 to 10.19 show typical coupling
'C(On)(mA>
Figure 10.15
Basic coupling characteristic for very small opto isolators and optoelectronic reading
heads. The possible "On" collector
current Ic(Onj 's illustrated as a function of the source-detector distance r for a typical pair combination with a GaAs diode TIL
24 and a Si photo transistor LS 600. 1000
mm mils = 1 inch = 24-4
characteristics for very small opto isolators and optoelectronic reading heads from selected source-detector component families, some of which have the same packages.
The components used for Figure 10.19,
the GaAs diode TIL 31 and the photo-
transistor TIL 81, have a metal
TO encapsulated in the transistor
46
metal can to which a glass lens is fitted.
Such optoelectronic components are
often operated without a heat-sink. This
has the result, that the highest GaAs
176
70
40 TIL 24
qOnXmA)
20
TIL 23 10
7
4
LS614
LS613
" "TIL 24 or
TIL 602
TIL 606
Ip = 50 mA TIL 610 or
IG = 25 OC TIL 614
:e,= 2v
-V = 25°c
2
1
0-7 0'4
0-2
01
01 002 004 007 1
0-2
N
N
\"
^
0-4
0-7 I
7
-If - 20 mA
'C(OnXmA)
4
TIL 78 Vce "SV-
"" >
\
0-01
002
04 00' 01
04
0-7 1
r(in)
Figure 10.16
Typical coupling characteristic for source-
detector combinations with GaAs diodes of types TIL 23, TIL 24 and Si phototransistors of types LS 613, LS 614, TIL 602, TIL 606, TIL 610 and TIL 614. The "On" collector current 1 ciOnj 'x shown as a function of the distance r
Figure 10.18 Typical coupling characteristic for a source-detector combination with the
GaAs diode TIL 32 and the SI phototransistor TIL 78. The "On" collector current iQOn) 's shown as a function of the distance r
500
1
'C(OnXM) 200
100
50
n
i
20
\\
10
Em
5
tU 25 OC
Deieclor
2
V C E = 5V;tc = *-25 0C||
1 lllllll
1 1 lllllll
01
0,05 0,1
0,5 1
i 1 i,
ii
1
1
1 II II
1 1 lllll
5 10
mm 100
\
C(OnXmA) 30
1
Emitter TIL 31
IF = 25mA H -25 0C
-^ |
\
1
Detector TIL 81
= 25 >C
i
\
j
01 02 03 04 05 06 07 0-8 0-9 I
r(in)
-
Figure 10.17 Typical coupling characteristic for source-
detector combinations with the GaAs diode TIXL 26 and Si phototransistors of Types LS 613, LS 614, TIL 602, TIL 606, TIL 610 and TIL 614. The "On" collector current Ic(On) is shown as a function of the distance r
Figure- 10. 19 Typical coupling characteristic for a source-detector combination with the GaAs diode TIL 31 and the Si phototransistor TIL 81. The "On" collector
current InOn) 's shown as a function
of the distance r
111
diode forward current Ip max for an ambient temperature tjj = 25 °C mav
only have a value, because of the loss
power, of about a quarter of the maximum
permissible value stated in the data sheet for a case temperature tQ = 25 C.
However in comparison, the diode forward
current If, max at an ambient temperature tjj = 25°c'for the GaAs diode TIL 32 (which is equivalent to the TIL 31 except for its plastic case) may only be about a fifth of the maximum permissible value stated for the TIL 31 at tg = 25°C. The relative radiant power <$>e re i as a
function of the ambient temperature tjj
for the GaAs diode TIL 31 can therefore also be read off approximately from the 4>e,rel = f(tij) graph of the GaAs diode TIL 32.
The absolute radiant power 4>e of the TIL 31 can be calculated in simplified form with this relative value and the absolute TIL 32 data-sheet value for any desired ambient temperatures tij. With a case temperature of tQ = 80 C, according to the TIL 31 data sheet, the diode forward current must have a value of
^tc.SO = ° mA -
Now, the current drift factor per C
ambient temperature, Alp/tij. can be
calculated for the GaAs diode TIL 31.
This current drift factor is the ratio of
the maximum permissible diode forward
^ current lF,max,tTj ?5 ^or e ambient
temperature tjj^s = 25 C to the ambient
m temperature difference Atij ax-25 between
the maximum permissible ambient
temperature t(j max and ty 25- In the
absence of a statement in the data sheet,
the maximum permissible case temperature tG.max 's to be used as the maximum permissible ambient temperature tij max , since tu,max may not be greater than tG,max- We have:
F _ lF,max, tij 25
lU Atu max-25
(10.29)
The values for the equations (10.29 and 10.30) can be taken from the data sheet for
the GaAs diode TIL 31. We obtain:
AtU,max-25 = tTJ,max - *U,25 = 80°C - 25°C = 55°C
(10.30)
~ -- iltF-,max,t* Tj25
**"
If,max,tr; ?5
>
200 mA 50 mA
(10.31)
AlF = lF,max,tu25 _ 50 mA m l U Atu, ax-25 55°C
mA
= 0-91
(10.32)
For the GaAs diode TIL 31, the maximum forward current lF,max,tu sn can now ^ e
calculated as an example for an ambient temperature of tij 50 = 50 C.
AtTj,50-25 = l U,50 - tU,25
o
50
C-25
o
C
=
25°C
(10.33)
lF,max,tij 5o " ^.max.tjj^S ;
A.I F
- (Atij,50 - 25 ·
)
tF^ax.tij^o = 50 mA
25°C.0-91 mA
=
= 27-25 mA
(10.34) (10.35)
In order to be able to use the graph, stated
for an ambient temperature of tij = 25 C for the GaAs diode TIL 31 and the phototransistor TIL 81, with the "On"
collector current as a function of the sourcedetector distance (see Figure 10.19), a correction factor must be calculated, corresponding to the different temperatures and the diode forward currents.
178
From the data-sheet graph <J>e rej = f(Ip) for the type TIL 31, a relative increase in
the radiant power of about
A<&e,rel = + 10%
is obtained, corresponding to this almost
linear function, for the calculated forward
mA m current iF.max.tu 50 = 27 " 75
'
comparison with the current lF,tij 75
mA = 25
used in Figure 10.19 at an ambient
temperature tyj = 25 C.
The ambient temperature difference
Atij 50-25 between the required ambient
temperature tjj 50 = 50 C and the temperature tjj 25 = 25 C stated in equation (10.30) is 50°C - 25°C = 25°C.
The effect of the ambient temperature difference tjj 50-25 on the change in the "On" collector current of the phototransistor TIL 8 1 can be read off from a function Ic(On) = f(tjj). Sufficient accuracy is obtained by taking the. average of the values read off from Figures 10.23 and 10.24. The change in the "On" Collector current is approximately
AkXOn) for tU = (50-25) = -5% (10.36)
The correction factor for the function IC(On) = f( r) which is shown with an ambient temperature of tjj = 25 C in Figure 10.19 is:
Kf = A$e,rel + I AlC(On)l for tu
= (50-25) °C
(10.37)
Kf=+10% + (-5%)=+5%
(10.38)
The consequent slightly higher "On"
collector current only changes the DC gain
B of the phototransistor TIL 81 to an
AB insignificant extent, so that
can be
neglected.
For the source-detector combination with
the GaAs diode TIL 32 and the photo-
transistor TIL 78, the calculation of the "On" collector current is simpler, since the
figures in the data sheet relate to the ambient temperatures and the coupling characteristic to the data sheet test conditions. With the exception of the
TIL 32/TIL 78 combination, almost every
representation of the coupling characteristic requires a separate evaluation for the individual needs. In addition, every
calculation should be confirmed by measurements. The coupling characteristics
stated apply in principle for all those Si phototransistors, which are in the same photocurrent sensitivity range as that of the types stated. Only phototransistors which differ greatly in actinic value form an exception. For combinations with a specified source-detector distance r or for ready-assembled combinations (e.g.,
optocouplers), the "On" photocurrent Ip(On) °r the "On" collector current
IC(On) °f the photodetector is of interest as a function of the diode current Ip of
the GaAs diode. In a graph, these two values are each shown as a function of the GaAs diode forward current Ip, in most cases for equal source-detector case temperatures of tQ = 25 C or an ambient temperature of tij = 25 C. The ratio of the "On" photocurrent Ip(On) or the "On" collector current Ic(On) to the GaAs
diode forward current Ip is designated as the transmission ratio or transmission factor:
^(On)
IF
(10.39)
kXOn)
If
(10.40)
The typical transmission ratio of a given source-detector combination, or an opto-
coupler can be determined for the desireti
GaAs diode forward current Ip from the graph Ip(On) = fdF) or JC(On) = ^F).
together with the equations (10.39) and (10.40). As an example, Figure 10.20 shows
the "On" collector current Ic(On) as a function of the GaAs diode current Ip
179
;VCE «0-5V: "tu - 25 >C ' 4
2
1
0-4
02
1
=4=
004
t~---
002
/
/
Figure 10.20
"On" collector current iQ(On) as a function of the GaAs diode current If for the TIL
138 optocoupler. This typical function corresponds to a separately constructed source-detector combination with the same cooling conditions, with the GaAs diode TIL 32 and the Si phototransistor TIL 78,
mm for a distance ofr = 3-2
-^^ IfYOnl
TIL102:u =
Ip
6-5 raA
=65% 10 mA
(10.42)
13 mA
TIL 103 : u =
=130%
10 mA
(10.43)
2 mA
TIL 112: u =
S20%
10 mA
(10.44)
i 100
IV-
40 =I B =
1
'c(OnXmA) ·U
10
4
1
04
004 31
3-Z:t
1
/ r-
'/
i
04
1
'
riL 103 1
TIL 102 :±
4
-I
4
10
40 100
'F (mA)
»-
for the TIL 138 optocoupler. For a GaAs diode current of Ip = 10 mA, the trans-
mission ratio is
Ir(On) °'45 mA
u= ( ' =
=4-5%
if
10 mA
(10.41)
The transmission ratio for optocouplers is also dependent on the transmission medium such as glass, epoxy resin or gas. Furthermore, the operating mode, which can sometimes be selected, plays a part. The transmission ratio in photodiode operation is about 1%, in phototransistor operation about 100% and in photoDarlington operation up to 500%. Figure 10.21 shows the "On" collector current Ic(On) as a function of the GaAs diode forward current Ip for the optocouplers TIL 102 and TIL 130. For
a GaAs diode forward current of Ip = 10 mA,
the transmission ratio of the optocoupler is:
Figure 10.21
"On " collector current IcfOnl as a function of the GaAs diode forward current Ip for the optocouplers TIL 102 and TIL 103
The transmission ratio and thus the "On"
receiver output current of a photodetector is temperature-dependent. At low temperatures the radiant power of the
GaAs diode increase, while at high temperatures the "On" collector current of the phototransistor rises. Depending on
the selected combination of components, the temperature drifts of the source and detector compensate for each other within a certain temperature range. In this,
only small changes in the "On" collector
current of the phototransistor occur. Very small optoisolators have almost equal source and detector case temperatures or equal source and detector ambient
temperatures. This makes it possible for
180
= VCE- sv; 40 :lB-o
1
qOnXmA) -tu-2
04 01
^p
X
1
2
4
10
20
100
Figure 10.22
"On " collector current Jc(On) as a function of the GaAs diode forward current If for the optocoupler TIL 112
16
14
lC(On),rel 1-2
the typical detector output current to be shown for the whole combination in one graph as a function of the component case temperature Iq or the component ambient temperature trj.
Figure 10.23 shows the normalised "On"
collector Ic(On),rel as a function of the case temperature tQ for source-detector
combinations of the GaAs diodes TIL 23, TIL 24 and Si photo transistors of the type families LS 600 and TIL 600. The normalised "On" collector current IC(ON),rel is shown as a function of the ambient temperature ty for the optocouplers TIL 102 and TIL 103 in Figure 10.24 and for the optocoupler TIL 112 in Figure 10.25.
1,4 lC(On),rel
1
VCE - 5 V
lF = 0mA
1,2
1,0
^
Emit er
Oelec tor L..600
5 3 Oder 1riL 24 - -TIL 601-TU 616-
if- 0mA
VCE V
-li -50 -25
50
75
100 125
'U (OQ
--
-75 -50 -25
25
50
75
100 125
tc( C) For emitter -
and detector
Figure 10.23
Dependence of the "On" collector current iQOnl °f a Si phototransistor from the LS 600 series or TIL 600, produced by the radiation of the GaAs diodes TIL 23 or TIL or TIL 24, on the source and detector case
temperatures tQ, which are equal. The
relative "On" collector current has been normalised to the value iQOnJ.rel = 1 f° r
a case temperature of tQ = 25 C
Figure 10.24
Normalised "On" collector current iQONI.rel as a function of the ambient temperature t[/for the TIL 102 and TIL 103 optocouplers
10.3.2 Contrast ratio
A very important criterion, related to
the detector output signal, is the contrast ratio. This is defined as the ratio of the
"On" collector current Ic(On),N ( as a result of useful radiation falling on the
181
1-4 'C(On),re!
VCE = 5 V I F =1 lB-0
Aperture and its arrangement between the source and detector,
Opacity of the material to be scanned (degree of opacity to radiation).
50
75
100 125
'U (°C)
--
Figure 10.25
Standardised "On" collector current IC(On),rel as a function of the ambient
temperature t\j for the optocoupler TIL 112
phototransistor through a marking hole in
a card or tape material) to the "On"
collector current Ic(On),S ( as a result °f stray or background radiation falling on the phototransistor through a card or tape
material).
For stable optical scanning conditions, the contrast ratio should be as large as possible.
A contrast current ratio IcCOrO.N/kXOrO.S
= 50/1 can still be simply evaluated by an
A electrical circuit. lower contrast current
ratio gives response levels in very narrow ranges, so that the circuit design becomes more difficult. The contrast current ratio is determined by several factors, which interact on one another:
Source and detector parameters,
Distance between source and detector,
Location of the punched card or tape between the source and detector,
Coupling medium (if any)
8 Cross-talk between individual channels.
For every application, the contrast current ratio should be calculated roughly with the factors listed and
confirmed by practical measurements. The GaAs diodes selected should still show an adequate radiant intensity for their energisation, for example with
TTL circuits, with a small forward current of about Ip =15 mA. The
phototransistor selected should still
produce a reasonable "On" collector
current Ic(On) f° r driving the
subsequent amplifier, TTL or interface
circuits.
To achieve optimum contrast current ratios, hole masks of opaque material,
card or tape material of high opacity and screening tubes for each channel to avoid optical cross-talk are very decisive factors. If a high contrast current ratio is specified for card or tape material with low opacity, then as the most effective protection against stray radiation and optical
cross-talk, each optoelectronic component will be installed in a spearate tube with a mask fitted to it..The tube and mask will advantageously be anodised or sprayed matt black. Figure 10.26 shows the contrast current ratio Ic(On),N/ I C(On),S of different card and tape materials as a function of the GaAs diode current Ip. The
source-detector distance r is the variable.
The data recorded relates to a mechanical arrangement, which had been equipped with separate tubes and hole masks, with an
aperture diameter of 0-9 mm, for each
182
Figure 10.26 Contrast current ratio IC(Onj,N^C(Onj,S of various card and tape materials as
functions of the GaAs diode current Ip,
with the source-detector distance as variable
channel, in order to eliminate cross-talk. Because of its low opacity, the white oil paper shows the lowest contrast current ratios. With black paper, contrast current ratios of greater than 1000:1 are
already achieved. The Mylar materials give the most favourable values.
The "On" collector current of reflection
photocell units is stated for a reflecting
surface of known reflectivity and its
distance from the photocell unit.
Among others, the following are used as
reflection surfaces:
Neutral white paper with 90% diffuse
reflection (e.g., Eastman Kodak),
Mylar magnetic tape (Registered Trade
Mark of Du Point),
Aluminium foil 0-001 inch thick, which is used as strips for the start and end of magnetic tapes.
For the TIL 139 reflection reading head,
the reflection surfaces listed are arranged
at a distance of r = 0-150 inch in each case.
The "On" collector current Ic(On) * s
mA measured for a current Ip of 40
flowing
through the GaAs diode TIL 32, a collector-
emitter voltage of V^E = 5 V applied to
the phototransistor TIL 78 and at an
ambient temperature of tij = 25 C. During
this, no radiation from the surroundings may
fall on the reflecting surface or directly on
the phototransistor. The "On"
collector current of the reflection reading head TIL 139 for the various reflecting surfaces is:
IC(On),min lC(On),typ
White Kodak
paper
10/M.
15//A
Mylar magnetic tape
5/iA
7/JA
Aluminium foil 100 /ZA
130 /iA
Very small optoisolators or optoelectronic reading heads should be rigid and mechanically stable, since the mechanical construction has a very decisive effect on the coupling characteristic and thus on the output signal of an electro-optical system.
Phototransistors are usually preferred as photodetectors because of their internal
183
gain. As a consequence of their slim glass rod housing, the phototransistors of the LS 400 family are suitable for installation in tubes. The phototransistors of the LS 600 type and TIL 600, and also the GaAs diodes TIL 23 and TIL 24 are obtainable in small pill housings. These pill housings permit easy mounting and convenient soldering on double-coated circuit boards, which, as usual, should be free of grease and sprayed with a soldering lacquer. These component
families contain small glass lenses, so that no additional fibre-optics or external lenses
are needed for focussing the GaAs diode radiation on the phototransistor.
10
.
Ie,rel
1
z
0-6
0-2
/
80o
40O
3
40°
800
10.4
Half-power and Half-value Points
The half-power point, HP, marks the two points on a distribution curve, where the power has fallen by 3 dB in comparison
with the maximum power Pmax-
max PHP :
(10.45)
For photodetectors, the half-value point HP
denotes the angle of incidence 4> with respect to the normal, at which the photo current sensitivity spjp has fallen
by 50% in comparison with the photocurrent
sensitivity s in the normal direction.
SHP :
(10.46)
The designation "half-power point" is
inappropriate for photodetectors, since
a drop in power by 0-5 times corresponds to a drop in sensitivity of only 0-707 times.
For radiation sources, the half-power
point HP determines the emission angle,
$, with respect to the normal, at which the radiant intensity I e j-jp has fallen by
50% in comparison with the radiant intensity
I e in the normal direction.
Figure 10.27 Radiant intensity distribution curve with
the half-power points HP at ip= 58-5 ,
in the Cartesian coordinate system for
the GaAs diode TIXL 06.
--Ie,o
Ie,HP=
(10.47)
Figure 10.27 shows the radiant intensity distribution curve with the half-power
points HP in the Cartesian coordinate system for the GaAs diode TIXL 06 and
Figure 10.28 shows the curve in the polar
coordinate system for the GaAs diode TIXL 16. The angle (/? is measured between
the half-power point, HP, and the normal. The angle between the two half-power points is also called the aperture angle 0.
0=2^
(10.48)
The aperture angles for the two GaAs
diodes are:
TIXL06: 0= 115° TIXL 16: 0= 150°
For emission spectra, the half-power point, HP, marks the wavelengths Apjp, at which
the spectral radiant power ^e^Hp only
shows the half value as compared with the
184
20°
ioo
\^\~
30°
5cxC\
V\ 40»
\\
50° r\ '
\x \ ^V*
J x" v \ V
60°
70°
yC^><HP^yCN^
80°
90° l^^53§l
Ie,rel
10°
20°
30O
%
90
§ 80 o
70 \
40° V' v
60-/
x/ > 5o/- /
500
40^.
'\
60o
30 _/
700
20 1
%/£AJJP-k
800
900
Figure 10.28 Radiant intensity distribution curve with
HP the half-power points at <p= 75 , in
the polar coordinate system for the
GaAs diode TIXL 16
*e,X,re!
maximum spectral radiant power <t>e \ max emitted at the maximum wavelength Xmax .
$e,X, max
*e,X,HP
(10.49)
Figure 10.29 shows the relative spectral radiation power $e,X,rel as a function of
the wavelength X for the Si-doped GaAs diode TIXL 16. The maximum wavelenght is Xmax = 933-5 nm.
The half-power points occur at
Xj-ip i = 956 nm
and Ajip 2 = 911 nm.
The bandwidth AX is:
AX=^HP1-^HP2 A\=956nm-911 nm = 45 nm
(10.50) (10.51)
For the lifetime of light-emitting or
luminescent diodes, the half-power point HP
is used to define the operating time, at which the total radiant power of an IRemitting diode or the total luminous power of light-emitting diodes has fallen to 50%, as
compared with the initial v^lue. One
definition relates the initial value to the
radiant or luminous power of unused
930 920 910 HP2
Figure 10.29 Relative spectral radiant power Qe,\rel as a function of the wavelength Xfor the
Si-doped GaAs diode TIXL 16. The half-
power points lie at X///>/ = 956 nm and A///>2 = 911 nm
Operating dme(HRS)-
Figure 10.30
Basic life-time curve of a GaAs diode. The relative radiant power ^?ere \ is shown as a function of the operating time. For the parameter Ip = 500 mA, the life is characterised by the half-power point at 30 000 hours
185
o
components. Other definitions which are often used omit the relatively high initial ageing (5% to 20%) during the first
20 - 100 operating hours. In these, the
initial value is related to an operating time after twenty to one hundred hours.
Figure 10.30 shows the basic life curve of a luminescent diode. The relative radiant
$ power e rei is shown as a function of
the operating time. The value of the forward
current Ip affects the lifetime considerably.
The half-power point for the device life-
time, with Ip = 500 mA and Iq = 25°C,
lies around 30 000 hours. After this time in operation, the luminescent diode is by no means destroyed, but still works satisfactorily but with correspondingly lower radiant power.
Figure 10.31 shows (with the forward
*'
50 *
S
s | '0 (V"
i:
|:
I:
l!
I.
j> -
.'' ' tt.
w ty \<s& H
1
±+
«yC^/ r\%
L-'
*>
Lr
j~
^
tti-
r"
t _T i-
|
ml
J
1*-^
lili
vt
1
K t c 25
3 13
10 3
II II 10*
10 s
Operating time (HRS)
»
Figure 10.31
Decrease in the radiant power of a "Low Power" GaAs diode as a function of the operating time, with Ip as a variable
10.5
Dynamic Data
The dynamic data is understood to refer to the switching time performance of electronic components. Figure 10.32 shows the theoretical test circuit for measurement of the switching time of
source-detector combinations. A GaAs
diode is operated in the pulsed mode. The pulsed radiation falls on a phototransistor. The input pulse and the output pulse can be observed on an oscilloscope to determine the actual switching time.
Input
--
®*(D1 Output
D RL»1 kii
VCE = 30V.
L
Figure 10.32 Test circuit for measurement of the swtiching time performance for the TIL 138 miniature optoisolator. The amplitude of the input pulse is adjusted
for an "On" collector current ofI(^fQnj = 500 yA. The pulse generator is to have a
specification of: Zout = 50Q tr <100 ns,
tf^lOO ns, mark-space ratio ^1:1
In Figure 10.33, the most important times are illustrated:
input pulse
current Ip as a variable) the percentage decreases in the radiant power as i function of the operating time. This very clearly shows the slight decrease in the radiant power for small forward currents Ip and the high decrease in the radiant power for large values of Ip.
Figure 10.33 Definition of switching times
186
Output pulse
The switch-on time t on is the time, in
which the amplitude rises from 0% to 90%
of the final value.
The switch-off time t ff is the time in which the amplitude falls from 100% to 10% of the final value. The delay time t^ is the time, in which the
amplitude rises from 0% to 10% of the final
value.
The storage time t s is the time, in which
the amplitude falls from 100% to 90% of
the final value.
The rise time t r is the time, in which the
amplitude rises from 10% to 90% of the
final value.
The fall time tf is the time, in which the
amplitude falls from 90% to 10% of the
final value.
The pulse width tw is given by the time
interval which lies between 50% of
the amplitude of the rising edge and
50% of the amplitude of the falling edge.
Separate measurement of the switching times of GaAs diodes or of photodetectors necessitate that, in the first case the measuring photodetector and in the second case the GaAs diode as the measurement emitter, must have considerably shorter switching times than the test device. Fast avalanche photodiodes are suitable as photodetectors for GaAs:Zn, GaAs: Si, GaAsP and GaP diodes. The GaAs-Zn diodes are suitable for use as measurement emitters for commercial Si photodiodes and Si phototransistors.
The switching time characteristics of non-amplifying junction photodetectors is determined, down to rise times of 10 ns,
by the RC performance. In practice, the
rise time will be determined by
R C tr = 2-2 .
.
(10.52)
In this expression, as an approximation the value of the total component capacitance from the relevant data sheet can be inserted for C.
Figure 10.34 shows as an example, the
. 100
:f-iMH 1G-25«C
]
1 1
»R(V)-
Figure 10.34
Total capacitance C as a function of the applied blocking voltage Vr for the
Si avalanche photodiode TIXL 56.
total capacitance C in pF as a function of
Vr the applied blocking voltage
for the
Si avalanche photodiode TIXL 56.
As a further simplification, the connected
load resistance Rl can be inserted for the resistance value R. From equation
(10.52), a photodiode capacitance of
100 pF and a load resistance of 1 k£l
give rise time of
tr = 2-2 . 100 . 10" 12 As/V . 10 3 V/A
= 220 ns
(10.53)
Commercially available Si photodiodes have rise times from about t r = 1 00 ns to 10 jUs. Si-PIN photodiodes, on the other hand, have rise times of about t r = 10 ns. Si avalanche photodiodes have even better values, their rise times lie around tr = 0-35 ns.
The switching times of phototransistors are mainly determined by the Miller capacitance between collector and base. This Miller
capacitance is dependent on the DC gain
187
B. Phototransistors have rise times of about 5 to 50 jUs.
The switching times of GaAs diodes are substantially determined by the lifetime of
the charge carriers in the crystal. Red-
emitting GaP diodes with the wavelength Xn = 690 nm, which have the PN junction formed by growing a Zn and O-doped
P-layer, in the liquid epitaxy process, onto a
Te-doped N-type epitaxial GaP layer (double-epitaxy diode), have average rise and fall times of (t r + tf)/2 = 200 ns to 500 ns. Components made by the vapour epitaxy process show shorter times.
The green-emitting GaP diodes with the
wavelength Xmax = 565 nm have average
rise and fall times (tr + tf)/2 = 50 ns. The same value applies for the yellow-emitting GaP diodes with the wavelength Xmax = 590 nm.
On the other hand, the red-emitting GaAsP diodes with the wavelength Xmax = 650 nm
t. VB -20
p pc 10
4
1
04
100
1 lB =
" '
'F(On) = 5 r
/ tw 100
mA 1 :(On)- u 51
, :!
t^
1000 R L(J2)
10000
Figure 10.35 Average switch-on and switch-off times (ton + t ffj/2 as a function of the load
resistance R^ for the optocouplers
TIL 102 and TIL 103. Curve 1 applies for phototransistor operation, curve 2 for photodiode operation
have a somewhat shorter rise and fall time of (t r + tf)/2 = 1 nsto 50 ns.
IR-emitting GaAs diodes with the wavelength X,max 890 nm, which have the PN junction formed by the liquid epitaxy process by counter-doping the melt with
zinc during the growing of the epitaxial layer, have average rise and fall times of (t r + tf)/2 = 2 nsto 10ns.
IR-emitting GaAs diodes with the wavelength Xmax = 940 nm, which have their PN junction formed by the liquid epitaxy process from the Si-doped melt, which behaves amphotically in the GaAs, have average rise and fall times of (t r + tf)/2 = 500 ns.
In general, it is also usual to state the average switch-on and switch-off times in
data sheets. As an example of this, Figure 10.35 gives the average switch-on and switch-off times as a function of the load
resistance Rl for the optocouplers TIL 102
and TIL 103.
Furthermore, the dynamic characteristic of a component is marked by the frequency limit fg. This denotes the frequency, at which the power has fallen to a half, as
compared with its maximum value. With
photodetectors, however, the sensitivity is stated, not with a power but with a current or voltage per incident unit of irradiance. Thus, on a constant load resistance Rj^, according to equation (10.54), a fall in power by 0-5 times corresponds to a voltage reduction of 0-707 times and a current reduction of 0-707 times.
VO^ 0-5 P = 0-5 U . J = y/51 U .
I
= 0-707 U . 0-707 I
(10.54)
The limiting frequency of a photodetector is thus that frequency, at which the sensitivity is only 0-707 times the
maximum sensitivity.
As the following transformation shows, the
188
limiting frequency fg and the rise time t r can be related to one another. Starting from
tr = 2-2.R.C The product RC is the time constant
T=R.C
(10.55)
The time constant T is the reciprocal of the angular frequency ca
1
T ="
CJ
(10.56)
The angular frequency uO corresponds to the
angular velocity of a rotating vector, in radians
27T
CO=
= 27Tf
T
(10.57)
The cycle duration Tg at the limiting
frequency fg is obtained from the equation
0-35
tr1 = 10 5 Hz = 3-5 Us
(10.63)
For a photodiode with a rise time of t r 0-1 /Lis, a limiting frequency of
0-35
fse = 0-1 JUS = 3-5 MHz
(10.64)
is obtained.
For Si-doped GaAs diodes, the limiting frequency fg lies around 1 MHz, while for Zn-doped GaAs diodes it is about 100 MHz. With Si phototransistors, a frequency limit between 50 kHz and 200 kHz can be expected.
For photodetectors and opto couplers the whole frequency response is usually stated in the data sheet. The sensitivity
s or the "On" collector current Ic(On) is shown in a graph as a function of the
m modulation frequency f of the
radiation falling on the photodetector.. Figure 10.36 shows the frequency response
F Tg=
(10.58)
If the equations from (10.54) to (10.57)
are inserted in the equation (10.52), then
with the cycles duration (period) Tg, the rise
time t r can be calculated:
2-2 1-1
-- R C tr = 2-2 .
. = 2-2 . r
=
.
T,,
8
U IT
(10.59)
tr
=
0-35
T
g
0-35 tr=-
(10.60) (10.61)
0-35
fg=-
tr
(10.62)
A phototransistor with a limiting frequency
of f,, = 100 kHz thus has a rise time of
vCE = sv
t1
4
Srel 2
1
0-7 0-4
0-2
01 007
004
T
002
--t
I
)fl
X
-RL = 1 kflS
SU
-
j
1
001
1
J
2
4
10 20 40
100
4<X)
f ,,.U 1
1000
Figure 10.36
Relative sensitivity srei as a function of
the modulation frequency fm of an
incident radiation for the types LS 600,
TIL 63 to 67, TIL 600, TIL 78, TIL 81 and for the optocouplers TIL 107 and TIL 108
189
of the phototransistor families LS 600, TIL 63 to 67, TIL 600, TIL 78, TIL 81 and the optocouplers TIL 107 and TIL 108.
1 l= = =vcc -iov:: -lB = -tu'CCOnKmA)
2
"^
:::= R
kb-:
L-475
Rl= loo a
40
100
400 10O0
rm(kHz)
--
Figure 10.37 "On " collector current as a function of
m the modulation frequency f of the
GaAs diode current for the optocouplers
TIL 111 and TIL 112
In Figure 10.37, the "On" collector current Ic(ON) ls shown as a function of
m the modulation frequency f of the GaAs
diode current for the optocouplers TIL 111 and TIL 112.
Figure 10.38 shows the measurement circuit for recording the frequency
response of GaAs power diodes. A load-
independent bias current Ip is applied
to the GaAs diode through a series resistance.
A low-impedance generator is capacitatively
coupled to the GaAs diode. The modulation current Ij^j through the GaAs diode is to
be kept constant during the measurement of the frequency response. Ij^ must not exceed twice the value of bias current Ip.
The modulation current can be monitored on the oscciloscope. In order to avoid possible electrical cross-talk between the emitter and detector, a large emitter-detector distance can be selected.
To focus the radiation from the GaAs diode, a lens can be arranged in the beam path between the emitter and detector. The Si avalanche diode TIXL 56 which is used
as the measuring photodetector is operated
ii
·
100)iF
1
Tantalum r~^ -S
d
ioo kn
.1
1.
47 nF
47 nF
JLf
, 1
S?
S>
1/
1
2
D>
-©
Figure 10.38
Measurement circuit for recording the frequency response of GaAs power diodes. The
m mA following values are to be set: a) GaAs diode: Modulation current I = 100
(peak to
mA peak), independent offrequency; bias current If = 55
constant, b) Photodetector
V^^170 TIXL 56: Dark current with applied voltage
V set to Ip = 0-5 [XA; reverse current
with the aperture open Ir = 10 /JA
190
with optimum gain. For this, with the
Vr * applied reverse voltage
70 V, a
dark current of approx. 0-5, /iA is set.
Although a further increase in the applied
Vr blocking voltage
gives a higher
amplification factor M, at the same time it
gives a higher current noise.
0-9 0-8 *e.rel
07 0* 05
0-4 0-3 0-2
01
1
^ :.
1
10
» fm(MH2)_
Figure 10.39
Frequency response of the GaAs diode
& TIXL 13. Relative radiant power e,rel
as a function of the modulation frequency
fm
'
t 0-9
*e,rel
07
N
0-5 0-4 0-3 0-2
01
1
1
10
Figure 10.40
Frequency response of the GaAs diode TIXL 16. Relative radiant power <$>e rei as a function of the modulation frequency
fm
'
\ 08
*e,rel 0-6
0-4
0-2
01
1
10
Figure 10.41
Frequency response of the GaAs diode TIXL 27. Relative radiant power 4>e rei as
m a function of the modulation frequency f
With the unmodulated radiation at first
falling on the avalanche diode through an
aperture set in the beam path, the reverse
« current Ir is to be set to Ir 10 jLtA. The
input voltage signal appearing on the
Rm measuring resistor
= 50 is amplified
and observed on the oscilloscope. The Input
voltage is a measure of the modulated
radiant power of the GaAs diode.
Figures 10.39 to 10.41 show the
frequency response of Si-doped GaAs power
diodes.
10.6 Reliability of optoelectronic semiconductor
components
The term "reliability of a component" covers component lifetime, changes which occur in the parameters during life and the
probability of the occurrence of a total
failure.
The life of a semiconductor component depends, among other factors, on the electrical loss power, the selection of the working point, the current and power density in the wafer (important for pulse operation), the junction temperature and the overall thermal loading. The life of optoelectronic components relates firstly to the operating condition and secondly to the storage condition. The reliability data should include all important parameters. (It is unnecessary to go further into the
191
reliability prediction of Si phototransistors, since it corresponds to that for ordinary
Si transistors).
The life test data on luminescent diodes
relate to the half-power life point (see Section 10.4). Thermal fatigue, resulting in
wafer breakage, occurs less often in GaAs
diodes than for other semiconductor devices.
Efforts are made to work with relatively low junction temperatures tj, so that a high radiation yield is achieved. The halfpower point of the lifetime is mainly determined by the junction temperature tj and the diode forward current Ip.
Different current densities can occur in the crystal, so that different local thermal loadings occur. With increasing forward current, the active region can displace itself more towards the contact point, which can also lead to failure of the luminescent diodes (see Figure 10.11). In addition, for pulse operation of luminescent diodes, the permissible current and power density in the bonding wire is important. High temperatures can cause changes in the
epoxy re-sin used for lens domes or
housings.
By improved technological processes, suitable component designs, careful component assembly and through the
selection of suitable system supports and plastics, the life of luminescent diodes has been considerably increased.
For protection against environmental
effects, GaP wafers and Si-doped GaAs wafers, made by liquid epitaxy, are mounted
with the P-junction side of the wafer system on the system support. The system support is either a metal base of the transistor case, a mounting bolt, in power luminescent diodes, or the end of a massive square connecting wire in plastic-encapsulated components. Since the electrical loss power mainly causes a heat build-up at the anode, the mounting of heat-sinks is only meaningful at the system support. For example, the mounting of the threaded bolt on a chassis is very convenient for
the power luminescent diodes TIXL 12 to TIXL 15, while the GaAs diode TIL 31, mounted in a TO-46 case, needs a heat-sink
which pushes onto the case.
It was possible to improve the life of Zn-doped GaAs diodes and GaAsP diodes considerably by use of the well-proven
planar technique.
With the data-sheet operating values,
the minimum life of luminescent diodes is about 50,000 hours. For many GaAs
diodes, the typical lifetime is over 100,000 hours. After this time, the radiant or
liminous power has fallen to 50% of the
initial value. Otherwise these components still work satisfactorily (see Section 10.4). With the present state of technology, the probability of the occurrence of a total failure of a luminescent diode is comparable with that of silicon components. The reliability of a component is considerably influenced by the manufacturing process.
The quality control department is responsible throughout for the component
yield. It provides for the necessary direct interventions during the manufacturing process, in order to guarantee the planned and required reliability and quality of the
components produced. Quality control includes the inspection of raw materials on receipt, the necessary inspections and tests during the critical production stages and the final testing and release of the finished product. The manufacturing steps and the frequent inspections and tests, which are carried out by the manufacturer of optoelectronic components, are illustrated in the flow diagrams in Figure 10.42 for the production of Si phototransistors of the pill-case family LS 600 and in Figure 10.43 for the production of GaAs diodes of the pill case types TIL 23 and TIL 24.
In the following sections, life tests on 11 000 phototransistors of the type family LS 600, with ten million component operating hours and on GaAs
192
Crystal wafer
·V- Case assembly
Lens and retaining ring
Meaning of symbols: Material receipt Manufacturing steps 100% inspection of production
o Quality control inspection or lest
o Visual inspection Electrical test
Electrical connections (made by bondingl
Visual inspection
o Visual inspection
o Visual inspection
Visual inspection
Visual inspection, electrical test
Visual inspection Main leak test
D Despatch
o
Electrical test, visual inspection
o Inspection
Figure 10.42
Typical production flow diagram for Si phototransistors of the pill-case type LS 600 family.
193
'V-
V- retaining ring
-o
o~
Meaning ol'iyinhfilv
\ / Material rtwipi
I
j ManufaL'iuriiijt si<p">
IOC.; inspeih.m »l product iun
O Qualify cum...! rupeciion ur k-si
O:
O
6 '""
O
D~
Production flow diagram for GaAs diodes of the pill-case types TIL 23 and TIL 24
diodes of the type families TIL 23 and TIL 24 with over two million component operating hours are described. It is
expressly pointed out, that only previously used components were used
for these. The tests listed are typical of TI sensor and emitter products. The test results have been summarised by means of graphs and tables. In addition, various mechanical and temperature stress tests are
listed.
Test 1 : Operational life test for phototransistors at an ambient temperature
of tjj = 25°C for 1000 hours.
The irradiance E e falling on every photo-
transistor, with the applied collector-
ernitter voltage of VqE = 10 V, gave a
total loss power for each phototransistor
of Py = 50 mW. The dark current Irj and
the collector current 1^ were measured in each case after 0, 250, 500 and 1000 hours of operation. As the criteria of failure
a maximum dark current of lDmax = 0-2 plA and a ± 20% deviation of the collector
current from the original limit were laid down. In all, 3210 phototransistors were
tested to these criteria, while 6 failures
were recorded. The samples tested were taken from batches, which contained
194
over 1,050,000 phototransistors.
Figures 10.44 and 10.45 and Table 10.2 show the evaluation of this operating life test at tjj = 25°C.
Test 2: Operating life test for phototransistors with an ambient temperature of
trj = 55°Cfor 1000 hours.
The irradiance E e falling on each phototransistor LS 600, with the applied
collector-emitter voltage of V^g = 1 V,
gave a loss power for each phototransistor
of Pv = 50 mW. The dark current Irj and
the collector current Iq were measured after an operating time of 0, 1 68 and 1 000
hours. As the criteria of failure a maximum dark current of lDmax = 0-2 jUA and a ± 40% collector-current deviation from the
original limit were laid down. In all, 3356 phototransistors were tested to these criteria, while 1 1 failures were recorded. The samples tested were taken from batches,
!'
Initial value
m
<C< J)
3
OV mW/cm1 '
OC
* / -S *?/
1
12 3
3
4
Final value '("(mA)
»
vct = 10 V
Ee = lU = 25oC
ff
H ' t f t / / t )
)
V
If
£9
Figure 10.44
Change in the initial collector current value after 500 hours operation for 3210
phototransistors of the LS 600 family.
During this operating life test, the ambient
temperature was t\j= 25 C, the applied collector-emitter voltage was V(jE = 10 V and the loss power for each phototransistor
wasfv = 50mW
Figure 10.45
Change in the initial dark current value after 500 hours operation for 3210 photo-
transistors of the LS 600 family. During
this operating life test, the ambient temperature was t\j = 25 C, the applied
collector-emitter voltage was Vq£ = 10 V and the loss power for each
mW phototransistor was Pv = 50
Units tested
3210
Unit hours
2 847 000
Total failures
Total 6
Failures due to parameter deviations
% Failure rate in per 1 000 h
60% probability 90% probability
Meantime
to failure
0-20
0-33
700 000 h
Table 10.2
Evaluation of the operational life test after an operating time of 1 000 hours for the phototransistor family LS 600. The ambient temperature was trj= 25 C.
195
which contained over 2,064,000 photo-
transistors.
The results and the evaluation of this
operational life test, carried out at
tij = 55 C are shown in Figurures 10.46
and 70.47 and in Table 10.3.
Test 3: High- temperature storage test for
the phototransistor family LS 600.
The phototransistors were stored in the oven at tij = 150 C for 500 and for 1000 hours. The dark current Itj and the collector current \q were measured after
" '"i 1 1 >C(n.A)
vCE = 10 V Ee = 20 mW/cm1
Mj = 55°C
/-
Z <V ·/
Pa /
'C <mA)
VCE = 10 V
Ee =
tU=55°C Initial Id(M)
Wyfi
Zr %/ 7$
x·
·H
'D(M)
Figure 10.46 Change in the initial collector current value after 1000 hours operation for
3356 phototransistors of the LS 600
family. During this operational life test, the ambient temperature was trj= 55 C, the applied collector-emitter voltage Vqj?
= 10 V and the loss power per photo-
mW transistor Pv = 50
Figure 10.47
Change in the initial dark current value after 1 000 hours operation for 3356
phototransistors of the LS 600 family.
During this operational life test, the ambient temperature was trj = 55 C,
the applied collector-emitter voltage Vce '' 10 V and the loss power per phototransistor Pv = 50 mW.
Units tested
3356
Unit hours
3356000
Total
failures
Total 11
Failures due to parameter deviations
% Failure rate in per 1 000 h
60% probability 90% probability
Mean time
to fallure
0-36
0-49
300 000 h
Table 10.3 Evaluation of the operating life test after 1 000 hours operation for the phototransistor
family LS 600. The ambient temperature was trj = 55 C
196
a storage time of 0, 250, 500 and 1000 hours. As the criteria of failure a
maximum dark current of iDmax = 0-2 /M. and a 20% collector current deviation from
the original limit were laid down. In ail, 1829 phototransistors were tested to these criteria, while six failures were recorded. The samples tested were taken from batches, which contained over 745 ,000
phototransistors.
The Figures 10.48 and 10.49 and Table 10.4 show the evaluation of the high-
temperature storage test carried out at tTj= 150°C.
Test 4: Long-term reliability test on phototransistors LS 600.
The long-term reliability of the phototransistor family LS 600 is illustrated by
r
value (mA)
VCE · 5 V tU-150°C
8f
«
9^
£·
Imml
value
IpiuA)
Vct - 30 V =0 fc c
tu = ] 50 >*'
4>
sS*'
c A <jIl >
'D(uA)
Figure 10.48 Change in the initial collector current value
after a storage time of 5000 hours for 1829 phototransistors of the LS 600 family. During storage, the ambient
temperature was t\j= 150 C
Figure 10.49 Change in the initial dark current value
after a storage time of 500 hours for 1829 phototransistors of the LS 600 family. During storage, the ambient temperature
was t\j= 150 C
Units tested
1829
Unit hours
963 500
Total
failures
Total
Failures due to parameter deviations
Failure rate in % per 1000 h
60% probability 90% probability
Meantime
t0 fallure
0-78
1-1
160 000 h
Table 10.4
Evaluation of the high-temperature storage test after a storage duration of 1000 hours.
During storage, the ambient temperature was t\j = 150 C
197
20%
Measured values
10%
^^Z-~^ AIC/%
+20
-
.
Projecte d values
·
'
Mean value
^--- " *
-10%
--
.
)
1000
10000
10OO0O
Operatinz time (HRS>
Figure 10.50
q Change in collector current A/ as a function of the operating time in a long-term reliability test at trj = 25 C for the phototransistor family LS 600
the change in collector current Al£ as a
function of the operating time, in Figure 10.50 for an ambient temperature of
tu = 25 C and in Figure 10.51 for an
ambient temperature of ty = 55 C. Here, all test results during a three-year reporting period were evaluated. The projected discrepancy limits are based on an exponential failure distribution.
Test 5: High-temperature reverse blocking
voltage test on LS 600 phototransistors.
This high-temperature reverse blocking voltage test was carried out for 1000 hours with an applied collector-emitter reverse
voltage of V^E = 45 V in a dark oven
with an ambient temperature of 150 C. The dark current Ir_>, the breakdown
AIc/%
Measured values
Projected values
Mean value
'-30%
Operating time (HRS)-
Figure 10.51
Chance in collector current Al£ as a function of the operating time in a long-term reliability test at trj = 55 C for the phototransistor family LS 600
198
Units Unit tested hours
Total
failures
3 320 3 320 000
Total 11
Failures due to parameter deviations
% Failure rate in per 1000 h
60% probability 90% probability
Mean time
t0 fallure
0-38
0-5
300 000 h
Table 10.5
Evaluation of the high-temperature reverse blocking voltage test on phototransistors of the LS 600 family after a storage time of 1 000 hours. During the storage time, the ambient
temperature was tjj= 150°C and the applied collector-emitter blocking voltage Vqe = 45 V
voltage Vc£o and the collector current
Iq during irradiation were measured after a storage time of 0, 168 and 1000 hours. As
the criteria of failure, a maximum dark current of Ijj) max = 0-2 JUA and a 20%
deviation of the collector current from the original limit were laid down. In all, 3320 phototransistors were tested to these criteria, while eleven failures were recorded.
Vce = 45 V Ee = 20 nH/cm" tU=150OC
'v"a'lV*u'e 'C(mA)
A
/
/ -7
/AT^
'eo*A)
Figure 10.52 Change in the initial collector current value after a storage time of 1000 hours for
3320 phototransistors of the LS 600 family.
During the storage time, the ambient
temperature was ty = 150 C and applied
collector-emitter blocking voltage Vqjt =
45 V
The samples tested were taken from batches, which contained over 2,027,000
phototransistors.
The results and the evaluation of this high-
temperature blocking voltage test, carried out at ty = 150 C, are shown in Figures 10.52 and 10.53 and in Table 10.5.
Test 6: Operational Life Test on GaAs
diodes.
The GaAs diode families TIL 23 and TIL
24 were also subjected to various operational life tests. The results of one operational life test are summarised in
Table 10.6. In this, the GaAs diodes
were tested at various ambient temperatures with different forward current Ip. The evaluation relates to the average percentage deviation in radiant power for the test conditions stated in the data
sheet (Ip = 50 mA, ty = 25 °C).
Test 7: Operational life test on GaAs diodes.
For the GaAs diode families TIL 23 and
TIL 24, further 1000-operating hour
life tests were carried out at ambient
temperatures of tij = 25 C and tjj = 55 C. The diode forward currents were: Ipj =
10 mA; Ip2 = 30 mA and Ip3 = 50 mA.
The radiant power was measured with a
solar cell after 0, 168, 500 and 1000 hours.
The diode forward voltage was also
measured at the same time intervals. No
significant deviations of the diode forward
voltages were detected. In all, 77 components
mA were tested with Ipj = 10
and 98
199
Quantity
Test conditions
Test time (hours)
A*e
Total failures
104
I F = 10 mA; tu = 25°C
104
lF = 30 mA; t\j = 25°C
104
lF = 50 mA; ty = 25°C
104
lF = 10 mA; tu = 55°C
104 lF = 30mA;tu = 55°C
104
lF = 50mA;tu = 55°C
100
lF = 10mA;tu = 25°C
100 lF = 30mA;tu = 25°C
100 lF = 50mA;tu = 25°C
100
I F = 100mA;tu = 25°C
1000 1000 1000
168 168 168 500 500 500 1000
+1-680% -0-420% -4-089% +1-488% -0-4761% +1-470% +1-856% -1-177% -1-167% -1-188%
Table 10.6
A% Summary of a life test for the GaAs diode family TIL 23 and TIL 24.
relates to the
average percentage deviation in radiant power for the test conditions stated in the data
sheet (I F = 50 mA, tu = 25°C)
m components with Ip2 = 30 A. In these
tests, no significant deviation of the radiant
power of greater than -10% was found. A total of 96 components were tested with the current Ip3 = 50 mA, among which four GaAs diodes exceeded a -20% deviation (maximum -27%).
Furthermore, operational life tests,
extended to 4000 hours, were carried out with 300 GaAs diodes. These confirm the extrapolated curves in Figures 10.54 to
10.59.
Initial 'D(mA)
value
VCE = 45V
Ec = tu= 150 °C
0,01
// ·
7&
0.0OO1
· ·/· j/
·
r
·r
ft ·
0,01
0,1
D Final value 1 ((jA)
^
Figure 10.53 Change in the initial dark current value after a storage time of 1000 hours for 3320
phototransistors of the LS 600 family. During the storage time the ambient temperature was tr/= 150 C and the applied collector-emitter voltage Vq£ = 45 V
200
_
-- +20 -
+ 10
--
(*)
_
-10 --
Mean value
-20 -
-
nun
i
i
i
-In
1
II
1
1 1 II III
10 000
100 000
Operating lime (HRS)
»-
nun
1 1
98 devices
nun
i
i
i
10 000
nun
i
i
Operating time (HRS) -
Figure 10.54
Percentage change in radiant power A4>e as a function of the operating time for the GaAs diode families TIL 23 and TIL 24. During the operational life test, the ambient
temperature was trj= 25 C and the diode
forward current Ipj = 10 mA
Figure 10.56
Percentage change in radiant power A4>e as a function of the operating time for the GaAs diode families TIL 23 and TIL24. During this operational life test, the ambient
temperature was t\j = 25 C and the diode
forward current Ifj = 50 mA
-
+20
+ 10
-
A*e
(%) °
=
-10
-20 -
~~
Mean value
-2a
98 devices
nun
1
1
nun
1 1
10000
Operating time (HRS)
11
100000
^
t 1 +20
_ + 10
1_ A*e
(%) °
__^fMva/ue
-10
-20
|
J 1 X)
~
^r^--~ ~27~-;::::
1 III lllll
1000
98 devices
MINI
!
1
1
1
1 1 1 llll
10 000
100 000
Operating time (HRS)
»-
Figure 10.55 Percentage change in radiant power A<J>e as a function of the operating time for the
GaAs diode families TIL 23 and TIL 24. DUring this operational life test, the ambient
temperature was trj = 25 C and the diode
forward current Ip2 = 30 mA
Figure 10.57
A$ Percentage change in radiant power e
as a function of the operating time for the GaAs diode families TIL 23 and TIL 24.
During this operational life test, the ambient
temperature was trj = 55 C and the diode
forward current If] =10 mA
201
+10
-10
_J^>_value
-20 ^~*> -
i i iiiiiii
98 devices
i i mini
10 000
mm
i
i
100 000
Operating time (HRS)
»-
Figure 10.58
Percentage change in radiant power A<J>e as a function of the operating time for the GaAs diode families TIL 23 and TIL 24. During this operational life test, the ambient
temperature was tjj = 55 C and the diode forward current Ip2 = 30 mA
\
1
20
10
4*c °
(«,)
-10
20
1
Mean value -2(7
1 ! 1 lllll
|
1000
<J8 devices
10 000
1
lllll
1
1
100 000
Figure 10.59
A$ Percentage change in radiant power e
as a function of the operating time for the GaAs diode families TIL 23 and TIL 24.
During this operational life test, the ambient
temperature was tjj= 55 C and the diode
forward current Ipj = 50 mA
Test 8: Storage temperature test on GaAs diodes.
In addition to the previous tests, a storage
temperature test was carried out for the GaAs diode families TIL 23 and TIL 24.
172 components were stored at an ambient
temperature of tjj = 85 C for 1000 hours. In this test, only two GaAs diodes showed a change in radiant power greater than -5% (maximum - 1 3%). The forward voltage
did not change in any component.
Test 9: Operational life test under extreme temperature conditions.
In some applications, it is not the life, but the reliability under extreme stress conditions which is the decisive factor for a component. For example, for semiconductor devices in motor vehicles, only operating lives of the order of 3000 hours are called for. However, the environmental conditions necessitate reliability over an extreme
temperature range of about tjj = -40 C to
tij = +120°C. Table 10.7 shows results on temperature shock tests from ty = -40 to +130 C on nine optical detectors with the GaAs diodes TIL 23 and the photo transistors TIL 81. In this test, the components were
operated outside the permissible data-sheet
conditions. The phototransistors and the GaAs diodes of the individual light-links were inserted in the holes provided in circuit boards. They were not specially optically aligned. The arrangement was of very stable construction. The source-detector distance
(lens to lens) was 6 mm. Only previously
unused components were used. The diode
m forward current was Ip jil 23 = 10 A,
and the applied collector-emitter voltage
VCE,TIL81 = 8V.
Test 10: Environmental conditions test.
The environmental condition tests carried out for the phototransistors LS 600 and the GaAs diodes TIL 23 are shown in Table 10.8. The product test specimens showed no total failures and only one parametric failure. Here it must be mentioned, that these test conditions imposed by customers do not represent the most severe component stresses which can be tolerated. In most cases, the component can withstand considerably more severe loading conditions.
202
c B
to
W s
1
Light-link No. and collector current Ic(On)
Temp. Time 1
2
(°C)
3
4
5
6
25 -40
+130
13.00 900 HA 545 MA
14.00 475 /(A 180 MA
mA 14.20 965 /iA 1-2
615 MA 1 mA 550 mA 102 mA
490 mA 800 mA 500 MA 690 MA 620 MA 940 MA 485 MA 990 MA
7
630 MA 500 MA 600 MA
8
9
1-25 mA 410MA 1-07 mA 260 MA
970 MA 550 mA
-40 14.40 460/iA 180 MA 490 MA 810 MA 500 MA 725 MA 495 mA 1-1 mA 275 mA 2 +130 15.00 960 /iA 118mA 610 MA 930 MA 480 MA 970 mA 590 MA 960 MA 540 MA
-40 15.20 460 //A 175 MA 485 MA 785 mA 490 MA 700 MA 475 MA 1-06 mA 265 MA 3 +130 15.40 955 mA 118 mA 610MA 930MA 480MA 970 MA 585 MA 960 MA 540 MA
-40 16.00 450MA 175 MA 490 MA 800 MA 490 MA 710/lA 480 MA 1-07 mA 272MA 4 +130 16.20 940 MA 116 mA 610 MA 930 MA 480 MA 965 MA 580 MA 950 MA 540 MA
-40 16.40 Not evaluate
M
5 +130 17.00 940 mA 1-16 mA 610 MA 930 MA 480 MA 965 MA 580 MA
25 8.00 820 mA 495 mA 610MA 970MA 530 MA 1-01 mA 600 mA
950 MA 540 mA
1-2 mA 410 MA
Table 10.7
Temperature shock tests of nine light-links with GaAs diodes TIL 23 and phototransistors
mm TIL 81. Parameters: If, TIL
23
=
10
mA >
VCE,TIL
81
=8
v >
r= 6
203
MIL-STD-750 Test methods
1051
Environmental condition test
Temperature cycles:
5 cycles, 30 min, +40 C to+100°C 5 cycles, 30 min, -65 C to+l25°C 5 cycles, 30 min, -40 C
to +100°C
5 cycles, 30 min, -65 C to+150°C
Quantity
Paramet ric changes and
total failures
LS600 TIL 23 LS600 TIL 23
2100 126 110 50
1056
Temperature shock:
5 cycles, Condition A
126
50
1021
Moisture resistance
126
50
2016
Mechanical shock test:
1000 g, 5 each axis, 0-5 ms
1500 g, Z\ axis, 0-5 ms
126 146
2056
2006 2046 1001
Vibration test, variable frequency: 10 g 20 g
Constant acceleration: 1000 g, 1 min 2000 g, 1 min Z\
126 146
126 146
Vibration fatigue: 10 g
126
mm Air pressure: 15
Hg, 45V 126
2026
Solderability: 240 C, 3 min
126
Airtight enclosure:
1071
Test condition E
2100 390
1
Table 10.8
Results of environmental condition tests for the Si phototransistors LS 600 and the GaAs diodes TIL 23
204
11
Parameters of Luminescence
diodes
11.1 11.2 11.2.1 1 1.2.2
11.2.3
11.2.4
1 1.2.5
11.2.6
1 1.2.7
11.3 1 1.4 1 1.5 11.6 1 1.7
Quantum efficiency
Thermal calculations
Basic prin ciples
Determination of the thermal resistance Rfh of luminescence diodes Loss power calculation for luminescence diodes in plastic packages Loss power calculation for luminescence diodes in metal cans with infinitely large
heat sinks Loss power calculation for luminescence diodes in metal cans without heat sink Loss power calculation for luminescence diodes in metal cans with heat sink
Calculation of the maximum permissible forward current Ip max
Radiant power Radiant efficiency r?e
Spectral radiant efficiency Electrical parameters Pulse operation
205
11 Parameters of Luminescence Diodes
11.1
Quantum efficiency
The quantum efficiency Qtj of a
luminescence diode is the quotient of
the number of photons npn emerging per unit time divided by the number of electrons n£ flowing through the diode
in unit time:
_
nPh .
"E _
nPh
° t" t nE
(11-1)
This shows, that the quantum efficiency of a radiation emitter, as compared with that of a radiation detector, which was defined in
equation (9.1), is defined in exactly the opposite way.
The quantum efficiency Qj) of a luminescent
diode, emitting almost monochromatically, is calculated in simplified form with the
maximum wavelength Xmax .
Number of photons:
_ *e,X-t nPh :
h.y
(9.2)
$e,X- \max . t
nph : h. c
Number of electrons:
(11.2)
I F .t E n :
(11.3)
The equation (11.2) and (11.3) are inserted in equation (1 1.1) and the constants are simplified as in equation (9.5):
_ $e,X ^max · * · e
Q(X) D
h F c .
.I
.t
6
10
A
IF
1-24 mW
(11.4)
Here,
the
determination
of
<$>
e
X
is
difficult, since the total radiation power
3>e X has to be included.
If
the
total
radiated
power
d> e
^
is
measured in accordance with the
arrangement described in Section 12.7,
with a Si solar cell, then the value of
4>e X must be calculated from the shortcircuit current Ip and the quantum
efficiency Q(X)gc of the solar cell by
means of the equation (9.4)
Ip h · c
Q(X)SC : d>e,X-Xe
IP
$f( e.X'X
1-24 mW
6
10
A
(9.4)
The quantum efficiency Q(X)d of a luminescence diode can be calculated as
the ratio of the short-circuit current Ip of the solar cell to the product of the forward current Ip of the luminescence diode and
the quantum efficiency Q(X)§q of the solar cell. For this, equation (9.4) is inserted in equation (11.4). The following is
obtained: ,
Q(X)D
Ip.X,max
mW 1-24
. 10° A
Q(X)sc-X-lF 10. A. 1-24 mW
IP
IF · Q(*)sc
(11.5)
Example:
The quantum efficiency Q(X)jj of the GaAs power diode TIXL 27 is to be determined
with the following parameters:
207
I F,TIXL27
=
300
mA :
!p,SC
=
10-5
mA
Q(X) SC = 70%
We obtain:
IP
Q(^)TIXL27 =
If · QOOsc
10-5 mA
= 0-05 = 5%
300 mA . 0-7
(11.6)
The quantum efficiency depends on the
construction of the luminescence diode. In this connection, especially, any total reflection inside the wafer system
considerably reduces the quantum
efficiency.
The semiconductors which are suitable for mass production of luminescence diodes have relatively high refractive indices. Gallium arsenide, for example, has a refractice index of n = 3.6. The radiation emitted from a point near the junction will be reflected, if the angle between the normal to the surface and the incident ray
is greater than the angle for total reflection.
By suitable wafer geometries, the angle of
incidence of a ray at the crystal-to-air boundary can be reduced and thus the total reflection can be decreased. (See Section 6.6). Flat wafer geometries have a small emission angle for each surface element, since, in accordance with equation (6.32),
the critical angle for GaAs is only Oq =
16-1°.
n Luft sinOg
1
0-278 -+<Xq = 16-1
nGaAs 3-6
(6.32)
Above the critical angle, total reflection occurs. Figures 11.1 to 11.6 show the
emission conditions for flat wafer
geometries. For diffused GaAs wafers, the geometry shown in Figure 11.2 has proved its worth. For example, the GaAs diode TIXL 06 is constructed from such
wafers. In the TIXL 06, an epoxy dome
is dispensed with, since as a high-quality
diode it has to have a very good emission characteristic and a high temperature
tolerance.
Contact *
m~ 69o
-,*pA
Contact
\
\
Lilerai emission
Figure 11.1 Flat epitaxial or diffused wafer. This
produces an unfavourable quantum efficiency, since, as a result of total reflection, the greater proportion of the radiation cannot leave the system. The radiation of the surface elements far away from the centre point undergoes almost
complete total reflection
i
/
^ /
/
N
v> *£~
v
'
Contact
P
Contact
Figure 11.2
Flat diffused wafer. Its quantum efficiency is somewhat better, as compared with that in Figure 11.1. Radiation is only produced in the central part. The losses of the
surface elements located at the outer edges are eliminated
208
:
Intermediate medium, e.g., epoxy resin or glass
:^w /. ·.··.·.·.·!
a:,K;^;,^w,^,.v,v|··>·· ····*
/
i
N
Si
Figure 11.3
Principle of construction of an optocoupler with diffused GaAs diode and diffused photodiode. The quantum efficiency of the GaAs diode is raised further by the
^ intermediate medium with n 7.5
With Si-doped GaAs wafers produced by liquid epitaxy, greater quantum efficiencies
are achieved, in comparison with diffused wafers. Figure 11.4 shows a typical Si-
doped GaAs wafer made by the liquid epitaxy process. Here, the bonds can also
be located in the outer regions. For
example, GaAs diode TIXL 27 has four
bonds, one in each corner.
For the diode TIXL 27, the critical angle OtQ has been improved, by setting an epoxy dome directly on the wafer. The epoxy
resin has a refractive index of approx.
n = 1-5. The new critical angle OtQi of a
radiating wafer surface is then:
sina^i
-- nEpoxyd _ 1*5 = 0-417 -+CtQi
nGaAs
3-6
= 24-6
(11.7)
Thus, for the approximately Lambertian radiation, the following improvement ratio
Ky is obtained
1-cosOqi
KV =
1-cosOq
l-cos24-6 1-cos 16-1
1-0-91 = 2-25
1-0-96
(11.8)
For high-quality GaAs power diodes, the quantum efficiency is improved by a more expensive procedure. The possibility of emergence of the radiation emitted from
a point near the junction is greater, if a ray strikes a spherical surface. Therefore the wafers are formed into hemispheres. In addition, the effective radiating area is restricted, by mesa etching, to a small central area. If the ratio of the radiating area to the total base area of the wafer is optimally designed, the "point radiation" undergoes no total reflection, at the hemispherical GaAs-air boundary.
A considerably greater radiant power
emerges from the crystal, since the surface angles of emergence of almost all rays are less than the critical angle CXq. Figures 11. 7 and 11.8 show the emission conditions of domed wafers.
Furthermore, the quantum efficiency of GaAs diodes can be improved, if an antireflection coating is evaporated onto
the wafer surface.
209
The quantum efficiency of red-emitting GaAsP diodes is affected, among other factors, by their own absorption. The
0.004" min 0,009" max
·-- . Bond
luminous power produced by recombination
is sufficient to re-excite charge carriers.
For good quantum and luminous efficiencies, therefore, the junction is formed very closely under the surface. As an example, Figure 11.9 shows a wafer of a GaAsP luminescent
diode, while the wafer of a monolithic
display element shown in Figure 11.10.
In the following, the typical quantum efficiency Q(X)rj will be calculated for various kinds of luminescence diodes by
the formula
Q(X)D
*e,X,typ ^max
if
6
1°
A
1-24 mW
(11.4)
In these calculations, the data-sheet test conditions were taken as a basis.
Gol J:GeiUoyon
pe lurfaa
T
0,005" - 0,010"
^_GaAs
substrate
I
\ Gold-. Zinc^Uoy/Nickel ©i
P-type surface
Materia : Amphotic grown, Si-doped GaAt
Grown junction
Figure 11.4
Flat, Si-doped GaAs wafer, produced by liquid epitaxy. For protection against environmental effects, the P-side of the junction is mounted on the system substrate. The N-connection is located in the centre of the upper surface of the wafer. This kind of wafer is used in many "low-cost" products. Often, an epoxy dome is also fitted. The quantum efficiency of these wafers, produced by the liquid epitaxy process is greater than that of diffused wafers
Luminescence diode TIXL 27. Si-doped GaAs wafer, flat wafer geometry, epoxy-resin dome, Xmax = 0-94 /Ltm.
20mW.0-94/im
6
10
A
Q(X)D =
300 mA
1-24 mW
5%
(11.9)
Luminescence diode TIXL 1 Si-doped GaAs dome wafer, with mesa etching, Xmax = 0-93 jUm.
50 mW. 0-93 Mm 10° A
Q(X) D =
300 mA
1-24 mW
12-5%
(11.10)
Luminescence diode TIL 3 Si-doped GaAs wafer, flat wafer geometry,
epoxy-resin dome, Xmax = 0-94 Mm
bmW. 0-94 Mm 10° A
Q(X) D =
100 mA
1-24 mW
= 4-5%
(11.11)
210
Figure 1 1.5
Flat wafer produced by liquid epitaxy. Improvement of the quantum efficiency by the convex mesa sides. The indirect radiation is reflected and some of it can emerge through the upper surface. The lifetime of the device is limited by the presence of the unprotected junction
Contact
Contact
Mesa etching
Figure 11.6
Flat wafer, produced by liquid epitaxy. For protection against environmental effects, the P-side of the junction is mounted on the system substrate
211
Figure 11.7
Diffused dome wafer, Its quantum
efficiency corresponds approximately to that of flat wafers produced by liquid epitaxy
Figure 11.9
Diffused wafer for GaAsP light -emitting diodes or Zn-doped GaAs diodes
^7^r -.
T.
)
'
-T
P
JT.Tr
^~t^
JT^- T^X/
P
N-CaAsP
P
Figure 11.8
Si-doped GaAs dome wafer, produced by
liquid epitaxy, with convex mesa junction sides. Through the liquid epitaxy process, the hemispherical wafer configuration and the optimum dimensioning of the effective radiating area in relation to the total base area, the quantum efficiency reaches the highest values so far attainable
GaAs
///////////////////////.//////////////
2
Figure 11.10 Diffused wafer for monolithic display elements
Luminescence diode TIXL 06 Si-diffused
GaAs wafer, flat wafer geometry, Xmax =
0-91 Aim
1-2 mW. 0-91 /im 10 6 A
Q(A)D =
500 mA
1-24 mW
0-18%
(11.12)
212
Luminescence diode TIL 24 Si-doped GaAs wafer, flat wafer geometry,
Vax = 093 Mm.
2 mW. 0-93 Mm 106 A
Q(X) D
5o0u mm/A\
1I-'2i4i ruranW
= 3%
(11-13)
f
Luminescence diode TIXL 20 GaAsP dome wafer, mesa etching, Xmax =
0-85 Mm
mW 15
. 0-85 Mm 10° A
Q(X)D
200 mA 1-24 mW
= 5-1%
(11-14)
g
Red light-emitting diode TIL 210 GaAsp wafer, flat wafer geometry, epoxy resin case, Xmax = 0-65 Mm.
25 MW . 0-65 Mm
6 10
A
Q(X) D
20 mA
1-24 mW
0-065%
(11.15)
Yellow light-emitting diode
GaAsP wafer, flat wafer geometry, epoxy resin case, Xmax = 0-59 Mm.
10 MW- 0-59 Mm 10" A
Q(X) D
50 mA
1-24 mW
s0-01%
(11.16)
Amber-coloured light-emitting diode
GaAsP wafer, flat wafer geometry, epoxy resin dome, Xmax = 0-610 Mm.
10MW.0-6lMm 10° A
Q(X) D :
50 mA
1-24 mW
-0-01%
(11.17)
Red light-emitting diode GaP wafer, flat wafer geometry, epoxy
resin case, Xmax = 0-69 Mm.
225 MW . 0-69 Mm 10° A
Q(X) D =
40 mA
mW ' 1-24
«0-3%
(11.18)
1
Green light-emitting diode GaP wafer, flat wafer geometry, epoxy
resin case, Xmax = 0-56 jJm.
14 MW. 0-56 Mm 10° A
Q(X D :
40 mA
1-24 mW
0-015%
(11.19)
m
Yellow light-emitting diode GaP, flat wafer geometry, epoxy resin case,
Xmax = 0-589 Mm.
5 MW. 0-589 Mm
6 10
A
Q(X) D
50 mA
1-24 mW
* 0-005%
(11.20)
The calculated quantum efficiencies apply
with metal cans for the case temperature tQ and with epoxy resin packages for the ambient temperature tjj. At higher
temperatures, the quantum efficiency falls,
but at lower temperatures it rises.
11.2
Thermal calculations
11.2.1 Basic principles
The mathematical treatment of thermal calculations with semiconductors is carried out with equations similar to those for Ohm's law in elementary electrical theory.
213
The higher the elctrical energy converted
into heat per unit time, the so-called loss
power Pt t, the more will the temperature of the component rise and give up its thermal energy to the surroundings by a
heat flow.
According to the laws of thermodynamics the amount of a heat transfer through a transmission medium depends both on the temperature gradient At and also on the characteristics of the medium, its thermal
resistance R^.
Analogy with Ohm's law gives the following
comparisons:
At=V;Ptot s i;Rth-R
(11.21)
Thus the "Ohm's law of thermodynamics' can be written as follows:
At Rtlr
Ptot
(11.22)
Figure 11.11 Thermal equivalent circuit for junction semiconductor
For the junction semiconductor, the total
thermal resistance Rth,J-G i s made up of
a series and parallel connection of individual thermal resistances. (See Figure 11.11).
In the following calculations, the following symbols will be used:
tj = Junction temperature tQ = Case temperature tK = Heat-sink temperature tij = Ambient temperature
Rth,J-G ,= Thermal resistance, junction
to case
Rth,G-K = Thermal resistance, case to
heat-sink
Rth,K-U = Thermal resistance, heat-sink to ambient
Rth,G-U = Thermal resistance, case to ambient
Rth,J-U = Thermal resistance, junction to ambient
With plastic-encapsulated luminescence diodes, the thermal resistance Rfh J-TJ *s °f
interest.
For luminescence diodes in metal cans
without heat sinks, the thermal resistance
Rth,J-G> Rth,G-U and Rth,J-U are
significant.
Rth,J-U = Rth,J-G + Rth,G-U
(11.23)
With luminescence diodes in metal cans and with heat-sinks, the thermal resistance
Rth,J-G> Rth,G-K. Rth,K-U and Rth,J-U
are of interest. In this case, the thermal convection and radiation resistance
Rth G-U can usually be neglected.
RthJ-U = R th,J-G + Rth,G-K + Rth,K-U
(11.24)
The individual thermal resistances will be
calculated in accordance with equation (11.22) as follows:
tj-tG R th,J-G =
Ptot
(11.25)
214
»nn lj-U
tG-tU
Ptot
tj - tu
Rth J-U =
Ptot
_ tG - tR Rth G-K
Ptot
_ tR - tu Rth K-U
Ptot
(11.26) (11.27) (11.28) (11.29)
11.2.2 Determination of the thermal resistance Rtfi of luminescence diodes
The thermal resistance of a luminescence
diode can be determined by measurement.
The method used is based on a correlation,
starting from the fact, that with constant
diode forward current, Ip, the diode forward
Vp voltage
varies inversely with the
junction temperature tj.
In an oven, with the luminescence diodes operated with a constant forward
current of Ip = 5 mA, the diode forward
voltage Vp is recorded as a function of the
junction temperature tj. For each luminescence diode, a characteristic curve
Vp = f(tj) is plotted. Following this, each
luminescence diode is mounted on a large heat-sink and the electrical connections are made to the thermal resistance measuring test set. The temperature of the heat-
sink is kept constant.
The measuring test set contains a circuit
to interrupt the diode forward current
for pulse durations of tp = 1 00 jUs
with a mark-space ratio of approx. 100:1.
During the pulse, the diode forward
voltage Vp is measured. The junction
temperature tj is determined by means of
Vp the prepared characteristic curve
= f(tj).
The junction temperature tj is determined for each diode for several values of the
Figure 11.12
Junction temperature tj as a function of
the loss power Ptot for a GaAs power
diode. The thermal resistance between the function and the case is approx. R[n j.q =
20°C/W
forward current Ip. Then t j. is plotted on a graph as a function of the loss power P tot . Figure 11.12 shows a function tj = f(Plo t) recorded by this method for a GaAs power
diode. In the linear part of this curve, the
thermal resistance R th,J-G can be calculated
in simplified form by forming the
differential quotient. In this example it is:
R tj,J-G
tJl -tj2
88 C-68 C
3W-2W Ptot.l - Ptot,2
= 20-
W
(11.30)
R^ The thermal resistance
J-G determined
with this measurement method must not be
used for the guaranteed operation of a
luminescence diode in accordance with the
data sheet. The maximum permissible loss
215
power Pmax f°r l°w temperatures up to
t = 25 C and the highest permissible temperature t max at Pmax = are stated in
the data sheet by the semiconductor manufacturer, after quantum efficiency and radiant efficiency measurements and after extensive life and reliability tests. Figure 11.13 shows the usual representation of the
IJU
max. iel(%)
Lim iting value fmax.QS 100
\
50
^ mitmg val ue \ · max.t =
L.,
t(«Q-
The maximum permissible loss power ^max t25 at t = 25 C can be obtained from
the data sheet as follows:
l*max,t25 = UF,max,t25 · lF,max,t25
(11.32)
The data sheets often contain the current reduction factor Dij.tj or Djj^j.
lF,max,t25 Dl
tmax " l 25
(11.33)
Dp = Di . U F)maX)t25
(11.34)
The reciprocal of the power reduction
factor Dp gives the mathematical thermal
resistance.
Rth,J-U D p,J-U
R th,J-G
1
D P,J-G
(11.35) (11.36)
Figure 11.13
Relative loss power Pmax,rel as a function
of the ambient temperature trj for the luminescence diode TIL 32, in a plastic package, and as a function of the case temperature tQ for the luminescence diode TIL 31 in a metal can
relative permissible loss power Pmax.rel as
a function of the temperature t for junction
semiconductors. From the limiting values Pmax,t25 at * = 25°C and P max ,t = at t max , the power reduction factor Dp j.tj
is calculated for plastic-encapsulated
luminescence diodes, or the power
reduction factor Dp j.tj is calculated for
plastic-encapsulated luminescence diodes,
Dpj^ or the power reduction factor
for
metal-canned luminescence diodes.
Pmax,t25
Dp
tmax " l 25
(11.31)
Example: The plastic encapsulated luminescence diode TIL 32 has the following data sheet
values:
VF,max,tTj25 = 40 mA tU.max = 80°C
Therefore:
Dl,J-U
I F,max,tU 25 = 40 mA
tU.max - tU25 80 " 25 °C
0-7272 mA/ C
(11.37)
Dp,J-U = DI,J-U · UF,max,tu25 = 0-7272 mA/°C . 1-6V = 1-1636 mW/°C
(11.38)
216
pmax,tu25 = UF,max,tu25 * I F,max,tu25
= 1-6 V . 40 mA = 64 mW
(11.39)
d
R th,J-U
1
Dpj.u
rc 11636 mW
The heat from diffused luminescence diodes in plastic packages (e.g., TIL 209A), can be led out through the cathode connection, while the heat produced from Si-doped GaAs plastic encapsulated diodes (e.g., TIL 32) can be led off through the anode connection. Here, a valid guide-
line value for the thermal resistance
R th,J-connection is 300°C/W.
= 859-3°C/W
(11.40)
11.2.3
Calculation of loss power for luminescence diodes in plastic packages
The maximum permissible loss power
pmax,tiT at an ambient temperature trj can be calculated by rearranging the equation (11.27).
11.2.4
Calculation of loss power for luminescence diodes in metal cans with infinitely large
heat sinks
The maximum permissible loss power
pmax,tr; at a case temperature tQ is calculated by rearranging the equation
(11.25).
"max,tij
tj-m R th,J-U
D pmax,tu = P,J-U · (tj - trj)
(11.41) (1 1.42)
PtTj25 - t|j 's tne difference in loss power
power between Pmax at ty = 25°C and pmax at an ambient temperature of tTj>25°C.
PtU25 - tij " Pmax,tu25 " Pmax,tu
(11.43)
*U - tU25 pmax,tTj ~ p max,tTj25 "
R th,J-U
(11.44)
The junction temperature tj can be calculated with the rearranged equation (11.27).
tj = (Rth,H-U · p tot) + tu
(11.45)
-p tot
ttj =
+tu
D P,J-U
(11.46)
"max,tQ
R th,J-G
W ^ pmax,tG = DP,J-G ·
-
(11.47)
01-48)
*G " tG25 pmax,tQ ~ p max,tG25
RthJ-G (11-49)
The junction temperature tj is obtained by means of the rearranged equation (11.25):
tJ = (RthJ-G " p tot) + to
(11.50)
ttj_=-ptot D P,J-G
+
tG
(11.51)
11.2.5
Calculation of loss power for luminescence diodes in metal cans without heat sinks
The maximum permissible loss power pmax,tu at an amD ient temperature tu
is calculated by combination of the equations (11.23) and (11.27).
217
tj-tu
:
"max.tTj
Rth,J-G + Rth,G-U
(11.52)
The junction temperature tj is supplied by equation (1 1.50). The thermal resistance Rth q.tj can be taken from the data
sheet. Otherwise the following guide-line
values apply:
for TO-18 and TO-46 cans:
R th,G-U = 450°C/W
(11.53)
for TO-39 cans: Rth,G-U = 250°C/W
(11.54)
A measurement of the case temperature with
a temperature probe on uncooled TO-18 and TO-46 metal cans is excessively distorted, since the contact probe draws off too much heat from the metal can.
11.2.6
Calculation of loss power for luminescence diodes in metal cans with heat sinks
The maximum permissible loss power
Pmax.tTj at an ambient temperature tjj is calculated by combination of the equations (11.24) and (11.27):
Pmax.tij
tj-tu
Rth,J-G + Rth,G-K + Rth,K-U
(11.55)
With GaAs power diodes, the anode
connection usually has to be cooled. In
the case of Si-doped GaAs power diodes,
this connection is the fixing bolt. With
insulated mounting with mica washers, thermal resistances up to Rfa q.\^ =
1-5 C/W are mentioned in the literature.
By the use of beryllium washers, hard-
anodised aluminium washers, silvered
mica washers or aluminium or copper
mm washers coated with 0-1
of Araldite,
and by coating the insulting layers on
both sides with an effective heat-transfer
paste, the thermal resistance can be
R^ reduced to about
q_k = 0-3 C/W. For
non-insulated mounting, the guide-line
value Rth,G-K = 0*3 C/W applies. The
thermal resistance Rth,K-U °f tne heat-sink is to be taken from the heat-sink
manufacturer's data sheet.
For luminescence diodes, the following
R^ minimum thermal resistances
k.tj
for the heat-sink are desirable:
Case
TO 18/TP 46
400 mW
TO 5 with bolt TO 5 with bolt j
or special
J
1W
R th,K
60°C/W
(Push-on heatsink)
10°C/W
5°C/W
Several GaAs diodes in an enclosed space
heat each other up, so that higher case
temperatures must be allowed for. The pill package GaAs diodes TIL 23/TIL 24, used in reading heads, are cooled through an optimally designed circuit board coating (anode connection).
11.2.7
Calculation of the maximum permissible forward current Ip max
The maximum permissible diode forward
current, Ip max for the calculated loss
m power P ax t can ^e determined exactly
with the characteristics Ip = f(Up) and Vp =
f(t). For luminescence diodes in metal cans,
the case temperature tQ has to be calculated
Vp to evaluate the characteristics
= f(t)
and <3>e = f(t).
Gt = tj - (R t h,J-G · ptot)
(11.56)
In simplified form, it is sufficient to
calculate Ip ;max,t W1th the diode forward
Vp voltage
>niaX) t25 at t = 25 C, stated in
the data sheet. The small error thus
produced is an additional safety factor in
constant current operation of the diode,
since Vp decreases with rising temperature.
218
Pmax,t
lF,max,t UF,max,t25
(11.57)
Example: GaAs diode TIL 31 in TO 18
metal can
Data-sheet values:
Diode forward voltage at tQ = 25 C:
^F,max,tG25 = * "^ ^
Diode forward current at iq = 25 C:
^=
^F,max,tG25
^
Maximum permissible case temperature:
tG,max = 80°C
Empirical values:
Maximum thermal resistance of the TO 1
can;
Rth,G-U = 450°C/W Thermal resistance with heat-sink fitted:
Rth,G-K = 1°C/W The calculation will be carried out:
1
With the data-sheet values at tQ = 25 C
With an infinitely large heat-sink at
tG = 50°C
Without heat-sink at tjj = 50 C
4 With a push-on heat-sink
Rth,K-U = 60°C/W at tu = 50°C
for the values:
a
m Maximum permissible loss power P ax
R Overall thermal resistance th
Power reduction factor Dp
d Cureent reduction factor D\
e
Case temperature tQ
f
Maximum permissible diode forward
current I F?max
la
pmax,tQ25 U F,max,tG25 - IF,max,tG25
W = 1-75 V. 0-2 A = 0-35
(11.58)
lb
1 _ tG.max " t G25
Rth,J-G :
Dp,J-G Pmax,tG25
80 C - 25 C = 157 C/W
0-35 mW
(11.59)
lc
-.
Dp,J-G
=
Pmax,t-r^;A251
tG.max " tG25
RthJ-G
=
5
= 6-363 mW/C
157 C/W
(11.60)
Id
U G'
_ "
I F,max,tG25
tCmax - tG 25
_
200 80°C
mA
- 25°C
= 3-636 mA/ C
(11.61)
le and If See data sheet values.
2a pmax,tG50
G tj - t
Rth,J-G
80°C - 50°C 157 C/W
0-191
(11.62)
219
2b to 2c Correspond to lb to lc
2e
Gt = 50°C = tij
2f
^.max.tGSO :
pmax,tG50
u F,max,tG25
191 mW 1-75 V
= 109 mA
(11.63)
3a rmax,tTj50
tj - tU50
Rth,J-G + Rth,G-U
80°C - 50°C
= 49-4 mW
157°C/W + 450°C/W
(11.64)
3b
Rth,J-U = R th,J-G + Rth,G-U
= 157°C/W + 450°C/W = 607°C/W
(11.65)
3c
DP,J-U
1
R th,J-U
1
607 C/W
= 1-747 mW/°C
(11.66)
3d
Dpj.u
Dl J-U = u F,max,tG25
1-647 mW/C
1-75 V
= 0-94mA/°C
(11.67)
3e
*G = tj - R th,J-G · p max,t U5 o
mW = 80°C - 157°C/W . 49-4
= 72-3°C
(11.68)
3f
mW _ pmax,tU5 o _ 49-4
^.max^uso uF,max,tG25
1-75 V
28-23 mA
(11.69)
4a ^max,tu50
tj-tu
Rrh,J-G + Rth,G-K + Rth,K-U
W-- (80°C - 50°C) .
5
5-
= 137-6 mW
157 C+l C + 60 C
(11.70)
4b
Rth,J-U = Rth,J-G + Rth,G-k + Rth,K-U
= 157°C/W + 1°C/W + 60°C/W
= 218°C/W
(11.71)
4c
D P,J-U
1
R th ,J-U
1
218 C/W
: 4-587 mW/°C
(11.72)
4d DlJ-U
D P,J-U
_ 4 -587 mW/°C
u F,max,tG25
1-75 V
= 2-62mA/°C
(11.73)
4e
tG = tj - R th,J-G · p max,t U5 o
mW = 80°C - 157°C/W . 137-6
= 58-4°C
(11-74)
4f
^.max^uso
pmax,tU5 o
u F,max,tG25
137-6 mW = 78-62 mA
1-75 V
(11.75)
11.3
Radiant power
The radiant power or luminous power produced from a luminescence diode depends on the quantum efficiency Q, the diode forward current Ip and the
temperature tjj or Gt . Red GaAsP diodes,
220
yellow and green GaP diodes and IRemitting GaAs diodes have, to a first approximation, a linear dependence of the radiant power 4>e on the diode forward current Ip. The linearity improves with small forward currents and low temperatures. Red GaP diodes have a non-linear
$ dependence of the radiant power e
on the diode forward current Ip. With higher diode forward currents, red GaP diodes reach a saturation point. Therefore they are unsuitable for pulse operation.
Figure 11.14 shows the luminous intensity Iy which is at present attainable in mass production, as a function of the forward current Ip for luminescent diodes. The temperature affects the quantum
Q efficiency and thus the emitted radiant
power ^e of a luminescent diode. With rising temperature, the quantum efficiency and radiant power decrease,
with falling temperature the reverse is
true. In practice, for GaAs diodes, the radiant power is decreased approximately
by the factor 0-5 with a temperature rise
of 80 C and it is increased approximately by a factor of 2 with a temperature decrease
of80°C.
Figure 11.15 shows, for comparison, the
$ relative radiation power e ,rel as a function
of the temperature t for GaP, GaAs and GaAsP luminescence diodes. It is very often advantageous to show the function
<I?e = f(t) with the diode forward current Ip as a variable. {Figure 11.16). For the function 4>e = f(Ip), as shown in Figure 11.17, the temperature t can be selected as the variable.
tu = 25 °c
'V.typ (mcd) 6
/
/ GaP (green) ,
GaP (red)
/^
\GaAsP(redJ GaP (yellow!
30 'F (mA) -
Figure 11.14 Luminous intensity Iv as a function of the
forward current Ipfor GaP and GaAsP
luminescent diodes
Figure 11.15
$ Relative radiant power e re i as a function
of the temperature t for GaP, GaAs and
GaAsP luminescence diodes. The forward
mA current is Ip = 20
in each case
In the data-sheet, the relative radiant
$ power e ,rel is usually stated for IR-
emitting luminescence diodes and the relative luminous intensity Iy rel for light-emitting diodes, as a function of the diode forward current Ip or as a function of the temperature
$ t. Here, the relative radiant power e re j
or the relative luminous intensity Iy re j relate to the absolute value of <&e or Iy
for the test conditions stated in the data sheet.
The loss power calculation shown in the
previous section 11.2, with the subsequent
determination of the maximum permissible
221
1"
1
e.iyp (W)
20
*max - 870 nm
Temp. " -198 °C^-
1-5
10
0-5 3
Temp. = 25 °C
1
1
1
1
1
10
2-0
30
40
50
60
Figure 11.16
Radiant power 4>e as a function of the heatsink temperature tf(for the GaAs power diode TIXL 12. The variable is the forward
current If
forward current Ip, is necessary, in order
to be able to evaluate the function
W $ ^e.rel = f and e>rel = fflF), shown
in the data sheet.
For the calculated maximum permissible
forward current Ip, the relative radiant
power ·J'reKI) can be read from the graph <&re{ = f(Ip) and the relative radiant power $^el(t) can be read from the graph "^rel = ft*) f° r the calculated temperature tjj or Xq with the parameter Ip stated for
the test conditions. Figures 11.16 and
11.17 show examples of such graphs. The
required absolute radiant power 4>t i s the
product of the values of 4^ ei(t) and
^Vel(I) reaQ off anc* the absolute radiant
power
4> t
25
stated
for
the
test
conditions.
Figure 11.17
Mean radiant power &e,typ as a function of the forward current If for the GaAs power diode TIXL 16; the heat-sink
temperature tf( is the variable
*t = *rel (I) · $rel(t) · $t25
(11.76)
For the example 4f listed in Section 1 1.2, the radiant power <f>e t will be determined with the data-sheet values.
mA Data: lF,tTj50 = 78 ' 62
'
o
tu = 50°C,t G = 58.4 C,
$e .typ,tG25 6 mW.
In Figure 11.18, for a diode forward
current of Ip «*78 mA, a relative radiant
power of ^reKI) = 0-75 is read off. For
a case temperature of \q = 58 C Figure
11.19 shows a relative radiant power of ^rel(t) = 0-7. For this working point, the typical emitted radiant power of the TIL 31 is:
$=
e,t u50 *e,rel(I) · *e,rel(t) *e,t G25
= 0-75 .0-7 .6mW = 3-15 mW
(11.77)
222
--1
7 tG-25 0C-
*e,rel
4
~j~
2
X
1
0-7
/
/
0-4
'
V
/
0-2
/
/
/
40
100
/
400 'F(mA)
1000 "-
Figure 11.18
Relative radiant power <J>e re\ as a function of the forward current Ipfor the GaAs diode TIL 31. The relative radiant power
has been normalised to the value <$>e re[
= 1 for a forward current ofIp=l 00 mA
at to = 25°
Figure 11.20 shows the absolute radiant
power "J>e as a function of the heat-sink
temperature tj[ for the GaAs diode TIXL
16. When using a heat-sink with Rth,K-U =
5 C/W and for a maximum ambient
o
temperature of tjj = 50 C, a diode forward
mA current of iF.tijsn = 990
and a case
temperature of Xq = 62-8 C were calculated.
*e (roW) 100
IF -01 A If « 0-01 A 1F -0-002 A
*e.rel
1 1
' IF" 150 mA
£W^ 00 mA--
01 ^-l'
007
002
001
SO
75
<G(»C)-
Figure 11.19
Q Relative radiant power ejel as a function
of the case temperature tQ for the GaAs
diode TIL 31, with the forward current
Ip as a variable
'K(°C)-
Figure 11.20
Radiant power 4>e as a function of the heatsink temperature tf^ for the GaAs power diode TIXL 16, with the forward current Ip as a variable
mA The parameter Ip = 990
is entered in
Figure 11.20. For tG «63°C and Ip = 990
mW mA, a radiant power of <l>e = 95
is then
read off.
11.4 Radiant efficiency Tfe
The radiant efficiency Tfe, like the quantum efficiency Q, depends on the construction of the luminescence diode and also on the chosen working point. By considerations of
radiant efficiency, it is possible to select
the optimum working point of a luminescence diode. The Si-doped GaAs
223
doped-wafer diodes produced by the liquid
epitaxy process have the greatest radiant
efficiency. It amounts to approximately
G G 8% Tfe in
at t
= 25°C. At t = -196°C, it
rises about Tfe = 25%. Figure 11.21 shows,
as an example, the radiant efficiency Y)e
of the GaAs power diode TIXL 12 as a
function of the diode forward current Ip,
with the temperature tQ as a variable.
According to example 4f)-'
$e .tU50
Ve
^.tUSO ' U F,max,tG25
3-15 mW
-=2-3%
mA 78-62
V . 1-75
(11.79)
According to the data-sheet test conditions, for the diode TIL 31:
= T?e
.
e <!>, >*G25
lF >tG25 · UF,max,tG25
6mW
= 3-43%
100 mA. 1-75 V
(11.80)
Figure 11.21
Radiant efficiency X\e as a function of the forward current Ipfor the GaAs power diode TIXL 12, with the case temperature tQ as a variable
The radiant efficiency TJe of a luminescence diode is calculated in accordance with equation (2.34). For the GaAs diode TIXL 12, it is:
<&P
$,e,min,tG25
T?e
^^25 P
· U F,typ,t G2 5
40 mW
= 9-5%
0-3 A. 1-4 V
(11.78)
This value is entered in Figure 11.21.
For the example 4f) listed in Sections 11.2 and 11.3, the radiant efficiency will be calculated. The radiant efficiency according to the data-sheet test conditions serves as a comparative value.
11.5 Spectral radiant efficiency
Luminescence diodes are selective emitters.
On the other hand, semiconductor laser diodes are decidedly monochromatic emitters. The spectral radiation distribution of a luminescence diode depends on the band spacing of the semiconductor. The maximum wavelength Xmax is calculated
from the typical band-spacing energy with equation (1.4).
X _ 1-24 yam. Wph
eV
(1.4)
Figure 11.22 shows the maximum wave-
length for different semiconductor radiating diodes. In addition, the relative spectral sensitivity of the eye, V(X), which is of interest for lightemitting diodes, and the relative spectral sensitivity s(X) rei of silicon photocells, which is of interest for IR-emitting
GaAs diodes, are shown as functions of the wavelength X. The bandwidth
AXjjp of the selective radiation of a luminescence diode is stated, in a unit of wavelength, between the spectral
224
,
A
V(X) s(\)rel
"TT
photocell
ii
800 900 nm 1000
X(nm)
»
Figure 11.22
Illustration of the various maximum wavelengths \nax for different
semiconductor radiation-emitting diodes.
In addition, the relative spectral
sensitivity of the eye, V(ty, which is of interest for light-emitting diodes, and
the relative spectral sensitivity s(*h) re i
of silicon photocells, which is of interest for the IR-emitting GaAs diodes, are stated as functions of the wavelength X
1
-
0-75
Xmax = 92s iTM
'
/
\ If -300 mA
/
1 t - 25 °C
half-power points HP\i and HP\2 (see Section 10.4). The average bandwidth of
luminescence diodes lies between 20 nm and
50 nm.
In Figure 11.23, the relative spectral
radiant power <£e \ re\ is shown as a function of the wavelength X for the Si-doped GaAs diode TIXL 12, produced by the liquid epitaxy process. The
bandwidth AXjjp has been specially
marked. It is AXj^p = 35 nm = 350 A.
Finally, Figure 11.24 shows <J>e rej \ = f(X) for the diffused red-emitting GaAs diode TIL 209A
r
0-8
1
1
IF - 20 mA
tu = 25 a
07
(\
0-6 '
0-5
0-4
03 02
01
SOO
620
640
660
68O
700
0-5
- 0-25
A-4XHP-35nm-4
980
1
960
1
940
1
920
1
900
880
X(nm)
*-
Figure 11.23 Relative spectral radiant power
*&e,\rel as a function of the wavelength X for the GaAs power diode TIXL 12
Figure 11.24 Relative spectral radiant power
4>e \ re[ as a function of the wavelength X
for the red-emitting gallium-arsenide-
phosphide diode TIL 209
The spectral radiation distribution of a luminescence diode is temperaturedependent. In the data sheet, the difference
between the maximum wavelength Xmax at t = 25 C and the maximum wavelength Xmax at any given temperature
225
t is stated. As an example, Figure 11.25
shows the difference in the maximum wavelengths Xmax as a function of the temperature t for the GaAs diode TIXL
06. The measurement of radiation wavelengths is carried out with a spectrometer calibrated
according to national standards.
Material Range
Maximum VF,typ
wavelength
GaAs GaAsP GaAsP GaAsP GaP
Infra-red 940 nm
Red
650 nm
Orange 610 nm
Yellow 590 nm
Green 560 nm
1-4 V 1-6 V 2-0 V 3-0 V 3-0 V
i 300 Ip = c instant
AXmax 200 (nm)
100
300 -75 -50 -25
25
50
75 100 125
'G (°C)
-
Figure 11.25
Variation of the maximum wavelength
£&fnax as a function of the case temperature tQ for the gallium arsenide
diode TIXL 06. The maximum wavelength
\nax has been standardised to the value
± nm for a case temperature tQ = 25 C
11.6 Electrical Parameters
Two of the most important electrical
parameters of luminescence diodes are
the diode forward voltage Vp and the
diode forward current Ip. The forward voltage is determined by the band
spacing of the. semiconductor, the diode forward current Ip and the temperature t. Under equal test conditions, therefore, different-coloured light-emitting diodes have
different forward voltages. The GaAs
diodes have typical forward voltages of
Vr= V 1-4 (sec Table 11.1).
Table 11.1
Forward voltages of various luminescence
diodes
The diode forward voltage Vp is shown as
a function of the diode forward current
Ip in Figure 11.26 for the GaAs diode TIXL 16 and in Figure 11.27 for the GaAs diodes TIXL 12 and TIXL 13, at a constant temperature t = 25 C. From this V-I characteristic, the dynamic impedance rs can be determined in simplified form for the normal working
range of a luminescence diode through the
differential quotient AVp/Alp. With a
straight characteristic section, as in the
Figures 11.26 and 11.27, the dynamic impedance r s can be stated directly. The dynamic impedance of GaAs diodes lies approximately between r s = 0-1 £2 and 2fl Modulation final stages for luminescence diodes are to be matched to the dynamic impedance r s.
The diode forward voltage of luminescence diodes is temperature dependent. The temperature coefficient is stated by the differential quotient dVp/dt,
i.e.
dUp TK
dt
(11.81)
GaAs diodes have smaller temperature coefficients than silicon components. For the normal working range of a luminescence
diode, the temperature coefficient can be calculated in simplified form through the differential quotient AV/At. For
GaAs diodes, temperature coefficients of
226
Figure 11.26
Forward voltage Vpas a function of the forward current If for the GaAs power diode TIXL 16
Figure 11.27
Forward voltage Vp as a function of the
forward current Ip for the GaAs power diodes TIXL 12 and TIXL 13
TK = -1-2 mW/C to TK = -l-5mV/°C
are obtained. Figure 11.28 shows the
function V F = f(t) for the TIXL 16 and
Figure 11.29 shows this function for the
TIXL 12 and TIXL 13. The forward current
Ip is the variable.
For the exact determination of the working point of a luminescence diode, the function Ip = f(Vp), which is often stated, is usually sufficient. In Figure 11.30, the function Ip = f(Vp) is shown for comparison, for various semiconductor luminescence diodes. As a further example, Figure 11.31 shows the diode forward current Ip of the
red-emitting GaAsP diode type TIL 209A as a function of the diode forward voltage Vp.
As well as optoelectronic applications, these light-emitting diodes can be used advantageously as reference diodes for low voltages, since they have a steep slope
Alp/AVp and a low temperature
Figure 11.28
Forward voltage Vpas a function of the heatsink temperature t% for the GaAs power diode TIXL 16; the forward current Ipis the
variable
coefficient of about TK = 2 mV/ C. Figure
11.32 shows the function Ip = f(Vp) for the GaAs diode TIL 23 with the temperature t(-j as a variable.
227
A
Figure 11.29
Forward voltage Vpas a function of heatsink temperature tfc for the GaAs power diodes TIXL 12 and TIXL 13. The forward current Ip is the variable
Figure 11.31
Forward current Ip as a function of the forward voltage Vpfor the red-emitting GaAsP diode TIL 209
Figure 11.30
Forward current Ip as a function of the forward voltage Vpfor semiconductor diodes of various materials
The temperature coefficient TK can
be calculated in simplified form through the differential quotient AV/At, its value isl-33mV/°C.
11.7 Pulse operation
In pulse operation, luminescence diodes are controlled by a pulsed current Ip. The peak pulse current amplitude depends on the construction of the diode and on
A the cooling. certain power- and current-
'F (mA)
!
'
/
tG = lOOoc/
/
/
/
//
/
h --55»C
K/ J
f«. /
25 »c.
yj J
1-3
1-4
IS
V F(V)
..
Figure 11.32
Forward current Ip as a function of the forward voltage Vpfor the GaAs diodes TIL 23 and TIL 24, with the case temperature tQ as a variable
228
f 90
-J 80
ITM
2 60 5°
| 1 «o 5
30
20
1
1
1
tC = 25 0C
\
'°00
200
400
700
1000
'F(mA)
*
Figure 11.33
D Mark-space ratio in % as a function
of the forward current If for the GaAs diodes TIL 23 and TIL 24
density must not be exceeded in the wafer and in the bond connection wire. Further, the peak pulse current depends on the
mark-space ratio D and the pulse width
tp. Figure 11.33 shows the mark-space
ratio D and Figure 11.34 the pulse-width
tp as functions of the diode forward current
Ip for the GaAs diodes TIL 23 and TIL 24
G at t = 25°C.
If such functions do not appear in the
data sheet, these illustrations can serve
as an approximate solution. For this, the
mA abscissa value of Ip = 100
is to be
replaced in each case by the maximum
permissible diode forward current Ip ^25
Figure 11.34
Pulse width tp is ms as a function of the forward current for the GaAs diodes TIL 23 and TIL 24
at t = 25 C for the corresponding GaAs diode. The other abscissa values are corrected with the ratio Ip,t25/100 mA.
Example: TIL 24: lF,max,tp25 = ^0 mA
m With lF,tr-25 = ^®® ^' t^ie maxuTlum
permissible pulse duration is tp = 40 (Js
and the maximum permissible mark-space ratio D=10%.
Example: TIL 31: lF,max,tp25 = ^0 mA
With Ip tc2<; = 1 A, the maximum
permissible pulse width is tp = 40 \l
and the maximum permissible mark-space
ratio D=10%.
229
12 Radiation
measurements
12.1 12.2 12.3 12.4 12.5 12.6 12.6.1 12.6.2 1 2.6.3
12.7
General Considerations Measurement of colour temperature of
A standard light
Measurement of radiant power with thermal photodetectors Measurement of radiant power of standard
light A radiation with the thermopile
Measurement of irradiance of a standard
light A radiation with Si photodetectors
Measurement of a luminescence diode radiation with Si photodetectors General measurement problems Measurements of relative spectral sensitivity with a monochromator Measurement of irradiance of a luminescence diode radiation with Si photodetectors Measurement of the total radiant power of a luminescence diode
131
12 Radiation Measurements
12.1 General Considerations
Every optical radiation has a given spectral distribution. This is taken into account through the spectral densities of the radiometric parameters. Every radiometric parameter can be related to a differential range of the wavelength \ or the
frequency v. The spectral radiant power is
linked with each of these spectral parameters. Therefore, for the sake of simplicity, the spectral distribution of an actual radiation source is related to the basic parameter, radiant power. Integration of the spectral radiant power over the whole wavelength range gives the radiant power (see Section 2.3.4).
$e= J*e,X
o
(12.1)
In general, every optical radiation is evalu-
ated by a photodetector. But also, every photodetector evaluates the optical radiation falling upon it in accordance with its spectal sensitivity. In the field of photometry, this evaluation is adequately illus-
trated in Chapter 5 and also in DIN standards 1301/1304/5031/5033. The visible radiation
from light-sources for illumination purposes or from displays for the representation of
information is evaluated by the human eye
as a "photodetector". It is related to the specified sensitivity curve of the eye, V(\).
X2 *V=J"*e,X-V(X).dX
Xi
(12.2)
Luminous fluxes are measured with photodetectors which are specially corrected to the spectral sensitivity of the eye. In comparison with measurements in the other optical wavelength ranges, photometry has two advantages: Every visible radiation is
related only to an obligatory spectral sensitivity of a photodetector or of the eye. The part of the radiation which can be evaluated is measured with simple measuring instruments in photometric parameters.
The calculation and measurement of an
optical radiation in the wider sense takes place in accordance with similar considerations to those in photometry. Firstly an optical radiation is measured as a whole, secondly the part of the radiation evaluated by a photodetector is determined selectively. (See Chapter 9).
12.2
Measurement of colour temperature of
A standard light
The measurement of the sensitivity of Si
photodetectors is at present carried out with
the radiation known as standard light A.
This kind of radiation is defined by the relative spectral radiation distribution S\ of
the Planck radiation between 320 nm and 780 nm at a colour temperatur Tf = 2856 K (see DIN 5033, Part 7). The radiation
functions S\ in the IR range which is of interest with an emitter temperature
Ty = 2856 K are not taken into account.
The scientific lamps (Wi-series lamps) which have previously been operated with a colour
temperature Tf = 2856 K show different
radiation functions in the IR range. This is due to the different absorptions of the glass and the different emitter temperatures between the filament fixing points and the central part of the filament. The newlydeveloped filament lamps Wi 40 and Wi 41 (Figure 12.1) therefore have self-supporting
filaments.
For exact radiation measurements in the IR range, the radiation function S^ of the
233
the Wi filament to be measured is formed by the objective lens in the plane of the filament of the built-in lamp.
Equality of radiance is set by varying the lamp current or lamp voltage, so that the filament cap of the built-in lamp no longer stands out from the incandescent material to be measured, but almost disappears. The voltmeter in parallel with the built-in filament lamp is calibrated with the colour
temperature.
Figure 12.1
Osram scientific lamps
filament lamp used and the difference factor in relation to the equivalent function S^ of a Planck radiator emitting at
Ty = 2856 K should be known. A recom-
mended Wi lamp to DIN standards and the
statement of exactly calculated conversion factors for the wavelength range of interest would, for example, be a solution.
The Wi lamp is to be operated in the previous way at Tf = 2856 K. The colour -temperature Tf can be measured, for example, with filament pyrometers, the principle of which
is illustrated in Figure 12.2.
4-
Red
Wi Eyepiece
Red
-+-0-- -- filter Lamp
lens
&
filter
&--
^=£
/s
Figure 12.2
Principle of the filament pyrometer
The filament of the built-in lamp is
observed with the eye through the eyepiece lens. The interposed red filter, with the spectral sensitivity of the eye, Vx, gives the
desired narrow spectral range. An image of
Furthermore, the colour temperature of the Wi lamp can be adjusted by a second method, in accordance with the lamp manufacturer's data sheets, by means of the lamp current or the lamp voltage. For this, calibrated lamps of types Wi 40 or Wi 4 1 are needed. The lamp manufacturer's conditions
must be observed. The Wi lamp will be operated vertically, with the cap downward. The internal glass frame, if any, is the side away from the detector. The screw ring is the anode. The Wi 40 lamp used then had
V V the following values: L = 32.0 (D.C.);
lL = 5.716 A.
Since discrepancies can occur through contact resistances between the lampholder and
base when the lamp voltage is adjusted, it is most convenient to work with the current.
It should then be possible to adjust the value of the stated lamp current exactly, to rather better than the third decimal place. Therefore, only finely adjustable and well stabilised mains supplies are suitable for the power supply.
Often, no precision ammeter is available to
measure the lamp current. In such cases, the
lamp current is determined through the
voltage drop in a precision series resistance.
For simplicity, the A-V-ft Multizet S, in
conjunction with the associated shunt
M955-A1 for current measurements up to
30
A,
-
is
used
as
the
series
resistance.
The
mV Multizet is to be switched to the 100
range. With a lamp current of 5.716 A, the
digital voltmeter shows a voltage of
19.0533 mV.
234
During such current and voltage measurements, the lamp leads must be connected to the current terminals of the shunt and the digital voltmeter to the voltage terminals of the Multizet. One measurement lamp Wi 40 should only be used as a calibration radiation source. Further, long-term measurements should be carried out with a second measurement lamp. The lamp Wi 40 should only be used as a calibration radiation source. Further, long-term measurements should be carried out with a second measurement lamp. The lamp Wi 40 can be operated for approximately 50 - 100 hours with the manufacturer's stated data without recalibration. Since this measurement lamp is in any case mounted on an optical bench with a home-made adapter, an expensively-made precision lampholder can be dispensed with. The current supply leads and the measurement leads for voltage measurement are soldered directly to the lamp. This arrangement also has the advantage, without great expense, that the current and voltage values recorded by the lamp manufacturer can be compared one with the other with two digital voltmeters. The tolerances of simple measurement resistors become visible and can thus be taken into account.
12.3
Measurement of radiant power with thermal photodetectors
A The standard light radiation has a great
spectral width, corresponding to the Planck radiation distribution. Such wide-band radiant powers are also measured with wideband, radiation-sensitive receivers. Since radiation detectors, which are based on the principle of the internal or external photoeffect, are, almost without exception, selective detectors, only the thermal types of detector are suitable for use as wide-band photodetectors. Thermal photodetectors are almost black bodies of small dimensions. They utilise the effect, that an absorbed incident radiant energy heats up their bodies.
The absorbing detector surface is blackened,
so that the degree of absorption is as great as possible, in accordance with Kirchhoff s law (equation 4.39).
In general, heat increases the vibrations of the crystal lattice of the absorbing substance. These vibrations are called phonons. Corresponding to the rise in temperature of the receiver, they cause a change in its electrical resistance, its contact potential or the product of pressure x volume of an enclosed quantity of gas in pneumatic detectors. For the greater part, the efficiency of thermal detectors is low.
Certain measuring instruments, which utilise the change in electrical resistance with a rise in temperature, are called bolometers. They are used for temperature and radiation measurement. For radiation measurement, a metal strip (e.g., platinum), coated with lamp-black, is used as a bolometer. The principle of a bolometer circuit is
shown in Figure 12.3. When not irradiated,
the Wheatstone bridge is balanced. Through irradiation, the change in resistance of the bolometer (R 1) unbalances the bridge, so that a current flows through the galvanometer. The current value is proportional to the temperature rise and thus to the absorbed radiant power. Changes in the
Figure 12.3
Principle of a bolometer
235
ambient temperature do not falsify the measured value, since the bridge is tem-
perature-compensated by R 2. The resistance R 2 consists of an identical, but
darkened bolometer. The battery must deliver a highly constant voltage. The newer semiconductor bolometers, equipped with thermistors, show an exponential change in resistance as a function of temperature.
Another detection principle which is often used is based on the change in the contact
potential of a thermocouple. A thermo-
couple consists of a small soldered or welded joint between two metal wires with different thermoelectric potentials. As is
shown in Figure 12.4 a constantan wire
uK \ Copper
Copper /
Constantan
©Ovf~
OOC
~^V]-©
d°c
Figure 12.4
Principle of construction of a thermocouple, composed of a constantan wire and the wires of the copper leads
forms a thermocouple with the copper measurement leads. For contact measurement, junction 2 is brought into good thermal contact with the substance to be measured. If, during this measurement, junction 1 is kept at a constant temperature of 0°C in melting ice, then the contact voltage is a measure of the temperature difference in relation to 0° C.
Figure 12.5 shows a thermocouple made
M M from the dissimilar metal wires 1 and 2.
The measurement leads are not used to form a thermocouple. In the same way, they must not form a new thermocouple at the con-
nection points A 1 and A 2. The metal wires
M M 1 and 2 are connected by the
Figure 12.5 Thermocouple with separate measurement
leads.
junction L, which is in good thermal contact with the blackened receiver surface E. Incident radiation is absorbed by the black
receiver surface E. The receiver E is heated
up, so that.the temperature of the thermo-
couple junction rises. To a first approxi-
mation, the thermocouple voltage varies in proportion to the rise in temperature and thus to the absorbed radiant power. The thermoelectric voltages of a single thermocouple are very small. Therefore, a number of thermocouples are often connected in series to form a thermopile. The thermoelectric voltage is thus multiplied according to the number of thermocouples. The greater mass of the thermopile gives longer response times, in comparison with thermo-
couples.
Figure 12.6 shows a thermopile from the American firm of Eppley Laboratory Inc.
The thermopile is usually located in an
evacuated housing, so that the heat losses through convection by the surrounding air
A are reduced. blackened receiver foil, with
the measurement junctions, is fitted behind the quartz window. The reverse side of the receiving foil is made reflective, so that its absorption and sensitivity can be neglected. Without irradiation, the measurement junctions are kept at the same temperature as the reference junctions. The reference junctions are screened from incident radiation. If radiation falls through the quartz window onto the blackened foil, the temperature of
236
Housing
ance is determined by dividing the measured value by the effective sensitive receiving
surface area.
Measurement output
Figure 12.6
Thermopile as produced by Eppley Lab. Inc.
the measurement junctions rises in comparison with the reference junctions. Changes due to the ambient temperature are compensated for by the reference junctions.
Radiation measuring instruments with
thermocouples and thermopiles sometimes
have calibration curves or calibration tables
for correction of the measured values. Such
radiation measuring instruments measure the
incident radiant power. They are usually
mW calibrated in mW,
or watts. The irradi-
12.4
Measurement of radiant power of the stan-
A dard light radiation with the thermopile
Radiation measurements are advantageously carried out in a separate laboratory room. The daylight entering through the window will be shut out with a closely-woven blackout curtain. The front of the curtain is
rubber-coated. Any further room lighting will be switched off. The undesired stray radiation from the actual radiation source must not fall on the measuring receiver through reflections from the room walls, the
work-bench surfaces and instrument surfaces. The room walls should therefore be painted matt black. The bench top and instrument surfaces will be covered with matt black material or cardboard. Measurement errors mainly occur with thermal photodetectors, if they are subjected to draughts of air. Therefore, the windows and doors are to be well closed and fans and air-conditioning installations switched off.
As shown in Figure 12. 7, the Wi lamp, the shutter and the thermopile are mounted on an optical bench at equal heights in the
Mask Shutter
TTiermopile
Figure 12.7 Measurement set-up on the optical bench for measurements of radiant power.
237
optical axis. The Wi lamp is operated with standard light A, as stated previously. The shutter is arranged approximately 10 cm away from the filament. The shutter aperture should be somewhat larger than the
maximum dimensions of the filament spiral or the window aperture of the Wi lamp. The
whole shutter must completely screen the thermopile, like a mask, against the lamp radiation outside the beam of radiation needed for measurement, since the longerwavelength IR radiation would heat up the quartz window of the thermopile unnecessarily. The matt black back of the shutter is towards the receiver. Its reflective front is directed towards the radiation source. The incident stray radiation from the filament lamp is very well reflected. The shutter can therefore not cause interference as a secondary radiation source. For the same reason, the lamp Wi 4 1 is also more suitable than the Wi 40. The radiation from the Wi 41 only emerges through a window in the lamp bulb. The remaining internal surface of the lamp bulb is coated with an opaque
film. This so-called skirt has the disadvantage, however, that it also acts as a thermal
radiator (T « 500 K).
For measurements of radiant power, Texas Instruments often use a silver-bismuth thermopile from Eppley Laboratory Inc. This is supplied already calibrated. This is supplied already calibrated. This firm can carry out this calibration in accordance with the rules of the National Bureau of Standards, USA, or of the National Physical Laboratory, England.
The quartz window of the thermopile must be clean. The response time of the Eppley thermopile is about 30 s. The measurement begins with the shutter closed. The value due
to the residual stray radiation is noted as the dark value. Following this, the shutter is opened. The distance of the thermopile from the Wi lamp is varied, until the desired radiant power is indicated. This measured value is called the exposure value. With the shutter closed, the dark value is checked once more. The true value of the incident
effective radiant power is calculated from the exposure value, corrected with the Eppley calibration curve, minus the corrected dark value. The actual value does not yet correspond exactly to the desired radiant power. The measurement is repeated, until the actual and the desired radiant power values agree. The actual purpose of this radiant power measurement is the subsequent calibration of a solar cell and the determination of the photocurrent sensitivity of Si photodiodes and phototransistors.
12.5
Measurement of irradiance of a standard
light A radiation with Si photodetectors
Irradiance measurements are necessary to
determine the photocurrent or the photo-
sensitivity of a photodetector. The photo-
sensitivity of Si photodiodes and Si photo-
transistors is related to given irradiance
values. Sensitive phototransistors are
measured with low irradiances such as
1
mW/cm 2 ,
2 mW/cm 2 , 5
mW/cm 2
and
9 mW/cm 2 , and less sensitive phototransis-
tors and photodiodes with higher irradi-
ances such as 20 mW/cm 2 .
From the preceding measurement of radiant; power with the thermopile, the irradiance is the quotient of the radiant power and the
effective receiving area. With the thermopile:, the required irradiance, corresponding to the data in the data-sheet of the photodiode or phototransistor typs, is determined exactly at the relevant distance from the radiation source. The photocurrent sensitivity of the corresponding photodiode or phototransistor is measured at the same position originally occupied by the thermopile.
For a series of measurements in the test department, such radiation measurements are very time-consuming. In addition, the large requirement for calibrated measurement detectors demands very inexpensive equipment designs. In the test department, Texas Instruments use calibrated Si solar cells as measurement detectors. The spectral
238
--
sensitivity of Si solar cells and the spectral
A emission of the standard light radiation
source can be regarded as constant. The proportion of the radiation which can be evaluated by the Si solar cell also remains constant. Therefore, the actual irradiance can be determined from the relative measurement with an Si solar cell.
The radiation measurements in the same arrangement with Si solar cells are less troublesome, since the longer-wavelength radiation from the filament lamp is not
evaluated. Also, the Si solar cells are
unaffected by their previous history, so that the photocurrent sensitivity has not been impaired by previous measurements with high irradiances. The value of the shortcircuit current of a calibrated solar cell can be reproduced exactly with the same irradiance. The short-circuit is measured with the same low-resistance microammeter. The voltage drop on the microammeter should only be 10 mV. If large-area solar
cells (A > 1 cm 2 ) are measured with irradiances above 5 mW/cm 2 , then a voltage
drop on the microammeter up to 100 mV
can be permitted. It is convenient, almost to short-circuit the solar cell with a low resistance (e.g., 10 fi). The short-circuit current is measured through the voltage drop across this resistance, with a digital voltmeter.
The calibration of the solar cell only needs to be checked infrequently, since it has high long-term constancy. The calibration itself is carried out with the thermopile, by measuring the irradiance for various distances
from the filament of the standard light A
radiation source. The thermopile is then replaced by the solar cell in the same position for each measurement point.
For each measurement point, the shortcircuit current of the solar cell and the irradiance, determined with the thermopile, are plotted in a graph. As an example, Figure 12.8 shows the short-circuit current of a solar cell as a function of the irradiance. If the radiation source can still be regarded as a point source from each measurement
Ip(mA) 10
,,'
i£ -* <
1
01
0-1
4~-
i
T M ^ e|ez^
* :g
-X-
J
h
:£:
/
i
1
A mm Mleasurement detector: Solar cell, = 360
1-
mitter:
Wj 40, Tf = 2856 K
:
M :
mW/cm' 2 138 mA -
1 -r mW/cm' 8 13-8 mA -
t
1
1
! 1 II
1
10
100
Figure 12.8
Short-circuit current Ip of a Si solar cell as a
E function of the irradiance e. If the short-
circuit current Ip of the solar cell is not a
E linear function of the irradiance e, then the
solar cell is defective or a measurement error has occurred
point and the solar cell has a strictly linear relationship between its short-circuit
current Ip and the irradiance Ee , then the
values of the measured short-circuit current Ip = f(Ee ) produce a straight line. Further measurement points can also be calculated with the inverse-square law.
Si photodiodes or phototransistors could also
be considered as calibrated detectors. Of
course, phototransistors can only be calibrated for fixed measurement points, since their non-linearity Ic = f(Ee) causes severe disturbance. Photodiodes are more suitable.
They have a linear relationship for Ip = f(Ee), at least over a wide range. However, because their sensitive areas are mostly very small, these photodetectors require very careful adjustment on the optical bench. The wafer should be centred in the component within close tolerances.
Built-in lenses are undesirable, in order to avoid possible misalignment effects.
239
Components with flat glass windows should be used, since glass windows can be cleaned without damage, in constrast to plastic windows. Furthermore, components with
metal cases are preferable for this purpose, in order to obtain a low thermal resistance between the junction and the case. The wafer of a photodiode, which is heated by the long-wavelength radiation from the lamp, is cooled better. Therefore measure-
ment errors due to the temperature rise of the wafer can be neglected with short measurement times.
Calibrated photodiodes or photo transistors are only used more advantageously for measurement and calculation of the tolerances of the photocurrent for prototypes in the laboratory. They do not form a good substitute for a radiometer. If necessary, a radiometer can be built with a commercially-available Si solar cell. As previously described, the voltage drop across the short-circuiting resistance of
R = 10 JZ is measured with a digital volt-
meter. This arrangement is quite suitable for relative measurements. For absolute measurements and in case of simp.le requirements, the Si solar cell can be calibrated with a calibrated photodiode. Regarding the
A previously-mentioned standard light
source as a point source, a calibration curve Ip = f(Ee ) can be prepared for the Si solar cell used with the inverse-square law. In this connection, it is explicitly pointed out, that all commercially available radiation measuring instruments with silicon photodetectors, themselves have to be calibrated to the
A standard light radiation with the methods
described.
12.6
Measurement of a luminescence diode radiation with Si photodetectors
12.6.1
General Measurement Problems
Radiation measurements can be carried out
with commercially available measuring instruments and also with simple measuring
detectors. The commercial measuring instruments mostly contain a PIN photodiode as a measurement detector, while commercially
available Si solar cells or, if necessary, Si photodiodes are suitable for use as simple
measurement detectors. Silicon photodetectors are used as measurement detectors for
the following reasons:
The selectively emitted radiation from current luminescence diodes falls in the spectral sensitivity range of Si photo-
detectors.
Si photodetectors are very sensitive, so that very small radiant powers can be measured.
Si photodetectors have high long-term constancy and are not dependent on their previous history.
Practical radiation measurements with Si photodetectors are simpler to carry out, as compared with those with thermal photo-
detectors.
The Si photodetectors of commercial measuring instruments are obtainable with
attached subtraction filters.
These spectrally-corrected Si photodetectors are divided into three groups (see Figure 12.9):
1
For measurement of radiation in radiometric values within the wavelength range
from 450 nm to 950 nm with tolerances of approximately ± 5% (see curve a). With
such photodetectors, the measurement of Si-doped GaAs diodes is difficult, since the
measuring detector no longer covers 1 00%
of the spectral radiation distribution of
the GaAs diode. With an exact spectral
240
.
(%)
/
/.
3^
7 *"^*
\^
A /
i
'
/
i ^ ,
\
/
/
/
d
/
A
1/
c
"3
* 5
·as
O
n K >6
_c
^^ ..-- """*
A \ \
A i \ \ \A. --
350
400
450
500
550
600
650
700
750
800
850
900
950
1000 1050
1 100
\(mn)
»-
Figure 12.9
Spectral sensitivities of various corrected Si photodetectors
sensitivity distribution curve and the exact
spectral radiation distribution of the GaAs
diode, a correction factor can be determined
graphically.
For measurement of mixed radiation in photometric units within the photometric sensitivity curve of the eye, V(X), corresponding to curve b. This mainly means use as a Luxmeter.
For measurement of monochromatic or selective radiation in photometric units for a small spectral range within the photopic sensitivity curve of the eye, V(X). With these photodetectors, for example, the selective radiation of luminescence diodes is measured. Curve c shows the spectrally corrected sensitivity for measurement of red-emitting GaAsP diodes. For comparison, curve d reproduces the relative sensitivity s(X) of an Si PIN diode without a filter.
The measurement of medium radiant powers
can be carried out with a neutral, calibrated grey filter placed before the Si photodetec-
tor. Very high radiant powers (> 1 W) are determined with an optical beam switch
connected before the Si photodetector, since grey filters heat up with high radiant
powers. A beam splitter, with a defined
division ratio, for example, is an optical
beam switch.
A measuring photodetector with a A£ sufficiently large photosensitive area
always measures the radiant power <D e or the luminous power 4>v of an incident, narrowly-focussed, parallel beam or radiation, e.g., from laser sources. From radi-
ation sources arranged as a point radiator, a photodetector only ever measures the
irradiance Ee or the illuminance Ev. The
radiant intensity or luminous intensity of a point source of radiation in the normal direction is calculated with the equation (3.25) (see Section 3.3), while with the measurement arrangements which
are usual in practice, cos^e can be made = 1
E.r'
u7
(12.3)
The radiance or luminance in the normal
direction is calculated according to equation (3.19).
--I
L=
AS
(3.19)
In most cases, the radiance or luminance can only be determined with very wide tol-
241
,
,
erances, because the effective emitting
area As of luminescence diodes is not
exactly known. Only in the case of highquality GaAs power diodes is the emitting
area As known exactly.
Luminescence diode arrays for illumination or irradiation purposes often have to be dealt with as surface emitters. To characterise surface emitters, the irradiance values are stated for planes at various distances.
12.6.2
Measurements of relative spectral sensitivity with a monochromator
The relative spectral sensitivity s(\)rei of an Si photodetector can be determined with
a monochromator. A monochromator can
select, from the wide-band emission spectrum of a Wi lamp at, for example, Tf = 2856 K, a narrow-band radiation in the wavelength interval A\.
In principle, a simple monochromator consists of the inlet slit, a prism, lenses or
mirrors for parallel alignment or focussing
or a radiation and the outlet slit. The incident radiation through the inlet slit is split up through the wavelength-dependent refraction of the prism. The arrangement of
the outlet slit permits selection of the radiation within very small wavelength ranges.
The monochromatic beam of the desired wavelength interval A\ can be determined exactly by measurement through its chopper frequency, in comparison with the beam which emerges after the first pass with an unwanted wavelength range.
The relative spectral sensitivity s(\)rei of a Silicon photodetector is determined in two
steps. First the spectral radiation distribution
N\ of the Wi lamp is measured at Tf = 2056 K.
With the radiation falling through the inlet slit of the monochromator and the thermopile arranged at the outlet slit, the spectral
radiation distribution N\ is recorded for each
wavelength interval A\. The thermopile used must have a flat spectral sensitivity curve
between 300 nm and 1 300 nm.
Following this, in the same measurement arrangement and with the same radiation N\, the spectral sensitivity s(N)\ of the Si photodetector used is measured for each wavelength interval AX. The ratio s(N)\
to N\ for each wavelength interval A\ gives
the relative spectral sensitivity s(\)rei of the Si photodetector.
The measured relative spectral sensitivity values for an Si solar cell are listed in
Table 12.1.
Figure 12.11 shows the function s(\)rei = f(\) of the Si solar cell used, derived from these values.
Texas Instruments often use the double monochromator, Model 99, from Perkin Elmer. Figure 2. 10 shows, schematically, the optical path of the incoming radiation:
M Inlet slit S 1 , parabolic mirror 1 M prism PR, Littrow mirror 2, prism PR, M M parabolic mirror 1 , mirror 3 , divided M M mirror 4, chopper CH, mirror 3, paraM bolic mirror 1, prism PR, Littrow M M mirror 2, prism PR, parabolic mirror 1 M mirror 5 and outlet slit S 2.
The chopper CH interrupts the beam for its
second passage through the monochromator.
12.6.3
Irradiance measurement for luminescence diode radiation with Si photodetectors
A luminescence diode radiation can only be
evaluated by measurement with an Si
photodetector, if the absolute spectral
sensitivity distribution s(\) = f(\) is known. However, among commercially available Si photodetectors, this is only stated in the data sheet for the large-area photodiodes and for solar cells. For radiation measurements, a large-area photodetector is usually
desired.
242
miiuuw Indirect path
^
J
/
\
"w
nI
\
figure 12.10 Construction principle of the double monochromator, model 99, from Perkin-Elmer
When the requirements are not severe, the
data-sheet characteristic s(\) = f(\) is also
adequate for measurement purposes. The selective radiant power of a luminescence diode, failing on the photodetector, can be read off from the function s(\) = f(X) with the measured photocurrent Ip of an Si photodiode or the measured short-circuit current Ip of a solar cell and with the known
maximum wavelength Xmax °f the selective
luminescence radiation. If the measurement detector has a photosensitive area of
A£ = 1 cm 2 , the incident radiant power is
equal to the irradiance for 1 cm 2 reference
area.
As a rule, the data sheets on small-area Si photodiodes contain the relative spectral sensitivity distribution s(\) rei = f(\). In the same way, only the relative spectral sensitivity distribution s(\) = f(\) is obtained initially if, for exact spectral measurements of large-area Si photodiodes or Si solar
cells, the spectral sensitivity distribution is
determined with a monochromator (see
previous Section 12.6.2). In principle, the relative spectral sensitivity distribution curve of a Si photodetector can be converted into the absolute spectral sensitivity s(\) with one absolute spectral sensitivity
measurement at a wavelength measure-
243
Column 1
\
nm
1-100 1-050 1-000
950 900 850 800 750 700 650 600 550 500 450 400
Column 2
s(\)rel
12-8 31-9 61-1 85-4 97-6 100-0 97-9 92-9 85-2 76-6 66-9 56-0 44-3 31.7 18-2
Column 3
s(\)
A/W
0-07 0-175 0-335 0-468 0-535 0-548 0-537 0-509 0-467 0-420 0-367 0-305 0-243 0-174 0-100
Table 12.1
Spectral measured values of an Si solar cell
Column 4
Qtt)
%
7-9 20-7 41-5 61-1 72-5 80-0 83-2 84-2 82-7 80-1 76-5 69-2 60-2 47-9 31-0
Column 5
s<Mth
A/W
0-886 0-845 0-806 0-766 0-737 0-684 0-644 0-604 0-564 0-524 0-483 0-443 0-403 0-363 0-322
n: tt
uv
s(X)re lQ<M
s(X),h s(l)
/sMrel ' f(X) \
i
A/W
0-8-
Si1(QWfM
N \
Vff rA
//
0'7-
//
J*\ \
//
S / /
S ^^ / /
J X^\ M > /
S 1
N \
rfXMlh-ffxA \
\ \\ Sl(k) = t(k)
\\
0-4-
v\
0-3-
\M
0-2-
v\
1-
\
0-
400 500 600 700 800 WO 1000 1100
A (nm)
»-
Figure 12.11
Spectral characteristics of a solar cell
ment ^Meas- ^n simplified form, it is sufficient to measure the absolute sensitivity with a spontaneously emitting GaAs diode with known radiant intensity Ie and maxi-
mum wavelength A.max . Somewhat more
accurate measured results are to be expected, if the spectral radiation distribution of the GaAs diode used is measured with a monochromator and the part of the radiation evaluated by the Si photodetector is then determined graphically.
It is advantageous to measure the absolute spectral sensitivity s(\) of the Si photodetector at several different wavelengths. Texas Instruments measure the absolute spectral sensitivity s(A.) of Si photodetector :s with an argon laser at \ = 514 nm, with a
HeNe laser at X = 632.8 nm and with a
spontaneously emitting GaAs diode at \ = 910 nm. The absolute spectral sensitivity
values of the Si solar cell used are listed in Table 12.1, column 3. Figure 12.11 shows the function s(\) = f(\) plotted.
244
12.7
Measurement of the total radiant power of a luminescence diode
The total emitted radiant power of a luminescence diode can be determined with an integrating radiant power meter. The Ulbricht spherical photometer, for measurement of total luminous flux, is well known from illumination engineering (see Figure 12.12). The spherical photometer belongs to the category of integral photo-
meters, since it integrates all partial lumi-
nous fluxes. The light emitted from the measurement lamp is scattered and reflected on the matt white interior wall of the sphere, so that every surface element has the same
remitted luminance. A small aperture in the
sphere permits the measurement, with a sensitive photometer, of the remitted luminance of the aperture area. Direct radiation from the light source onto the aperture is prevented by a screen B. For the measurement of the radiant power of IR-emitting
GaAs diodes, spherical photometers are made from small aluminium hollow spheres.
ion (y)DVM
Silicon photocell
Figure 12.13
Measurement of the total radiant power of a luminescence diode with an Si solar cell
A luminescence diode is arranged opposite a
large-area Si solar cell. In addition, this diode is located in a special reflector, so that the total emitted radiation falls on the Si solar cell. With the measured short-circuit current Ip and the absolute spectral sensitivity value (A/W) read off from the characteristic
s(\) = f(\) at the maximum wavelength \max of the luminescence radiation, the
radiant power can be calculated in accordance with the rearranged equation (9.25).
$,
s(X)
a 2.4)
The quantum efficiency Q(A) of the luminescence diode, which is also of interest, can be
calculated according to the' equation (11.5):
IP
Q(X)D = IfQ(^)sc
(11-5)
QMsc The quantum efficiency
of tne
solar cell, which is necessary for this, can be
determined by means of equations (9.9)
and (9.4):
Figure 12.12
Construction principle of a spherical photometer
For the serial measurement of luminescence diodes, another measurement method has become established (see Figure 12.13).
^r Q(X)sc=
(9.9)
Q(X)SC
IP
$e,X- X
1-24 raW
6 10
A
(9.4)
To complete the spectral parameters of the
solar cell used, the values for s(X.) ln and Q(\)SC are therefore entered in the Table 12.1 and shown in Figure 12.11 as
characteristic curves.
245
13 Optoelectronic couplers
13.1 13.2 13.3
13.4 13.5 13.6
13.7
Direct Couplers Reflected optoelectronic couplers Couplers with non-stationary source emission Simple examples of couplers
Opto couplers with lenses Opto couplers with unmodulated optical
radiation
Opto couplers with modulated optical
radiation
247
13 Optoelectronic Couplers
Optoelectronic couplers or source/detector
systems are used in many varied forms in
almost all branches of electronics. For example, the so-called twilight switches are simple examples of such systems. Here, the natural daylight serves as the source radiation. The actual twilight switch is the detector. Optoelectronic rangefinders, for example, are more complicated source/ detector systems. Spontaneously emitting
GaAs diodes or lasers are used as the
radiation source.
The small couplers described in Section 10.3
relate to very short source-detector dis-
tances, which lie below the critical photometric distance. Those for short, medium or
long ranges relate to source-detector distances above the photometric critical distance. For the radiation calculation of couplers or systems, the fundamental prin-
ciples worked out in Part I are needed, with the parameters of optoelectronic semiconductor components described in Part II and with practical radiation measurements.
13.1 Direct Couplers
By the term "direct couplers", it is under-
stood that the photodetector is irradiated directly by the radiation source. In most practical applications, the normal to the source coincides with the normal to the detector. This means, that the photodetector receives the radiation in its major direction of reception and this is emitted in the major direction of the source. Figure 13.1 shows the principle of direct couplers. The
irradiance Ee falling on the photodetector
can be calculated for non-parallel source emission by equation (3.25), where cosipe
is made = 1
U E-=- Go
r
(3.25)
If, as shown in Figure 13.2, the source S is
aligned at an angle </><; to the detector E, then its radiant intensity Ie ^ can be calcu-
lated with the value of Ieo in the normal to
the source and the function f(^s) stated in the data sheet. The angle ipg is formed between the normal to the source and the
Figure 13.1
Principle of a direct coupler:
a) Single coupler b) Double coupler
Figure 13.2 Direct coupler with a source turned through the angle ips an-d a detector turned through the angle <pg.
249
direction from the source to the detector.
Ie,<p =I e,o- f(<0S)
(10.12)
The simplest method of determining the radiation falling on the photodetector from a parallel beam of radiation from the source is by measurement.
The photocurrent Ip of a photodetector working in the linear part of the characteristic Ip = f(Ee) can be calculated through the
rearranged equation (9.25).
Ip = s . Ee
(13.1)
For photodetectors working non-linearly,
the amplified photocurrent I\f or Ic is read off the characteristic curve I = f(Ee) in the data sheet.
If, as in Figure 13.2, the detector E is
aligned at an angle i^E to tne source, then its sensitivity s^, can be calculated with the value of s in the normal to the detector and the function f(vE) stated in the data sheet. The angle i^e is formed between the normal to the detector and the direction from the detector to the source.
Figure 13.3
Principle of a reflection coupler
A beam of rays from a source can also be
reflected onto the photodetector through several mirrors in series. Mirrors reflect an incident ray in accordance with the law of reflection. (See Section 6.4). They must be adjusted exactly. Their sensitivity to vibration causes serious problems.
Instead of direct couplers, reflection
couplers can also be used by the backreflection method. Figure 13.4 shows the
s$ = s · f(VE)
(13.2)
13.2 Reflection Optoelectronic Couplers
Reflection couplers differ from direct systems in that the radiation from the source falls on the detector via a beamdeflecting device. Figure 13.3 shows the principle of a reflection coupler. To deflect the beam from the source, mirrors, triple reflectors, diffuse reflecting surfaces or any desired materials are used. The reflectivity <p of commercially available reflectors can be
taken from the data sheet, that of known
substances from tables or characteristic curves. The radiance returned diffusely from a reflector was calculated in Section 2.3.3. Diffuse reflection was described in Section 6.4.
Figure 13.4
Principle of a back-reflection (autocolli-
mationj coupler
principle of this variation, known as the
back-reflection or autocollimation coupler.
The source and the detector are mounted in one housing. Triple reflectors (back-
reflectors) are used instead of mirrors to
A reverse the beam. triple, reflector throws
the radiation back to its point of origin. In comparison with a mirror, it can be misaligned by a few degrees. The triple reflector is an exact corner of a cube, made of a transparent medium, e.g., glass. The backreflection takes place almost without losses through total reflection. Figures 13.5 and 1 3.6 show the principle of a triple reflector.
250
$J>
Figure 13.5
Principle of back-reflection with triple reflectors, in two-dimensional
representation
1,
E
EC
Dividing prism
Mirror in part of the beam path
<f *
fi
Wedges in beam path
·T
Figure 13.6
Principle of back-reflection with a corner of a cube, to be considered as reflective, in three-
dimensional representation
A further important optical component in
back-reflection couplers is the optical switch. It is needed to separate the emitted
and received radiation, since these both occupy the same or almost the same radiation space ("light tube"). The optical switch then prevents an optical short-circuit, so that the source does not irradiate the photodetector directly. Radiation losses always occur at an optical switch
{Figure 13.7).
Symmetrical lens system
VE=
^^r\ r ~~^^~\
---07
E -=
--"-E
---S
Figure 13.7
Principle of optical switches
251
13.3 Couplers with non-stationary source emission
Reflection couplers on the back-reflection principle are also operated with nonstationary source emission. This means that the beam from the source is usually moved
periodically over a given deflection range. Figure 13.8 shows the principle of a backreflection coupler with non-stationary source
emission. The focussed radiation from the
source strikes the triple reflector via the rotating mirror and the parabolic mirror.
The radiation thrown back from the triple
reflector passes in the opposite direction
along the same beam path to the optical switch and then to the photodetector. In
this case, the optical switch is a partially-
transparent mirror. The rotating mirror guides the emitted beam periodically over the preset deflection range (moving beam
A principle). light beam is produced, and
can be used, for example, for accidentprevention on machines or for measuring the length of materials.
/
//
Emitter
\V j
Rotating mirror
W GaAs diode X^j
-,1 '11
jjf- Detector lenses
rfoi > Si pholodiode
Optical switch, in this case half-silvered
^v Parabolic mirror
Triple reflector
selected according to the operating conditions needed, the most suitable constructional form for the particular application, the quality and quantity supplied and the purchase price.
The radiant power and radiant intensity of a luminescence diode, and the sensitivity of a photodetector are determined according to radiometric considerations and calcu-
lations.
In Chapter 3 and Section 10.1, the solidangle and radiant intensity calculations of luminescence diodes with approximately spherical radiation distributions were described. For this, the relationship
J^k = 27T(l-cosi£) . fio
(3.4)
applies. With the minimum radiant power
mm 3> e
of a luminescence diode, stated in
the data sheet, the radiant intensity
m Ie
j n is calculated.
$e,min
e,o,min
(10.16)
The radiant intensity I e of luminescence diodes with narrow emission peaks is determined more advantageously by means of an irradiance measurement in the normal to the source and with equation (2.40). The photometric limiting distance of the measurement detector from the luminescence diode is to be maintained as a minimum.
Figure 13.8
Principle of a back-reflection coupler with
non-stationary radiation
13.4 Simple examples of couplers
Optoelectronic components are generally
AE
(2.40)
The radiant intensity tolerances of luminescence diodes can be determined roughly, equally well with the tolerance calculation described in Section 10.2 or by batch measurements. In Table 13.1 , the radiant intensity values of various GaAs diodes are
listed.
Photodetectors show different sensitivities
252
Type
TIL 23 TIL 24 TIL 31 TIL 32 TIL 06 TIXL 12 TIXL 13 TIXL 14 TIXL 15 TIXL 16 SL1183 SL1278 TIXL 26 TIXL 27
Radiant intensity Ie in mW/sr
Calculated Typical min. values measured values
0-21 11 5-5 16-5 8-3
43 1300
1
0-26 3-9
1
2 25 0.5
15 7 22 11
46 1600
3
5
Table 13.1
Radiant intensity table for GaAs diodes
(For test conditions, see data sheets)
according to the type of incident radiation.
The sensitivity of Si photodetectors is stated, in accordance with the data-sheet
test conditions, for the tungsten filament
lamp radiation with a colour temperature
of TF = 2856 K or 2870 K (2856 K =
Standard light A). The data-sheet characteristic curves Ip = f(Ee ) and Ie = f(UcE). with the variable Ee , can only be evaluated for this type of radiation down to a mini-
mum irradiance of approx. Ee = 1 mW/cm 2 .
For photodiodes and photocells, the stated absolute sensitivity s or the characteristic curve Ip = f(Ee ) can usually be extrapolated
linearly to smaller sensitivity values (see
Section 9.4). The absolute sensitivity sjy[ of a phototransistor is calculated approximately
with the corresponding typical characteristic
curve Brei = f(I^) (See Section 9.6).
The sensitivity s for different radiation to
that stated in the data sheet, e.g., radiation Z, is calculated for photodiodes and phototransistors with the relevant actinic value ae (Z) and, for phototransistors, also
with the corresponding typical characteristic
B rel = fdc) ( See Section 9.8).
For exact tolerance calculations, the characteristic B rei = f(Ic) and the actinic value ae of the radiation used are to be determined for the photodetector selected.
For practical ra'diation calculations, the
m typical sensitivity s(LIR)93Q n for tne m luminescence radiation LIRa3on of a
Si-doped GaAs diode has been determined
with various Si photodetectors. Table 13.2 shows the typical photocurrents or collector currents for s(LIR)930nm anc* for various
irradiances Ee . The numerical value or 1^ at Ee = 1 mW/cm 2 corresponds, for
example, to the typical sensitivity s(LIR)930nm in
/XA
mW/cm2
The expected typical collector current 1^ of a detector in an opto coupler will be calculated with the values from Tables 13.1 and 13.2.
Example:
A coupler is to be designed approximately
with the GaAs diode TIL 31 and the phototransistor TIL 81, for a distance r = 6 cm. From Table 13.1,v/e obtain Ie typ,TIL31 = 25 mW/sr. The irradiance
;
falling on the phototransistor amounts to:
Ee
--2I e-
_
Oq
25 mW.sr
*
sr.6.6.cm
= 0.694 mW/cm 2
(13.3)
According to Table 13.2, the interpolated collector current is
mA ^.TILS.l = 4 -43
-
In an accurately adjusted coupler, with a pair, TIL 31 and TIL 81, selected according to the data, this value can be reproduced
fairly accurately.
253
«
_
Type
^^ 1 = s(LIR) typ .Ee MA
H62
TIL 64 TIL 67 TIL 78 TIL 81 LS613 LS614
Photodiode Phototransistor Phototransistor Phototransistor Phototransistor Phototransistor Phototransistor
1 mWcm' 2
44 mA 77 mA 1300 uA 1030 nA 6218 mA 258 nA 342 nA
Irradiance Ee
0.5 mWcm" 2
0.1 mWcnT 2
22 MA 36 mA 645 mA 505 mA
M 3235
116 mA 157 mA
4.4 MA 6 nA 120 mA 90 nA 610 nA
M 20
28 uA
Table 13.2
Typical photocurrents or collector currents for s(LIR)gsonm °fSi photodiodes and Si
phototransistors.
13.5
Opto couplers with lenses
The distance to be spanned by an optical coupler is primarily dependent on the
optical aids used. Suitable reflectors, lenses
or lens combinations in the beam path of the source and the detector cause a considerable increase in the radiant power falling on the photodetector.
It is only a secondary factor that the radiation source must have a sufficiently high radiant power, in order to produce an adequate irradiance for the photodetector with the optical aids selected.
Lastly, the photo-current sensitivity of the detector must be mentioned. It can be compensated for within wide limits by the subsequent amplifier stage.
1
hG
/»
-4T"%
T"
.7 ^~**
B
--
--
Figure 13.9
Formation of image of an object by a
double-convex lens
Figure 13.9 shows an object, an image of which is formed by a double-convex lens. The focal length f of the lens permits the calculation of the most important relationships with relation to the position and size of an object and the position and size of the image.
The general equation for a lens reads:
11 1
fbg
(13.4)
The magnification ratio will be calculated:
V
=
-B
=
b -
Gg
(13.5)
In these: f = focal length, b = distance to image, g = distance to object, h = height of incident ray, B = image size, F = focus on
V G image side, = size of object and = mag-
nification ratio.
For the formation of a useful image, single lenses are only suitable with very small aperture ratios. The aperture ratio is the ratio of the diameter d of the inlet aperture to the focal length f of the lens. The diameter d of the inlet aperture is determined either by the diameter d of the beam of radiation or by the diameter d of an aperture mask which limits the beam (see
254
Figure 13.12). The full aperture corresponds to the diameter d of the lens (see Figure 13.13). The quality of the image from single lenses is only sufficient for demands of low sensitivity, even with a
very small aperture ratio.
Riy from emitter
Figure 13.11 Parallel radiation through a plano-convex lens; a) for an incoming received ray, b) for an outgoing emitted ray.
source. Similarly, in the ideal case, an image of a radiation source at an infinite distance should be formed on the photodetector.
Figure 13.10 Occurrence of spherical aberration
With a larger image height h of the incident radiation, the image points move on the optical axis away from the focus F* (see Figures 13.9 and 13.10). This aperture error is called spherical aberration; it can be reduced to a certain degree by the shape of the lens, if the refraction of the rays is divided as equally as possible between the two lens surfaces. The lenses corrected in
this way only have an aberration minimum
for a single direction of incidence of the radiation and only for a single magnification
ratio.
With a plano-convex lens, the refractions are most favourably distributed, if rays parallel to the axis fall on the convex side of the lens (see Figure 13.11).
Ideally, the emitted radiation of a coupler should form an image of the radiation source
at infinity. In this way, a beam of rays parallel to the axis is obtained from a point
The radiant power of the image of the source
on the photodetector is firstly proportional
to the effective receiving lens area Ajr, which is limited by the aperture or by the crosssection of the beam of rays from the source,
and secondly it is inversely proportional to
the square of the image distance b. The
Ag effective receiving lens area
is pro-
portional to the square of the diameter d of
the inlet aperture. With a source at an
infinite distance, the image distance b is
equal to the focal length f.
For this case, the radiant power 4>e of the image of the source falling on the photodetector is proportional to (d/f) 2 . The
fraction d/f represents the relative aperture
or the aperture ratio. The aperture ratio d/f is expressed by twice the tangent of the half aperture angle a. The half aperture angle a is called the aperture. The reciprocal of the aperture ratio d/f gives the stop number of
the lens:
ad 2 tan - = -
2f
(13.6)
_f
d (13.7)
Figure 13.12 shows the relative aperture of a lens, fixed by an aperture stop. In Figure 13.13, the relative aperture of a lens with full stop setting is shown. Finally,
255
Figure 13.14 shows lenses with equal rela-
tive aperture at full stop setting.
Figure 13.12 Determination of the relative aperture with an aperture stop; the angle a. is the aperture angle
J
The front lens to suit an optoelectronic component depends in each case on the specifi-
cation and on the aperture angle ©hp of the
emission or reception peak. When selecting a
lens, the aperture ratio relates at first to the
full aperture of the lens. Narrow peaks have a low ray divergence, so that planoconvex lenses with small aperture ratio are con-
veniently selected.
If an image of the source is to be formed, relatively accurately, on the photodetector, irrespective of its emission peak, then planoconvex lenses with small aperture
ratio are also needed. Spherical maxima
have a wide divergence, so that aspherical
condenser lenses are used to advantage. A
large aperture ratio of the aspherical lens is to be selected, in order to cover a large part
of the spherical peak. The whole spherical maxima can be covered with suitable concave mirrors, but not with lenses. Since an exactly parallel beam of radiation cannot be produced with a single lens, the lens for optoelectronic components with flat windows is to be adjusted to form an image of the wafer on the distant object plane (see Figure 13.151
Figure 13.13 Determination of the relative aperture at full stop setting
Figure 13.14 Lenses with equal relative aperture at full stop setting
In practice, if requirements are not very severe, both the production of the parallel
beam and its focussing on the photodetector are carried out with only one lens in each
case.
Figure 13.15 The front lens for an optoelectronic com-
ponent with a flat window is adjusted to form an image of the wafer on the distant
object plane
For an optoelectronic component with builtin lens, the attached lens is either to be adjusted to form an image of the wafer or of the edge of the lens on the distant object plane. (See Figures 13.16 and 13.17.)
In general, large aperture ratiosd/f produce a beam which also has a large diameter.
Opto couplers designed in this way are
uncritical in their construction and can be
256
Figure 13.16 Here, the front lens for an optoelectronic component with built-in lens is adjusted to
form an image of the wafer on the distant
object plane
Figure 13.17 Here, the front lens for an optoelectronic component with built-in lens is adjusted
to form an image of the edge of the lens on
the distant object plane
V S ) C\
^fe»
Source: GaAs diode TIL 31
mA Parameters: Iq = 25°C; If = 100
Main peak: l£ = 21 mW/sr
mW Total emission: e <t> = 3-72
Source lens
Dimens ons in mm
d = 42 f = 50
Double convex
d = 22.4 Aspherical
f = 18
condenser lens
d = 22.4 Aspherical
f = 18
condenser lens
d = 31.5 A erical
f = 25
condenser lens
d = 22.4 Aspherical f = 18 condenser lens
d = 42 , _ ,,,
Planoconvex
Distance
r
m
1.20 2.50
d = 31.5 Aspherical
f = 25
condenser lens
3.00
f = i go Double convex
3.00
f - 1 90 Double convex
3.50
f = 190 Double convex
f = too Double conves
The forward current Ip =100 mA mA was raised to Ip = 200
(permissible at tQ = 25° C)
5.25 11.0 11.0
+>(J ' E
Detector: Phototransistor TIL 81
s(PA)2870K = I 4 - 7 mA/5 mW/cm 2
Ee,TIL31 = 0-1 mW/cm 2
\q=
0-72 mA
0-5 mW/cm !
2-9 mA
VCE = 5V 1 mW/cm 2
8-3 mA
Receiver lens
mm Dimensions in
d = 31-5 Aspherical
f ~ 25
condenser lens
d = 22-4 Aspherical
f = 18
condenser lens
d = 31-5 Aspherical
f = 25
condenser lens
d = 31-5 Aspherical
f = 25
condenser lens
d = 22-4 Aspherical f = 18 condenser lens
d = 42 f - 50
Planoconvex
d = 31-5 Aspherical
f = 25
condenser lens
f - 1 *^0 Double convex
d = 315 Aspherical
f = 25
condenser lens
I . ,,, Double convex
f
? - ISO Double convex
Collector current of the
TIL 81
!C,TIL81
mA
1.2 0.066 0.165
0.190 0.043 0.200 0.025 2.2 0.450 8.0 3.0
6.5
Table 13.3
Examples of direct opto couplers with a calibrated GaAs diode TIL 31 and a calibrated Si phototransistor TIL 81
257
very easily adjusted. Small aperture ratios d/f
also give a beam of small diameter. Depending on their cost, these couplers are sensitive
to vibration and need precision adjusting
devices.
The examples listed in Tables 13.3 and 13.4 were constructed with lenses from Spindler and Hoyer. This firm is one of the few which supplies individual lenses, even in small batches. The well-proven GaAs diode TIL 3 1 was chosen as the source. With a
housing temperature tQ = 25° C and a diode
forward current Ip = 100 mA, a radiant
intensity Ieo = 21 mW/sr in the normal
direction and a total radiated power
mW e <i> = 3.72
were measured on the speci-
men. As the detector, the phototransistor
TIL 81, which matches the TIL 31, was
used. The housing temperature was
tQ = 25° C and the collector-emitter voltage
VCE = 5 V. The specimen used had the
following sensitivities:
m"
,*3,85
TIL 31 tc = 25° C; Ip = 100 mA
mW Ie = 21 mW/sr; e <l> tot = 3-72
--D{T] r = 2, 7s
New domestic
mirror
TIL 81 tG = 25°C; VCE = 5V
s(PA)2870K = I 4 '? mA/5 mW/cm 2
s(LIR)TiL3l = 8-3 mA/mW/cm 2
Source lens
mm Dimensions in
Detector lens
Dimensions in mm
Collector current 1q TIL81
mA
d = 62
d
f
=
Double convex 190
f
d = 62
d
f
=
Double convex 190
f
22-4 Aspherical
1
condenser lens
31-5 Aspherical
25
condenser lens
0-05 0-15
d = 62
d
50 Aspherical
f
=
Double convex 190
f
40 condenser lens
0-5
S \ TIL 31 as above
Table 13.4
Examples of reflection opto couplers with a calibrated GaAs diode TIL 31 and a calibrated Si phototransistor TIL 81
258
---- -- 14.7 mA
s(PA)2870K =
:
and
.
5 mW/cnr
=_--___ 0.72 mA
s(LIR)Ti L3i
and
0.1 mW/cnr
2.9 mA
8.3 mA
or
0.5 mW/cm3 1 mW/cm3
The radiant power and the sensitivity values of the specimen used lie below the typical values for these components.
13.6
Opto couplers with unmodulated optical
radiation
Unmodulated opto couplers are still relatively new at the present time. In the consumer sector, in particular, they are used more and more frequently. For example,
they are used in tape recorders for automatic tape-switching. In comparison with modulated optical opto couplers, they have a considerable price advantage, since they need considerably fewer components, both on the transmitter and the receiver side. At the source end, the radiation source is either connected to a constant-current source or, in very low-priced versions, to a simple power supply. At the receiving end, a DC-coupled amplifier is usually driven through a photodiode, a photo transistor or a photoresistor. Such detector circuits have one important disadvantage, since they cannot distinguish between the actual radiation from the transmitter and radiation from external sources. Very sensitive detectors respond both to background radiation and to reflected ambient radiation. For example, during interruptions of an unmodulated optical coupler, light-coloured material can reflect the radiation from the
surroundings onto the detector. No response
signal can be given. For an adequate contrastcurrent ratio, the desired useful photocurrent Ip(useful) should be greater, by a factor of 10, or better still by a factor of 50, than the photocurrent Ip(interference) Pro-
duced by the unwanted maximum stray
radiation from outside sources. It therefore follows that: Unmodulated optical couplers should have a powerful source and a relatively
insensitive detector.
The source will be arranged in a position
with the darkest possible background, so that the detector only records the useful radiation and none or only little of the background radiation.
By skilful mechanical installation, the photosensitive sensors can be protected from background or ambient radiation. Tubes are mainly used to screen this unwanted radiation. The interior walls of these tubes
can reflect residual ambient radiation onto the sensor. Therefore, the interior walls are sprayed with matt-black photographic lacquer or synthetic lacquer. Matt black surfaces absorb a very large proportion of the optical radiation. The consumer industry often uses black plastic tubes made by injection moulding. The plastic used should be particularly tested for its transmission in the near IR range. For example, black coloured polystyrene has a very good transmission in the near IR range.
Further protection from background and ambient radiation is offered by very narrowly focussed reception characteristics of the photodetector. If the sensitivity is highly dependent on the angle of incidence of the rays, this has the effect that background radiation is only received in the major direction. Detection of ambient or background radiation from other angles is thus effectively suppressed. The radiometric characteristics of the photosensitive sensor are greatly narrowed with optical reflectors or lenses. With correct arrangement of these optical aids, the apparent photosensitive sensor area is magnified several times. The incident irradiance is thus also increased by a similar factor. With very closely focussed detector characteristics, the interfering background radiation can sometimes be shaded by the actual source
structure.
259
Opto couplers are subjected to various environmental effects. They are protected from mechanical stresses, such as shock, impact or vibration, by very rigid housings
in cast iron or glass-fibre-reinforced plastic.
The housings are given an appropriate paint
coating against atmospheric oxidation or corrosive substances such as acids and alkalies. The electrical components and wiring are protected against moisture and condensation by a special, thin sprayed-on film of lacquer. Regeneratable desiccant cartridges absorb the moisture in the housing. Water-cooled housings give protection against great heat. Objective lenses are protected against the accumulation of dust by dust-protection tubes (Figure 13.18). Between the built-in baffles, the dust collects in so-called dust chambers. In addition, the lens can be cleaned with compressed air. The flow of air passes from the lens, through the dust-protection tube, to the exterior. The accumulated dust is carried away to the open air.
small animals (in the tropics). By suitable
construction, opto couplers can be safely used in explosion-hazard areas.
13.7
Opto couplers with modulated optical
radiation
The criteria listed in Section 13.6 also apply substantially for couplers with modulated radiation. The transmitter of the coupler contains a stage to modulate the forward current of the luminescence diode. The modulated radiation produced falls on a photodetector. The receiver amplifier is usually RC-coupled. Either amplitude, frequency or pulse modulation can be selected. Pulse modulation has the advantage, that the luminescence diode can be loaded with considerably higher pulse currents, as compared with continuouswave operation. At the receiver end, a more
favourable signal/noise is obtained.
r\c c clc c Objective
\i lens
V
I tlC C\ c
1
Comp ressed air
x
Dust chambers
c,
Baffles
Figure 13.18 Dust-protection tube
Misting of the lenses and front windows is prevented by a heating system. Electrical heating wires can be cast into the front windows. In tropical designs, all electrical components, including the wiring and the
chassis, are given a coating of a fungicide lacquer, as a protection against mildew. All necessary openings, e.g., for air circulation
and cooling, must be closed by fine-meshed
tropical screens. In general, the housings should be closed and sealed against dust, moisture, water splashes, oil, insects and
The relatively large spectral sensitivity range of junction photodetectors partly coincides with the spectral emission distribution of possible interference radiation
sources. Non-selective coupler receivers therefore respond to modulated interference
radiation. Filament lamps on the A.C. mains have an emission modulated at 100 Hz with relatively few harmonics. Gas discharge lamps, e.g., fluorescent lamps on the A.C. mains, have an emission modulated with 100 Hz and also with relatively strong harmonics up to 3 kHz and weaker harmonics
up to 10 kHz. A remedy is provided by
optical filters, placed in front of the photodetectors. Band-passes with interference filters would be an elegant but expensive solution. But the spectral sensitivity distribution of the photodetector can also be narrowed with inexpensive black glasses or edge filters. Filters cause the signal-to-noise ratio to be affected by the reflection and attenuation losses.
An improvement in the ratio of useful
signal to interference and also in the
260
signal/noise ratio is achieved, if the bandwidth of the receiver is narrowed and a
m modulation frequency f od> 10 kHz is
selected.
With suitable circuit design, coupler receivers with photodiodes or photocells can still be used safely, even if the interference irradiance is several order of magnitude greater than the modulated useful irradiance. It must be remembered, that the interference photocurrent Ipfint) produced in a photodiode causes a high current noise and thus worsens the signal/noise ratio. Photodiodes with small dark currents are used in sensitive coupler receivers. This
results in a low current noise and a low NEP value. When phototrarisistors are used, both
the useful irradiance and the interference irradiance are to be measured exactly. In principle, the working point of the phototransistor should be determined by means of the useful irradiance with the base current iP(useful) (photocurrent of the collectorbase photodiode). The useful irradiance
must have a minimum value, so that the
phototransistor still reliably amplifies the incident useful radiation in complete darkness (see data-sheet graph 1^ = f(Ee) ).
The phototransistor is shielded from high
interference irradiance levels, since the phototransistor, if driven into the saturation range, can neither detect nor amplify the incident useful modulated radiation.
261
14 Operation of luminescence diodes with direct current
14.1 14.2 14.3
Operation through series resistances Operation from constant-current sources Drive with logic circuits
263
When designing circuits with luminescence
diodes, account must first be taken of the fact that these have a very low differential internal resistance of only a few Ohms. In addition, the tolerances on the forward voltage, Vp, due to scatter between devices, and its temperature-dependence must also be taken into account. For these reasons, these diodes should only be fed from circuits, which have a high internal resis-
tance. In the simplest case this is achieved
by selecting a correspondingly high supply voltage Vfc and setting the desired diode
Ry current with the series resistance (Figure 14.1a). A more elegant method,
however, is to feed the diode from a constant-current source (Figure 14. lb). The diode variables previously mentioned can then be disregarded.
14.1 Operation through series resistances
A voltage source, e.g., a battery, in series
with a resistance is a simple current source. In this case, however, fluctuations in the
14
Operation of Luminescence Diodes with Direct Current
working voltage cause a variation of the
forward current Ip and thus of the radiant
power of the diode. Figure 14.2 shows the
Ry effect of the series resistance
on the for-
ward characteristic of luminescence diodes.
High working voltages and thus large series
resistances cause relatively smaller variations
of the radiant power in case of voltage
fluctuations.
Ry In practice,
is determined by the pre-
determined working voltage in the equip-
ment. Figure 14.3 shows two circuits with
series resistances for two different working
voltages.
14.2 Operation from constant-current sources
It is more advantageous, to operate lumi-
nescence diodes with a constant-current source. In this case, fluctuations in the working voltage have no effect on the forward current If and thus on the radiant power of the luminescence diode. Simple constant-current sources can be built both
Figure 14.1
Principle of operation of luminescence diodes, a) through series resistance, b) from constant
current source.
265
:
Since Iq > Ig, the calculation can be
simplified:
Vb-VcEsat-Vp
if
RV
5 V-0-3 V-l-6 V
=
180 n
IF = 20-6 mA
In order to drive the transistor as far as possible into the saturation range, the calculation is only based on a current gain
hFE > 30. Thus RB 3-9 kft. The input
current Ijl of the circuit thus lies below
1 mA; this corresponds to a fan-in = 1.
In the same way, circuits which are compatible with High-Level-Logic families can be built with discrete components. Figure 14.10 shows a circuit, which is
designed to drive circuits of the HLL family
"300". Since large fluctuations of working voltage are permitted here (V D = 10.5 to 16.5 V), it is not advisable to adjust the diode current Ip through a series resistance. In Figure 14.10, therefore, a circuit similar to that in Figure 14. 7 is used.
Vb = 10.5 ... 16.5V
V V V V = 0.7 + 0.7 + 6.2 - 0.7 = 6.9 V
The maximum possible diode current is determined by the loss power in the transistor T 2 (Pvmax = 0.8 W) and is calculated
at:
PV!max
:-max
Vbmax - Vre - Vf
W 0.8 56 mA
16-5 V -0-7 V-l-6 V
(tu = 25°C)
The emitter resistance Rp is then
RE = VBE1
0-7 V
12.5 ft
If 56 mA
In the same way, luminescence diodes can
be driven directly from TTL circuits. The types SN 7416N and SN 7417N, which can deliver an output current of Iql = 40 mA,
are particularly suitable for this. In this case,
the current is determined once again by a
series resistance {Figure 14.11). This is calculated in accordance with the formula:
, Vcc-Vol-VF 5V-0-7V-1-6V 2-7 V
RV
RV
RV
2-7 V RV =
if
Figure 14.10 High-Level-Logic gate to drive luminescence diodes
The Zener diode D 3 in the input circuit matches the threshold voltage Vjh at the
input to the corresponding values of the logic family and is calculated at:
vTh = VrE + VB E2 - VD 3 - VD l/2
1/6SN7417N
Control input
o Vcc = 5 V
RV
T
Figure 14.11
Driving of luminescence diodes by TTL
circuits
In principle it is also possible, to connect the luminescence diode between the integrated circuit output and ground, if the circuit in question has an inverting (totempole)
270
When designing circuits with luminescence diodes, account must first be taken of the
fact that these have a very low differential internal resistance of only a few Ohms. In addition, the tolerances on the forward voltage, Vp, due to scatter between devices, and its temperature-dependence must also be taken into account. For these reasons, these diodes should only be fed from circuits, which have a high internal resistance. In the simplest case this is achieved by selecting a correspondingly high supply
voltage Vb and setting the desired diode
Ry current with the series resistance (Figure 14. la). A more elegant method,
however, is to feed the diode from a constant-current source (Figure 14. lb). The diode variables previously mentioned can then be disregarded.
14.1 Operation through series resistances
A voltage source, e.g., a battery, in series
with a resistance is a simple current source. In this case, however, fluctuations in the
14
Operation of Luminescence Diodes with Direct Current
working voltage cause a variation of the
forward current Ip and thus of the radiant
power of the diode. Figure 14.2 shows the
Ry effect of the series resistance
on the for-
ward characteristic of luminescence diodes.
High working voltages and thus large series
resistances cause relatively smaller variations
of the radiant power in case of voltage
fluctuations.
Ry In practice,
is determined by the pre-
determined working voltage in the equip-
ment. Figure 14.3 shows two circuits with
series resistances for two different working
voltages.
14.2
Operation from constant-current sources
It is more advantageous, to operate luminescence diodes with a constant-current
source. In this case, fluctuations in the
working voltage have no effect on the forward current Ip and thus on the radiant power of the luminescence diode. Simple constant-current sources can be built both
Figure 14.1
Principle of operation of luminescence diodes, a) through series resistance, b) from constant
current source.
265
IF*TM*)
Typical diode characteristic
l F -«VF)
Vb (V)-
Figure 14.2
Typical forward characteristic of luminescence diodes with and without series resistance
Or
or
Figure 14.3
Operation of luminescence diodes through
series resistance
with bipolar transistors and with field-
effect transistors.
Figure 14.4 shows the output characteristics IDS = f(VDS) °f an N-channel junction FET, with the voltage between gate and source
VgS as a variable. In the left-hand part of
the graph, the resistance range, the output current Ids is strongly dependent on the voltage Vtjs applied between drain and source. In the right-hand part of the graph, the current saturation range, the output current Itjs only varies very slightly as a
function of the applied voltage Vds- With
circuits of this kind, care is to be taken, that the transistor is operated in this range under all conditions. For the circuits shown in Figure 14.5, field-effect transistors with a
slope of 5 to 20 mA/V and a pinch-off V voltage of about 5 to 7 are needed. With operating currents of Ip = 5 to 40 mA,
which are necessary for low-power luminescence diodes, the necessary gate bias voltage V(js is then to 5 V. In both circuits in Figure 14.5, the necessary gate
bias voltage Vgs is produced automatically
through the resistance in the source lead. The required diode current can be adjusted exactly with the 250ft potentiometer. The necessary working voltage for these circuits is determined in accordance with the following scheme:
Gate bias voltage V(}S Drain-source voltage Vrjs
to 5 V >4V
Forward voltage of the diode Vp
Minimum working voltage Vfc
V 1- 5 10-5 V
Because of the voltage drop across the drainsource path, which is sometimes very high,
266
lDS-f(VDS) Vqs = const.
ids
s-J©
Vgs
Saturation range
-4V v Ds(v>-
Figure 14.4
N Output characteristic curves of a self-conductive -channel FET
r
T lOOuF
lF = 20mA
With clip-on heat-sink
r
I00(i
T
I lF-20n
-o-12V
With clip-on heat-sink
Luminescence diodes: TIL 23/24;TIXL'26; TIL 31 ;TIL 32 01 similar
Figure 14.5
N Operation of luminescence diodes with FET constant sources, a) with -channel FET,
b) with P-channel FET
267
the loss power in the transistor becomes very large, so that these circuits can generally
only be used up to currents of approx. 40 mA.
diodes can be connected in series. The diode current is calculated in accordance with the
formula:
Constant-current sources can also be constructed with bipolar transistors. In these, however, a separate bias voltage, which is
stabilised - as shown in Figure 14.6 - with
-OVb = 24V
IN 754 (6-8V)
270 n
Figure 14.6 Series connection of luminescence diodes with a constant-current source.
a Zener diode, must be supplied to the base of the transistor. At the same time, with all circuits of this kind, several luminescence
IF = ic - IE
Vz-VbE 6.8 V -0.7 V 22-6 mA
Re
270 ft
When designing these circuits, care must be
taken, that the transistor is not operated in
the saturation range (Vqj < Vbe)-
The maximum possibly number, n, of
diodes in the collector lead is then calculated in accordance with the formula:
Vb > n.VF + VcEmin + Ve
n< Vb-VcEmin-VE 24 V-0-7 V-6-1 V
UF
1-6 V
n< 10
Further, simple current sources can be constructed with two transistors {Figure 14.7). In this case, the current is again determined by the emitter resistance Re- The transistor
&
Re
>
T1,T2 = BC213
U-7kn
y(F
Luminescence diodes: TIL 24; TIXL 26; TIL 32 or similar
4-7kn
®-
)
T3,T4 = BC 183
RE
12 J2
Figure 14.7 Constant-current source with transistors
268
I
T 1 , the base-emitter voltage Vgg of which
serves at the same time as the reference voltage, measures the voltage drop on the emitter resistance and then drives the tran-
sistor T 2. In this case, the diode current is
calculated in accordance with the formula:
VBE1 IF « E2 =
Re
0-6SV 54 mA nil
If luminescence diodes are operated in equipments in which large fluctuations of the working voltage are to be expected, it is advisable to stabilise the supply voltage for the diodes (Figure 14.8). In this case, the
«Vb «-!0. .-20 V
Figure 14.8
Operation of luminescence diodes from a
constant-voltage source; the supply voltage for the luminescence diodes is stabilized with a Zener diode
diodes are operated in parallel. In order to
ensure accurate current sharing, each diode
has its own series resistance. The diode
current Ip is determined by the emitter
voltage of the transistor and the series
Ry resistance
and is calculated with the
formula:
Vz-Vbe-Vf
If
RV
6-2 V- 1-5 V- 1-6 V
120 ft
IF = 25-8 mA
Since, in this circuit, all the anodes of the luminescence diodes are at ground potential, they can, if required, be mounted on a heatsink, without the need for special measures for insulation.
14.3
Drive by logic circuits
In digital systems, luminescence diodes
often have to be switched on and off by
digital signals. In this case, the circuits must
be so designed, that digital signals can carry
out the desired functions directly. The
circuit in Figure 14.9 can be driven directly
from TTL circuits. As well as the main-
tenance of the required diode current Ip, it
must also be ensured, that definite currents
flow in the diode with the levels supplied
from TTL circuits. The diode is therefore
connected in the emitter circuit. Therefore,
a voltage of at least Vfo = Vp + Vgg = V V V 1 .6 + 0.7 = 2.3 must be present at
the base of the transistor, so that a current
flows through the diode. The corresponding
input voltage before the diodes D 1 and D 2
Vd V is then Vj. = Vb -
V = 2.3 - 0.7 = 1.6 V.
Since, with TTL circuits, ViLmax < 0.8 V
and ViHmin > 2.9 V, the circuit is
TTL-compatible at this point.
o Vb - 5 V
Input 1 o-- Input 2 o--
-TO. 31, TIL 220
Figure 14.9 TTL-compatible control circuit for luminescence diodes
The diode current is calculated by the
formula
If = ic + Ib
269
:
Since \q> Ib, the calculation can be
simplified:
Vb-VcEsat-Vp
if
RV
5 V-0-3 V-l-6 V
180 n
IF = 20-6 mA
In order to drive the transistor as far as possible into the saturation range, the cal-
culation is only based on a current gain
hFE > 30. Thus Rb 3-9 kil The input
current Ijl of the circuit thus lies below
1 mA; this corresponds to a fan-in = 1.
In the same way, circuits which are compatible with High-Level-Logic families can be built with discrete components. Figure 14.10 shows a circuit, which is
designed to drive circuits of the HLL family
"300". Since large fluctuations of working voltage are permitted here (V^ = 10.5 to 16.5 V), it is not advisable to adjust the diode current Ip through a series resistance. In Figure 14.10, therefore, a circuit similar to that in Figure 14. 7 is used.
«Vb =l0,5... 16,5V
= 0.7 V + 0.7 V + 6.2 V - 0.7 V = 6.9 V
The maximum possible diode current is determined by the loss power in the transistor T 2 (Pvmax = 0.8 W) and is calculated
at:
PV;max
iFr
Vbmax - Vre - Vf
W 0.8 56 mA
16-5 V -0-7 V-l-6 V
(tu = 25°C)
The emitter resistance Rp is then
n VBE1 0-7 V
RE =
if
i2.5
56 mA
In the same way, luminescence diodes can
be driven directly from TTL circuits. The types SN 7416N and SN 7417N, which can deliver an output current of Iol = 40 mA,
are particularly suitable for this. In this case,
the current is determined once again by a
series resistance (Figure 14.11). This is calculated in accordance with the formula:
VcC- vOL-Vf 5V-0-7V-1-6V2-7V
lF =
RV
RV
RV
2-7 V RV =-
Figure 14.10 High-Level-Logic gate to drive luminescence diodes
The Zener diode D 3 in the input circuit matches the threshold voltage Vxh at the
input to the corresponding values of the logic family and is calculated at:
VTh = Vre + VbE2 - VD3 - Voi/2
Figure 14.11
Driving of luminescence diodes by TTL
circuits
In principle it is also possible, to connect the luminescence diode between the integrated circuit output and ground, if the circuit in question has an inverting (totempole)
270
output (Figure 14.12). The current is then determined by the internal organisation of
the integrated circuit.
-0VCC -SV
°--
Control input
I>
£
Figure 14.12
Driving of luminescence diodes by TTL
circuits
r~
rn
(T
ioonl
j
i
TI
J
vT~
i
t"~
h r=r
Figure 14.13
Circuit diagram for determination of the diode current when a gate of type
SN 74LS37N is used.
Figure 14.13 shows the part of the circuit
of the gate SN 74LS37N which determines the output current. The current through the luminescence diode is now calculated
from the formula:
Vcc " VcEsat 1 - VBE2 - Vf
If= £
5 V-0-3 V- 0-7 V- 1-8 V 24 mA
100 ft
However, two points must be noted with
this circuit: firstly the tolerance of the
resistance R is + 30%, so that reproducible
values can only be achieved with difficulty.
Secondly, the maximum permissible power
mW dissipation of the I.C. (Pvmax = 60
for
a 14-pin packager) must be observed.
Considerably higher currents can be
achieved with the integrated interface
circuits of the SN 75400 series. The maxi-
mum permissible output current Iql for
mA SN 75450 - 454 is about 300
per output,
so that when both outputs are connected in
mA parallel, a diode current of 600
can be
achieved (Figure 14.14). Of course, a
separate resistor has to be connected in each
collector lead, in order to achieve exact
current sharing. The necessary series resis-
tances are then calculated according to the
formula:
R _ Vcc-VQL-VF _ SV-0-7V-l-6V _ J ,,
lOL
300 mA
-- Inpui ! o
_l_^
\[
TIXL 16 ^
9 1 Si
I
T
k
i! '* '
'
L
X S\7S451S
Figure 14.14 Driving of luminescence diodes using interface circuits
271
15 Photodetector
circuits
15.1
Principle of operation
15.2
Detector circuits for two-pole junction
photodetectors
15.2.1 Direct relay control with phototransistors
15.2.2 Photo-Darlington circuits
15.2.3 Control of thyristors and triacs with
phototransistors
15.2.4 Driving of transistor and operational
amplifiers with phototransistors, photo-
diodes and photocells
15.2.5 Driving of multivibrators with
phototransistors
15.3
Detector circuits with three-pole
phototransistors
15.3.1 Operating modes of three-pole
phototransistors
15.3.1.1 Photocell and photodiode operation of the phototransistor TIL 81
15.3.1.2 Advantages of the phototransistor TIL 81
' in photodiode and photocell operation
15.3.1.3 Phototransistor operation of the TIL 81
15.3.2 Driving of amplifiers with three-pole
phototransistors
15.3.3 Phototrigger and photomultivibrators
15.4
Simple couplers with filament lamp and
two-pole phototransistors
15.5 15.6
Logical circuits with phototransistors
Photodetector circuits to drive TTL
integrated circuits
273
15 Photodetector Circuits
15.1 Principle of operation
Photodetectors are used to measure and evaluate radiation. The simplest optoelectronic couplers contains any radiation source as the emitter and a sufficiently sensitive photodetector for the direct control of a switching stage. Figure 15.1 shows
Radiation source
t·hotodetector
K
Trigger stage
*?/
Figure 15.1
Simplest opto coupler for unmodulated
radiation
the schematic diagram of a very simple coupler. The photodetector makes a Yes-No statement:
Yes = Sufficient radiation
No = Insufficient or no radiation
The measurement of grey steps or contrast current ratios is not possible by the above
principle.
Figure 15.2 shows the block diagram of a coupler for more severe requirements, where the photodetector circuit has been considerably extended. The photodetector
is selected for a desired sensitivity range.
The amplifier which follows is designed for a defined voltage or current gain. The threshold switch ensures definite switching points. In many applications, a large
hysteresis is chosen, so that in operation at the limit of sensitivity, the modulated ambient (interference) radiation from fila-
ment and fluorescent lamps does not cause triggering in time with the modulation. The
Radiation source
-w-
Constant -current source
00
7~\ Photodetector
X
V
Amplifier
~r--
s
Threshold switch
n
,
V
Power amplifier
~r~
Trigger srage
Figure 15.2
Principle of a coupler for unmodulated
radiation
power amplifier control the switching stage, which consists, in the simplest case, of a
switching transistor or -- if electrical isolation is necessary at the output - of an
optocoupler or a relay.
Figure 15.3 shows the principle of measurement of radiant power. For this, photodetectors with a linear relationship between the output signal and the incident radiant power are needed.
The output signal of the photodetector is
fed to a DC voltage amplifier. The amplified DC voltage then controls a measuring
instrument, on which the received radiant power can be read.
275
*
t>
1
Linear phoiodetector
\>
D.C. voltage amplifier
Measuring instrument
Figure 15.3 Principle of analogue measurement of optical radiations
IC>5mA
A [-7*1
Figure 15.4 Simple relay control with the sensitive phototransistor TIL 81
15.2 Detector circuits for two-pole junction photodetectors
Two-pole junction photodetectors are understood to mean photocells, photodiodes and phototransistors without base connection.
15.2.1 Direct relay control with phototransistors
Sensitive phototransistors already deliver relatively high collector currents. With adequate irradiances, low-current relays
can be controlled directly. The phototransistor TIL 81 has the following typical
parameters:
Ic = 20 mA at Ee2 856K = 5 mW/cm 2
These values are sufficient, as is shown in
Figure 15.4, for a direct relay control with
this phototransistor. The light-current relay is reliably turned on with an irradiance
of Ee = 5 mW/cm 2 , if the response current of the relay is below 5 mA. The diode BA 180 protects the phototransistor from
induced voltages when the phototransistor
is turned off.
15.2.2 Photodarlington circuits
Phototransistors can be extended with subsequent bipolar transistors to form Darlington circuits. Such circuits are more sensitive in proportion to the current gain B (= hpg) of the subsequent transistor. The total current gain Btj of the Darlington configuration is:
BD = B Phototr.-Bbip.Tr.
On this basis, simple but sensitive relay
controls can be produced. As an example,
Figure 15.5 shows an NPN phototransistor with a subsequent NPN transistor as an NPN
photodarlington in a collector or emitterfollower connection. Figure 15.6 shows a
complementary circuit with an NPN phototransistor and a subsequent PNP transistor as an NPN photodarlington. In both circuits,
the relay pulls in when the phototransistor
is irradiated.
In Figure 15. 7, the circuit of a simple Luxmeter with a photodarlington can be seen. With average illuminances and the design data stated, the Luxmeter has a sufficiently linear characteristic Ie = f(Ee ). The circuit can be calibrated by comparison with another Luxmeter.
276
.
®^>
kn
.1
vCEsat vCEsat TIL 81 * VBE BC 184
..E
fh
Figure 15.5
NPN Photodarlington as an emitter follower
Figure 15.6
NPN Photodarlington in a complementary
circuit
15.2.3
Thyristor and Triac control with
phototiansistors
In the same way, thyristors can be driven by phototransistors. Figure 15.8 shows the
theoretical circuit for this. If radiation falls
on the phototransistor with sufficient intensity, a current flows in the gate and
R fires the thyristor. The resistance 1 and the capacitor C 1 prevent the firing of the
thyristor by rapidly rising leakage currents and interference voltages. In addition, the firing threshold of the circuit can be varied
with R 1
Figure 15.9 shows a thyristor equivalent
circuit driven by a phototransistor. The
collector of the PNP transistor drives the
base of the NPN transistor and conversely
NPN the collector of the
transistor drives
the base of the PNP transistor. Through
the back-to-back coupling of the two tran-
sistors, a bistable circuit is obtained. The
phototransistor initiates the trigger process
through the base of the NPN transistor
(cathode gate). Thyristors and thyristor
equivalent circuits fire unintentionally
below their firing threshold, if the ignition
current or the anode voltage rise very
A rapidly (rate effect). remedy is provided R by the resistance 1 and the papacitor C 1
at the cathode gate and the capacitor C 2
at the anode gate.
=©
llOOuA
+1C V
TIL 67 TIL 78
^(^
t Rlfl Ik Si [J
+\
CI
rlnF
r
A
*TIC44
g/ K
Figure 15.7
Simple Lux meter
Figure 15.8 Thyristor control with a phototransistor
277
.
BC 18:
10k n RL
j-
Figure 15.9 Control of a thyristor equivalent circuit
formed from two transistors
Very sensitive thyristor drive circuits can be constructed with a photodarlington arrangement. In Figure 15.10, the thyristor fires when the phototransistor is irradiated, while in Figure 15.11 the thyristor is fired when the phototransistor is darkened. Furthermore, as well as thyristors, triacs can also be controlled. In this case, the phototransistor has to be adapted for operation on an alternating voltage. Figure 15.12 shows a simple circuit for A.C.
operation. A bridge rectifier is connected in
series with the phototransistor, in order to ensure that the voltage on the transistor has the correct polarity. In Figure 15.13, the characteristics Ic = f(VcE> of tr| is "A.C.
phototransistor" are shown. The family of characteristics, which are already known,
in the 1st quadrant, is repeated in the 3rd quadrant.
A further circuit for operation on alternat-
ing voltages is shown in Figure 15.14. During a positive half-wave, the photo-
T transistor 2 is short-circuited by the diode D 2 connected in parallel with it. The phototransistor T 1 receives its collector
voltage with the correct polarity, fed
D through the diode 2. Conversely, during a
negative half-wave, transistor T 1 is shortcircuited by diode D 1 and the phototransistor T 2 receives its collector voltage through the diode D 1
Figure 15.10
Thyristor control with an NPN
phototransistor-Darlington
BC 182
\
------fy* -~(Vbk*')
/_TJa
(r \
Ah TIL 78
)
·
HlOk.T []rl
- *0\
In i-i
'tic 45
7
= lnF
Figure 15.11
Thyristor control with an NPN
phototransistor-Darlington
Figure 15.12 Simple phototransistor circuit for A.C. operation
278
This circuit serves for the brightness control
or regulation of a filament lamp by phase-
angle control. Here, the triac, in series with the filament lamp which acts as the load,
forms the main circuit. The gate of the triac is driven through the trigger diode from the
RC phase-shifter network. When the conR trol voltage on 1 and C 1 reaches the
breakdown voltage of the trigger diode, this turns on and fires the triac. The triac is
turned off again when the AC voltage passes
through zero.
The A.C. phototransistor" is in parallel with
the phase-shifter capacitor C 1 . When the
phototransistors T 1 and T 2 are darkened,
the conduction angle in the triac is deter-
mined by the time-constant R 1 x C 1 . If radiation falls on the phototransistors T 1 and T 2, part of the current through R 1 is
diverted through the transistors and thus the charging of the capacitor is slowed down, so that the breakdown voltage of the trigger diode is not reached until later. Thus the conduction angle in the triac is reduced and the brightness of the filament lamp decreases. Conversely, with low irradiances, the conduction angle in the triac increases, as does the brightness of the lamp.
^I<
1
'
E«4
'
Figure 15.13
E Characteristic curves Ic~ f( vCE) °f a so-called A.C. phototransistor, with the irradiance e
as variable
279
TIL 66
r-
»
»
/
I T )T2 V
"
f
k
T1 )
TIL 66
Load
r£°9^
\ Rl
AT
BA 187
T
1
/ TIC 226D
HWI
^3ln,D2
TIC 56
o
s
CI
*01«F
A
BA 187
Figure 15.14
NPN Phototransistor circuit for A.C. operation with two diodes and two
phototransistors
for lighting control with filament lamps
15.2.4
Driving of transistor- and operationalamplifiers with phototransistors, photodiodes and photocells
Two-pole radiation-sensitive semiconductors can control a D.C. amplifier by various
principles.
Since a phototransistor or a photodiode behaves like a constant current source, variations in the working voltage cause little or no variation in the output current.
The output current of a phototransistor or a photodiode either controls an amplifier directly, or the amplifier is controlled by the voltage produced on the working
resistance of the photodetector. If the output impedance of the photodetector is high in comparison with the input impedance of the amplifier, the term "current control" is used. In the converse case, the term "voltage control" is used.
Figure 15.15 shows the current control of
transistor amplifiers with NPN phototran-
-o+Ub
rs
rc MOk a
3 Output
·-)
Mn r4jioon
a)
o+ub
Output
mLS '4022
-o+Ub
-0tUb RB 22k a
Rs Mikn
BC214
Output
> Output
RB 22k RC 10k n
S® RC 10k a
Figure 15.15
NPN Current control of transistor amplifiers with
phototransistors
280
0*
-o+V b -o Output
>
Rb
Q*
-o-V b --o Output
)
Rb
Figure 15.16 Current control of transistor amplifiers with photodiodes
sistors. The collector current of the photo-
transistor controls the base of an amplifier
transistor, which is operated in common
emitter connection. Interfering leakage
currents are led off through the base .resistance Rg. In addition, the sensitivity and
in some cases the switching time of the
phototransistor can be varied with a
variable resistance in the base circuit.
Scatter between individual components can
sometimes be compensated for by a feed-
Rg back resistor
in the emitter circuit. If
very high irradiances fall on the phototran-
sistor, a protective resistor is to be provided
to limit the loss power. If radiation falls on
the phototransistor in circuits a) or c), the
output voltage decreases in each case, while
in the circuits b) and d) it increases.
Figure 15.16 shows the current control of
transistor amplifiers with photodiodes, while
the latter, together with the transistor, form
an NPN phototransistor in circuit a) and a PNP phototransistor in circuit b).
In comparison with phototransistors, photo-
diodes have smaller photocurrent tolerances, since the effect of the current gain B is eliminated. Also they have a linear characteristic Ip = f(Ee). In this, Ip is the photo-
current and Ee the incident irradiance. If
one of these two characteristics is needed,
the use of photodiodes is recommended, despite the greater expenditure on ampli-
fication.
Figure 15.17 shows the voltage control of
transistor amplifiers with NPN phototran-
sistors. The collector current of the phototransistor produces a voltage, which controls the subsequent amplifier transistor, on the working resistance R^. The amplifier transistor is operated as in emitter-follower. Scatter between individual phototransistors is compensated for by a variable working resistance Rj±. With this, the switching times of the phototransistors are affected considerably more than with the resistor
Rg in Figure 15.15.
In Figure 15.18, the voltage control of an operational amplifier with a photodiode is shown. The inverting input receives its input voltage through the feed-back resis-
tance R 1 from the output of the opera-
tional amplifier. If radiation falls on the photodiode TIL 81, a current flows through its working resistance K\. There is a positive voltage drop across the working resistance R\, and this drives the non-inverting input of the operational amplifier. The voltage gain is determined by the resistance ratio
R 1/R 2. The offset voltage can be balanced R with the resistance 3. The circuit is
281
-o+vb
t ° +vb
-o-Vb
-o-Vb
a©
K "-f-- JBC184
Output
3 Output
ReU
\lv
80214
J
i®
Output
RE 4-7k S!
5® ~®
3 Output
RAfl Re|,
47k S2 4-7kJ2
Fz^wre 15.1
NPN Voltage control of transistor amplifiers with
phototransistors
excellently suited for sensitive Luxmeters. It can equally well be used in optoelectronic couplers or as a detector for modulated radiation. The upper frequency limit is determined firstly by the operational amplifier and secondly by the junction capacity of the photodiode and its working
resistance R^.
Power matching of a photodetector to the following circuit is mainly applied with solar cells for power generation. Basically,
a photosensor represents a generator. It
can work in the no-load, the matched or the short-circuited mode. In the no-load mode, the sensor must not be loaded, so that an amplifier with a high input imped-
ance or a high-resistance measuring instrument must be connected to it.
Figure 15.19 shows the no-load voltage
Vp l as a function of the irradiance Ee for
a photosensor. The characteristic curve has
an exponential form, so that saturation
R A MiMn
Figure 15.18 Photodiode D.C. amplifier
-o Output Gain »44
fBO = 1 ! kH;
282
w"
,
v. I.
6O0-
400-
V P,L
.
200/
/^
.10
mA
-8
-6
-4
2
Ec
mW/on1
Figure 15.19
No-load photo-voltage Vp i and short-
£ circuit photocurrent Ip as functions of E the irradiance e for a photocell
occurs at high irradiances. The no-load operation of a photosensor has little prac-
tical importance.
In Figure 15.20, the short-circuited
operation of a small-area photosensor is
represented. In this case, the internal
resistance Rj of the sensor is large in pro-
portion to the load resistance R^. With
large-area photosensors, the load resis-
n Ra tances
have values of approx. 0.1
Figure 15.20 Short-circuited operation of the photosensor TIL 81
to 100 n, while all small-area photosensors can be used with load resistors of over 1 kfi. In contrast to the no-load voltage
Vp l, which has an exponential charac-
teristic, there is a strictly linear relationship
between the short-circuit photocurrent
k Ip and the irradiance Ee . This can also be
seen from Figure 15.19. Therefore, photosensors are usually operated in the short-
circuit mode practice.
Figure 15.21 shows an operational amplifier driven by two TIL 81 photosensors, connected in opposing parallel. The two input resistors Rj\/2 of the amplifier form,
_ + * ~3 --
-f
TIL 81
3+
X 100(1 F I
-o Output
Ra/: Dikn RAr-D ,kn
100 uF
Figure 15.21 D.C. amplifier with photosensors connected in opposing parallel
283
together, the load resistance of the photosensors. If an equal radiant power falls on both sensors, then the voltage at the amplifier input, and thus also at its output, is zero. If the two devices are irradiated unequally, the difference signal is amplified, by a factor of 100. This circuit is used in measuring instruments, for example, for relative measurement of the photocurrent sensitivity (with one photosensor as reference and a second photosensor as the test device), as a direction-dependent detector for couplers and, where requirements are not severe, as a distance /voltage transducer for follower controls.
RA I Ik n
Qr
H®
--o Output
Figure 15.23 Phototransistor cascade circuit with improved dynamic characteristics
replaced by a common base connected NPN
transistor. The input impedance rgE °f tne base circuit is very low and is approximately:
rBE * VT
ic
26 mV
'26 J2
1 mA
V-p is the thermal voltage. In accordance with the equation (10.62)
Figure 15.22 Equivalent circuit of an A.C. phototransistor for position control
Figure 15.22 shows the principle of a follower or position control with an "A.C. phototransistor" circuit. With equal irradiance on the phototransistors, the
capacitor C is charged up during both A.C.
voltage half sine waves through the transistors, with the same charge but opposite polarity. The resultant voltage is therefore approximately zero. With unequal irradiance on the phototransistors, the difference signal is amplified, to energise a motor, for example.
Figure 15.23 shows the principle of a phototransistor cascade circuit to improve the dynamic range. In this, the working resistance of the phototransistor has been
1
6
A 2 7T.R .C
a high limiting frequency is achieved, since the working resistance of the phototransistor is only about 1 /l 00th of the original R-A value. This very low resistance has the effect, that the voltage gain and thus the dynamic collector-base capacitance (Miller capacitance) remains small. The voltage gain Vjj of the whole cascade circuit corresponds to that with an emitter-connected phototransistor.
Figure 15.24 shows a further cascade circuit with a photodarlington transistor. It also serves to improve the dynamic characteris-
tics. The resistance Rg has been inserted
between the base and emitter of the tran-
sistor T 2, in order to divert the leakage
currents of the transistors.
284
--o Output
Figure 15.24 Circuit to improve the dynamic characteristics of a photodarlington transistor
15.2.5 Control of multivibrators with phototransistors
Figure 15.26 shows the control of a Schmitt trigger by a phototransistor. If a sufficiently high irradiance falls on the
phototransistor TIL 65, the transistor T 1
becomes conductive. Thus the base current and also the collector current of the tran-
sistor T 2 is reduced. As a result, the
potential at the emitter rises and the
T current in the transistor 1 increases
further, until the circuit suddenly trips over, so that the output is switched off. This circuit can also be constructed with
NPN transistors. It is used as a simple
optoelectronic detector or "twilight switch"
100k Jl
270 n
sctr M -- Til
v'
2?k n
>ikn
Figure 15.25 Temperature compensation for phototransistors without base connection, with a selected pair of phototransistors
Figure 15.26
Control of a PNP transistor trigger circuit
with an NPN phototransistor
In Figure 15.25, a method for temperature compensation of two-pole junction semiconductor devices is shown. For this, matched pairs of transistors are used. With
rising temperature, the leakage current of
the phototransistor T 1 is compensated for
by the leakage current of the (non-
irradiated) phototransistor T 2, which
flows in the same direction. Instead of phototransistors, photodiodes can also be
used.
In Figure 15.27, a somewhat more com-
plicated circuit for a photodetector is
shown. The output signal of the photo-
transistor T 1 is amplified further in the
subsequent Darlington stage and fed
through a decoupling amplifier T 4 to the input transistor T 5 of the Schmitt trigger. The output transistor T 6 of the Schmitt A trigger controls the relay directly. When
the phototransistor is irradiated, the relay
is pulled in. The sensitivity of this circuit
can be varied with the variable resistor R 1
285
fjun
Tl
L«
^3 -- >
·
BC 184
!47nF
moon
n
jt y
r2
T6 !22jiF
Vbz^l
nD
22° n
p
Figure 15.27 Optoelectronic detector
:©
Rs
lkn
il
-w-
)
3--?-f BC 183
CE
CI
,
47 F -(j
470 JiF
|2-2kfi 4-7kSi
|IOkS!
|
R4
470 n
Twilight switch with flasher
Figure 15.28 shows a twilight switch combined with a flasher. With the onset of dusk, the collector current of the photo-
T transistor 1 is no longer sufficient to
keep the transistor turned on. The transistor
T 2 blocks, whereupon the base voltage and also the emitter voltage on T 3 rise.
This voltage change also acts, through the
capacitor Cg, on the emitter of T 4, so that
the current in this transistor decreases. The
collector voltage of this transistor is fed
back through the diode BA 180, so that the
circuit suddenly trips over to the new state.
The capacitor Ce is now discharged through R T the resistor 4, until the transistor 4
again becomes conductive and thus blocks
the transistor T 3. This process is repeated
periodically, so that the luminescence
diode TIL 220 flashes.
In Figure 15.29, a more sensitive Schmitt trigger, with the operational amplifier
SN 72741, is shown. In this application, the
operational amplifier only needs one supply
voltage.
286
TIL 63 TIL 78
R2 1MSI
{F)bC,83
Rl 1MSI
R3
1MB
fr L
Figure 15.29
More sensitive Photo-Schmitt-Trigger with the operational amplifier SN 72741
The inverting input of the amplifier is set to half the supply voltage by a symmetrical
voltage divider R 2, R 3. The non-inverting
input of the amplifier is driven by the
R^ voltage produced on the load resistance
of the phototransistor. For Schmitt trigger operation, the operational amplifier has a
feed-back, through the resistance R 4, from
the output to the non-inverting input.
When the phototransistor TIL 78 is irradi-.
ated, the potential at the non-inverting
input rises. When it reaches half the supply
voltage, the Schmitt trigger switches its out-
put from the LOW state (approx. V) to the HIGH state (approx. Vb).
The hysteresis of the Schmitt trigger is
determined by the ratio R4/RA, while R^
is given a value according to the required
sensitivity. The sensitivity of this circuit increases with decreasing supply -voltage, since at the same time the threshold voltages of the Schmitt trigger decrease and thus a
S(f)
D iMr
c, 100 |iF'
Mill
IMS!
Figure 15.30
Parking-light switch with the phototransistor TIL 67
287
smaller current is necessary through R^-
This circuit is used as an optoelectronic detector or twilight switch. Figure 15.30 shows an extended circuit for use as a parking light switch for motor vehicles. The
capacitor C slows down the response of the
circuit, so that it does not operate with short-term variations in the light.
15.3 Detector circuits with three-pole phototransistors
Three-pole phototransistors are understood
to mean phototransistors in which the base connection is brought out and used.
15.3.1
Operating modes of three-pole photo-
transistors
Phototransistors with base connections can
work in various operating modes.
15.3.1.1
Photocell and photodiode operation of the photo transistor TIL 81
Both the collector-base junction and the base-emitter junction can be operated either as a photocell or as a photodiode. For the base-emitter junction, these two modes are of no practical importance, since firstly their photosensitive area is very small and secondly the high emitter doping does not permit a high reverse voltage in photodiode operation (max. 5 V). Also, the area of the collector-base junction is about ten times as large as that of the base-emitter junction. Therefore, the former is generally used as the photodetector for photocell and photodiode operation.
When an NPN phototransistor is operated
in the photocell mode, a negative potential is obtained at the collector and a positive
potential at the base.
In the diode mode, the collector-base diode
is biassed in the reverse direction. Negative potential is applied to the base and positive potential to the collector.
The maximum reverse voltage V^bO oi" tne
collector-base diode of the phototransistor
TIL 81 is 50 V. Figure 15.31 shows the
typical current-voltage characteristics for a
silicon P-N junction. The 3rd quadrant applies for photodiode operation with the function Ip = f(VR). The incident irradiance Ee is the variable. For a constant incident
E E irradiance e i, e> 2 . . . , the working
point A of the photodiode can be deter-
mined by drawing in the working resistance R^. The working resistance Ra,3> f°r example, has a higher value than Ra,4The 4th quadrant shows photocell operation. The points of intersection of the individual characteristic curves with the negative ordinate show the short-circuit
k current Ip for the relevant incident irradiance Ee . The values of the shortcircuit currents Ip k almost correspond to
the currents Ip in the diode mode. Here, the dark current in diode operation has been neglected. The points of intersection of the individual curves with the positive
abscissa characterise the no-load voltage
Vp l for the incident irradiance Ee in each
case. For a constant incident irradiance
A Ee i , Ee 2 · · - , the working point of the
photocell is determined by drawing in the working resistance R^. Here, the working resistance R^\ \ has a higher value than
^A 2- The 1st quadrant shows the typical
forward characteristic of a Si photodiode. During exposure to radiation, the forward
voltage Vp increases slightly. This forward voltage Vp corresponds to the no-load voltage Vp l in photocell operation.
15.3.1.2
Advantages of the phototransistor TIL 81 in photodiode and photocell operation
The universally applicable junction photodetector TIL 81 is built into a modified
TO 18 metal case. It can therefore be fixed
288
.
Figure 15.31
Current-voltage characteristics of a silicon photosensitive P-N junction
mechanically very simply by a press fit or clip, in guide tubes or transistor plug bases, or by mounting on a printed circuit. In comparison with unencapsulated or largearea photocells and photodiodes, firstly mechanical fixing and adjustment is simpler, and furthermore, an additional lens or objective can be fitted more conveniently, in accordance with the laws of optics, to the small lens area of the TIL 81
The absolute sensitivity of this phototransistor in photodiode or photocell operation
170 m A
S2856K : 25mW/cm 2
This value can only be used for comparison of the sensitivity of exactly equivalent photodiodes and photocells.
A comparison of the quality of different
photodiodes and photocells can be made
with the values NEP (Noise Equivalent
Power) and specific detectivity described in Chapter 9. In most cases, these values are not stated for simple junction photodetectors. Here, approximate calculations or measurements of the signal-noise ratio can help. The noise level of a junction photodetector is very dependent on its dark current, so that as a simplification, the dark current serves as a further comparative criterion for different Si photodiodes and
photocells.
Solar cells made by the epitaxial technique
cannot be used as sensitive detectors, since they have very high dark currents and also, in the diode mode, have very low breakdown voltages. Also, large-area photocells or
photodiodes made by the epitaxial or planar
289
:
technique are less suitable for the detection of modulated radiation, because of their high junction capacitance. Furthermore, the high sensitivity of a large-area cell, as
compared with the TIL 81 is the photocell mode, is no advantage, since in short-
circuited operation, the load resistance of a large-area photocell can only have 1/1 00th to 1/1 0th of the value for the small-area
photo transistor TIL 81.
In photodiode operation, the data-sheet of
the TIL 81 guarantees a maximum dark current of iDmax = 10 nA at tij = 25° C.
Typical dark currents lie two to three orders of magnitude lower, i.e., in the'picoampere
range. Therefore, this phototransistor shows a low intrinsic noise level in photocell and diode operation, so that it can be
used as a detector for very weak incident radiations. As an example, let us mention an IR telephony unit with the TIL 81 in the photocell mode as detector and with the low-power GaAs diode TIL 31 as the transmitter. With these two, a good
signal/noise ratio is still attained at dis-
tances greater than 10 m. With the use of additional lenses, distances of several hundred metres can be spanned. Without external optical components, the irradiance falling on the TIL 81, under the conditions
for the TIL 31 : <P e = 5 mW, Ig = 20 mW/sr;
source-detector distance r = 10 m, has the
following value
^e
~U mW ^o :
m ·
20
. sr
sr . 100 2
20 . 106 nW
nW
= 20
cm 1 . 10 6
2
cm 2
The lower limit of detection of radiation for the TIL 81 is still considerably lower. For the reception of HF-modulated radiation, the TIL 81 is operated more favourably as a photodiode. The applied
reverse voltage reduces the junction capaci-
tance. The collector-base capacitance of the TIL 81 in photocell operation is approx. C-CB = 40 pF, and in photodiode operation
V with a reverse voltage of 15 it is about
CCB = 12pF.
Both in photocell and photodiode operation, the phototransistor TIL 81 shows little scatter between individual specimens, while in transistor operation the scatter is determined by the differing D.C. gain B.
Photocell and photodiode operation of the phototransistor TIL 81 is clearly identified by circuit symbols (see Figure 15.32).
a) Base
b) Base
c) Collector
Collector
Collector Photocell
Collector Photodiode
<b
Emitter Phototransistor with unused base connection
%>
6
Base
Phototransistor with base connection in use
Figure 15.32 Circuit symbols for the Si junction photodetector TIL 81 in various operating modes
290
15.3.1.3
Phototransistor operation of the TIL 81
Figure 15.32 shows, under c), the circuit
symbol for the TIL 81 with the base connection unused and, under d), that with the
base connection in use. Photqtransistors with base connection, in emitterconnection, and phototransistors without base connection are used as detectors for unmodulated radiation.
Phototransistors with an external base connection can be driven independently of any
incident radiation. Among other uses, they
are used as reference transistors in differential amplifiers. The external base
connection also permits the elimination of side-effects, for example, such as reduction of the collector-base dark current, improve-
ment of the limiting frequency and control of the working point. The phototransistor TIL 81 with a base connection can be used
very economically in multivibrator circuits, since only one additional transistor is needed.
Reception of modulated radiation with phototransistors requires stable working-
point conditions. In common emitter-
connection, the working point is mainly determined by the incident irradiance. For this, Figure 15.33 shows the modulation and working-point conditions of the photo-
ic = f («B)
IC = f(VC E)
U 7*&\j£~*-J?f-
ic
Ee^ lSmW/cm^-JB^y.
Vh = 20 V RA
-o +Vb
VCE
Figure 15.33 Modulation and characteristics of a TIL 81 phototransistor
291
transistor TIL 81. If a very weak modulated radiation signal and a very intense interference radiation fall simultaneously on the
detector, the interference radiation drives the phototransistor into voltage saturation.
The modulated radiation signal will not be amplified. If a very intense modulated radiation falls on the phototransistor, the
useful radiation drives the phototransistor into voltage saturation; the output signal
will be distorted. If, on the other hand, only a very weak modulated radiation falls on the phototransistor, then the photocurrent produced by the collector-base
diode is insufficient to drive the transistor.
A remedy is provided by a defined external
irradiation or an additional base control current, to fix the working point. In practice, a strong interference radiation is usually to be expected, so that the working point has to be shifted according to the
interference intensity.
Phototransistors in common emitter con-
nection or without base connection are therefore usually unsuitable for reception of modulated radiation.
The emitter follower offers an alternative,
if a fixed potential from a high-impedance voltage source is applied, as shown in Figure 15.34, to the base of the
phototransistor.
The working point of the transistor in the dark condition is determined by the base
voltage divider R 1, R 2, which forms a V voltage source of 3.7 with an internal resistance Rq = 50 kn. The collector
current is then
Vb-Vre 3-7 V- 0-7
Ic ~ IE
Re
io kn
= 0-1 mA
The input impedance rj of the emitter
follower is obtained, with hp£ TIL 81 > 100, as
Mn r, » RE-b-FE = 10 kn - 100 = *
Thus the load resistance of the photodiode is primarily determined by the resistance Rq, through which the useful and the interference photocurrents ip and Ip flow. Since this resistance is relatively low and the voltage gain of the subsequent emitter
V follower is u « 1, even high interference
t
° vb
«0 sok a
©»*
'1 dr-®
bquivalent circuit
Figure 15.34 Phototransistor used with an emitter follower for reception of modulated radiation
292
currents Ip are unable to overload the
whole circuit. The output impedance of the circuit is calculated from the formula
r
= V-T±+-R^- = 26 mV + 50 kn IC hFE 0.1 mA 100
= 760fi
Since the voltage gain of the circuit is
V < u 1 , the effect of the Miller capacitance
is also slight. The upper frequency limit is
determined by the collector-base capaci-
tance Cqq and the resistance Ro-
15.3.2 Driving of amplifiers with three-pole phototiansistors
Figure 15.35 shows a cascade circuit to improve the dynamic characteristics of the phototransistor TIL 81. In addition to the circuit in Figure 15.23, the base of the
phototransistor T 1 is driven through the
R R feed-back resistors 2, 3 from the collector of the NPN transistor. The D.C. feedback through R 2, R 3 has the effect that
the dark current of the phototransistor is
now hardly affected by its current gain.
The quiescent current of the circuit is set
R with the resistor 3.
lkf!
Output
Figure 15.36 Circuit to improve the dynamic characteristics of a photodarlington arrangement with connected base
The photodarlington cascade in Figure 15.36
is of similar construction to the circuits in Figure 15.24 and Figure 15.25. This circuit is very sensitive. With high irradiances, the photodarlington transistor, consisting of
T 1 and T 2, and also the base-connected NPN transistor T 3, go into saturation. For R current limitation, a series resistance is connected in the base lead of T 3 and the
base is capacitatively earthed. With rising temperature, the collector-base leakage
current IcBO increases. It acts as the base
RL IkS
®
R3 T>3'3M a
m- Tl TIL 81
-o Output
Figure 15.35 Circuit for improvement of the dynamic characteristics of a phototransistor with base connection
RC 10k n --o Output
Rb 100k -1M
Figure 15.37 Temperature compensation of the leakage
current IcBO for a phototransistor by the
Rg base resistance
293
h®
^T
Q>
j
22k n
S-6k n
~o Output
Figure 15.38 Phototransistor D.C. amplifier with temperature compensation
control current and is further amplified, according to the D.C. gain B of the phototransistor. This effect is particularly undesirable with low irradiances.
The dark current is reduced, by connecting a base resistor Rg, which diverts the collector-base leakage current to ground, between base and ground {Figure 15.37), At the same time, the sensitivity of the
phototransistor is reduced, since part of the photocurrent Ip of the collector-base photodiode is also diverted.
15.3.3
Phototrigger and photomultivibrator
circuits
Phototrigger circuits can be constructed very economically with a phototransistor and a bipolar transistor. Figure 15.39 shows
a photothyristor circuit with the NPN phototransistor TIL 81 and the PNP transistor BC 212. The collectors of the
two transistors each drive the base of the
Figure 15.38 shows the temperature compensation of phototransistors in a difference
stage. It contains the phototransistor T 1 as
photodetector and an obscured phototran-
sistor T 2 as reference transistor. The tem-
perature response of the two phototransistors has no effect, since the subsequent operational amplifier is only driven by a difference signal. Scatter between specimens of the phototransistors and of trie operational amplifier are compensated by
the potentiometer R 1 . The desired gain is set with the resistance R 2. This circuit is
used as a scanning amplifier and as the
detector in optoelectronic couplers.
Figure 15.39
NPN Photothyristor, simulated with
phototransistor and PNP transistor
294
vA
^
Forward blocking range
Hold-on current
"V
Reverse blocking range
Reverse breakdown voltage
Firing threshold with
an irradiance £ of
^ Emin-
vA
Figure 15.40 Theoretical characteristic of a photothyristor
other transistor. With sufficiently high irradiance, the photo transistor turns on
the transistor T 2. Through the feed-back
to the base of the phototransistor, the latter remains conductive, even if the irradiance
decreases again. The response sensitivity of the thyristor is primarily determined by the
base resistance R 1 , which diverts part of the photocurrent, while the capacitor C and the resistance R 2 prevent undesired
firing of the thyristor in case of rapid
changes in the anode voltage. The thyristor can be extinguished again by switching off
the supply voltage.
The characteristic in Figure 15.40 shows, in the 1st quadrant, the characteristic of the thyristor in the forward range and, in the
3rd quadrant, the reverse characteristic.
-o Output
Figure 15.41 Phototrigger circuit with reset button
295
I.,
i.
csp
47m
i
2 )
Figure 15.42
Phototrigger with reset button and NPN-Darlington Power stage
This circuit is used as a simple detector for A.C. operation, in firing circuits for power thyristors and for the storage of nonrecurring optical events.
As a further example, Figure 15.41 shows a phototrigger circuit with a reset button. In
the, dark condition, the transistor T 2 is driven through R 1 and is conductive. If radiation falls on the phototransistor T 1 the base control current of T 2 is reduced
by the amount of the collector current of
T 1 . The collector voltage of T 2 rises, so
that the base of the phototransistor TIL 81
is driven additionally through R 3 and R 2.
The circuit trips to the new state, which is
maintained, because of the feedback, even if the phototransistor is no longer irradiated.
The diode B A 1 80 prevents a part of the
photocurrent from flowing away through
the feed-back resistance R 3. The original
operating condition can be re-established
by the reset button.
In Figure 15.42, the transistor T 2 has
been replaced by the NPN-Darlington transistor TIP 110.
The circuit in Figure 15.43 is constructed as a phototrigger in a complementary circuit. If the phototransistor TIL 81 is
irradiated, both transistors T 1 and T 2
-o Output
Figure 15.43 Phototrigger in complementary circuit; in the quiescent state (without radiation), both transistors are non conducting
change to the conductive state, which is maintained until the reset button is operated. In the circuit in Figure 15.44,
T 2 has been replaced by the PNP-
Darlington transistor TIP 115. in order to be able to drive high-power loads. The applications of such photo-trigger circuits
include optical threshold switches, twilight switches, sensors in alarm installations etc.
A further two-state circuit is the mono-
stable multivibrator shown in Figure 15.45.
The output transistor T 2 drives a counter
relay. In the steady-state condition, radi-
296
I=2A
csp
± -0
c 1.
n A
>
Figure 15.44 Phototrigger with PNP-Darlington power stage
r
X
T
® )
ft
39k a
-o Output
Figure 15.45 Phototrigger circuit, as monostable multivibrator with a counter relay
ation falls on the photo transistor T 1. The
output transistor is blocked and the counter
relay is de-energised. If the radiation falling
on the phototransistor TIL 81 is briefly
interrupted, T 2 becomes conductive. The collector voltage on T 2 decreases.
The base of the phototransistor receives a negative potential through the feed-back
capacitor C 1, so that T 1 changes to the non conducting condition and T 2 further
into the conductive state. The capacitor is
now discharged through the resistors R
297
R and R 2 || 3 and, if exposed to light,
through the photocurrent of the transistor
T 1 . The drop-out time of the relay there-
fore depends on the irradiance falling on
the phototransistor, while high irradiances give short drop-out times.
Figure 15.46 shows a timing switch with
the N-channel junction FET Type BC 264 A.
The circuit also works as a monostable multivibrator. If radiation falls with sufficient intensity on the phototransistor TIL 81, the phototransistor switches to the
conductive state and the FET is turned off. After the discharge of the capacitor C 1 through R 1 and R 2, the FET becomes
Jrokn
M47kn
<XyJO/
0-1 jiF
R2| |33MD
AtVGS(ofO"
2v .
r<*R.C
Figure 15.46
Photo-time-switch with FET
conductive again, irrespective of the radiation falling on the phototransistor. The high input impedance of the fieldeffect transistor does not load the timing circuit, so that the time constants can be constructed with very high resistances and small capacitor values.
Finally, in Figure 15.47, the circuit of a sensitive photo-Schmitt-trigger with a low dark current is shown. With very low irradiances and high operating temperatures, the not inconsiderable residual current of the collector-base photodiode is to be
diverted through a base resistor. The resultant loss of sensitivity is compensated
A R by the variable resistor 1 . Schmitt trigger (T*3, T 4) is driven through the decoupling amplifier T 2. The circuit is
used in sensitive optoelectronic couplers.
15.4
Simple couplers with filament lamps and two-pole phototransistors
1
Construction with few, inexpensive components.
Little mechanical adjustment work.
o Output
Figure 15.47 Photo-Schmitt-trigger with low dark current
298
The couplers are to be designed optically and electrically with very high safety
factors.
A command signal is initiated by a power
switch directly at the detector output
Simple couplers only make a Yes-No decision. They are used to control various
functions, such as vane-wheel sensing in anemometers, sensing and monitoring the
rotational speed of rotating parts, cam
sensing, film frame alignment in film projectors and end-of-tape switching in audio and video tape recorders.
Such simple opto couplers need a powerful radiation source. Thus the amplifier cost at the receiving end is kept low and a good signal/interference ratio is achieved against
A stray radiation. filament lamp satisfies
these conditions. The type of filament is mainly determined by the distance from source to detector. The filament lamps used can include types with and without lenses, selected and non-selected miniature lamps, dial lamps, torch bulbs, telephone indicator lamps and small and large special filament lamps for optical applications. Allowance must be made for their sensitivity to vibration, temperature-dependence and wide tolerances on radiant intensity. Colour temperature data is only obtainable for the
optical filament lamps.
The distribution temperature which is actually needed is seldom stated. The filament of the radiation source used should be as straight and evenly wound as possible. Where possible, the lamp will be operated with D.C., since A.C. would cause undesired modulation of the radiation. To improve reliability, the lamps will be operated at about 10 to 20% below their rated voltage. Depending on the type and whether the lamp is a gas-filled or vacuum type, the colour temperature then falls to about 2000° K. The spectral radiant power distribution (emission function) of the lamp is
then still just adequately matched to the
spectral sensitivity distribution of silicon
junction detectors. However, the very large
proportion of thermal radiation can cause
an unwanted heating of the photodetector.
RGN Therefore, an IR filter (e.g.,
9 or
RG 38) may sometimes have to be placed
in front of the photodetector.
As a sensor for simple couplers, for
example, the two-pole, plastic-encapsulated
phototransistor TIL 78, which has a typical
aperture angle of ± 20 , is particularly
suitable. With an irradiance of Ee 2856K
= 20 mW/cm 2 , it delivers photocurrents of
mA at least 1
and typically 7 mA. This high
sensitivity for a plastic phototransistor also
makes it possible to keep the expenditure
on amplification low. Taking account of the
stated tolerances, the distances stated for
the couplers described below contain a high
safety factor. With exact adjustment, greater
distances can be spanned.
As a first example, Figure 15.48 shows a
mm simple coupler for r = 10
with a dial-
light lamp and a photo-Darlington transistor.
When the phototransistor TIL 78 is
irradiated, the relay pulls in.
12V/3W
dial lamp
t>
CSpwn
Figure 15.48
mm Opto coupler for r = 10
The coupler in Figure 15.49 is designed for r = 15 mm. The phototransistor is followed by a voltage amplifier, which drives the collector-connected power stage. The relay drops out, when the radiation beam is
interrupted.
299
W 2 1 V/3
miniatuie dial lamp
k
1
2©
56 n|l
)'
rvrxm
Figure 15.49
mm Opto coupler for r = 15
W 1 2 V/3
miniature dial lamp
10k S!
:© 1
rr 90nr~7n
A 1N914
)
Figure 15.50
mm Opto coupler for r = 15
with threshold switch (Schmitt trigger)
In the coupler in Figure 15.50, also for
r = 15 mm, the phototransistor TIL 78 drives a Schmitt trigger. On irradiation, the
relay pulls in.
In Figure 15.51, an opto coupler for r =
mm 30
is shown. The phototransistor drives
a sensitive two-transistor amplifier. On
irradiation, the relay pulls in.
Finally, Figure 15.52 shows a coupler for an alarm installation. If the radiation falling
on the phototransistor is interrupted or the contact mat is trodden on, the relay pulls
in. The radiation source is a pocket torch
RG with an IR filter
9 attached. The
detector is mounted at the focus of a flash-
RG light reflector. An IR filter
38 is also
fitted in front of the phototransistor TIL 65.
m With this arrangement, distances of r > 5
can be spanned. Despite the IR filter, the
sensitive detector of the coupler must be
shielded against stray radiation.
300
12V/3
mbiiilurc (
dial lamp
KD
90nCZ3
XlN914
>.q
>
)-
47k n
-CZ3-
F&we 75.57
mm <5pfo coupler for r = 30
® l(pl \/ 1R filteis
1N9MJJ
CZ_1
ft
IN914
I
\*y£>
1N9M
-w-
IR phototransistor detector
Figure 75.52 Simple opto coupler with contact mat for an alarm installation
301
V' -/
15.5 Logic circuits with phototransistors
Logic circuits can also be constructed with phototransistors. They are used for the logical evaluation of one or more radiation sources directly at the measurement point. High potential differences can exist between the radiation sources and the photodetector logic. The logic levels are characterised as follows
Optical logic level
L (low) = Phototransistor not irradiated H (high) = Phototransistor irradiated
Electrical logic level
L (low) = Voltage almost zero
H (high) = Almost supply voltage
Figures 15.53 to / 5.56 show the four basic
logic circuits.
Figure 15.53:
AND gate consisting of two TIL 78 photo-
transistors in series.
Figure 15.54:
NAND gate consisting of three LS 61
phototransistors in series.
Figure 15.55:
OR gate consisting of three LS 613 photo-
transistors in parallel.
Figure 15.56:
NOR gate consisting of three LS 612 photo-
transistors in parallel.
1 1 Ok 12
|
Figure 15.54
NAND gate with phototransistors (3 inputs)
3©=®
-o Output Y 10k n
Y»A+BtC
Figure 15.55
OR gate with phototransistors (3 inputs)
--o Output Y
& & &^
Figure 15.53
AND gate with phototransistors (2 inputs)
Figure 15.56
NOR gate with phototransistors (3 inputs)
302
1
A further logic circuit is illustrated in
Figure 15.57. Here, the output switches to
L-level, if both photo transistors are
NAND irradiated. Thus the circuit is a
gate.
The diode in the emitter circuit and the
T base diverter resistance on transistor 3
prevent the transistor from being turned on
by residual currents of the phototransistors.
Figure 15.59 shows a circuit to give L-level at the output, when the phototransistor
T 1 is irradiated and phototransistor T 2 is
not irradiated.
o*10 V
lOkn
<
o Output
T2
IS 615
TM
tk/ 80 83
=dp
-o Output
> T3 BC 184
n~
look n
47k n
Figure 15.57
Logic circuit with two phototransistors and
NAND NPN an
transistor; it represents a
gate
The circuit in Figure 15.58 delivers an
H-signal at its output, if the phototransistor
T 2 is irradiated and the phototransistor T 1
is not irradiated.
Figure 15.59
Logic circuit with two phototransistors and
NPN NOR an
transistor, it fulfils the
function
In the same way, Exclusive-OR gates can be constructed with phototransistors and additional active components {Figure 15.60). If neither of the two
phototransistors are irradiated, the transis-
tors T 3 and T 4 are non conducting
because their bases and emitters are almost at the potential of the supply voltage. If
only the phototransistor T 1 is now irradiated, the transistor T 3 becomes con-
ductive, because its emitter potential falls
=®
6«°
<B-
-o Output
MlOkn
Figure 15.58
Logic circuit with two phototransistors and
OR PNP a
transistor; this is an
gate
Figure 15.60
NOR Exclusive
gate with phototransistors
303
OV almost to
and it receives a base current
supply through the collector resistance of
the phototransistor T 2. If the photo-
transistor T 2 now also turns on, then both
transistors T 3 and T 4 are again non con-
ducting, because both their bases and their
V emitters are connected to O through the
phototransistors.
15.6
Photodetector circuits to drive TTL
integrating circuits
In industrial control installations and in punched tape and card readers, Si junction
photodetectors are used to drive TTL
integrated circuits. TTL-compatible photo-
detectors can drive a TTL circuit directly.
Si phototransistors are TTL-compatible, if the collector current produced by the incident radiation is at least as great as the
necessary input current of the TTL
circuits. For this, they must have a
sufficiently small saturation voltage
VcEsat < 0.4 V.
Input currents and voltages of the SN54/74 Series Schmitt triggers
Non-TTL-compatible photodetectors need a matching or interface circuit for the cor-
responding TTL circuit. Also, TTL circuits
only work reliably, if the control signals have a rise and fall time of tr , tf <200 n sees. The output signals from phototransistors do
not satisfy this requirement. Indeed, the optical turn-on and turn-off of a radiation falling on the phototransistor usually takes several milliseconds or longer. Therefore it is necessary, to speed up the edges of the signals from the photodetector with a
Schmitt trigger, in order to prevent oscil-
lation of the subsequent TTL circuits.
The input levels and currents of the most
commonly used TTL Schmitt triggers are
summarised in the following table.
In this:
Vx+ = Positive response threshold, Vt_ = Negative response threshold, V[h = H-level input voltage,
VlL = L-level input voltage,
IjH = H-level input current, IlL = L-level input current.
D- n
H® LS 600 Series
LS6I0Series TIL 600 Series
SN 7413 SN 7414 SN 74131 SN74S13I
Figure 15.61
Direct control of TTL Schmitt triggers
with a phototransistor
Figure 15.61 shows a phototransistor inter-
face circuit to drive TTL Schmitt trigger
circuits. With a collector voltage of
VcE = Vj-min- tne irradiance falling on
the phototransistor should produce a mini-
mum collector current of I(]£ = Ijl. Under
these conditions, the TTL output will be H safely switched to the logical level. The maximum value of the resistance R 1 is cal-
culated as follows:
Type
'13 '14 '132
'S132
vT+max
(V)
2-0 2-0 2-0 1-9
vT-min
(V)
0-6 0-6 0-6 1-1
V IH a t
(V)
IlH
(mA)
2-4
0-04
2-4
0-04
2-4
0-04
2-4
0-05
VlL a t
(V)
IlL
(mA)
0-4
-1-6
0-4
-1-2
0-4
-1-2
0-5
-2-0
304
vb - vT+max
Ri max JCE " ! IL
Unused TTL inputs are connected together
and through a pull-up resistance (1 kfi) to the supply voltage. This an H-level is ensured at these inputs. Because of their small mechanical dimensions, the LS 600, LS 610 and TIL 600 phototransistors are particularly suitable for punched tape and card readers.
Figure 15.62 shows a drive circuit with
inverted output function. When the photo-
transistor is irradiated, an L-level is obtained at the output, if a voltage of at least
VT+max is present at the Schmitt trigger
input.
the Schmitt trigger. For this, Figure 15.63
shows a phototransistor interface circuit
and Figure 15.64 shows a photodiode
TTL interface circuit to drive
circuits.
Figure 15.63 Phototransistor interface circuit for driving
TTL Schmitt triggers
LS 600 Serin L56tOSerfct T!L6O0Serie»
Figure 15.62
Direct control of TTL Schmitt triggers with
a phototransistor
"
The necessary collector current Icmin OI
the phototransistor is calculated as:
Vfe-ViL
k^min
~
+ 4H
R
l
The resistance R 1 is obtained as
vT-min
Rl
IlL
If the radiation falling on the phototransistor is weak, a matching amplifier is necessary between the phototransistor and
Punched-tape and punched-card reading heads for a large number of reading channels can be constructed more economically, if the signals from the phototransistors or diodes are led directly, through
any necessary interface circuits, to the TTL
gates. In this form, however, the output signals would not be usable, since the gate oscillates during the turn-on and turn-off of the phototransistor, because the signal edges are too slow (Figure 15.65). In order still to be able to evaluate the read signals, an additional marker channel, for which the drive perforation track can be used in the case of punched tape, is needed. This signal must then be fed through a Schmitt trigger,
which delivers a satisfactory TTL signal at
its output. With this strobe or interrogation signal, the read amplifiers are then
only conducting at a time when definite logic levels can be relied on at the input of
the amplifier.
Figure 15.66 shows the corresponding
circuit. Since the SN 75450 circuit already
contains the necessary transistors, this circuit can be constructed very economically.
305
TIL 81.
H38
or similtf
lkft
3
Figure 15.64 Photodiode interface circuit with operational amplifier
m 1/2 SN 7413N
Output of phototnuisbAor
{y-
Output of
g_
Figure 15.65
Oscillation of a TTL gate when driven with signals with insufficiently fast rise and fall times
306
+SV O
--
Rnd unpUfkr 1
+SV o
1.
PhotolmMfefort LS600 LS610
lSknl [
fate InF
Figure 15.66 Sensitive phototransistor interface circuit with strobe signal through a marker channel
307
16
Modulated
transmitters with luminescence diodes
16.1 16.2 16.2.1 16.2.2 16.3
The simplest modulator circuits for
luminescence diodes Sine-wave-modulated transmitter with luminescence diodes Modulation operation with bias voltage sources Modulation operation with constantcurrent sources Pulse-modulated pulse transmitter with luminescence diodes
309
16
Modulated transmitters with luminescence diodes
16 Modulated transmitters with luminescence diodes
The radiation from a luminescence diode can be modulated. Basically, both sine-wave and pulse modulation are possible. If a luminescence diode is to be controlled with a
sinusoidal alternating current, then its working point has to be set with a bias current Ip.
The necessary thermal calculation for luminescence diodes was described in
Section 11.2. In pulse operation, lumines-
cence diodes need no additional bias
current (see also Section 11.7).
Figures 16.2 and 16.3 show two further
theoretical circuits for modulcators. Through a switch S, the current through the luminescence diode is increased in Figure 16.2, or short-circuited in Figure 16.3, in time with the clock frequency.
16.1
The simplest modulator circuits for lumi-
nescence diodes
Figure 16.2 Simple modulator circuit with periodical
keying of the GaAs diode by the switch S
Figure 16.1 shows the simplest modulator
circuit for luminescence diodes. The GaAs diode TIL 32 is controlled with a sinusoidal alternating current. The -Si diode connected
in parallel protects the luminescence diode from high reverse voltages. Because of the non-linear forward characteristic of the
GaAs diode, the signal emitted will be distorted.
If the contact S in Figure 16.2 is replaced
by a phototransistor (e ;g., TIL 81) opposite the luminescence diode, and a moving punched tape is inserted between the luminescence diode TIL 31 and phototransistor TIL 81, to produce a timing frequency, then the coupler shown in
Figure 16.4 is obtained. Here, an optical
'Fmax = 20 mA
RV
D= I
" TIL 32
*<
t
Figure 16.1
Control of a luminescence diode with alternating current
311
Control current Radiated power
10 V-1.6 V-0-3 V 24-5 mA
330 a
Figure 16.3 Simple modulator circuit with periodical
short-circuting of the GaAs diode with the switch S
I
ll
' lp2 = 25mA
h®
Rv2
Mfi 330 n
mA 1 Ifl = 2.5
--o Output
Rvi
hU 3 3kS!
Figure 16.4
Opto coupler with optically -closed feedback loop
feed-back loop is used to key the luminescence diode. If the phototransistor is obscured, the forward current through the luminescence diode TIL 31 is determined by the series resistance Ryi-
_ Vb-VF,TIL31 _ 10 V- 1-6V
IF1
R
3-3 ktt
= 2-45 mA
If the feed-back loop is closed, so that the radiation from the luminescence diode falls on the phototransistor, the current increases, in relation to the coupling characteristics, to the value Ip2. This is calculated as
= vb - VF/TIL31 - VcEsat,TIL81
!f2
RV2
Then the total forward current iFtot through the luminescence diode TIL 31 is:
mA mA JFtot = !f1 + JF2 = 2-45
+ 24-5
= 26-95 mA
Figure 16.5 shows an oscillator circuit with optical feedback. In this, the contact S in Figure 16.3 has been replaced by a phototransistor TIL 81 opposite the luminescence
diode TIL 31. When the phototransistor is irradiated, the GaAs diode is short-circuited. The radiation from the luminescence diode is thus switched off. The phototransistor, which is now not irradiated, increases in
resistance, so that a forward current again
flows through the GaAs diode. The process is repeated periodically. The frequency of
this oscillator lies in the low-frequency range. It is determined by the storage times of the phototransistor and thus depends on the degree of overload of the transistor, which is affected by the coupling charac-
teristic of the coupler (distance). The output signal can be amplified further with an LF
amplifier. This circuit is suitable for simple couplers and for alarm installations.
Figure 16.5 Modulator circuit for luminescence diodes using a coupler with optical feed-back
16.2
Sine-wave modulated transmitters with luminescence diodes
The steady-state current of a luminescence
312
diode in sine-wave modulated operation
can be produced either from a voltage
Ry source through the series resistance
or
through a constant current source.
16.2.1
Modulated operation with bias voltage
sources
If the steady-state current of the lumines-
cence diode is taken from a voltage source,
Ry the series resistance
is calculated from
the formula:
\V-Vp
Ry
if
In order to avoid modulation distortion, the
maximum modulation voltage during the
negative half-wave must not reach the lower non-linear part of the characteristic curve Ip = f(Vp). In Figure 16.6, the working points Aj and A<j> for low-distortion modulation of luminescence diodes are
plotted.
The luminescence diodes can be modulated by coupling the modulation frequency in
parallel with the bias voltage supply. In the circuit b in Figure 16. 7, additional measures have been taken (choke and filter capacitor) in order to prevent interaction of the modulation voltage of the voltage source which produces the steady-state current.
It is, however, more usual to connect the modulation voltage generator and the
Figure 16.6
Selection of the working point for low-distortion modulation of luminescence diodes
313
Figure 16.7 Modulator circuits for luminescence diodes (parallel feed)
steady-state supply in series (Figure 16.8). This method, with which both functions are combined in one circuit, will be used in almost all the applications to be shown
later.
An amplitude-modulated transmitter cor-
responding to the theoretical circuit in Figure 16.7 can be constructed with few components. As an example, Figure 16.9 shows a low-frequency-modulated transmitter for the transmission of music pro-
grammes. The transmitter receives the modulation signal from the loudspeaker
Figure 16.8 Modulator circuit for luminescence diodes (series connection)
output of»a radio receiver. The photodetector can be connected, for example, to the preamplifier of a tape recorder. If additional lenses are fitted in front of the
luminescence diode TIXL 27 and the
m photodiode TIL 81, distances of r > 10
can be spanned.
In Figure 16.10, a series connection of the modulation amplifier and steady-state current source is used. The steady-state current is supplied by the transistor, which
works as a current source. The steady-state current is calculated from the formula:
lF,R :
Rl VhD.
Rl + R2
Re
Vrtift-
10 V.
2-2 kft
-0-7 V
2-2kft+2-2kft
220 ft
20 mA
In Figure 16.10, the modulation voltage is fed in to the base of the transistor. Because
314
r
From loudspeaker -i
--500 ^F
°
output
|5W
lOOkU
39 n
1
XI
> N
i2v ; 1
-- if
TIXL27
TIL 81
*
JL *"
BF 247A
1
Amplitude-modulated IR trinsmitter
47k O
22k O
y
IR detector
IMF
figure 76.9
Simple IR LF transmission link
of the current feed-back, the alternating current in the transistor is proportional to the modulation voltage.
Figure 16.10 In this modulator, the amplifier and steadystate current source are connected in series
The circuit in Figure 16. 11, in which the
luminescence diode TIXL 27 is operated with a current Ip = 300 mA, is of similar
construction. Because of the loss power transistor TIP 33 is used as a driver, and is preceded by an additional emitter follower, in order not to load the modu-
lation source.
Figure 16.12 shows an amplitudemodulated transmitter with the operational
lOOjiF!
1=
Input O--
IdF
c
)
ion 10W
Figure 16.11
Modulator circuit for the GaAs power diode TIXL 27
315
amplifier SN 72741P. The working point is R set by the voltage divider 1, R 2, which at the same time determines the gain. The resistance R 4 determines the steady-state
current Ip:
Vb -
if
R4
15 V 22 mA
680 ft
In order to keep the loss power in the
operational amplifier low, the greatest part
of this current is supplied through the resis-
tance R 3. The operational amplifier permits a modulation variation of approx. ±10 mA.
Thus the necessary modulation voltage
Vmo(i at the input of the amplifier is cal-
culated as:
10 mA.680 R2
Vmodrms :
Rl +R2
22 mV
Figure 16.13 shows an HF-modulated transmitter to carry a television picture. The modulator input is driven directly by the
HF output signal of a television camera.
The HF power transistor XB 436 works on
an oscillator circuit tuned to f = 50 MHz. The luminescence diode is connected in
A series with the oscillator coil. silicon
diode, connected in opposing parallel,
protects the GaAs diode from high reverse voltages. It must be noted, that silicondoped GaAs diodes, such as the TIXL 12 used here, only have a low radiant efficiency at a frequency f = 50 MHz. Of course, the useful radiant power produced
is sufficient here, to drive the photodetector.
As the photodetector, the very sensitive Si
avalanche photodiode TIXL 56 is used, it
has a very high limiting frequency. Its working point must be adjusted so that the high internal "avalanche gain" is utilised
with a good signal/noise ratio (see
Chapter 9). The avalanche diode TIXL 56
drives a broad-band circuit tuned to f = 50 MHz. The parallel-tuned circuit consists of the junction capacitance of the avalanche diode, the circuit capacitance
and the coil L 1 . The output signal, coupled out at low impedance through L 2, is fed to the 60-Ohm aerial socket of a
television receiver switched to Channel 2.
c
"mod
c)
1
IkiJ
-'
SN 72741P
\sy>
R2
I
I
Rl
220k n
+vb °I5V
r
~ TIL 23/24
f
- TIXL 26
-
TIL 31/32
i
n R4
M680S2
*
° 15 V
Figure 16.12 Amplitude modulated transmitter with an operational amplifier
316
$
pX
\
For connection
to a TV camera
40 nH
D ik "
240 p F BA 18177 J-
IR HF modulator fo=50H/
"D lMn|
HFdemodulalo
]
TIXL 56
120 pF
KLI
1 1 turns
--60 n
Vogt construction
set D41 -2165
For connection
to a TV set
J. TIXL 12
IR detector
frj = 50 MHz
Figure 16.13
Simple IR HF-link for transmission of a TV picture
Figure 16.14 shows an amplitudemodulated IR transmitter, which is used, together with a suitable receiver (see Figure 17.10) as an optical-link telephone. For this reason, the frequency range of the amplifier is designed for a range of 300 .. . 3000 Hz, but can be extended without difficulty, by changing the appropriate components, to a range of 50 ... 60 000 Hz. The signal coming from
the microphone is first amplified with a low-noise transistor amplifier. The voltage gain is calculated from the formula:
Vu Rl . R2 C * I Rl + R2 26 mV
18 kft. 100 kft 0-1 mA
59
18kft+100kft 26 mV
10 M F=±=
Cl 3-3n F
220 »F
-o-Vb
m
R2j/h00kn
b
150k n
10kS2 S2 '"
,C4
01 hF
GaAs diode:
Vb = -6V: T1L31 "I
with heat sink in each case
V b = -9V; TIL 27 /
Figure 16.14
Amplitude-modulated IR transmitter with integrated LF power amplifier
317
The upper frequency limit is determined by
the time-constant R l.C 1. The gain of the
whole circuit, and thus the modulation level,
can be adjusted with the potentiometer R 2. There follows an SN 76001 AN integrated audio power amplifier. The resistance R 3
V fixes the gain u = 200 = 46 dB, while the
capacitor C 2 in series with it determines
the lower frequency limit fu = 300 Hz. The
capacitors C 3 and C 4 compensate the
phase and frequency response of the integrated amplifier and thus prevent oscillation of the circuit.
Since the positive input is grounded, the
GaAs diodes can be mounted on large-area
heat-sinks, without the occurrence of insulation problems. The steady-state current Ip of the diodes is set by the series
Figure 16.15 Amplitude modulation of the radiation from luminescence diodes with current
sources; a) series connection and b) parallel connection of current source and modu-
lation generator
resistance R 4. With an operating voltage
of 6 V, for the diode TIL 31, its value is:
-- -- -- -- -- 2
I Fo =
= 3V-1---5V = 0-1 A
R4
15fi
and with an operating voltage of 9 V, for
the diode TIXL 27:
--4-5 V- 1-5 V
lFo=
=0 - 2A
r7TT-
In conjunction with a sensitive photodetec-
tor (Figure 1 7.10), the range r is more than 10 m, if the TIL 31 is used as the trans-
mitting diode. If additional lenses are fitted
to the transmitter and the receiver to focus
the beam, several hundred metres can be
spanned. In this case, however, exact align-
ment of the transmitter and receiver is
necessary (tripod). As a protective measure
against irradiation by direct sunlight, IR
RG filters
38 are to be fitted in front of
the diodes, since otherwise the focussed
solar radiation would destroy the wafer of
the photodiode or the luminescence diode.
16.2.2 Modulation operation with constant-current sources
The operation of luminescence diodes with
330 a
Modulation input
Rs 50 SI
Figure 16.16 Current sources as modulation amplifiers for luminescence diodes
318
current sources has already been described in Chapter 14; therefore it does not need to be discussed further.
Figure 16.15 shows the principle of the modulator circuit in conjunction with the current sources. In Figure 16.15a, the current source itself is modulated, so that the diode current varies with the modulation voltage. Circuit b shows the modulation voltage fed in in parallel.
In Figure 16.16, constant-current sources
work at the same time as modulation amplifiers. The modulation signal is fed in in
V parallel with the input voltage e of the
constant-current source. The steady-state current is determined by the resistance Ry,
R together with the variable resistance s .
-0+Vb
UU
D,
CD
1
2N3244
01 jiF
l_Li> TIXL 13 T1XL 27
Figure 16.17
HF drive of luminescence diodes
Figure 16.17 shows how high-power luminescence diodes are driven with HF. The transistor T 1 works in Class B. Its load
impedance is a parallel tuned circuit located in the collector lead. The GaAs diode is connected in series with the coil of the tuned circuit. During a negative halfwave on the base, a high collector current flows in the tuned circuit. The coil current through the GaAs diode is higher than the current in the transistor, by the Q-factor of the tuned circuit. In linear modulation operation, the working point of the GaAs diode is determined by the constant current. If the GaAs diode is to work in pulsed operation, the constant-current source is omitted. The luminescence diode has to be protected from high reverse voltages.
Therefore, a Si diode BA 1 87 is connected
in opposing parallel with it.
As a further example, a sine-wave modulated IR transmitter is shown in Figure 16.18. The oscillator consists of a three-stage amplifier with a field-effect transistor jn the input stage. Through the feed-back branch
(Wien branch), consisting of the RC networks R 1, C 1 and R 2, C 2, the frequency
of the oscillator is determined and can be calculated from the formula:
1
2rr VR1R2.C1.C2
M2-2kn
82k nM
Tl BF264B
01 nF
-I--
:l00(iF
^V I--J C2
R2
' ISnF 100k n
W^l .Wr HS.
3-gTI []«*«,
Figure 16.18 Sine-wave-modulated IR transmitter with Wien-Robinson oscillator
319
A non-frequency-dependent feed-back
(Robinson circuit), through the resistances
R 3, R 4 and the filament lamp, stabilises
the amplitude, since with increasing oscillator amplitude the resistance of the filament lamp, and thus the degree of feedback, increases.
This circuit ensures an excellent frequency and amplitude stability, which is particularly required, if narrow-band-pass amplifiers are used in radiation links, in order to suppress interfering radiation.
A current source, the steady-state current lp
of which is determined by the base voltage
R 5, R 6 and the emitter resistances R 7, A R 8, serves as the modulator. steady -state
current of
R6 Vb - R5 + R6 -VfiE
lF : R7+R8
33 k ft
9 V-
V ·0-7
10kft + 3kft
15kft+120ft
6-2 V
If =
-- = 46 mA
135 ft
In order to achieve a large modulation
excursion with a small oscillator amplitude (Vosc rms = 400 mV), part of the emitter resistance is shunted capacitatively, so that
only the resistance R 7 determines the
degree of modulation.
iFp-
500 mV.2 .s/2 94 mA
15 ft
Because of the loss power which occurs, the luminescence diode TIL 31 must be mounted on a heat sink. This is possible without insulation problems, since the positive pole of the supply voltage is grounded.
m The range is about 5 and can be increased
further with additional lenses.
16.3 Pulse-modulated pulse transmitter with luminescence diodes
For the pulse modulation of luminescence diodes, the two basic circuits shown in Figure 16.19 are used. With these, the modulation of the diode current can be achieved in the following ways:
1
Periodic interruption of the forward current lp-
Figure 1 6.1
Theoretical circuits for the pulse modulation of luminescence diodes; a) pulse modulation by short-circuiting switch in parallel with the luminescence diode, b) pulse modulation by switch in series with the luminescence diode
320
Figure 16.20 Simple pulse modulator circuit to drive luminescence diodes
Periodic short-circuiting of a biassed luminescence diode
Pulse drive of the modulation amplifer.
Discharge circuits with four-layer semiconductors.
As an example, Figure 16.20 shows a simple modulator circuit for the luminescence diode TIL 32. The bias current Ip only
flows through the GaAs diode during the
pulse duration tr>
Ip = Vb -VF = 5V-1-5 = 23-3 mA
Rv
150
During the interval, the forward current Ip
flows through the transistor T 1 , which is connected in parallel with the GaAs diode. The GaAs diode TIL 32 is then short-
circuited. In Figures 16.21a and b, the base
of the amplifier transistor T 1 is driven by
the modulation signal. As in Figure 14.9,
r-TLTL
ovTLTL
Figure 16.21
NPN Pulse modulators for the control of luminescence diodes; a) with
transistors, bj with
PNP transistors
321
the transistor T 2 serves to limit the forward
current Ip. This is calculated at:
V BE,T2
iF
R B,T2
For the generation of high pulse frequencies, discharge circuits with four-layer semiconductors are suitable. Figure 16.22 shows an oscillator circuit with the unijunction
transistor TIS 43. The capacitor C 1 is charged through R 1. When its potential
reaches the response threshold of the unijunction transistor, the resistance of the emitter-base path suddenly falls and the capacitor discharges through this path and
the luminescence diode. The peak current
Ip can be adjusted with the resistance R 2.
Figure 16.22 Narrow-pulse generator with unijunction transistor for the control of luminescence diodes
100k (2
--0+40V
+10Vo-
IOjiFi
Figure 16.23 Narrow-pulse generator with unijunction equivalent circuit to drive luminescence diodes
Figure 16.23 shows an equivalent narrowpulse modulator with a unijunction equiva-
C lent circuit. In this, the capacitor 1 is
discharged through the emitter-base diode
T of the transistor 2, the collector-emitter path of T 1 and the luminescence diode TIXL 27. The maximum pulse current of
this circuit is determined by the permissible
base current of T 2. If very fast -acting power transistors are used for T 1 and T 2, then with a working voltage of Vg = + 40 V, a pulse current of Ip p = 8 A is achieved
with t p = 0.5 ms. Discharge circuits can also be constructed with trigger diodes. Figure 16.24 shows
two narrow-pulse generators with the
iookn
0<2jiFa
5 J TIC 56
A
TIL 31
0-lfiFi
2 f TIC 56
1
TIXL 27
lF«2A;tD»2jn
Figure 16.24 Pulse modulators with trigger diode
Ip = 3 A; to = iw 322
trigger diode TIC 56. The anode connections of the luminescence diodes are mounted directly on a grounded heat-sink. Figure 16.25 shows a further narrow-pulse modulator with the thyristor TIC 106. The
Trigger circuit
MRv
--U
i
7-
-o+Vb
if<ih 1h = HHoolding current
of thyristor
I
Figure 16.25 Narrow-pulse modulator with thyristor for high pulse currents
gate of the thyristor receives the trigger
signal from the trigger circuit. The capacitor
C 1 discharges through the thyristor and
the luminescence diode TIXL 27. Following
this, the thyristor turns off, when the
Ry current Iy through the resistor
is less
than the holding current Ijj of the
thyristor. The capacitor C 1 charges up
again. The peak current lp through the
GaAs diode TIXL 27 must not exceed 4 A.
The pulse width tp must then be less than
1 ms and the mark-space ratio must be less
than 10%.
Finally, Figure 16.26 shows a pulse trans-
mitter for the GaAs diodes TIXL 27 and TIL 31. The pulse generator consists of an emitter-coupled multivibrator (T 1 , T 2)
with short rise and fall times. The driver
transistor T 3 drives the power stage T 5, in
which a Darlington power transistor is
M lkn
||470fi
rUjkfi
fjiOOfi
®7
Tl
luF
470 n
-LZ3-
m T2 BC183
0*
T5 TIP 110
f= 1 klfctD = 90|is
Luminescence diude
TIL3I
Re = 0.68 Si:
A = I
1
TIXL 27 R E = 0,22 £2; I - 2.5 A
Figure 16.26 1R pulse transmitter with high power rating
323
used, so that a low control power is needed. The transistor T 4 measures the voltage drop on the emitter resistance of the final stage and regulates the current to the preselected value. The circuit is again so
designed, that the anode connection of the luminescence can be mounted directly on a
heat-sink connected to ground. When the TIL 31 is used, the range of the transmitter, without additional lenses, is more than 15 m.
324
17 Photodetector
circuits for
modulated radiation
17.1 17.2
Circuits with phototransistors
Circuits with the TIL 81 as photodiode and
as photocell
325
For the demodulation of a modulated
optical radiation, phototransistors,
photodiodes and photocells can be used.
background radiation can be substantially
suppressed by an IR filter (RG 830).
17.1 Circuits with phototransistors
The operation and the working-point conditions of emitter and collectorconnected phototransistors have been dealt
with in detail in Section 15.3.1.3.
As a first example, Figure 17.1 shows a sensitive phototransistor receiver. The output signal of the phototransistor T3 is
amplified in a two-stage transistor amplifier, with feedback through the resistor Rl, to stabilise the working point and the gain.
I
22k n
»
»
o+24V
120k n
2-2kf2
o Output
-- s OlyuF
5L
®"
TIL 63 -67 TIL 601 -616
Figure 17.1 Receiver amplifier for pulsed radiation
A very weak incident signal will hardly be
amplified at all by the phototransistor, since with very small collector currents this only has a low current gain, while with very large irradiances the phototransistor, and then also the following amplifier, will be overloaded. Therefore this circuit is primarily
17 Photodetector circuits for modulated radiation
suitable for the reception of pulsed radiation, since distortion (limitation) of the signals is then only of subsidiary importance. Interference effects from In Figure 17.2, a receiver circuit with an operational amplifier is shown. Through
the D.C. feedback Rl and R2, the output
Figure 17.2
Simple phototransistor amplifier for mod-
ulated radiation voltage of the amplifier in the steady-state condition is held at V, while the A.C.
voltage gain is determined by the resistance R3. The output voltage is calculated by the formula
V = R 3 · Ip
where Ip = photocurrent.
The circuit is suitable for the reception of LF-modulated radiation, if the working point of the phototransistor is set by the
incident irradiance. Let us dispense with further circuit examples, since firstly Figure 15.34 has been explained in detail and secondly circuits with photodiodes and
photocells are generally more favourable.
327
17.2
Circuit with the TIL 81 as a photodiode and
as a photocell
Photodiodes and photocells are suitable for the demodulation of a modulated optical radiation, even under unfavourable reception conditions. Photodiodes with reverse bias or short-circuited photocells have a linear
characteristic Ip = f(E e). The mode of
operation and the setting of the working point for the TIL 81 as a photocell or as a
photodiode have been described in Sections 15.3.1.1 and 15.3.1.2.
Figure 17.3 shows the demodulation of a weak useful optical radiation with and without unmodulated interference radiation.
In practice, the working points A\ and A2
can lie several orders of magnitude apart. Without interference radiation, the lower limit of detection of a radiation source depends on the noise level, or in simplified form, on the dark current of the photodetector.
&P, interference signal
Ee, interference signal
Figure 17.3 Demodulation of a modulated optical radiation
328
The possibility of evaluation of a weak useful radiation with a simultaneous more intensive, interference radiation depends on
the required ratio of useful to interference signal. Intensive interference radiations produce high interference photocurrents Ip interf. an(* tnus high input noise.
The sensitivity of a photodetector circuit
depends firstly on the absolute sensitivity s
of the photodetector and secondly on the
load resistance Ry\.
The frequency limit is determined by the
R^ load resistance
and the junction
capacitance Q:
1
f,,=
A Q R B
2tt .
·
(10.62)
Die mximum value of the load resistance RA can be calculated by rearranging the
equation (10.62).
1
Ra =
Q 27T . f,, .
For pulse operation, the rise time of the photodetector is calculated as:
0-35
(10.61)
Figure 1 7.4 shows the use of FET broad-
band amplifiers, driven by the photodiode TIL 81. The high-impedance load resistor
RA is not loaded by the subsequent source
follower. In addition, the field effect transistor is suitable for automatic gain
control. The control voltage is fed to the gate through the resistance Rg-
As a comparison, Figure 17.5 shows the use
of FET broad-band amplifiers, driven by the
photocell TIL 81. In comparison with diode
operation, the coupling capacitor to the gate
of the FET can be omitted in this case. The
irradiated cell produces a negative potential
on the gate. The shift in the working point
of the FET is unimportant, since the maximum no-load photo-voltage is 0-6 V. The gate resistance R \ of the source-
follower serves at the same time as the load resistance of the photocell.
The operating mode of the photocell is determined by the value of the load resistance R a- If very high interference
radiation levels are to be expected, the photocell must work in the short-circuited mode, in which case the load resistance
must lie between Ra = 1 kfi and R \ = 5 \&l. With low useful and interference
radiation levels, the resistance can be
i.0l u F
HI
1-
iookn
Input sjage unregulated
f
°*vh
[J 1 00k
z=Z&
Input stage regulated
BC :64B O Output
IT'
lOkSI
IMil
" CZD iluF
T
° vR
4ljF
Figure 17.4 Voltage drive of broad-band amplifiers with photodiodes
329
0-1 n F
BF 247A
o Output
RA 4-7k n
hokn
TIL 81
h-1
Input stage unregulated
Input stage regulated
3+Vb
* l|iF
hokn
-o Output
--owR
I/iF
Figure 17.5
Voltage drive of broad-band amplifiers with photocells
increased to 10 k£2 to 100 kSl The
interference radiation falling on to the
photocell is to be reduced by masks and IR
filters. If the load resistance has a value of
1 M£2, for example, the photocell works in
the no-load mode. With medium irradiances,
its working point is already in voltage
Vp^ saturation
*** 0-6 V. The useful signal
will either be totally suppressed or severely
A distorted. remedy is provided by variable
load resistances. Against high unmodulated
interference radiations, a Schottky diode
can be used as an additional variable load
resistance. With high irradiances, the working
point of the photocell is then determined by
the forward voltage of the diode.
With very high and severely fluctuating useful irradiances, it is advisable to use a field-effect transistor, the forward resistance rdson °f which is adjusted by a control voltage, as the load resistance (Figure
17.5b).
In Figure 1 7.6, two photodetector circuits
for modulation frequencies in the LF range are shown. The original FET source-follower
has been replaced by a source-connected
circuit. The Miller capacitance of the FET
reduces the upper frequency limit still further. In these applications, it is
advantageous to use low-noise FET types such as BC 264A - D or BF 805.
OluFi
n 100k I
22k n
-o+V b --O Output luF
fj.Mn
F
J°""
M.ookH
-- ~W -I
F:
-o+Vb -o Output
MnM 22k nM luF
Figure 1 7.6
Photodetector for modulation frequencies below 10 kHz; (a) with photocell TIL 81, (b) with photodiode TIL 81
330
-o Output
u.
a
loon §
u.
a
7I
1 1' 0*°
Rs · 1% Stabilised
+«-
0-1(1 F J- TIXL 56 TIXL 57
|
^ > RF
Output
Figure 17.7
Current drive of the broad-band amplifier TIXL 151; (a) with photodiode TIL 81, (bj with avalanche photodiodes TIXL 56 or TIXL 57
Figure 17.7 shows the broad-band amplifier
TIXL 151 driven by photodiodes. The base-
V emitter voltage Vbe ** 0-6 of the input
stage can be used as the reverse voltage for the photodiode. Circuit (a) is suitable for use as a measurement receiver. With it, for example, the modulation of the optical radiation from filament lamps, fluorescent tubes or luminescence diodes can be shown on an oscillograph. Also, the frequency response of these radiation sources can be investigated. With circuit (b), optical radiation with sine-wave modulation up to
fmod = 40 MHz can be demodulated. The
reverse voltage of the avalanche photodiode is to be adjusted exactly for the desired avalanche gain and stabilised to one part per thousand. In this process, the temperature coefficient of the avalanche diode is to be taken into account (see Section 9.1.1.).
With low irradiances, the safety resistance
RS can be omitted. It is to be so designed,
that the permissible loss power of the avalanche photodiode is not exceeded with
the maximum incident irradiance.
The avalanche photodiode TIXL 56 has the
typical relative spectral sensitivity of silicon junction photodetectors. Its reverse voltage
is approximately Vr = 170 V.
The avalanche photodiode TIXL 57 has the
typical relative spectral sensitivity of
germanium junction photodetectors. Its
Vr reverse voltage is approximately
= 40 V.
In both circuits, the gain of the TIXL 151 is
approximately 38 dB.
In Figure 1 7.8, examples of the driving of selective photodetector circuits are shown. The load impedance of the photodetector TIL 81 is a parallel tuned circuit, tuned to the modulation frequency in each case. The Q-factor of the tuned circuit determines the band-width b of the receiver detector circuit.
*.M fres _ ,,
/cj + CP
Q
L
Rv = Loss resistences in the input circuit.
A high-Q circuit gives reduced band-width
and thus reduced noise in the input amplifier. These circuits also have a number of other advantages. Since the tuned circuit only has a high impedance at the resonance frequency, all interference radiations except for the resonance frequency are shortcircuited, which also improves the usefulsignal/interference ratio. Furthermore, neither the junction capacitance of the photodetector nor the input capacitance of
331
«+vb
a)
9 j
nD
o-f+H Ibc 2
^_
sjf c
*^ Z
IMJi| 10kli|
J
|
T1L81
-o+V
b Output
-o+V
b
D
Figure 17.8 Driving of selective photodetector circuits; (a) with photodiode TIL 81, (b) with photocell TIL 81
the amplifier has an adverse effect, since they are included in the capacitance of the tuned circuit.
the condition, that R \ = R.2 = 2 · R3 and C\ = C2 = 0-5 . C3, the resonant frequency
of the filter is calculated by the formula
In selective receivers for LF-modulated radiation, it is advantageous to use active filters {Figure 1 7.9). The photocell TIL 81 works in the short-circuited mode. The subsequent source-follower drives the active filter, which consists of a two-stage amplifier, with feedback through a double-T network, with which the feedback reaches a
minimum at the desired frequency. Under
'res
1
27T Ri C\
With the design values stated in the circuit diagram, a resonant frequency fres = 1 kHz is obtained. The filter components should
have a tolerance of 1% maximum, in order to obtain a narrow passband. The filter is
47k n 2 2kn
4-7kn
SchottkyDiode
2-2k
n
l«F
3 Output
[JLr^IJH
ikn
y
56ka
y
Rl = R2= 10k n
R3
= 5k
C1 = C2 = ISnF
C3
= 30n F
Figure 17.9 Photodetector with active filter
332
a
184
mbi:47a ,,
I On F
-n--
BF 247A h ,
!IOO(j F
--o Output
hookn
22k a
! 47n F
M 1
S2
33k 12
47n F
10k Si
Figure 17.10 Receiver for optical telephony unit
followed by a thiee-stage amplifier with
emitter-follower output, with a voltage gain
Vu ~3000, so that even weak signals can
still be evaluated. In order to avoid
interference from fluorescent tubes, it is
RG advisable to fit an IR filter (Type
830)
in front of the photocell.
Figure 17.10 shows a sensitive receiver for optical telephony units. The photocell TIL 81 works into a high load resistance (R = 100 k£2), so that a relatively large signal is obtained at the amplifier input. Care must be taken, by the use of IR filters, that the cell is not driven, by interference radiation, into voltage saturation and that the useful signal is not thus suppressed. The voltage gain of the three-stage amplifier is
approximately 4000.
The upper frequency limit f is determined
by the load resistance of the photocell, the internal capacitance of the photocell and
the input capacitance (Miller capacitance)
of the amplifier; in the circuit shown it is
* f
7 kHz. If a source-follower is
connected in front of the amplifier (see
Figure 1 7.5a), the upper frequency limit is
increased to 20 kHz, since the effect of the
Miller capacitance of the first amplifier
stage is eliminated. The lower frequency
limit fu is approximately 350 Hz. It is determined by the 47 nF capacitors in the
source circuit of the field-effect transistors.
By changing these components, the frequency limit can be altered in a wide range.
The amplifier shown in Figure 17.11 can be
used together with the photodetector of Figure 1 7.10 as a monitor amplifier. This circuit can also be used as a measurement receiver or receiver for A.C- telegraphy.
Figure 17.11 Monitor amplifier for connection to the
photodetector shown in Figure 17.10
In Figure 17.12, an automatic gain control is used to compensate for fluctuations in irradiance. The control amplifier T5 is driven through the T4 output through a voltage divider. The output signal taken
from the collector of T5 is rectified with a peak rectifier. The control voltage thus
obtained is freed of interfering modulation components with a low-pass filter and then
controls the MOS transistor, which works as
a variable resistance. With an increasing
333
useful signal, the load resistance of the photocell becomes less, so that the cell is not driven into voltage saturation.
The photoreceiver circuits in Figures 1 7. 9, 17.10 and 17.12 are very sensitive. When
evaluating weak signlas, an 1R filter should always be placed in front of the photodetector. Depending on the application and construction, it must be noted that even undesired reflected useful radiations will be
evaluated.
--2~ _,
\ TIL 81
BC 264B
r\
3N 160
22nF
-H
\
100kJ i a
( · r~\
MTU' 00"" D
SchottkyDiode
UmF
Figure 17.12
A Pho to- . C. -telegraphy receiver
~% o Output
~1 w
|
luF«t
1M!!
3I
334
18
Practical measurement
of the photocurrent
sensitivity of Si phototransistors
18.1 18.2
Theoretical circuit of test equipments Test equipment for measurement of the relative sensitivity of phototransistors
335
The user tests optoelectronic components, not only in accordance with the data-sheet test conditions, but also in accordance with his own application specifications. The user's own selection measurements on optoelectric components are usually carried out in the goods-inward inspection. In these,
firstly the sensitivity tolerances of the phototransistors delivered are selected into various groups and secondly, criteria related to the application can be taken into account.
Criteria related to the application include:
1
Installation of masks, lenses, objective,
filters or light-guides in the beam path between the radiation source and photo-
transistor.
The radiation source used can show a radiation function S\ which differs greatly from the radiation source stated
in the data sheet.
18
Practical Measurement of the Photocurrent Sensitivity of Si Phototransistors
the same way, only the relative radiant power of GaAs diodes can be measured.
The values measured with these test equipments cannot necessarily be compared with the data-sheet values, since the datasheet test conditions do not usually agree
with the individual test conditions. Nevertheless, the values found in these
measurements are, in most cases, more
informative for the user than the pure data-sheet values, since the appropriate construction of the test equipment and the mechanical and optical characteristics of the whole circuit are covered.
18.1
Theoretical circuit of test equipments
Since all tests of this kind involve comparative measurements, the test equipment normally includes a bridge circuit (Figure 18.1). For the measurement of the
The selected working point for the phototransistor test device does not need to correspond with the working point stated under the test conditions in the
data sheet (e.g. VqE = 5 V). The working
point will be determined by the function I^ = fCVcg) with an irradiance value related to the application as the parameter and with the selected load resistance.
In many cases, therefore, a test equipment
to be built will have almost the same circuit, the same optoelectronic components and an equivalent mechanical construction, corresponding to the system in the finished equipment. With test equipments of this kind, the relative photocurrent sensitivity of phototransistors and diodes is measured. In
Figure 18.1
Theoretical circuit of a test equipment for the measurement of the photocurrent sensitivity of Si phototransistors
relative radiant power of GaAs diodes, the bridge circuit contains two phototransistors (or photodiodes) with the same photo-
current sensitivity in each case. Since the
337
absolute sensitivity is only of minor importance here, the user can very easily measure two devices with identical
R characteristics. The resistance will be so
selected, that the same current flows through the test specimen as in the final circuit. The potentiometer P can then very easily be calibrated directly as a percentage of the relative radiant power, since the radiant power is proportional to the current.
For the measurement of the relative photocurrent sensitivity of the phototransistors, the circuit is to be so designed, that both transistors are irradiated with the same radiant power. Instead of the fixed resistance, a potentiometer will then be connected in series with the test specimen and will again be calibrated as a percentage of the relative sensitivity.
Basically, the calibrated values can be measured with high accuracy with the circuit shown. The measurement errors will be primarily determined by the mechanical tolerances of the arrangement.
18.2
Test equipment for measurement of the relative sensitivity of phototransistors.
This test equipment has been developed, to record the squint effect of a phototransistor in a given mechanical arrangement. The squint effect is caused by the fact, that the optical major axis (normal) of optoelectronic components does not coincide exactly with the mechanical axis. This has a particularly adverse effect, when, for example, phototransistors are combined with thin light-guides, since the latter are mounted, for constructional reasons, in the mechanical and not in the optical axis of the phototransistor. Therefore only the sensitivity of the phototransistor in the major axis is decisive for the whole arrangement, while the reception aperture angle, and thus the relative sensitivity of the phototransistor, can be varied over the diameter of the light guide.
Figure 18.2 shows the circuit of the test equipment. It consists of two source-
4
Dli TIL 31 A
US"
<&
OT1
LS400I
Rod light-guide
'
I
0-355 mm0
-- tTest specimen
1R2 iok a
>
_o
approx. -8V stab
0-
D2
, TIL 31
Figure 18.2
Test circuit for measurement of relative photocurrent sensitivity
338
detector combinations, while the phototransistor under test, Tl, and the reference
phototransistor T2 form a bridge circuit with the collector resistances Rl and R2. To suit the application, two GaAs diodes TIL 31 are
A used as radiation sources. light guide is
located in the beam path between the GaAs diode Dl and the phototransistor Tl under
test, as in the practical application.
The bridge circuit compensates the temperature-dependence of the phototransistor under test, if the same type is
selected for both photo transistors Tl and T2. In addition, both phototransistors Tl and T2 should have approximately the same photocurrent sensitivity. The indicating instrument in the bridge shows the deviation
of the photocurrent sensitivity of the test specimen from that of the reference
phototransistor. The indicating instrument has a high internal resistance. The bridge is balanced by varying the source-detector
distance of the reference system. If a changing polarity on the indicating instrument is not desired, the bridge will be
set off-balance.
339
19
Light measurement
with Si
phototransistors in electronic flash units
19.1 19.2 19.3 19.4
Principle of an electronic flash unit Types of exposure control Circuit of a flash unit Si phototransistors for the automatic flash exposure control
341
^
--
19
Light measurement with Si phototransistors
in electronic flash units.
The advances in recent years in the miniaturisation of electronic components, combined with a reduction in costs, have led to the production of small but efficient electronic flash units. They thus helped the breakthrough of flashlight photography into the amateur field. But one problem
remained: the choice of the correct exposure time.
Determination of the aperture value by the formula: "Aperture = Coefficient/Distance"
only gives a very inexact result. Firstly, the exact determination of the distance of the subject is quite complicated. However, a
much more important fact, is that with this
formula the environmental conditions, such as the reflectivity of the surrounding walls, are not taken into account at all and thus lead, especially in colour photography, to considerable exposure errors. It is therefore worth while to measure the radiation reflected from the subject and to control the emission time of the flash unit from this. This problem can be solved in a simple manner with the aid of optoelectronic components.
19.1 Principle of an electronic flash unit
The supply unit, which consists of an accumulator or a number of dry batteries and the voltage converter, charges the main
V capacitor to a voltage between 300 and
500 V. Firing is triggered by the synchronising contact of the camera.
The flash discharge tube is fired through the igniter electrode by a high-voltage pulse from the ignition circuit. In the flash discharge tube, the xenon is partially ionised, so that the discharge of the main capacitor commences. The plasma is excited into intense radiation by very high currents of several hundred amperes. (Figure 19.1.)
Supply unit
·
o
a.
--
** 1
2 «
3-
g
I
1
c ^--1
1
i
Ignition
circuit
\f*^\
hZ
Figure 19.1
Theoretical circuit of a modern flash unit
The various parameters, such as the spectral radiation distribution, luminous flux, luminous efficiency, flash duration and
half-value time of xenon-filled flash discharge tubes depend firstly on the construction of the flash tube and secondly
on the design of the flash unit. The
variation of the flash discharge with time is characterised by the light-emission/time curve. Figure 19.2 shows a typical emission/ time curve with a steep -rise and slow fall. In
this curve:
tl
Contact or delay time: Time from the
triggering of the flash until the
commencement of light emission.
t2
Start-up time: Time between the
triggering of the flash and the time
when the half maximum value of the
luminous power is first reached.
= *3
Peak time: Time from the triggering of the flash until the maximum luminous power is reached.
t4 =
Half-value emission time: Period between the first and second times
that the half-value of the maximum
luminous power is reached. This time
is a significant criterion, since,
343
JIOCK
1
*rcl
SOW
--
14
1
i m !>,,-r|
"
n
1
,J
'
'
i
,5 '
i
Figure 19.2 Emission/time curve of a flash discharge
together with the maximum luminous
power, it determines the photographic
efficiency.
t5 =
Mid-point time: Period between the triggering of the flash until the midpoint of the light emission time.
t£ =
Light-emission time: Duration of the
quantity of light emitted between the
values e ,rel =10% and e rei = 90%.
Other, similar definitions are also
common.
The discharge circuit determines the shape of the emission/time curve, mainly with the anode voltage, the capacitance and the total resistance. The total resistance is composed of the internal resistance of the flash tube, the series resistance of the main capacitor and the lead resistances.
The efficiency of a flash discharge lamp is stated as the luminous efficiency. It is the ratio of the total quantity of light produced to the electrical energy consumed.
<J>. t[lms]
Vv
P. t[Ws]
(2.35)
In amateur flash units, the luminous efficiency
T?v lies between 35 lms/Ws and 50 lms/Ws. The luminous efficiency mainly depends on
the power and the maximum peak current.
It is calculated from the formula:
C. IT
W=-
(19.1)
W = Energy (Ws), C = Capacitance (F or
As/V)
Example:
C = 500 (IF; V = 360 V; T? v = 40 lms/Ws
W 500 .
10
6
As .
(360 V) 2
:
V-2
64-83 AsVz
2V
32-4 Ws
(19.2)
The quantity of light emitted is:
W W v = <£ . t = TJ .
(19.3)
40 1ms . 32-4 Ws
w=
. = ! 296.4 ims
Ws
(19.4)
19.2
Types of exposure control
In exposure control circuits, a distinction is
made between a switching device in parallel with the flash tube {Figure 19.3) and a
switching device in series with the flash tube
(Figure 19.4). The control circuit remains the same in principle in both cases. The
switching device in parallel with the flash tube consists of a low-resistance short-
circuiting circuit. When the optimum exposure
is reached, the remaining energy in the storage capacitor is shorted by the lowresistance parallel circuit, so that the flash
radiation is interrupted. In more recently
developed flash units, the low-resistance shorting circuit in parallel with the flash tube is replaced by an interrupter circuit in series with the flash tube. In this case, only the
energy necessary for the optimum exposure is drawn from the main or storage capacitor. The energy which is not needed remains stored in the main capacitor. For battery
344
and accumulator-operated flash units, this saving of energy gives a larger number of flashes and a more rapid flash repetition
rate.
7
\i /
)
Control circuit
Phototransistor
Figure 19.3 Principle of exposure control with a shortcircuiting circuit in parallel with the flash-
tube
Control circuil
-@~
I'hototransisto
Figure 19.4 Principle of exposure control with an inter rupter circuit in series with the flash tube
19.3
Circuit of a flash unit
Figure 19.5 shows, as an example, the circuit of an amateur flash unit with a switching device in parallel with the flash discharge tube. With the onset of the flash discharge, the operating voltage for the automatic exposure control is produced at the same
time through the capacitor C 3. Thus the
automatic exposure control remains unaffected by other flashes. After the flash has been triggered, the phototransistor receives the reflected flashlight. The output signal of the phototransistor is integrated up
to a threshold value, the optimum exposure value. -The capacitor C 1 serves as the integrating element. The pulse released through the thyristor TIC 47 to the primary
side of the ignition transformer Tr 2 is stepped up on the secondary side, so that the quench tube fires. The tolerances on the response sensitivity of the thyristor and on the photocurrent sensitivity of the phototransistor are eliminated by calibration of
the automatic exposure control with R 1.
The quench tube, when fired, has a
considerably lower resistance than the flash
discharge tube. The storage capacitor C 2
therefore discharges through the quench tube, so that the flash emission is interrupted.
Figure 19.5 Circuit of an amateur flash unit
345
2'--
§
19.4 Si phototransistors for the automatic flash exposure control
Photodetectors for automatic exposure control are selected according to their absolute sensitivity, spectral sensitivity distribution, their sensitivity as a function of the angle of incidence, the turn-on time and sometimes the case dimensions. For reasons of cost, sensitive phototransistors are
preferred.
Any proportionality error between the collector current Ic and the irradiance Ee
can usually be neglected in practice.
The spatial sensitivity s = f(ip) can be adapted to the requirements by an aperture mask. Phototransistors with relatively narrow reception peaks represent an optimum
solution, since their sensitivity distribution s = f(i£) corresponds approximately to the desired radiation conditions.
In Figure 19.6, the relative sensitivity srei is
shown schematically as a function of the angle of incidence for the Si phototransistor TIL 78 with an aperture mask. The
measurement area A of the subject
illuminated by the flash is covered almost exactly by the reception peak of the TIL 78. Phototransistors with very narrow reception peaks cover the reflected radiation from the
measurement area A with greater errors. In
addition, the installation tolerances of the phototransistor would have to lie within very close limits. The turn-on time of the
phototransistor TIL 78 is less than 1 lis. Thus even very short flash times can be
safely dealt with.
The spectral emission distribution of an ideal flash radiation source and the spectral sensitivity distribution of an ideal photodetector should be linear. Their spectral limits should correspond with the spectral sensitivity limits of the film. In practice, xenon flash tubes have become established
/CN \ / \
/
/
rD
\
\
\
\
V^/ \V
/)
Measurement area A, the mean outline area of the subject
^
**
·%,
3--
1
IS
«
**
^
I*
.--
oO
<s *-
^~--*'
*s
5 "5. **
--
e zc>
=3 « H a o>
--
<
--
c
-'
s
^5
-- 3 ~~
3
Z-.
^ ^
**
t
^
fc*
tS
E&
tS
E?
ooo o
o
0\
00
r-
'O
vi
/ / //
// / /
//
fl
«
S
^§
--
/""^
y-
Phototransistor TIL 78
-^^^
Fixed aperture
.. ^~~~~-in flash unit
S
-""
\ ^ o>
--
^
\
Sensitivity peak of the phototransistor TIL 78
.--
--
1
--
r 3'^0 m
m
£= = 7-5°;2
15°
v>
I
Figure 19.6
Determination of the relative sensitivity srej as a function of the angle <p of the returned incident flash radiation for the phototransistor TIL 78
346
as flash radiation sources and silicon phototransistors as photodetectors.
Figure 19.7 shows the spectral emission distribution S(XB)\ of the xenon flash discharge, the spectral sensitivity s(X)jil 78 of the phototransistor TIL 78 and the graphic determination of the spectral sensitivity s(XB)X,TIL 78 of the photo-
transistor TIL 78 for the xenon flash
radiation XB (seel also Section 9.8). The
relative spectral sensitivity V(X) of the eye is shown for comparison. Despite the
difference in the spectral sensitivity of the
film, the spectral sensitivity s(XB)X is suitable for a relative flashlight exposure measurement, since most substances have almost the same reflectivity in the visible and in the near IR range.
1-2
1-0
»(*>rtl
1
0-8
/>
V(X)
1
0-6
S(XBh
^·^ /Wl /V(X)
\
1L78,»1
S(X) 0-4
0-2
\
/
'\/«> »KTIL78^el
VJ
VK
^J
\ L
·3
0'4
5
C ·6
0-7
0-8
9
(
10
1I
1-2
Figure 19.7 Graphical determination of the relative spectral sensitivity of the TIL 78 phototransistor for xenon flash radiation (see Section 9.8). The symbols denote: S(l) = Radiation function of a radiation s(X) = Spectral sensitivity of the phototransistor TIL 78, S(XB) = Radiation function of the xenon flash radiation, s(XB) = Spectral sensitivity of the phototransistor TIL 78 for the xenon flash radiation, V(X) = Relative spectral sensitivity of the eye
347
20 Circuits with
light-emitting
diodes
20.1 20.2 20.3 20.4 20.5 20.6 20.7
Simple indicators Diode tester Logic tester Polarity and voltage tester Large format seven-segment display unit Analogue indication of digital values
Analogue measuring instruments with LED
indication
349
20 Circuits with Light-Emitting Diodes
This chapter describes the application of luminescence diodes, the radiation from which lies in the visible range. These
components can be used in many ways as optical indicators and are replacing filament
and discharge lamps to an increasing extent.
A significant advantage of these light-
emitting diodes is firstly the low working voltage, as a result of which they can be driven directly by almost all semiconductor components (i.e. transistors, thyristors and
A integrated circuits). further advantage to
be mentioned, in comparison with filament lamps, is that no current peak occurs at switching on, as is the case with filament lamps because of the low resistance of the cold filament. If the circuit is designed with
insufficient care, this very phenomenon can
easily lead to its destruction.
20.1 Simple indicators
Luminescence diode can be used in many
cases for operational indication in equipments, replacing conventional filament
lamps. With low operating voltages, in particular, these diodes consume considerably less power. Figure 20. 1 shows the corresponding circuit. The series resistance
Rv can easily be calculated from the formula
Vb- yF
Rv
if
For the diode TIL 220, with an operating
V voltage of Vfc = 5 and a current Ip = 30 mA:
Rv
=
5V-1-6V
30 mA
«120nH
In the same way, these diodes can be driven
directly from TTL circuits (Figure 20.2).
Care is to be taken, however, that the
Figure 20.1 Supply voltage indicator
outputs of the T.T.L. can deliver the
necessary current. The drivers SN 7416N and SN 7417N, which can deliver an output current up to 40 mA, are particularly
suitable for this purpose. The series
resistance Rv is calculated from the formula:
Vcc-vol-Vf RV
In order to indicate changes of state in an equipment as clearly as possible, it is often advantageous, to use indicators of different colours. With the following circuit, the red
diode lights up when Hrlevel is present at the input, and the greed diode with L-level (Figure 20.3). It must be noted, that green-
emitting diodes have a higher forward
voltage Vp and thus the value of the series
resistance becomes less.
351
I> RV
VCC
TIL 220
CHC=J RV
TIL 221
I
1
Figure 20.2
Control of luminescence diodes by TTL circuits
Input O--
WXW TIL 220 (red)
7
120 n
L>
, 1
TIL 222 « W
(green) .*.
1 »V CC
.. ».
82 n
l>
1/3 SN 7416N
Figure 20.3 Indication with colour change on change of logic level
20.2 Diode tester
specimen; diode defective.
The circuit in Figure 20.4 shows a diode tester, with which the polarity and function can be tested in a simple manner. According
to the polarity of the half-wave of the alternating voltage, the current flows either
through the green or the red diode. The
following four conditions are possible:
1
Neither diode lights up: Open circuit.
Red diode lights up: Anode of the test
specimen to connection 1 ; diode in order.
Green diode lights up: Anode of the test
specimen to connection 2; diode in order.
Both diodes light up: Short-circuit in test
20.3 Logic tester
The testing of digital circuits in complete
systems often causes difficulty. Several
inputs En and the corresponding outputs A = f(En) must be observed simultaneously,
which in most cases is not possible even with multichannel oscillographs.
The logic tester shown in Figure 20.5, with which all 16 pins of the integrated circuit to be tested are contacted, is suitable for measurements of this kind. Through a logic
circuit, the V^c and GND connections are
automatically determined and the actual test circuit is supplied with operating current. At every IC connection there is an amplifier, the input configuration of which corresponds
to that of the TTL or DTL circuits. The
output of the amplifier dirves a luminescence
352
Bell transformer
Figure 20.4 Simple diode tester
V cc .Bus
Figure 20.5 Circuit of the logic tester
diode, which lights up, if H-level is present at
the amplifier input.
As can be seen from the circuit of the logic
D tester, the diodes 1 and D 2 search for the Vqq and the GND connections. In
considering the mode of operation of the
V^C circuit, it is at first assumed, that the
connection carries the highest positive and
GND the
connection the lowest negative
VqC potential. If, for instance,
potential
(+5 V) is present at connection 16, then the
diode D 1 here is conductive and feeds the
VCC bus- AU other diodes D 1 are connected
to the inputs or outputs of the test specimen
and, in all cases, a more negative voltage is
353
6
present at these points, so that these diodes
are non-conducting. The diodes D 2, which
GND search for the
connection and supply
current to the GND bus, work in the opposite
direction.
Because of the forward voltage Vp2 of the D diode 2, there is a voltage of 0-7 V,
relative to the ground potential of the logic
GND system, on the
bus. In order that the
transistor becomes conductive, the voltage
V on its base must be at least 2 x 0-7 more
positive (because of the diode D 4). Therefore
> V GND a voltage Vj 1-4 (relative to the
connection of the circuit being tested) must
be present at the cathode of the diode D 3.
This value corresponds approximately to the
threshold voltage of the DTL and TTL
circuits. The design of the other components
mA is very simple. For a diode current Ip = 10
mA 10
through the luminescence diode
TIL 209A, the series resistance is calculated
from the formula:
Ry VCC " Vfi - VF5 - VcEsat - VF4 " Vf2
so that the tester does not represent any
significant load.
20.4 Polarity and voltage tester
The test device in Figure 20. 6 permits rapid testing of the polarity of unknown voltage sources. The polarity is indicated by a red and a green light-emitting diode.
Dl IN 914
D2 IN 914
RV = V V V V V V 5 - 0-8 - 1-6 - 0-3 - 0-7 - 0-8
10 mA 82 mA
If it is assumed, that the current gain of the
^ transistor hpE(sat) 50, then the base
resistance is calculated as:
RB = VCC " VF1 " VfiE " VF4 " VF2
hFE
ic
RB V V V V V 5 - 0-8 - 0-7 - 0-7 - 0-8
50 10 mA
= 10 kfi
The input current Ijl, with which the test specimen is loaded, is thus less than 200 [dA,
Figure 20. Polarity and voltage tester
Such devices have to be required to work in a large voltage range. Therefore it is not
possible to stabilise the current for the luminescence diode with a simple series
resistance, because with low voltages the diode would only light up weakly or not at all, while with high voltages the current would rise so far that the diode would be over stressed. For this reason, in polarity and voltage testers, the current is regulated to a constant value by a stabilising circuit. For this, the transistor Tl (T3) measures the voltage drop on the resistance Rl (R3). If the voltage becomes greater than 0-7 V, which corresponds to a current of approxi-
mately 20 mA, the transistor lowers the base voltage of transistor T2 (T4), so that the
current through the luminescence diode remains constant.
354
7
A separate current source is provided for
each polarity, driving a red and a green luminescence diode respectively. The diodes
D D 1 and 2 short-circuit the part of the
circuit which is not in use because of the
polarity. The maximum permissible input
voltage is primarily determined by the
maximum permissible loss power of the transistors T2 and T4.
--· o*12 V
F
35 x TIL 220
or 7 x TIL 265 with Ry = 1 80S2
20.5 Large-format seven-segment display element
If numerical display units have to be read
over long distances, then sometimes the
figure size of the display units at present
obtainable (e.g. TIL 302) is not sufficient.
In this case, a seven-segment display element
of any desired size can be constructed from
individual diodes. Figure 20. 7 shows the
circuit. As the luminescence diode, the type
TIL 220 is used, while 5 diodes are connected
in series for each segment. A figure height of
mm approximately 60
is thus obtained.
Because the diodes are connected in series,
the segment current is only 30 mA, which
makes it possible to drive this unit from a
normal SN 7447AN TTL seven-segment
decoder. It is only necessary to provide a
12V supply for the luminescence diodes.
20.6 Analogue indication of digital values
It is often necessary to indicate signals, which originate from digital measuring instruments, but where the absolute measured value is of no interest, but only the relative value, e.g. the filling level in a tank: Full, %, Vi, V4, Empty; or the difference of a motor speed from the nominal value: Positive - Zero -
Negative.
Although digital display units are easily read they do not always give the required accuracy. In these cases it is advisable, to display the measured values in analogue form. With a row of luminescence diodes,
Figure 20. Large-format seven-segment display for a
mm figure height of 60
simple scales, which display the measured value in quasi-analogue form, can be constructed. Figure 20.8 shows a simple circuit, in which a digital 4-bit value is decoded in a 1 -from- 16 decoder and is then displayed with 16 luminescent diodes. One of the 16 diodes lights up, according to the incoming digital value.
Since the TIL 209A, which already emits sufficiently brightly at 15 mA, is used for the diodes, the SN 74159N is adequate for
355
SN 74I59N
A
B
C
D
Digital input
TT
2
13
14
15
Figure 20.8 Simple analogue indication of digital values
the decoder and driver. Because only one diode ever lights up at a time, only one
series resistance, common to all diodes, is A used. less favourable feature of this circuit,
however, is that the value read off always has to be compared with a scale, which has to be arranged beside the diodes. In order to avoid this disadvantage, several diodes can
be made to light up, according to the value
to be displayed, so that a luminescent strip is produced. In comparison with the circuit in Figure 20.8, additional drivers are to be connected between the individual outputs (Figure 20.9). Through a "Wired-OR" relationship, these switch all lower-valued outputs to L-level, so that with the input
H information 8 (= L L L), the diodes . . .
light up.
nTTTTTTn
h^ o -a <} (y<\ <}
«o
Decoder Buffet
SN 7414SN SN 7417N
Digital input
LEDs
10 x TIL 220 li TIL 270
U Series resistance Rv = 120
Rv " 150 ft
Figure 20.9
Representation of digital values by a light strip
356
Scales can also be constructed, with the zero point in the centre instead of at one end, so that both positive and negative deviations from a nominal value can be indicated. For the representation of negative numbers there are two basic possibilities, which are shown on the two linear scales in Figure 20.10. Which of the two numerical representations is selected, depends on the particular
application. From the circuit point of view, neither of the two codes has any advantages
or disadvantages for the construction of the
scales.
Decimal
-7 -6 -5 -4 -3 -2 -1
1
2 3 4 5 6 7
Code A
CodeB
Sign C B A Sign C B A
H HHH H LLH H HHL H LHL H HLH H LHH H HLL H HLL H LHH H HLH H LHL H HHL H LLH H HHH
L LLL L LLL
L LLH L LLH
L LHL L LHL L LHH L LHH L HL L L HL L L HLH L HLH L HHL L HHL L HHH L HHH
Table 20.1
Code table for negative and positive numbers
Table 20.1 shows the two codes which are used in the circuits in Figure 20.1 land Figure 20.12. In both cases, positive numbers are represented in the same binary code. In both cases, the necessary sign bit is "Low". With code A, negative numbers are characterised by the amount of the number and an H-signal in the sign bit. With code B, on the other hand, the negative number is represented by the unit complement and an
H-signal in the sign bit.
In both circuit diagrams, the value of the number is again represented by the length of the luminescent strip and the number of diodes which are lit. In order to be able to distinguish positive and negative numbers easily, positive values are indicated by green and negative numbers by red luminescence diodes. The two circuits only differ in the connection of the outputs, which arises from the significance of the respective decoder outputs in the two numerical codes.
20.7
Analogue measuring instruments with LED
indication
Up to the present, electromechanical
measurements are indicated almost exclusively in analogue form. These measured values permit a high resolution,
which is, however, not required in many
cases. Therefore, in the following section, a
number of circuits will be described, by means of which analogue measured values are indicated by a row of luminescent diodes
1
r
-i
-5
-4
-3
i
r
t
1
~r
i
1
r
4
5
6
-t
1
--1
n
r
2
3
4
5
6
Figure 20.10 Representation of positive and negative numbers
357
,,y
\y
SN 74145N
A
B
Vj
S 8 x 68 SJ /8 x 82
! x TIL 222/ 1 x TIL 278
\J \J SN7417N
SN 74 MSN
C
D
Figure
<1
Figure 20.11
A Luminescent strip with centre zero; numerical representation in accordance with code in
Table 20.1
\y-\y
SN 74145N
A
E
8x68fi/
8x82Q L
8 x TIL 222/ I x TIL 278x-"
'"<y<h(y- SN7417N V.
SN 74145N
C
D
Figure
^3
Sign
Figure 20.12
Luminescent strip with centre zero; numerical representation in accordance with code B
(units complement) in Table 20.1
- while these simulate a scale -- similar to
the circuits described in the previous section.
It is a requirement of such circuits, that,
according to the voltage (current, resistance)
at the input, a number of diodes proportional to the input value lights up or a given diode lights up in a row.
358
4 x TIL 209A or 1 x TIL 264
3-
I
Figure 20.13 Simple voltmeter with luminescence diode indication
Figure 20.13 shows a simple circuit, which
is applied in principle - but of course in considerably more extensive form -- in the
integrated circuits described later. If the input voltage Vj is less than 0-7 V, then all
transistors are switched off. Thus none of
the diodes light up. If the input voltage
V exceeds the value 0-7 (Vjje of the
transistor Tl) the first transistor becomes conductive and the luminescent diode in its
collector circuit lights up. Since there are
one or more diodes in series with the bases of the other transistors (Dl - D3), these at first remain switched off. With an input
> voltage Vj 1-4 V, the diode Dl now
becomes conductive, so that the transistor T2 turns on and the second luminescent diode lights up, and so on until with an
> V input voltage Vj 2-8 all four diodes
light up.
Disadvantages of this circuit are firstly the low input resistance and secondly the inexact switching thresholds, which are determined exclusively by the forward voltages of the diodes and the base-emitter voltages of the
transistors.
More suitable for applications of this kind is
the integrated level indicator SN 16889P,
which has been specially developed for the
purpose. It has a high-impedance input
amplifier. The switching thresholds for the individual amplifiers, which drive the luminescence diodes, are stabilised by an internal temperature-compensated reference voltage. The switching thresholds lie at 200,
400, 600, 800 and 1000 mV and can be
enlarged as desired by a voltage divider at the input. Figure 20.14 shows the circuit of this level meter. The operating voltage of this
module may lie within the range V = 10 to
16 V. In order to keep the brightness of the luminescent diodes constant, they are fed from stabilised current sources.
It is useful, to indicate separately when a value exceeds or falls below a given level. This is done in the circuit in Figure 20. 15 in such a way, that when the input voltage falls below the lowest threshold value
< (Vj 200 mV), that is, when output 2
switches off, this output is periodically switched on again by feedback, so that the luminescence diode at this output begins to
flash.
If the input voltage falls below 200 mV, then output 2 switches off. The voltage at
this point rises almost to the supply voltage. Through the 5 M.Q, resistor, the 50 /iF
capacitor is now charged up to 200 mV, so
359
that the output 2 switches on again, until the capacitor has been discharged through
the resistances in the input circuit and a new
charging process begins.
-ov
b
=10
...
16-5
V
IN 754 (68 V)
S \ TIL 209A r 1 x TIL 265
Si sQ-
Figure 20.14 Level indicator
T
-O+10 ... 16-5 V
50>lF
4""
3
5
SN 16889P
T
Figure 20. 15
Level indicator with flasher circuit to indicate when input falls below a limit
360
21
Numeric and
alphanumeric
display units
21.1 21.2 21.3 21.4 21.5
Seven-segment display units Multiplex operation of display units Numerical display units with integrated logic Monolithic display units 5 x 7-point matrix display units
361
21 Alphanumeric display units
21.1 Seven-segment display units
In recent times, the so-called seven-segment
display units are becoming more and more
widespread. In these units the figures or
symbols are composed of individual lightemitting bars (segments). With a total of
seven segments, the figures to 9 and, for special cases, the letters A, C, E, F, H, J, L,
U P, S, can be represented. The basic pattern
is formed by an upright rectangle, divided in the middle by a horizontal line. (Figure 21.1.)
There are no problems in the driving of
these luminescent diode or LED display
systems. For current limitation, it is only
necessary to provide small series resistors
(Figure 21.3). The decimal point can be controlled through a switch, a normal gate or an inverter (SN 7400N, SN 7401N, SN 7404N, SN 7405N). These circuits are capable, without difficulty, of delivering the
required diode current of 20 mA, if an
> output voltage VqL v ° 4 is permitted,
Display units of this kind are produced by Texas Instruments under the type numbers TIL 302, 303, 304, 312 and 320. The
individual segments consist of light-emitting
or luminescent diodes. These displays are characterised by high brightness and good
legibility, even at flat viewing angles. As a further advantage, the low power consumption
should be mentioned. Since the operating
V voltage is only a few volts (typically 1-7 or
3-4 V), these units can be used directly with
TTL integrated circuits. As decoder and
driver, the integrated circuits SN 7446AN
(30 V/40 mA output) and SN 7447 AN (15 V
mA (15 V/40
output), which have an open-
collector output configuration, are available
(Figure 21.2).
SN 7446/47AN
I
Figure 21.
Circuit of the output stage of SN 7446AN
and SN 7447AN
n
I_l
u D i ~i
i_ l
c J J I
I
i_i
u i '-' ·-!
i
'
o r c c u p i
i
c,
ij
J u II l_ l_ I
i
I
I- i
-·
Figure 21.1
Representation of figures and letters with seven-segment display units
363
--
o
which is of no importance for applications of
this kind.
On
RBI
i son
g TIL 302 7 x68fl
Figure 21.3 Control of display units with light-emitting diodes
The seven-segment decoder drivers also have a number of additional inputs and outputs, with which numerous further functions can
be carried out. First, the "Lamp Test" (LT)
input should be mentioned. It serves to test the display unit. If this connection is at
"Low" level and there is not a "Low" signal
at the same time on the "Blanking Input" (BI) connection, all seven segments light up,
i.e. the figure 8 is displayed.
The blanking out of figures ("Blanking" and "Zero-Blanking") can be carried out very
simply, since the necessary electronics for these functions are already contained in the
circuit: If there is a "Low" level at the input
"Ripple Blanking Input" (RBI), all output transistors will be switched off, if the inputs
A, B, C and D are also "Low", i.e. if a A decimal zero is present at the inputs.
further connection "Blanking Input" and "Ripple Blanking Output" (BI/RBO) works simultaneously as an output and an input. If
a "Low" signal is applied to this connection
through a switch or an IC with opencollector output, the display is extinguished. Thus, for example, it is possible to attach a
TJL 302 j| DP a bp c du e fI gB |
TIL 302 -/
_/ op | ab.c d e f g
--
vec
IkR
;
\
150
7 x68
17 x68
SN 744 7N
a bcde f I
rtrtflfn'inn
i bed e f g
RBI
RBO >*-0 RBI
RBO
ABCD LT
ABCD LT
cS
4>
OdIIoIn I
kSI ]SKl'7447N
V(T )SN 7447N
&J ibcdeft
BC 182
RBI RBOpl
AB-CD"" LT
9
a bed e f g
RBI
RBO
ABCD LT
t
All
i inverters SN 7405N
++-
6666 ABCD
6666 ABCD
--DP v Data
a 66
ABCD
1
DP
QVcc
o
IT x kir*
AT
6666 ABCD
Jin.
Brightness modulation
Figure 21.4 Multi-digit display system with seven-segment display units
364
brightness control to this point. If this
connection is working as an output, it
becomes "Low" if the inputs A, B, C, D and
"Ripple Blanking Input" are also "Low". With this function, automatic zero-suppression can be carried out over any desired number
8-digit display, the 8 figures to be displayed are then switched on one after another. The switching of the individual figures takes place with such a high frequency, that because of
the slow response of the human eye, a
stationary image still appears.
of digits.
Figures 21.6 and 21. 7 show the circuit for
Figure 21.4 shows the circuit of a display unit, in which the facilities described above are utilised. This system can be used, for example, in a digital frequency-meter. By switching the decimal point, numbers can be displayed in the range from 00-00 to 9999,
while zeros before the decimal point are
blanked off, so that instead of the number 00-35, only the value -35 is displayed. Thus reading errors, which easily occur with longer numbers can be avoided. If the switch S is opened, the zero-blanking is switched off. The function of the display units can be tested, with the button T. When the button
is pressed, an 8 is displayed in all digits. Through the brightness modulation input, it is possible to control the brightness of the
such a multiplex operation. Despite a
somewhat higher component cost, this circuit is hardly any more expensive than 8
individual SN 7446A or SN 7447A decoders. A considerable advantage lies in the fact that,
with this method, only 8 data lines and 2 power supply lines are needed between the control and display sections, in contrast to a total of 34 lines with a conventional configuration. This is of particular interest, if long distances have to be spanned between the data source and the display system for remote indications. With displays with more than 8 digits, this method is always advisable, since in these cases the system described here gives price advantages in every
case. The circuit works as follows:
display continuously with a square-wave
signal of variable pulse-width. The circuit shown in Figure 21.5 is a suitable pulse
generator.
A free-running oscillator (SN 741 3N) with a
frequency of approximately 1 kHz drives the counter (SN 7493AN). The outputs of
the latter lead to the 4 multiplexers
(SN 74151N), which switch the inputs
Output
C8 Al Bl CI A8, . . .
. . . B8,
...
and
Dl . . . D8 in sequence to the lines Y^., Yg,
Yc and Yt> which finally drive the seven-
segment decoder SN 7448N. Its outputs are
connected with 7 current sources of type
SN 75450N (with the decimal point there
SN 74123N
are 8 current sources), which serve to feed the GaAs luminescent diodes. Because of
L
J
Figure 21.5 Clock generator for brightness control of
semiconductor display units
the low operating voltage (5 V), current limitation by the use of resistors is not advisable, since slight variations in working voltage would result in considerable current variations. The outputs of the counter also
lead to the decoder SN 74145N, which, in
21.2 Multiplex operation of display units
synchronism with the selected data address in each case, switches on the corresponding display unit (TIL 302) through additional
Display units constructed with luminescent
diodes can be operated in the multiplex mode
without difficulty. For example, with an
power drivers (BD 736). Each time the
counter is advanced, the oscillator also
triggers the SN 74122N monostable flip-flop
365
6
O Al O A2 O A3 O A4 ID-
AS
O A6 O A7 O A8
DO S Dl D2 D3
D4
D5 D6 D7
4xSN 74I5IN
YA
TE
ifczz
BLANKING |
SN 7493N
+5V
4,7kn L/50 kfi
SN 74122N
Figure 21.
Multiplex operation of LED display units, control section
366
Figure 21. 7
Multiplex operation of LED display units, display section
367
which drives the blanking input of the sevensegment decoder. With the 50 kfl
potentiometer, the "On" time of the
luminescent diodes can be varied, in order to control their brightness without power
loss.
The design of such multiplex circuits will be explained by means of the following example:
mA A mean segment Ip = 15
is required.
Since a total of 8 units are driven one after
another, the peak current per segment is
If=Tf- 8= 15 mA.8= 120mA
The current in the segment driver is calculated from the formula:
0-7 V
1= "Re"
Thus the emitter resistance Rg is obtained
0-7 V 0-7 V Re
if 120mA
5-6 ft
The digit driver must be capable of
producing the current for the seven segments and the decimal point. Thus the digit current works out to:
mA mA I D =
F I
.
8
=
120
. 8 = 960
was possible, from the middle sixties, to solve the same problem with only three
integrated circuits: counter SN 7490AN, store SN 7475N, decoder and cold-cathode tube driver SN 74 14 IN. Thus an area of only about 15 cm was needed. A significant advance came with the development of the
visible light emitting diodes (VLEDs), which, together with the highly complex logic
circuits which are now available, make it
possible to combine the assembly described above into a 16-pin Dual-in-Line case. The
space requirement is now only about 2 cm .
At the same time, because of the smaller number of components and connecting leads, the reliability of the whole circuit has increased considerably. The latest development from Texas Instruments are the two highly-complex circuits TIL 306 and TIL 307, which each contain not only the counter, store, and seven-segment decoder/ driver, but also the display. The two circuits differ only in the arrangement of the decimal point. While, with the TIL 307, the decimal point is arranged to the right of the figure and with the TIL 306 it is on the left.
The circuit used in the counter has a number of special features which permit universal application of this device. The counter can
be operated either asynchronously or synchronously with serial carry or with
"carry look ahead". The simplified logical circuit is shown in Figure 21.8.
Therefore, a power transistor, which still has a small saturation voltage, even with a collector current of 1 A, is used as the
digit driver.
21.3 Numerical display units with integrated logic
Advances in semiconductor technology are making it possible to construct ever more complex assemblies in the smallest space. Around 1960, to build a counter decade with store, decoder and cold-cathode tube
driver, a circuit board of about 150 cm
area was needed. With integrated circuits it
As well as the "Clock" input, the counter has two "Enable" inputs: one for serial carry ("SCEI" = Serial Carry Enable Input) and one for parallel carry ("PCEI" = Parallel
Carry Enable Input). A signal at the clock
input can only take effect if these two inputs are on "Low". Care is to be taken, that the level at the enable inputs can only be changed when the clock input is "High",
since otherwise malfunctions can occur.
Further, the counter has a carry output
("MC" = Maximum Count Output), which becomes "Low" when the counter has
reached the figure 9 and if at the same time the serial carry input is "Low". This output
368
the flip-flops in the individual decades work
synchronously, while the carries are
produced serially. Although the maximum
possible counting frequency (approximately
18 MHz) is not affected by this, the response
time of the counter is. Since the propagation delay for each counter is approximately 1 2 ns, with a 4-digit counter it takes about 36 ns, until the clock signal reaches the most significant decade and about 80 ns
until this decade has also reached its new
state.
Figure 21.8 Carry circuit in TIL 306/307
delivers the necessary carry to the following
decades. Thus there are three possible ways of operating the counter: in asynchronous
operation, the output MC is connected to
the clock input of the following decade.
The enable inputs SCEI and PCEI are always "Low" {Figure 21.9). With this arrangement,
In synchronous operation, the enable inputs SCEI and PCEI are connected in each case to the carry output of the preceding decade.
The clock signal is fed in common to all
counter stages {Figure 21. 1 0). With this operating mode, all relevant flip-flops switch simultaneously with the positive-going edge of the clock signal. Thus a very short response
time is obtained. The maximum possible
counting frequency is, however, limited by the counter length. Before a new clock pulse can be processed, all carry operations must have been dealt with. But since this takes place serially, these signals must run through
LSD
~i_r -
"LOW" -
>
PCEI
n SCEI
MC
>
-- PCEI
SCEI
MC
-- PCEI
1
>-- SCEI
MSD
MC
>
MC
-- PCEI
SCEI
Figure 21.
Asynchronous counter operation with TIL 306/307
"LT -T
PCEI
-- "LOW 4-0 SCEI
MC 3 . c PCEI
t:
--U MC D ·--O PCEI
MC 3 » O PCEI
t: SCEI
MC
Figure 21. 1
Synchronous counter operation with TIL 306/307
369
LSD
IT
"LOW
>
SCEI
i--C PCBI
>
MC
SCEI
MC
r-<-
i
PCEI
>
SCEI
MC
PCEI
MSD
:*
SCEI PCEI
Figure 21.11 Synchronous operation with carry look ahead
all the circuits from the least-significant to the most-significant decade. The time needed for this naturally increases with the
counter length and determines the minimum permissible interval between two clock pulses, and thus the maximum counting
frequency.
In synchronous operation with carry look ahead {Figure 2 1.11), the clock signal is again fed to all decades in parallel, so that all flip-flops in the counter switch synchronously. The response time of the counter is thus again only about 33 ns and is independent of the counter length. The carry from the first counter is fed in parallel to all other counters through the PCEI inputs. The serial carry enable input SGEI of the second decade is always "Low", so that this
decade always switches, when the first
decade switches from 9 to 0. As with all subsequent decades, the carry output is connected to the carry input of the next decade. If the counter now reaches the position 9990, then the serial carries run through the whole counter, for which the time of 9 clock pulses is available, and thus preset the more significant decades. When
the counter finally reaches the position 9999, all stages.are released ("enabled") by the carry from the first decade. Thus the
critical propagation delay is now only
determined by the first decade; it is thus again independent of the counter length.
Figure 21.12 shows the complete block diagram of the TIL 306/307. As well as the inputs already mentioned, the counter also has an asynchronous reset input ("Clear").
A "Low" signal at this input sets all flip-
FCEI SCEI
Decoder * driver
Display
// //
« o oe
ec
Figure 21.12 Block circuit of the TIL 306/307
370
Outputs
A
QA
B
QB
C
QC
D
OD
DP.
ODP
Decoder + driver
Blanking
Lamp Test
Display
//
Figure 21.13 Block circuit of the TIL 308/309
flops to zero, irrespective of the signal at
the "SCEI", "PCEI" and Clock inputs. If a "Low" signal is applied to the "Strobe"
input, the existing counter state is transferred to the store, until the strobe input again
becomes "High". The stored number can then be taken off from the outputs QA,
QB, QC, QD.
QB, QC, QD and QDP. During a "Low"
signal at the "Strobe" input of the store, the latter accepts the data at the inputs A, B, C,
D and DP. A "High" signal at the "Blanking"
input switches off the display, while a
"Low" signal at the "Lamp Test" input
causes all diodes in the display section to
light up.
The decoder/driver substantially corresponds in its logical circuit with the SN 7446/47 AN,
so that is is possible, here too, to construct a circuit for zero blanking without additional
expense. The "Blanking" input is not, however, combined with the "Ripple-
Blanking" output, so that brightness control is simpler to carry out than in Figure 21.4. Finally, the driver for the decimal point has also been integrated, so that the latter can
be switched on or off with a standard TTL
signal.
The TIL 308/309 {Figure 21.13) each contains
a store, which, as well as the four input bits A, B, C, D, also stores the signal for the
decimal point. The outputs of the store are brought out through the connections QA,
21.4 Monolithic display elements
As well as the above-mentioned display devices, in which the segments consist of individual GaAs diodes or combinations of these, Texas Instruments also manufacture
monolithic display units, which, because of
their small dimensions, are particularly
intended for pocket calculators, digital
watches and similar applications. Numerous
circuits are available to drive these units,
which bear the type designation TILD 100.
Each circuit always contains several
Darlington amplifiers, which can deliver
mA output currents up to several 100
with
only small input currents.
371
:
V Vss = 6-5 . . . 8-8 O
VSS
--I
TMS 0100
ur similar
?
1J vSs
H>
i
l>
Kt
Output
SN 75493N
-A/
UDD
Rl-
r~
Seg. A
A-
T
Vcc =3-2...8-8 V v Cc ~~i
Output
Rl
I
1
T^ TILD 100
T^
II-
vss
H ^^-i
DH
H>
2> SN 75494N
vVdd
T~
I
vCc
DH
Figure 21.14
12-digit display unit for pocket calculator with the TILD 100 monolithic display devices
Figure 21.14 shows the circuit for a 12-digit display unit in pocket calculators, with which all circuits which have a multiplex sevensegment output and can deliver an output
current of 0-5 mA, such as the TMS 0100,
for example, can be used as the calculator
circuit. As the digit driver, two SN 7549N are used. The segment drivers SN 75493N
form a special feature. With these, the output current is not set by a simple series resistance; this would result in the segment current and thus the brightness being greatly .dependent on the working voltage. Instead, each output of this circuit has an additional reference input, which measures the output
current through an external resistance Rl
and regulates it to a constant value. In the circuit shown, the segment current should
be about 10 mA. The corresponding resistance Rl is then calculated from the
formula
0-7 V 0-7 V
« RL =
'see
=
10 mA
68 ft
With the circuit shown here, the supply
voltage \qq for the display unit can then fluctuate between 3-2 V and 8-8 V, without
variation of the brightness of the
luminescent diodes.
372
vcc- TTTTT
300 SI
J 300 Si
I 1
VcEat
12 3
4
5
6
7
X
Figure 21.15
Display unit TILD 100 driven with TTL circuits
373
But the TILD 100 display devices can also be used together with TTL circuits {Figure 21.15). As the multiplexer, the same circuit as in Figure 21. 7 is then used. As the digit driver, the demultiplexer SN 75145, which
is capable of delivering an output current of
80 mA, is used. As segment drivers, NPN transistors of type BC 183, which are driven
directly from the seven-segment decoder
SN 7448N, are used. The segment current is
calculated here from the formula:
VCC " vOL - VF - VcE:sat
I,seg
vcc - Vol - Vf - vbe
|
2kfi
Then, for a segment current of 10 mA:
5 V-0-5 V-l-6 V-0-3 V 10mA=-
Rl
5 V-0-5 V- 1-6 V- 0-7 V
2kft
2-6 V 10 mA -1.1 mA
Rl
n R L
=
2-6 V 8-9 mA
~
300
21.5
5 x 7-point matrix display units
The display units so far described were only capable of representing figures. By advances in semiconductor technology, it is now possible, to produce devices which are suitable for the display of any desired letters,
figures or symbols. In these, the desired character is formed from a matrix of 5 x 7 points {Figure 21.16).
In such a display unit (e.g. TIL 305), every point in a matrix is represented by a lightemitting diode. In each case, all cathodes of a line and all anodes of a column are connected together, as is shown in
Figure 21. 1 7.
Column D.P.
12
3
4
© © Q ® ® i>
©- :_c:_£:Lje_jr
>
2 @3®-
4 <4>-
5 ®6 ®^ ®-
C_iTLi
r r £T
r jt
1
f-
r.
r-j^ 1
1
d_c r f^-j^1
r r c j r: J--
r r r
X_
Figure 21.17 Arrangement of light-emitting diodes in display unit TIL 305
Figure 21.16
Example of the representation of the A letters to O in a 5 x 7 point matrix
374
With this circuit arrangement, of course, multiplex operation is again necessary for the drive. Figure 21.18 shows the complete circuit with which all letters, figures and a number of special characters can be displayed. The form of the display is shown in Figure 21.19.
An oscillator, constructed with the Schmitt trigger SN 741 3N (f = 50 kHz), {Figure 21.18), drives the counter SN 7493 AN, in
which only the first three flip-flops (outputs A, B, C) are used. Through the decoder/
driver SN 7445N, the cathodes of the luminescent diodes are now selected a line at
a time. At the same time, the associated column data is read out through connections
Jl, J2, J3 of the constant store TMS 250 INC and is fed to the display unit through 5 NPN transistors (e.g. type BC 183).
Selection of the total of 64 possible characters is made through the leads J4 to
J9 and CS1, the individual characters being coded in accordance with the UASCII code.
Construction of a multi-digit display using the circuit shown in Figure 21. 18 is not economical, since in this case a multiplex control, including the character generator, is
A necessary for each digit of the display.
more suitable circuit is described below for a
16-digit display {Figure 21.20). It consists of a store (16 characters of 6 bits each), comparator, clock generator, character
generator TMS 2501, a driver stage, the
decoder, character and line counters, a timer stage and the control and synchronisation.
The position, in which the character
corresponding to the data is to be written, is marked by the address. The strobe signal causes the synchroniser to deliver a write signal to the store. With a clock frequency
of 1 MHz and a 100 jus pulse from the timing
r
2O 3>
A BCD
T
vcc o-
Column 3
SN 7493N
hhhhh
01 02 03 04 05 06 07 08 09 010 TMS 2501NC
J4 J5 J6 J7 J8 J9
66 666 6
CS1 CS2
T
6
Figure 21.18 Drive circuit for a 5 x 7-point matrix, using the TIL 305
375
J4
1
J5
J6 "o
ri
1
_0_
1
fo"
)
1
]
1
1
]
1
~r IT ~0~
1
1
"6"
"
o"
1
1
1
1
~T~ ~T~
J7
J8 HP
pP IT
i
1
1
1
1
1
1
J9
(SI _1_
\
1
\
\
1
1
1
1
1
1
1
1
1
1
1
1
111111
J4 1
1
1
)
1
1
1
1
J5
1
1
1
]_
J6
o"
1
1
1
1
1
1
1
1
1
1
1
1
J7
_0
1
1^
1
1
1
1
J8 1
1
i J9 _0_
C SI
1
1
±
l'
1
1
1
1
1 1
1 1
1 1
1
1
1
1
1
1
1
1
1
1
1 1
J4 u5-
1
J5
J6 3-
J7
"o
51 1
1
1
1
H ~6"
o
JS
_0_
J9 1
1
1
1
1
1
(SI 1
1
1
1
1
II
1
1
IT
1
1
1-
o
1
T-1
1
1 1
CD l°j Cl] l°D ll
^ rc
r°^ ITj tr
m a m t tM,n
n_ i_i
i
i
m,
;
J
0_
1 1
1 1
1
1
rf
1
a a 1
1
k es s
iiiniiiiii
Figure 21.19 Code table and representation of characters
376
> Address > Strobe
Store, 16 x 6 bits
1
Display
16 x TIL 305+ line store (16x5 bits)
i
Rl-7
14-9
1 from 8 decoder + driver
.
Comparator
1
1 from 16
Line counter,
1
<
decoder
3 bits
A=B
Write
Synchroniser
Load Count
,
Character counter, 4 bits
,
Qa-Qd
Timing stage 100 ms
.
Inhibit A
Character generator + driver
.
Clock generator
1MHz
--
Figure 21.20 Block circuit of a multi-digit alphanumeric display with TIL 305
stage, the strobe signal must be present for at least 116 /us. in order t0 be able t0 be evaluated. The circuit is so designed, that the 7 lines of the 16 display elements controlled are switched on one after another, the associated column information is output from the character generator and temporarily stored in the drive.
During a clock cycle, the synchroniser delivers a write signal (only in conjunction with a strobe signal), a load and a count signal, one after another. With the load signal, a signal is given to the drive through the decoder and the column information corresponding to the particular character is stored. The counting pulse advances the character counter and the load process is repeated. After 16 counting pulses, the whole of the column information has been read out, the timing stage is triggered and the inhibit signal (100 Ms) switches on the relevant line of the display. Following this, the line counter is advanced and the process described is repeated for the next line. Because of the divider ratio of the line
counter, a mark-space ratio of 1 8 : is obtained for the display.
If a character is to be erased, this can only
take place during the read-out cycle. When
the input address corresponds with the address of the character counter, the
comparator gives the signal A = B to the
synchronisation. If the strobe signal is
present at the same time, the information for the new character, at the data inputs, is taken into the store with the write signal.
With the clear signal, the whole display can
be erased. This is done by writing the
character "Space", with which none of the
LEDs are lit, into all 16 store spaces
(Dx-6 =
1). The clear signal must
be applied throughout a complete read-out
cycle, that is, for a minimum of 116 Ms, in
order to overwrite the whole contents of the
store.
The whole circuit of this multi-digit alphanumeric display is as shown in Figures
21.21 to 21.23.
377
--
SN 7489N -D3
SI 2-- S2PS3f
\--CME
A Br D
/
tore
\
^ H>
51 D52 >-
S3
D >C >B >A >--
> Strobe
fc-
Compafavoi SN 7485N
1
04
(--C mi:
S4
VVI >-
A B< U
A>B
A=B
A=B
A<B
|--C K -A-c s:
A>
l>
Clock jwnctaiw
1/2 SN 74I20N
K
330 II
2-2 «F
Figure 21.21 Store, Comparator, Synchroniser and Clock Generator
378
-- I4
-- is -- 16 -- "7
---18
-- lo
-«-QB
tr^^^>
l^y^y
J
Figure 21.22 Character generator, line and column drive
379
Rl
Rl
S3
R2
R3
R3
R4
R4
R5
R5
R6
R6
R7
. R7
CI
1Q ID
3
20 3Q 4Q SN 74175N
2D 3D 4D
5Q 5D
i ka
+5 V
OS
Figure 21.23 Drive for the TIL 305
380
22 Direction-dependent
photocell units
22.1 22.2 22.3 22.4
Principle of operation Direction-dependent counter Direction-dependent optoelectronic couplers Digital rotary knob
381
1
1
22 Direction-dependent photocell units
Direction-dependent photocell units make it possible to recognise the direction of motion of moving or rotating objects. With these
photocell units, forward-or-reverse counters can then be used to evaluate the signals. Their applications include rotational speed measurement with detection of direction, counting people with detection of direction,
counting of goods on conveyor belts, frame counters in film cutting machines, tape counters in audio tape recorders, measurement of angles with the aid of incremental angular
position transmitters, etc.
22.1 Principle of operation
Direction-dependent photocell units contain
two individual optoelectronic couplers. The
photodetectors Tl and T2 are arranged as shown in Figure 22.1, staggered with respect
to one another. During a switching operation, the photodetectors Tl and T2 operate in sequence, but overlapping in time. Thus it is possible for the direction to be recognised by the following logic. Figure 22. illustrates the correct sequence of switching of the staggered photodetectors Tl and T2. Depending on the application, the photodetectors are operational either when irradiated or when the irradiation is
interrupted. A direction-dependent photocell
unit for short ranges (up to 10 cm), needs
a GaAs diode with built-in lens, e.g. types TIL 31 or TIL 24, as the radiation source and two silicon phototransistors with built-in lenses, e.g. types LS 400, LS 600 or TIL 81, as the detectors.
Segments
Figure 22.
Correct switching sequence of the staggered phototransistors
383
Forward travel Phototransistor Tl
Pliot otransist or T2
Reverse travel Phototransistor Tl
Phototransistor T2
22.2 Direction-dependent counter
Figure 22.2 shows an optoelectronic counter, which can take the place of mechanical counters, for example in tape
A recorders. separate tape roller has
transparent windows. The radiation from the GaAs diodes TIL 24 can only fall on the phototransistors LS 4022 through the transparent windows. Depending on the spacing of the windows, the tape length can be counted in centimetres or other units. Depending on the direction of tape travel, the phototransistors Tl and T2 are turned on in sequence and overlapping in time. During forward running, the phototransistor
Tl operates earlier than the phototransistor T2, and conversely, during reverse running
T2 operates before Tl. The output signals from the two phototransistors are converted
by the Schmitt triggers U 1 A and U IB into
TTL-compatible signals.
The pulse diagram in Figure 22. 3 shows the
U signal at the output 1A leading the signal at output U 1 B during forward travel and U the signal at the gate output 1 B leading the signal at the gate output U 1A during
reverse travel. The sianals from the two
Schmitt triggers are differentiated in each case. The negative-going pulses are inverted
by U 2A and U 2B respectively and drive the NAND gates U 2B and U 2D. During forward travel, the gate U 2B is turned on by the signal from U IB, so that the up-down
counter is driven by the forward counting
pulses. The gate U 2D remains off through U the low level signal at 1A. During reverse travel, the gate U 2D is unblocked by the U high level signal at the output 1 B, so that
the reverse counting pulses drive the
reversible counter. The SN 74F92 counts in the 8421 BCD code and drives the optical display TIL 308. On the SN 74192, the
unused "Load" input is connected through 1 kftto +5 V. The "Carry" output, for
cascading the forward counting pulses, drives the forward-counting input of the next decade. The "Borrow" output, to cascade the reverse-counting pulses, drives the reverse-counting input of the next decade.
The circuit in Figure 22.2 only works satisfactorily if the phototransistors Tl and T2 are turned on and off consistently on every counting operation. If, for example, the phototransistors are first only turned on in the sequence Tl, T2, corresponding to the
220pF
SN7413N ~]X^
o
1-N.
U2B
/_/
TS A B C D
QA QB QC QD
>U,, >Down
Carry Borrow
Load
Clear
qj
X 220pF
L-L/
U2D
+5 v "N\\
Figure 22.2 Circuit of a direction-dependent counter
384
v,» V IB
V-,*
V2B V2D
1 1
|
1 1
_l
\ Forward
L
1 1
1 |
1
..
/
Backward
Countin 1 erru r on eversal (R verse coun ting Dulse only).
1
--|
V
Figure 22. Pulse diagram of the direction-dependent counter
forward direction, and following this they are turned off in the reverse direction, then although the original tape-length condition has been restored, a reverse counting pulse is obtained. In the counting of tape lengths, this case, which occurs infrequently, can be
disregarded.
22.3 Direction-dependent optoelectronic coupler
The circuit in Figure 22. 4 no longer has this
defect. It is therefore suitable for use in incremental angular position transmitters.
The phototransistors are in a staggered
n iokn iokn 1A
U IB
D
Q
U3A
U2A I Forward T_r
D
Q
U3B
t>
Q
V ID
U1E
J-N
^---ju
U U2B
HH
330 a
47nF
Figure 22.
Improved evaluation circuit for direction-dependent optoelectronic couplers
385
arrangement, for example, as in Figure 22. 1. As amplifiers and to produce the necessary fast pulse edges to drive the subsequent
TTL circuits, each phototransistor is followed by a Schmitt trigger (U 1A and U 1C). The Tl -signal is inverted again by the inverter U IB and is connected to one input of each of the gates U 2A and U 2B.
The T2-signal is divided by two subsequent
U flip-flops 3A, U 3B into time-delayed
signals. Through the appropriate logical
Q relationship of the flip-flop output signals Q and with the Tl-signal, the negative-going
forward-counting pulses are obtained at the
output of the gate U 2A and the negative-
going reverse counting pulses at the output
U of the gate 2B. The flip-flops receive their
clock signal from the oscillator, consisting of
the two Schmitt triggers U ID, E, with
* f 25 kHz.
When the transistor T2 is turned on, the
U output of the Schmitt trigger 1C is at a
high level. With the next clock pulse, the
U flip-flop 3A is set, and the following pulse sets the flip-flop U 3B. When the photo-
transistor T2 is turned off, the flip-flops ar,,*
reset in the same order. Thus two states of the two flip-flops, which characterise the turning on and off respectively of the
phototransistor, are obtained:
Turn-on = Qu3A · QU3B
Turn-off = Qtj3a · QlJ3B
Since the output gates U 2A, B only allow
signals to pass, if the phototransistor Tl is
turned off, the following relationships apply:
Forward = Tl · Qu3A · QU3B Reverse = Tl · Qtj3A · QU3B
Accordingly, a pulse, the width of which corresponds to the cycle time of the oscillator (in this case 40 /is), is obtained in each case at the output of the gate. This arrangement ensures, that no false signals are given on reversal of direction: if the reversal takes place from forward to reverse, when Tl has not yet turned on, first a forward signal is given and then a reverse signal after the reversal of direction, so that the total in the subsequent counter is unchanged.
|t
iffi
r
urn
ISOpFl
-- --* J
\ Ul: SN 74132N U2: SN49713N U3/U4: TIL 138
v\c
r
220pF
HI-
Figure 22.5 Digital rotary control knob
386
22.4
Digital rotary knob
A further application for direction-dependent
optoelectronic couplers is the "digital control knob", which is used together with a counter to generate digital set-point values. In contrast to decade switches, the digital
control knob has the advantage that the numerical range which is of interest can be traversed continuously and that there are no steps on carrying from one decade to the next, which can never be completely avoided with the usual decade switches.
The heart of this digital control knob, in
Figure 22.5, is the oscillator U 2B, the
frequency of which can be varied over a wide range by means of the potentiometer P. Because the start and end of the potentiometer track are connected together, the oscillator generates the lowest frequency when the potentiometer is in the central
position. A disc with a cut-out is mounted
on its spindle, so that when the slider is in
the central position, both phototransistors are turned on and thus both outputs of the
U Schmitt triggers 1A, B are at the logic high level. Through the gate U 1C, the
oscillator is thus switched off. If the
potentiometer is now turned in one or the
other direction, first one of the two phototransistors is covered. Through the
feedback between the gates U 1 A and U ] B,
this "Latch" adopts a stable position and
stays in this condition, even when the second phototransistor is covered. Now, the further the potentiometer is turned from the
centre position, the higher the frequency of the oscillator becomes and the faster the counter advances. Through the gate
U 2A, the position of the flip-flops U 1A, B
is transmitted to the counter and thus the
required counting direction is set. The RC network at the input of the gate U 2A delays
the negative edge at the output by about 0-5 jUs and thus prevents an error in the counting direction, if the potentiometer is turned back to the central position.
387
23
Optoelectronic rangefinder
23.1 23.2
Phase measurement as a measuring principle Practical circuit of the rangefinder
389
1
1
23 Optoelectronc rangefindei
The optoelectronic rangefinder described here was originally developed for film
cameras. A connected control circuit then
had to set the appropriate focal distance automatically for the measured distance of the subject. For the circuits described below, after appropriate modifications, there are
numerous other possible applications. To
mention one further application, for example, a highly-sensitive proximity switch in alarm systems. This equipment can equally well be used as a highly-accurate level detector in silos or tanks, while it should be emphasised, that the measurement takes place without contact with the contents.
23.1
Phase measurement as a measuring principle
This short-range rangefinder, the principle of which is illustrated in Figure 23. , is suitable for measuring the distance from fixed and moving objects in the range from
m 1 to 15 m. The phase-measurement
principle is applied. The emitter delivers
modulated infra-red radiation, which is optically focussed onto the object. The radiation is modulated with a crystalcontrolled oscillator with an accurate frequency of 4433 kHz (this frequency was chosen, because low-priced crystals are available). As the IR source, a GaAs luminescence diode is used.
In the IR receiver, which forms part of a fixed assembly with the source, the IR radiation is first demodulated, to recover
the modulation frequency. A fast-acting Si
photodiode serves as the photodetector. According to the propagation time of the IR beam, the distance travelled by which is equal to twice the distance r to the object, the received signal shows a lagging phase displacement i£. Since, through the relationship with the velocity of light, c, i/> is proportional to r, the measurement of the distance r can be obtained directly from the measurement of ip.
-- For the propagation time: t = 2
c
(23.1)
IR radiation
Signal 1
Moving object
IR transmitter
/S^ Direct
v IR radiation
Mirror
1
IR receiver
Beat -frequency oscillator
Signal 2
1
Discriminator circuit
indicator unit
Figure 23. Construction principle of an optoelectronic rangefinder
391
f
and for the phase displacement:
V= mi . t 4m f .
Therefore: y
(23.2) (23.3)
At a frequency of 4-433 MHz and a distance
of 10 m, the phase displacement obtained is:
47T-
10 m. 4-433.
6
10
s
l
3
.
8
10
-1
ms
0-5911 . 7T = <£ = 106
(23.4)
Because of the circuit selected, phase
measurement is possible up to an angle
ifi = it (= 180°). At a frequency of 4,433 MHz, the maximum measurable
distance r r
is then
<f . c
rmax ~'~
47T.f
m = 16-9
IT. 3 . 10° ms
47T.
4-433
.
10 6
_1 s
(23.5)
The measured result is completely independent of the signal amplitude. The signal received by the photodetector and then demodulated is amplified, so that phase comparison with the modulation signal is possible. At the same time, exact measurement of the phase displacement is only possible, if the stability of the circuit
fulfils strict requirements.
By using the method, which is often used in phase meters, of superimposing an auxiliary frequency from a second oscillator, the received electrical signal is transposed to a lower frequency While retaining the same phase displacement if. For reasons of
stability, this beat-frequency oscillator is also a quartz crystal oscillator. Its frequency
is 80 Hz higher than that of the transmitter.
A difference frequency of 80 Hz is then
formed in the mixer stage. Following the mixer stage, this low frequency is amplified further with relatively simple low-frequency amplifiers and is freed of interfering noise
components with non-critical and economical
RC filters. To improve the interference
rejection further, the signal is limited in its amplitude in subsequent limiter stages. This is permissible, because the distance information is present in the time-displace-
ment of the zero crossing point voltage of the signal and not in its amplitude. Simple and cheap transistor circuits are adequate for limiting, while, because of the low
frequency used, the transistor storage time does not yet have an interfering effect.
Furthermore, through the frequency
conversion from 4-433 MHz to 80 Hz, the A system accuracy is significantly increased.
phase difference which is still detectable, of Asp = ±1/4° at 80 Hz corresponds, from equation (23.2), to a time displacement of
Aip
At = 360°
.
1
At = ±-
- «= ±8-6 jus
4 . 360 . 80 s"
(23.6)
Since the phase displacement is carried over in the frequency transposition, it corresponds to a change in the propagation time of the high-frequency signal of
At = ±
1
« r-3i ±138-8 ps
4 . 360 . 5 . 10° s '
(23.7)
According to equation (23.1), this change in propagation time is equal to a range tolerance of
138-8. 10 12 s-. 3 .10 8 ms ' r=±
±2. 10 i m
(23.8)
In order that the phase comparison can be carried out with at least the same accuracy, a reference signal with good time stability must be available.
392
2
Figure 23. 2 shows the block diagram of the rangefinder. Through the circuit selected, a reference signal of high stability is produced.
Two IR receivers of identical construction
are used. Each receiver has a photodetector at its input. The IR radiation from the transmitter is directed through an optical lens system on to the object to be measured and is fed, as the reflected component, delayed by the propagation time, to the IR receiver 1. In front of the photodetector, this also has a lens system with a narrowly focussed radiation
A characteristic, directed at the target.
small part of the transmitter radiation is taken off before or after the transmitter lens system and fed by a short, direct path to the IR receiver 2. This can be done by deflection
of the beam at the inner wall of the equipment housing or by a small reflecting
This method with two IR receivers
eliminates the effect of interfering phase shifts in the stages of the transmitter, since the paths of the signals for receivers 1 and 2 pass through identically-constructed stages and have the same phase shift.
The signal photocurrent from the photodiode and the original current from the beat-
frequency oscillator meet at the HF bandpass and pass together through the HF
amplifier to the mixer stage, in which the
product of mixing -- the difference frequency - is produced. Since the two
frequencies have little difference between them, the effect of phase shift on both signals is practically equal, so that the lowfrequency signal is formed without a recognisable phase error. Since the following
LF amplifiers are of similar construction, residual phase displacements, which may occur in the RC band-passes and in the LF
amplifier, cancel each other out between signals 1 and 2.
The last assembly in the rangefinder is the
discriminator circuit. This has the function
of an exclusive-OR gate. As shown in the
truth table
Signal 1
Signal 2
Output
L
L
L
L
H
H
H
H
L
H
L
H
HiiHodeleclj
M « u7u2
Bj ml paii
^
Compiled NF jinplitici
t>
Mi,.,
KC ba mi -pas
%/ --
35
1
cTM,,,,i «.l,,f.
1
ainirul vi.luge amplifier
<
Li amplifier
O
o^ r ,,
M
Limiier amplifier
YL
Lmuci
\L
L ,,,,«,,,,, de
JK
M
1IKcin.uK
%
HF power
<
Trj ,,,,,,,«,,,,,,,,
^
teat -frequent: oscillator U-rystdl)
%
Dnelac 1R-
W
Strahlung
HF-Bandpah
HF-VersUrkef
Mitcher
RC-Bandpab
SB
NF-Verstarkcr
r>
- VL \L
Figure 23. Block diagram of the rangefinder
H< M
393
43
n
signal
Figure 23. Phase comparison of the two received signals
a high level signal appears at the output, whenever the signals 1 and 2 are different
from one another. Figure 23. 3 shows two signals 1 and 2, limited to square-wave form,
while signal 1 has a phase delay in relation
to signal 2. The output signal shown below has a pulse width which is equal to the time displacement between signals 1 and 2. Both rising and falling edges are used to form the
output signal. For this reason, the maximum
phase displacement which can be evaluated is 180°. The relationship between the output pulse width and the distance is linear.
An appropriately calibrated moving coil
instrument can indicate the distance directly
in metres. By integration with an RC
network, an analogue voltage, the value of which is a measure of the distance, is obtained. For control purposes, the output signal can be fed directly to a comparator stage, in order to control a servo-motor.
23.2 Practical circuit of the rangefinder
In Figure 23. 4, the IR transmitter circuit is
shown, with its design values. A frequency-
stabilised quartz crystal oscillator drives the
modulator stage T16 through the emitter follower T15. The parallel-tuned circuit in the collector lead of T16 is tuned to the oscillator frequency of 4-433 MHz. The
82 a 7 6/30 " *
R2
r
[Ji-Skn
Dr
4> " j47nF " 7nF
X T 3 '3nF
j47nF
.
Dr4
3-3nF
""·
S-- --o Bias current
'-ffso
T 1
DJZll
6-8nF
1 ±-
i tixli2
Figure 23.
IR transmitter circuit of the optoelectronic rangefinder
394
6
fy From Q-- *-
LS900I
i] «F2«
T1XLM CM
/*7*N
s, v -
1U4» U22knUiSk!ij!2kn UlSkn jv/kN rV
T T
I
Tf
2:5 V T3
-*--
BF224
.
UiknLJ»7kni.
LI:
TT
Control voltige
iwi (HBOI o
n «»n R14
in »i2f] cl = ,
CJ1
IW U2 2k«j2 2k n UlOk njlk n
T
Tr
9c»
R*
--Ufe--J_ 1MB 2-2 |iF
°i
0-1 ill
-w-
8
iU2l2kn
4--1
HiLl2 2kn
I
Measurement point »nd working point.
Figure 23.5
IR receiver of the optoelectronic rangefinder
working point of the modulator stage is set
with the potentiometer R5. The GaAs diode is connected in series with the coil LI of the tuned circuit. Thus the modulation current
(coil current) is higher than the total current
in the leads by the Q factor of the tuned
circuit. GaAs diodes have a low breakdown voltage. Therefore the TIXL 12 is protected against excess negative potentials by a IN 914 silicon diode, connected in opposing
parallel. The GaAs diode can be operated
with or without bias current, as desired.
Direct and indirect cross-talk from the transmitter to the receiver is prevented by various measures. The transmitter is well screened to prevent electromagnetic pick-up. In the same way, the main receiver, the references receiver and the auxiliary oscillator are each well screened as individual
R58 27k a
"
OljiF
4-
]27pF
-- ^~ 3> 5
^li/BpF
^150 pF
'Ml2kft
Q2. f= 4433080 H/
Figure 23. Circuit of the auxiliary oscillator
395
-- -- 1 J »- To main receiver -- (»~*i m* To reference receiver
-o J.
modules. The supply leads inside and
outside the modules contain effective HF
filters. In general, all leads are kept short and some are screened. Each module has only one ground connection. Thus undesirable earth loops can be simply avoided throughout the system.
Figure 23.5 shows the receiver circuit for the main and reference receivers. The difference between the two receivers lies in the fact that the main receiver contains a highly sensitive Si photo-avalanche diode and the reference receiver contains a Si photodiode with low dark current and a high limiting frequency. The reverse voltage of approximately 165 - 170 V, applied to the photo-avalanche diode, must be adjusted to the most favourable signal/noise ratio. In
addition, the reverse voltage is stabilised, at least with an accuracy of 0-1%.
The load impedance of the photodiodes in each case is a parallel-tuned circuit, tuned to the modulation frequency. The input stage
of the subsequent HF amplifier is an FET
source-follower. With this, the damping of the tuned circuit is still relatively slight. The
working point of the FET is determined by
the control voltage, according to the received signal. For this, the positive half-waves of
the LF output signal are amplified by T8. From the T8 output signal, the control
voltage is obtained through a peak-path rectifier. The time constant of the subsequent control voltage filter chain is so designed, that large changes in the reception level, occurring in relatively rapid succession, are well compensated for by the control
voltage. Otherwise, the mixer stage T4 will be overloaded. The transistor T4 in the
mixer stage is the low-flicker-noise type
BC 384. The LF amplifier contains a band-
pass tuned to the mixer output of 80 Hz.
Figure 23. 6 shows the auxiliary oscillator circuit. The phase angle can be controlled with the 100 fi potentiometers.
Finally, in Figure 23. 7, the evaluation
section is shown. The LF output signals
riR32 IJ56kS2
Limiter Amplifier I
MR34 IJ22kn
f]R36 (| R37 M4-7k!2 |j4-7k 12
LF input from
receiver I
-T
C26 4-7 uF
ta
T8
T9
BCW64 BCW64
n^
T10 BCW83
R39 10k
nifl]
R38
10k n
v!v ...
t|1
R30
R33
39k a 47k n
I
C27 B
10|iF
C28 »
22uF
LF input from
receiver II
-f
i
Umiter amplifier II (circuit as above)
Figure 23. 7
Evaluation section of the optoelectronic rangefinde'
396
!220uF Exclusive-OR gate
from the main receiver and the reference receiver are formed into symmetrical squarewave pulses in the relevant limiter amplifiers. The operating voltage is stabilised with a value of 5 V. The limiter amplifiers are
TTL-compatible, because a TTL circuit,
connected as an exclusive -OR gate can carry out the phase comparison. The measured distance is indicated by a
moving-coil instrument.
397
24 Data transmission with optocouplers
24.1 24.2
24.2.1 24.2.2 24.2.3 24.2.3.1 24.2.3.2 24.3 24.4
24.5 24.6 24.7
Interference on transmission links Construction and characteristics of
optoelectronic couplers Current transfer ratio Mechanical construction
Dynamic performance Photodiode operation
Phototransistor operation Simple transmission links
Improvement of the switching performance of optocouplers Optocouplers in photodiode operation Duplex operation with optocouplers
Common-mode suppression of optocouplers
399
24 Data transmission with optocouplers
24.1
Interference on transmission links
The transmission of data in a digital system over long distances proves to be very difficult, if low error rates are to be achieved. The reasons for this are interference voltages, which are induced in the transmission line and introduces errors to the useful signal.
The simplest forms of interference to deal with are those which originate from adjacent signal leads. The interference voltage produced through cross-talk is about 15% of
the signal voltage, if twisted pairs are used.
Even with a large number of pairs of leads in
the cable harness, cross-talk increases to a
maximum of 20%, because only lines in the
immediate vicinity of the affected line determine the cross-talk level. Since the interference rejection of a logic system is about 30-45% of the signal amplitude, no problems are to be expected on this score.
Interference, which originates from other sources, such as relays, motors, etc. is far more difficult to deal with. In this case, the interference voltages reach amplitudes of up to several kilovolts. If it is assumed, that these interference voltages are induced from a cable running parallel to the signal line, a cross-talk level of about 15% can again be expected. These voltages are then large enough in any case, to interfere with the data transmission, even with signal voltages
V of 1 2 or more. Further, it must be noted,
that the cross-coupled interference voltages
are now present on the relatively low line
impedance (approximately 100 £1), so that the interference energies can very easily destroy the transmitter or receiver circuits, if special protective measures are not taken. Since undisturbed data transmission is impossible under these circumstances, the signal leads should always be screened, in order to keep the effects of interference sources, external to the system, as small as possible.
Different earth potentials between transmitter and receiver are the third source of interference. As a basic rule, a balanced transmission line would help guard against this (push-pull output stage in the transmitter and differential amplifier in the receiver input). Of course, the available integrated circuits for symmetrical operation are only capable of suppressing common-
mode voltages up to about 15 V. In
practice, however, considerably higher potential differences occur between transmitter and receiver, so that these circuits can only be used to a limited extent.
The only possibility, of safely suppressing
interference of this kind, consists of electrical isolation of the transmitter from the
receiver.
For this, optoelectronic couplers are
excellently suitable. On the one hand, they
are capable of isolating potential differences
up to several 1000 V between the input and
output. Of course, the practicable value is only a few hundred volts, since because of the small spacings between conductor tracks on the printed circuit, higher voltages cannot be permitted in most cases, unless special
constructional forms such as the coupler
TIL 109 are used. On the other hand,
optoelectronic couplers can be driven directly from integrated circuits and can themselves drive integrated circuits directly, so that relatively simple and low-priced interface circuits are produced.
24.2 Construction and characteristics of optoelectronic couplers
24.2.1 Current transfer ratio
Optoelectronic couplers, or optocouplers for
401
O
1
O 1
i
o
o
O 1
i
O
X y.
'
o
I
o
1
J
L
_o
O
1
3 T^ : o
i-
-o
Figure 24.1 Circuits of optoelectronic couplers
short, contain a luminescent diode as the radiation-emitting part and a photodiode or phototransistor (Darlington transistor) as the radiation detector {Figure 24.1).
Optocouplers, which only contain a photodiode as the detector, are not commonly used. For application, in which this
operating mode is necessary, couplers in
which the base connection of the phototransistor is brought out are generally used, and the collector-base junction is used as a photodiode. The emitter of the transistor is then unconnected.
The current transfer ratio IfAp is of the
order of approximately 0-001, i.e. for an
mA input current of 10
(luminescent diode),
about 10 //A of output current is obtained
from the photodiode or photocell (this is a
typical value). The actual current transfer or
conversion ratio is taken from the relevant
data sheet.
The phototransistor amplifies the photodiode current according to the current gain B(hp£) of the transistor. The current gain of the transistors is about 50-500, so that in this operating mode a current transfer ratio of about 0-1-0-5, corresponding to 10-50%, is obtained. Darlington phototransistors, with current gains of about 10 , give a current transfer ratio of from ·
1 to 10.
24.2.2 Mechanical construction
Figure 24.2 shows, in simplified form, the mechanical construction and the resultant parasitic coupling capacitance between input (luminescent diode) and output (photo-
transistor). Because of the method of
construction, high insulation resistances, of
11
typically 10 fi between input and output, can be achieved, so that in practice the actual
V^7 r^¥\ o_^=
Figure 24.2 Construction of optoelectronic couplers in the Dual-in-Line package
402
3
insulation resistance is primarily determined by the circuit board material and the surface contamination which inevitably occurs in service.
However, the coupling capacitances between the luminescent diode and the phototransistor are more likely to limit performance in normal operation. Although the capacitance, depending on the coupler type, is only between 0-5 and 2 pF and is thus considerably less than the coupling capacitance of other isolating devices (line transformers, etc.), under extreme conditions it can still have a disturbing effect. In this respect, the capacitance between the anode connection of the luminescent diode and the base connection of the phototransistor is particularly critical, while the effect is increased still further by the capacitance due to the conductor tracks on the printed
circuit.
The base connection is the most sensitive part of the coupler, since interference currents coupled in here reappear at the output, amplified by the current gain factor.
If the greatest possible interference suppression is required, it is therefore advisable to use couplers, in which the base connection of the phototransistor is not brought out.
Under these conditions, photodiode operation with a subsequent highly sensitive amplifier is not advisable, since in this case even small interference voltages between the input and output of the optocoupler can interfere with the following circuit.
measured. In practice, the subsequent circuit reacts on the optocoupler, so that in this case the data sheet values can only be used with reservations to determine the switching performance of the circuit.
24.2.3.1 Photodiode operation
The switching performance of the optocoupler in this operating mode is generally
Figure 24. Measuring circuit for the dynamic performance of optocouplers (photodiode opera-
tion}
measured with a load resistance Rl = 1 k£2
(Figure 24.3). The delay times are then determined by the following parameters:
The rise time of the radiation from the
luminescent diode.
The junction capacitance of the photodiode (it should be noted that the capacitance of the diode depends on the
applied reverse voltage).
24.2.3
Dynamic performance
The switching data given in the optocoupler data sheets must always be taken into consideration in connection with the relevant measurement circuits. The latter are selected, so that they have little or no influence on the coupler, and so that in fact only the performance of the coupler itself is
The value of the load resistance, which,
together with the capacitance of the diode, detemrines the time constant of the output circuit.
Since photodiode operation will only be used, if short switching times are required, significantly higher load resistances will not be used in practice, in order to keep the effect of the photodiode capacitance and the
403
capacitances of the subsequent circuit as small as possible. Thus, in this case, the speed of the subsequent circuit can be determined directly from the data-sheet values. For the TIL 103 optocoupler, for example, rise and fall times for the diode current of tr , tf = 150 ns are stated. Thus
the duration of a cycle at the maximum input frequency is T = 2 . tr , tf = 2 . 150 ns, which corresponds to a frequency of about 3 MHz,
a value which can be achieved without
difficulty in practice.
In practice, however, load resistances of several kilohms are mostly used, as a result of which the voltage gain of the circuit is 10 to 50 times greater than in the measuring circuit. Secondly, the transistor is turned on as far as saturation; but wi^h these small collector-base voltages, the capacitance of the photodiode (= collector-base capacitance) rises further by a factor of 3 to 4, which further increases the rise arid fall times
(Figure 24.5). This explains why trans-
mission frequencies of only about 5 to
10 kHz are achieved with simple circuits.
24.2.3.2 Pltototransistoi operation
Figure 24.4 shows the measurement circuit to determine the switching times in phototransistor operation. Basically, the statements
Figure 24.4 Measuring circuit for the dynamic performance ofoptocouplers (phototransistor opera-
tion)
made in the previous section also apply here,
since the phototransistor can be visualised as
a combination of a photodiode and an NPN
transistor. In addition, however, the following
points must be noted. The photodiode
capacitance is in parallel with the collector-
base junction and thus acts as a Miller
capacitance. With the collector voltage
> V Vjj 10 used in the measuring circuits,
firstly this capacitance is relatively small and
secondly the voltage gain of the measuring
Rl circuit, with a load resistance
= 100 £2,
is only small. The effective Miller capacitance
CM is calculated from the formula:
CM = CCB (Vu + 1)
Figure 24.5 Collector-base capacitance of an optocoupler as a function of the collector-base voltage.
24.3 Simple transmission links
Optoelectronic couplers are easily interfaced
with TTL. i.e. they can be driven directly
from TTL circuits and are in turn capable of driving TTL circuits directly. Figure 24.6
shows two circuits, which are adequate for many applications.
To determine the output current of the gate and thus the current in the luminescent
diode, in Figure 24. 6, the internal circuit of the gate must be used. Figure 24. 7 shows the equivalent circuit, in which the
resistances Rli and Rl2 represent the line
resistances.
404
6
R \>
RV
R {>
?)-- IT
i
i
fr- R
Figure 24. Simple transmission links
"1 v Cc
1
T
° iV
'
nM
Ri
rMii3on|
3fc>> !
SN7404N |
1 I
^--r
RL2
,
,
r
1
1
W VF
!
l
I
1
9
Oplocoupler
Figure 24. 7
Circuit for determination of the output
current
The current is calculated from the formula:
VCC " VcEsatTl " VD - VF
IF
Ri+Rli+Rl2
mm If a twisted pair, each of 0-4
diameter,
is used as the line, then with a line length of
Ry 100 m, the resistance
= Rl2 = 14 f2.
Thus the input current Ip of the opto-
coupler is:
_5 V-0-3 V-0-7 V-l-2 V
if
130 £2+ 14fi+ 14 ft
17-7 mA
For type TIL 117, with a minimum current
transfer ratio of 50%, an output current
mA lOL = 9
is then obtained.
In order to take account of the process variations which affect the components
within the SN 7404N, of fluctuations in
working voltage and of ageing of the optocoupler, the calculation is continued with only half this value (Iol = 4-5 mA). The
R resistance in Figure 24. 6a is selected to be
as low as possible, in order to achieve short switching times. Thus:
R=
VCC
I OL" I IL7414
= 1-5 kfi
5V 4-5mA-l-2mA
The component values for the circuit shown
in Figure 24.6b can be calculated accordingly.
405
-^<
l£
[V
Ccb
M
n
Figure 24.8 Equivalent circuit for the time-constant components in the optocoupler
24.4 Improvement of the switching performance of optocouplers
The time which primarily determines the
maximum possible transmission frequency,
is the turn-off time of the phototransistor.
Cm While the Miller capacitance
= Cqb
(V + 1) is charged up relatively rapidly
during turn-on by the photodiode current
Ip (Figure 24.8), when the transistor is
turned off, this capacitance must be
discharged through the high-resistance base-
emitter junction, which can take up to about
100 jUs unless additional measures are
taken. If, on the other hand, a resistance Rg
is connected in parallel with the base-emitter
junction, then an additional current, which
discharges the Miller capacitance, is delivered
from this path. Naturally, the lower the value
of this resistance, the more it shortens the
switching time.
60% of the photocurrent will be diverted to
ground and thus the current transfer ratio will be reduced by the same percentage. Against this, there is a reducation of the turn-off time by about 50%, which often makes up for this disadvantage.
It is true that this very simple circuitry improves the switching performance of the
phototransistor, but the actual cause - the Miller capacitance - is not affected. This is
only possible, if the voltage gain of the
A circuit is reduced. low-value load resistance Rq (approximately 100 £2) would
however, also reduce the output amplitude
to the same extent. A circuit which avoids
these disadvantages is the cascode circuit,
well known from high-frequency work,
although it was used there for other reasons.
In this case, the phototransistor works into the low input irnpendance (approximately 20-100 fi) of the base-connected transistor Tl, the working point of which is set by the diode D. Thus the voltage gain of the phototransistor falls to very low values (1 to 10). At the same time, its collector-base voltage
V is prevented from falling below 2 and thus
causing an excessive rise in the collector-base capacitance (see Figure 24.5), which also improves the switching performance. This
operating mode of the phototransistor is
very similar to the measurement conditions
stated in the data sheets, so that a direct
comparison can be made from these values
on the performance of the circuit in Figure 24.9.
The subsequent transistor T2 provides an adequate output signal so that TTL circuits
can be driven directly. Since this transistor ensures short storage times and short rise and fall times, even in saturated operation, a Schmitt trigger is unnecessary in most cases.
On the other hand, it also diverts a part of
the photocurrent Ip and thus reduces the
current transfer ratio of the coupler. With a
mV base-emitter voltage VgE = 600
on the
phototransistor, a photocurrent Ip = 10 to
20 A jU and a resistance Rg = 1 00 kfi, up to
The mode of operation of the circuit in
Figure 24. 1 corresponds to that in Figure
24.9. By using the SN 75450N interface circuit, however, the number of discrete components is considerably reduced. The
circuits in Figure 24. 9 and 24. 1 permit
406
n 1 5-6i
Nun
Inpul
}
I
I
TIL HI ur similar
--
-o Output
Sf-®-
Figure 24.9 Cascode circuit to improve the switching performance of optocouplers
m
0+5 V
ayH SN 75450N
I
Input
1
O--
> '
i
i
TIL III or similar
Figure 24.10
Cascode circuit with the SN 75450N interface circuit
407
to
-o Output
maximum transmission frequencies of 100-300 kHz and are thus about 10 times
faster than the circuits in Figure 24.8.
On the other hand, the circuit in Figure 24.11 permits transmission frequencies of up to about 100 kHz (typically 50 kHz). Here,
the phototransistor works into the relatively low input resistance of an emitter-connected transistor (approximately 1 to 5 k£l). Because of the input characteristic of the amplifier, the collector-base capacitance of the phototransistor is again prevented from becoming excesssively large due to the small
collector-base voltages. An additional
resistance between base and emitter allows the Miller capacitance to be rapidly
discharged.
moved up into the range above 1 MHz. Since the photodiode only delivers small currents, but on the other hand only low values of load resistance may be used to avoid large time-constants, the received signal must be
interfaced to the subsequent logic circuits by highly sensitive amplifiers.
For this, the comparator SN 72710 or
similar circuits are particularly suitable. (Figure 24.12). In this case, the load
A resistance of the photodiode is 1 kfi. bias
current, fed in through a high resistance, defines a specific working point in the quiescent condition, this bias current being chosen to be about half the value of the opposing current of the photodiode (Ip «s 40 £JA).
24.5
Optocouplers in the photodiode mode
As was shown in the previous section, the physical characteristics of the photo-
transistor limit the maximum transmission
frequency. If only the photodiode of an optocoupler is used, the undesirable effects of the Miller capacitance and the storage time of the phototransistor are avoided. The frequency limit of these circuits can then be
At the same time, a number of measures have been taken here, firstly to prevent line reflections and secondly to protect the circuit from damage due to coupled inter-
ference voltages. This is easily accomplished at the end of the line, i.e. at the input to the optocoupler. The luminescent diode can handle aperiodic forward current pulses of up to 1 A. With a 100-12 line, this corresponds to an interference voltage of
100 V. A clamp diode D3 is provided, solely
to limit negative voltages.
Input
i
l
A-
TIL III or similar
hzd-
^
-O Output
24.11
Amplification circuit for transmission frequences up to approximately 100 kHz
408
In the case of the line driver, however,
further measures are necessary. The diodes
Dl and D2 limit the interference voltage and
the resistance R3 limits the current to non-
BAV critical values. The
24 diode, which is
used, permits aperiodic pulse currents up to
4 A. Together with the resistance R3,
interference voltages on the line, up to
400 V, can then be safely suppressed. In the
case of the circuit used as a line-driver,
however, further measures must still be
taken. With the above-mentioned currents,
the forward voltage of the clamp diodes is
more than 1 V. However, the voltage at the
output of an integrated circuit must not
become more negative than the component
substrate, since otherwise damage can occur.
In the case of the SN 75450 driver, the substrate is brought out separately. A voltage
V of --2 to --15 is applied to this connection,
in order to avoid an incorrect polarity.
Finally, the resistances Rl and R2 protect
the output transistors from excessively large
collector currents (> 300 mA).
The driver circuit used delivers a no-load
^ voltage Vqh 4 V, while the output
impedance is only a few ohms, both at High
and Low level. With the aid of the resistance
R3, the internal resistance of the driver is then matched to the line impedance, so that
line reflections are safely prevented.
Naturally, the measures described here can be used successfully with the circuits stated previously.
24.6 Duplex operation with optocouplets
With long transmission distances, the cable costs play a substantial part. Since the data generally has to be passed in both directions between two stations, it is worth while to do this over the same line. Parallel connection
of the two stations - as is generally usual with TTL circuits -- is not possible here,
since the luminescent diodes in the optocouplers have a very low resistance and therefore do not permit definite current
sharing in the two receivers. It is, however, possible to connect the two stations in series,
so that they handle a common current. The
two logic levels are then represented by turning the current on or off.
Figure 24.13 shows the circuit for such a
transmission system. In the rest condition,
the two enable inputs are at low level. The
current source formed by the transistor Tl
mA now sends a current of about 20
through
the phototransistor, which is also turned on,
in coupler I and the luminescence diode in
coupler II. For a data transmission from
Rl 1
Dl
BAV 24
R3
Input
/VWW 100 ft 1
Substrate -2 . . -I5V
R2
D2
isn
BAV 24
T
Figure 24.12
MHz Data-transmission link for transmission frequencies in the
range
409
Optocoupler II
fr
-- Data output
Data output
j/VWWV k=L Id-
Optocoupler t
-- /^T o DiU input
~"
\J--o Enable II
STATION I
Figure 24. 13 Duplex data transmission with optocouplers
STATION II
station I to station II, the input enable I is
switched to a high level. The information at
the data input now modulates the current in
the line. The coupler II transmits this to the Schmitt trigger, at the output of which the information is again available in TTLcompatible form. Data transmission from station II to station I takes place by interruption of the current path by the coupler 1 (the input enable I must of course be "High"). The voltage at the emitter of the transistor Tl thus collapses. Finally, the signal is coupled out again through a Schmitt
trigger.
With a line length of 1000 m, this system permits transmission frequencies up to 10 kHz. Here, the limiting frequency is not so much determined by the speed of the coupler, as by the slow level changes, originating from line reflections, on the line, since the latter is not correctly terminated at its two ends.
24.7
Common-mode suppression of optocouplers
The advantage of the high common-mode
suppression of optocouplers is mainly due to the low coupling capacitance between the GaAs luminescent diode and the phototransistor. The coupling capacitance depends on the insulation material, the distance between the luminescent diode and the phototransistor and also on the base area of the transistor.
Investigation of common-mode suppression
for the optocoupler was undertaken by means of the test circuit shown in Figure 24.14. Voltage pulses, with different amplitudes and rise times are applied to the input of the coupler. The rates of rise of voltage (dv/dt) at which the flip-flop is set are considered as the criterion for common-
mode suppression. The measurements were
carried out with an input diode current Ip = and with an Ip value, which corresponds to the typical Ic(on) value of the data book in each case. In order to eliminate the effect of the brought-out base in types TIL 102, 111 and 113, these connections were clipped off.
With the TIL 102, 108 and 109 optocouplers no reaction occurred with voltage pulses
V with an amplitude of 1000 and rates of
410
----
+SV
+5V
F^
lOnF.
Test specimen
r
~
1
O
r lkn y
1
"Fault" indication
---
1
i
fffV--
T1L 209
S<T
220 n
IN 914
1
1
1
1
V kV »
... I
tr = 0-5...1O0tis
-- Wh* tSV o- i
\j i
i
'
D
Reset button
Figure 24.14
Test circuit for determination of the common-mode suppression of optocouplers
rise of 400 V/jUs, either when the current was or was not flowing.
V(V) 800
Flip-Flop set
On the other hand, the TIL 113 coupler
reacts, because of its large current transfer ratio, even to voltage pulses of low amplitude and slope. Measurements without current
V were discontinued at values of V = 1 80
and dv/dt = 1 V/jUs.
--r-
50 80
100 120 140 160
du/dt WyS)
When carrying current, the couplers TIL 1 11 and TIL 113 showed no reaction up to the
values V = 1000 V and dv/dt = 500 V//is.
Figure 24. 15
Common-mode suppression of the optocoupler TIL 111 when not carrying current
411
25 Light exposure
switch for
photographic
enlargers
25.1
25-2 25.3
Principle of construction of a light exposure switch The timing system Practical circuit of the light exposure switch
413
1
25
Light exposure switch for photographic enlargers
The amateur photographer determines the optimum exposure time for photographic papers, either by test strips or by means of an exposure meter. In an exposure switch, the exposure measurement is coupled to a time switch for switching the lamp in the enlarger on and off. The time constant of
the timing circuit is varied according to the irradiance falling on the photodetector.
Simple exposure switches work with a trigger circuit, e.g. a monostable multivibrator. The timing circuit usually consists of an electrolytic capacitor and a photo-
resistor. In this case, either the charging or discharging of the electrolytic capacitor through the irradiated photoresistor serves as a measure of the necessary exposure time.
The two components which determine the time have several disadvantages. The capacitance of electrolytic capacitors is
temperature-dependent. In addition, they have a relatively large leakage current. Photoresistors are also temperature-
dependent, they show fatigue and ageing
effects and their function R = f(Ee ) is not
exactly linear. Also, photoresistors with a high dark resistance must be selected. Generally, these have a very slow response. Therefore low irradiances (small lens apertures) are used, so that an exposure time
of about 5 to 10 seconds is obtained. Some
of the above-mentioned disadvantages of simple exposure switches can be avoided with modern semiconductor components.
25.1 Principle of construction of a light exposure switch with Si photodiodes
Figure 25. 1 shows the block diagram of the exposure switch. The exposure process is prepared by pressing the normally-closed push-button. After the button is released, the Schmitt trigger receives a D.C. voltage level which is high in comparison with its threshold value; this causes the light source
J^
Starter circuit
--r 3
1
^NS--^*
Integrator
Photc diode
I
Schmitt trigger
JL.
Trigger circuit
Figure 25.
Block circuit of an exposure switch
415
* ---®
Power switch
Opal lamp
in the enlarger to be switched on through the switching amplifier and the triac. The enlarger projects an image of the negative on to the photosensitive paper. The photodiode receives the radiation reflected from the paper. The output voltage of the integrator decreases in proportion to the light exposure of the photodiode. After the optimum exposure, it reaches the turn-off level of the Schmitt trigger. The Schmitt
trigger terminates the exposure.
The maximum distance r from the photo-
diode to the projected image is determined by the reception peak of the diode. The photodiode should only evaluate the central picture content A. Evaluation of darkened or over-exposed areas is then prevented. In a greatly simplified form, the solid angle £2 can be determined with the half-value points of the reception peak. Here, the angle ip is the half aperture angle of the photodiode.
£1 = 2ir (1 - cos yp) fi
(3.4)
'a
(25.1)
A The effectively evaluated area is greater,
since the aperture angle only takes account
of the decrease in sensitivity down to 50%.
For photodiodes with very narrow reception
peaks, this error is negligible, since the effectivity evaluated area lies within the
illuminated image. The photodiode TIL 81
satisfies this condition.
The reception characteristic of photodiodes with large reception peaks is narrowed, either by an aperture mask or an external lens. With an aperture mask, a more uniform sensitivity is obtained over the solid angle which is of interest. At the same time, the absolute sensitivity of the photodiode decreases according to the degree of masking.
When the enlargement scale is changed, the
evaluated solid angle should remain constant, in proportion to the projected image size. The distance r is then to be corrected accordingly.
25.2
The timing system
The photodiode, the normally-closed pushbutton and the integrator represent the timing system of the exposure switch. If a photodiode with a reverse bias voltage is irradiated, a photocurrent flows. The photocurrent is independent of the applied reverse
voltage. It is calculated with the equation
Ip = s . Ee
(9.26)
The capacitor is charged or discharged by the diode current. The voltage on the capacitor, which is proportional to the stored
charge, serves as the evaluation criterion.
Figure 25.2 shows the circuit principle of the
timing system. The amplifier is connected as
an integrator. Therefore the capacitor lies
between the drain and gate of the 2N 3822
N-channel junction FET. The photodiode
TIL 81 is connected between gate and
ground. The normally-closed push-button is
located between the storage capacitor, the
FET load resistance and the drain. In the
^ V quiescent condition, a voltage Vtjs
is
present at the gate.
+20 V
C 1 2-2nF
rz> i |[ TIL 81
[TlOk 12
p*tv
2N 3822
Figure 25.2 Circuit of the timing network
If the drain circuit is interrupted with the button U, the storage capacitor charges up
through the load resistance R and through
the parallel-connected diodes, the photodiode and the gate-source diode, until it
416
reaches the working voltage. When the
U contact is closed, the voltage at the drain falls, since the transistor is now
conducting again.
The decrease in voltage at the drain is transmitted with negative potential, through the capacitor C, to the gate. The gate voltage
Vqs falls and reduces the drain current
until no further voltage change takes place at the drain. The negative gate voltage serves as a reverse voltage for the photodiode
TIL 81. If radiation now falls on the photodiode, the capacitor C will be discharged by the photocurrent. A subsequent emitter
follower (see Figure 25.3) reduces the out-
put impedance of the integrator. The voltage
on the load resistance R now decreases
linearly with the exposure. The exposure error is calculated from the sum of the dark current \\y 25 an(i tne residual gate current lGSS,25 :
iLeak = Id,25 + lGSS,25
(25.2)
Because the ambient temperature in the darkroom is between 20°C and 25°C, as required for the temperature-sensitive photographic baths, the leakage currents of the semiconductors also only have to be taken into account at this temperature. The
maximum dark current of the photodiode
TIL 81, for the measurement conditions
V Vcb = 10 at tij = 25°C, is about
iD,25,max = 10 nA. The typical values are two to three orders of magnitude lower. For
the measurement conditions V^g = 3 V at
tij = 25°C, typical values of lD,25,typ = 20 pA to 50 pA are obtained. For the measurement condition Vgs = -30 V at tij = 25°C, the maximum residual gate current of the 2N 3822 FET is about lGSS,25,max = 100 pA
For the measurement condition Vgs = -- 3 V
at tij = 25°C, the typical value is lGSS,25,typ = 15 pA. Thus, in accordance with equation (25.2), the leakage current 'Leak ' s obtained as
iLeak = 50 pA + 15 pA = 65 pA
The measurement condition V(^g = 3 V or Vgs = -- 3 V was chosen, because the
typical pinch-off voltage of the 2N 3822
FET is about Vcs.typ = -1 V to -3 V. The
capacitance of the storage capacitor is
determined by the shortest exposure time needed, t, and by the photocurrent sensitivity of the photodiode TIL 81. The capacitance
of the capacitor C was determined
experimentally at 2-2 nF. The time constant, determined by the leakage currents, is
calculated in accordance with the following equation:
C.U
tLeak : !Leak
2-2.
10
y
As.
3V
A V tr eak = "
rp1>2
65 . 10
.
= 1 01 s
(25.3)
With a permissible exposure error of 10%, the exposure time must therefore not exceed t = 10 s. The value tLeak increases to 25 minutes, if the photodiode TIL 81 and the
FET 2N 3822 are selected for their leakage
currents. In simplified form, the residual
gate current of the FET without the photo-
diode connected can be calculated in accordance with equation (25.2). For this, it is necessary to measure the discharge time t\ due to the gate residual current in a wellinsulated circuit. The discharge time t2 is measured with the photodiode installed but
completely darkened. By means of the time
difference t\ - 12, the dark current of the photodiode can be calculated.' It is better to measure leakage currents with an electro-
meter. If the FET is selected for minimum gate leakage current, then the BF 805, BC 264B and BC 264C can also be used. If,
in addition, the pinch-off voltage is also
restricted to the range between Vgs = 1 ^
and 3 V, the BF 245 FET can also be used.
Even more severe requirements can be
satisfied with the photodiode TIXL 80; it
has a more favourable ratio of photocurrent to dark current. In this case, the capacitor
is enlarged accordingly.
417
c
-t-
<&m
IN 757 .
(9 1V).
I
WTOHlOOkn
Screened cable with external insulation
X Ceramic stand-off insulator
Figure 25.3 Circuit of the light exposure switch
'y^
S 1 Continuous lighting S 2 Off switch
25.3 Practical circuit of the light exposure switch
Figure 25.3 shows the circuit of the light exposure switch with the TIL 81 photodiode. The performance of the timing circuit is considerably affected by insulation resistances. This problem will be eliminated, if all connection of the gate circuit are taken to a ceramic support or if the printed circuit (on glassfibre reinforced epoxy material) is sealed with insulating spray varnish and
baked out. A screened cable is to be used as
the lead to the photodiode.
The exposure time is adjusted to the sensitivity of the photographic paper by Rl.
This varies the charging voltage of the storage capacitor. Switching of the capacitance of the storage capacitor permits
variation of the exposure in wide ranges.
Through the emitter-coupled Schmitt trigger, the integrator drives an oscillator, which fires the triac during the exposure time and thus switches on the opal lamp.
W The 15 darkroom lamp is connected in
series with the opal lamp of the enlarger. In the stand-by condition, the darkroom lamp is lit, while the opal lamp still remains dark. In the exposure condition, the darkroom lamp is short-circuited by the triac and the opal lamp is thus switched on. Thus a false exposure through the darkroom lamp is prevented. The light can be switched on continuously with the switch SI. S2 serves as an emergency switch. The supply voltage is produced through a capacitor as a series impedance, is rectified with the IN 4006
V diode and is stabilised with a 22 Zener diode.
418
26
Optoelectronic couplers as switches for analogue signals
26.1
Semiconductor switches and potential
26.2
isolation
The phototransistor as a switch in the
optocoupler
26.2.1 Steady-state performance
26.2.2 Dynamic performance
26.3
Application of optocouplers in a digital
voltmeter
26.3.1 Measurement method used
26.3.2 Practical circuit of the digital voltmeter
26.3.2.1 Analogue section
26.3.2.2 Digital section
26.3.2.3 Voltage converter
26.4
D.C. voltage amplifier with chopper
26.5
Line tester
26.5.1 Test principle
26.5.2 Practical circuit of the line tester
419
26
Optoelectronic couplers as switches for analogue signals
26.1
Semiconductor switches and potential
isolation
Optoelectronic couplers, have found a wide field of application in digital techniques for the transmission of digital signals, while at the same time the transmitter and receiver are electrically isolated from one
another. In many cases, however, a switching
device, in which the control and switch sections are electrically isolated from one another, is also needed for analogue signals. This problem cannot be solved with normal bipolar transistors, since the control circuit (base-emitter path) has a physical connection with the switching circuit (collector-emitter
MOS path). Better isolation is achieved with
field effect transistors, in which, because of the extremely high input impedance of the transistor, a quasi-isolation between the control and switching sections is achieved. The disadvantage is that the potential difference between the gate and source connections affects the forward resistance of the transistor, i.e. the voltage to be switched must lie in a certain potential range in relation to the control voltage.
With the optoelectronic coupler, the conditions are considerably more favourable. The phototransistor, which works as a switch, is controlled by the radiation
emitted by the GaAs diode, so that there is no longer a physical connection between the
control and switching sections. Therefore, potential differences of 1000 volts or more can exist between the control and switching sections. The low coupling capacitance, of only 1 pF, between the diode and transistor, should also be emphasised. The semiconductor switches just mentioned are compared in Figure 26. 1.
26.2
The phototransistor as a switch in the
optocoupler
26.2.1 Steady-state performance
A switch is required to have a blocking
resistance in the open condition which is practically infinitely high, while the
Bipolar Transistor
MOS l--|-"T
Transistor
Figure 26.1 Comparison of various semiconductor switches
421
conduction resistance in the closed condition
should be as small as possible. The required conduction resistance depends on the
particular application. If only small currents are to be switched in high-resistance circuits, a conduction resistance of a few hundred
Ohms is often of no importance.
The blocking resistance of a phototransistor
is determined by the collector-emitter
leakage current IcEO It s about 1 nA with
V a collector-emitter voltage of V^g = 10
and 25°C ambient temperature. This
corresponds to a blocking resistance of
10
10
O.
It is, of course, considerably more difficult to achieve a low conduction resistance in
the phototransistors in optocouplers than with normal bipolar transistors. In the stated applications, the latter are operated with a base current of several milliamperes, so that the control current is several orders of magnitude greater than the load current.
Because of the low efficiency of the optocoupler, these conditions cannot be achieved here. The effective base current in the phototransistor is only a few times 10 MA. The result of this is that the forward resistance of the switch only reaches a few tens or a hundred Ohms, but in many cases this is of subsidiary importance. As an example, Figure 26.2 shows the forward characteristic of a phototransistor in an optocoupler. Each curve relates to a different forward current Ip in the diode which emits the radiation. In inverse operation, the transistor very quickly comes into the current saturation range, because of its low current gain, so that in this case only currents of a few microamperes can be switched with reasonable efficiency. In Figure 26.3, the forward resistance is shown as a function of the diode current. If a low conduction resistance is required both for positive and negative currents, two phototransistors can be connected in opposing parallel (Figure 26.4). It should be noted,
Figure 26.2 Characteristic of the phototransistor in the TIL 111 optocoupler
422
however, that in this case the breakdown voltage of the switch will be determined by the breakdown voltage of the base-emitter diode of the phototransistors and only amounts to about 7 V.
26.2.2
Dynamic performance
In the applications described here, the turn-
on and turn-off times of the phototransistor will be affected less by the transit frequency of the transistor than by the collector-base capacitance (in inverse operation the emitterbase capacitance) and the internal resistance of the circuit (Figure 26.5). The latter
V determines the apparent voltage gain of
the transistor and thus also the actual
effective capacitances.
The effective Miller capacitance is calculated
as follows:
for normal operation:
DC Cm = (Vn +
Cb
b for inverse operation:
CM = VN I + DCeb
In order to achieve a high radiation sensitivity of the phototransistor, it must have a large base area. This leads automatically to a large collector-base capacitance, which lies approximately between 20 and 100 pF. Since the transistors still cause a
RD(S2)
\
\ \
\ \ \
V \ \
1
1
I \ \
\ \ \
IC = 10 (iA
normal
^^
n verse
Figure 26.3 Conduction resistance of the phototransistor
423
'F (mA)
14
Figure 26.
Anti-parallel connection of phototransistors to reduce the conduction resistance
Figure 26.5
Equivalent circuit of a phototransistor for dynamic operation
very large voltage gain in high-resistance
Cm circuits, the effective Miller capacitance
reaches values of several nanoFarads. This then leads to switching times, which can amount, in the least favourable case, to several milliseconds. The conditions are
considerably more favourable during
inverse operation of the phototransistor. Here, firstly, the emitter-base capacitance is considerably less than the collector-base capacitance, Secondly, because of the low current gain of the transistor in this operating mode, a smaller voltage gain is obtained and thus a considerably smaller effective Miller capacitance.
The switching times were determined with the measurement circuit shown in Figure 26. 6. The results are summarised in
Table 26. 1.
As was to be expected, the switching times of the phototransistors in inverse operation are considerably less than the values in normal operation. Therefore, if short switching times are required, it is advisable to operate the phototransistors in reverse. If both positive and negative voltages are to be switched, two transistors are to be connected in series with opposite polarity. Since one transistor then always works in reverse, short switching times are always, obtained, irrespective of the polarity of
the voltage.
R[kft]
V =1V i
10 100
Vj = -1 V
10 100
Vj = 1V !)
10 100
ton (*&)
1-3 1-2
2-0
1-3 1-0
toff (/*0
130 810
21
130 830
' Two transistors connected in opposing parallel Two transistors connected in opposing series
Table 26. Switching times of phototransistors (TIL 111)
424
Vi = 1V2)
10 100
-
1-4
20
V i= ±l V ( lp = 40 mA
Figure 26.
Measurement circuit for determination of switching times
26.3 Use of optocouplers in a digital voltmeter
Digital voltmeters are replacing analogue
voltmeters to an increasing extent. The advantage of these instruments is, that the measured results are presented in a form, which can be processed further by datalogging systems or process computers without intermediate treatment, which is of great importance for process automation.
can be eliminated by isolation of the measurement section from the evaluation
section.
In the following paragraphs, a simple digital voltmeter, in which isolation is achieved between the measuring and evaluation
sections in a very simple way, by use the of optocouplers as switches for analogue signals and as transmission devices for analogue and
digital signals.
It is often an important requirement with measuring instruments of this kind, that the analogue input and the digital output are isolated from one another, because the measurement source and the digital signals have different reference potentials. There is often a danger, that the measured values will be falsified or the digital output signals will be interfered with through transient currents through undefined earth loops. Such errors
26.3.1
Measurement method used
The unknown voltage is measured by the
"Dual-slope" method. In this, the voltage
V x is integrated over a given time ti
(Figure 26. 7). After the time t^ has elapsed,
V the voltage x is switched off and a reference
VR voltage
e f is switched to the integrator.
425
Vx0-
v^
R
-cm-
-vref
Integrator
Figure 26.7 Theoretical measuring circuit of a digital voltmeter
Comparator
Thus the capacitor C is discharged again. When the output voltage of the integrator
reaches zero, the comparator operates. Figure 26.8 shows the variation of the
voltage.
Figure 26.8
Variation of
integrator
voltage
with
time at the
The following relationships apply:
1
Vci=R--C Vx .ti
and
As can be seen, the external time constant
of the integrator (R and C) has no effect on
the accuracy of the measurement. Since
only the ratio of the times tj and t2, but
not their absolute value, is of interest, this
source of error can also be easily eliminated.
An oscillator, which serves as a time-base,
only has to have adequate short-time
constancy. The absolute value of the
frequency does not affect the accuracy of
the measurement. On the other hand, the
VR reference voltage
ef has a direct
influence of the measured result. Therefore,
high-stability Zener diodes with low
temperature coefficients must be used to
generate it.
In the above calculation, the performance of the operational amplifier in the integrator has been disregarded. However, the operational amplifiers which are currently available have such small offset voltages and currents and such an excellent temperature response, that errors of this kind can be neglected in the circuit described here.
---1
Vc 2 = RC (-VRef)t 2
V V With
+
C][
C2 = 0, then:
-- -- 1
1
Vx tl+
(-VRef)t2 = 0or
=~ Vx l2 vRef
tl
26.3.2 Practical circuit of the digital voltmeter
26.3.2.1
Analogue section
As can be seen from the overall circuit in Figure 26. 9, the analogue section consists of the integrator Ul, the comparator U2, the
426
Wl: WWdt-.O-JCul W3: l6Wd»..0-ICul W3 *OWd(..0! Cui W4 MWdl-.CMCul
Fi&ure26.9 Digital voltmeter, overall circuit
reference voltage source and a total of four analogue switches, which are formed by photo transistors. During the time t\, throughout which the switch U14/16 is
closed, the unknown voltage Vx reaches the
integrator Ul through a 100 kQ, resistor and charges up the feedback capacitor. Since both positive and negative voltages have to be switched here, two phototransistors are
arranged with opposite polarities, so that one transistor is always operated in the
inverse mode and thus short switching times are ensured. The offset voltage of the
operational amplifier is balanced with the
5 MQ, potentiometer and thus the zero
point of the voltmeter is set. the subsequent
operational amplifier U2 serves as the
comparator. The amplifier has no feedback, so that it works with its full no-load gain and switches over suddenly when the voltage at its input passes through zero. The output signal from the comparator is coupled out
through the optocoupler U21 and. is fed to
the digital section. During the time tj the capacitor is discharged again, either by a positive or negative reference voltage,
depending on whether the voltage Vx was
negative or positive. The reference voltage is produced by a temperature-compensated Zener diode, which is fed with a constant current from the operational amplifier U23. The mid-point of the voltage divider across this Zener diode is at zero potential, so that
a negative reference voltage is obtained at the anode and an equal positive reference voltage, of about 3-1 V, is obtained at the
cathode. The full-range value for positive and negative input voltages is balanced with the two 20 k£2 potentiometers. After completion of the measurement, the integrator is reset to zero through the switch U19/20. Since the comparator is
involved in the discharge of the integration capacitor, its offset voltage is compensated.
427
26.3.2.2 Digital section
When considering the functional sequence in
the digital section, it is best to start from the
fact, that the monostable U3 has been set at
the end of the last measurement. Thus the
output of the gate U4C is at a low level and
the switch U19/20 is turned on (see Figure 26.10). At this time, the analogue section is at rest. After about 100 ms, the monostable resets and thus initiates the next measurement.
V The voltage x is switched to the integrator
through the gate U8C. At the same time, the
clock generator U5A, B, C starts and,
delivers, at the gate outputs U6E and U6F,
two pulse sequences, displaced by about 180 to one another, with a frequency of approximately 100 kHz. The counter,
consisting of Ull, 12, 13 now counts from
000 to 999. Through the subsequent carry,
V the flip-flop U7B is set and the voltage x is
disconnected from the integrator. At the same
U7A time, the carry signal sets the flip-flop
according to the polarity of" the output
voltage of the integrator, detected by the comparator. Thus the appropriate reference
voltage is switched through the gate U8A or U8B to the integrator. When the output of
the integrator again reaches zero voltage, so
Output U! -
Output U2 0/U3
I
I
I
I
I
I
I
I
I
I
II I
Figure 26.10 Pulse diagram of the digital voltmeter
428
that the comparator switches over, there is then a high level signal at the D-input of the flip-flop U10A.
The next clock pulse sets the flip-flop, so that the counter is stopped through the enable inputs. The next clock pulse transfers the contents of the counter to the store. At the same time, the position of the flip-flop
U7A is interrogated and is transferred to the
flip-flop U10B. The signal just mentioned
also triggers the monostable U3. Now the
"Ready" signal carries a low level and thus reports that the new measured value is ready at the output of the instrument. At the same time, the counter and flip-flop are set to their initial state through the gate U4D.
26.3.2.3 Voltage converter
As already mentioned, the requirement existed, for the analogue and digital sections to be electrically isolated from one another. In the control of the analogue switches and the return signal from the comparator, this was achieved by the use of optocouplers. In order to make it possible for the digital
voltmeter to be operated from only one 5 V
supply, a voltage converter has been provided, to deliver the supply voltages for the operational amplifiers and the reference voltage. Blocking converters are voltage converters with a good efficiency, but their output voltage is very dependent on the load.
A Therefore, stabilisation must be provided.
conventional stabiliser circuit at the output of the converter cannot be used, since this would unnecessarily reduce the efficiency of the circuit.
It is more economical, to regulate the output voltage of the converter by varying the mark-space ratio on the transistor T3. For this, the transistor Tl measures the deviation of the output voltage from the voltage of the reference diode (2 x IN 753). The transistor T2, the collector current of which determines the mark-space ratio in the converter and thus the output voltage, is then driven through the optocoupler U22.
Basically, optocouplers are only suitable with restrictions for the transmission of analogue signals, since firstly the temperaturedependent output current is not exactly proportional to the input current; and secondly there is a variation in the current transfer ratio from device to device. In the application described here, however, this only has a slight effect on the performance of the circuit, if care is taken that the control gain is high enough, even in the worst case, to ensure the required stability of the output voltage.
The components used in the digital voltmeter are summarised in Table 26. 2.
U 1: SN72307P U8:
SN7437N
U2: SN72741P U9:
SN 7403N
U3: SN7412N U10: SN 7474N
U4: SN7437N U 11-13: TIL 306
U5: SN74132N U14: TIL 304
U6: SN49703N U 15-22: TIL 111
U7: SN 7474N U23: SN 72741P
Table 26.2 Parts list of the optocouplers used
integrated
circuits
and
26.4 D.C. voltage amplifier with chopper
In instrumentation, D.C. voltage amplifiers are often needed, to amplify very small signals, originating, for example, from thermocouples, so that they can be processed further by subsequent equipment. Generally,
severe requirements are imposed on these amplifiers, such as a) high input impedance or low input current, in order not to load the signal source, b) defined gain, c) low zero drift, d) constant parameters with variations in temperature and operating
voltage.
With conventional D.C. amplifiers, these requirements can usually not be met, or can only be met at great expense. Therefore it is often simpler, to chop the D.C. voltage signal, then amplify it with a conventional A.C.
429
Input ~y--
sT
+~TnI
·
C Outpm
J
Figure 26.11 Theoretical circuit of a chopper amplifier
voltage amplifier and finally rectify it again. The principle of such an amplifier is illustrated in Figure 26.11. The amplifier concerned here is a conventional A.C. voltage amplifier. The desired gain is set by an internal current or voltage feedback. The no-load gain of the circuit should be at least 10 times greater, so that variations in operating voltage and temperature do not
affect the performance of the amplifier. An
oscillator controls the switches at the input
and output. At the input, the direct voltage to be measured is chopped, so that a squarewave voltage, the amplitude of which corresponds to the voltage to be measured, is obtained. At the output end, the amplified square-wave voltage is rectified again with a switch and is smoothed with the capacitor C.
Figure 26.12 D.C. voltage amplifier with chopper
430
Figure 26.12 shows the complete circuit of a simple amplifier with a chopper. In the first stage, a field-effect transistor ensures a high input impedance. The second stage has a current source, formed by a transistor, as its load impendance. In this way, a gain of 2000 is achieved with an amplifier with only two stages. The amplified signal is fed out through an emitter follower. The resistances
Rl and R2 form a feedback loop, so that
the gain of the amplifier is reduced to 100. At the same time, the working points of the transistors are stabilised by this. For the switches, the phototransistors in optpcouplers are again used, with two transistors connected in series with opposing polarity in each case, in order to achieve short switching times, both with positive and
A negative input voltages. multivibrator
with a frequency of approximately 1 kHz drives the diodes in the optocouplers. The
150 ^2 resistors in the collector leads limit
the diode current to about 40 mA. It should
be noted, that the feedback capacitors are not connected to the collectors, but to the anode of the diodes. This ensures, that the base-emitter junctions of the transistors are not driven into breakdown.
26.5 Line tester
The very many cable connections which are
needed in electrical system, necessitate a simple test system, with which the cables can be tested for defects, such as shortcircuits between two conductors and open circuits. The following report describes an instrument, with which multicore cables can be tested in a simple manner.
26.5.1 Testing principle
In the testing of cables, the following defects must be reliably detected:
a
Short-circuit between two cores,
Open-circuit in a core,
Cross-over of two cores.
With an appropriate test circuit, the last defect can be detected together with that mentioned under b).
{>" Amplifier
Table 1
Cable i
Figure 26.13 Theoretical circuit of the tester
Figure 26.13 shows the theoretical circuit of the tester, while for the sake of simplicity, the circuit for four conductors only is illustrated. Testing commences with all
switches SnA closed, except for the switch SI A. The switches SnB are open. Now, if
there is a short between conductor 1 and any other conductor, the input of the amplifier is at ground potential. This voltage level characterises the defect named. Then the switch SIB is closed. If conductor 1 has continuity, then again the amplifier input is at ground potential. If conductor 1 has an
open circuit or - what produces the same defect - if two conductors have been
wrongly identified at one end of the cable, the amplifier input is not at ground potential, which identifies the fault in this
case. Then the switch S2A is opened; all other switches SnA are closed and the test
for conductor 2 is carried out as described above. Thus, for testing a four-core cable, the switch diagram shown in Table 26. 3 is obtained.
431
' I6X
220 n
-cm-*
b l»-0
A.AtJaA&JtJiA^A A A A A A A
1 2 3 4 5 6 7 8 9 10 11 12 13 14 IS
Inverter
SN7404N
SN74154N AB CD
Gl G2
IT
-d
QA BQBOCQD
ft-i-
Start
T1L311
II S
BL
16x ,-tZH,
IN 914
> N-i
-- £: > N--> W--> w- J^L,
: Hl°1
£1
11 ASAAAaaAAAAAAA
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
SN 74159N ABC D
Gl G2 TT"
2 x SN 7472N
I
J
Q
-c >
Kc Q
Open Circuit
n. 220
Shon Circuit 220n TIL210
KQ
1
Figure 26.14 Complete circuit of the line tester
432
Switch position (x = Switch closed)
S1A S2A S3A S4A SIB S2B S3B S4B
X
X
X
X
x
X
X
X
x
X
X
x
X
X
X
X
X
X
X
X
X
X
X
x
X
X
X
X
Table 26.3 Switch diagram
Test
Short circuit 1 Open circuit
Conductor
1 1
2 2
3 3
4 4
26.5.2 Practical circuit of the line tester
Figure 26.14 shows the circuit of a cable tester, which tests multicore cables with a
maximum of 16 cores. By appropriate extension, cables with more than 16 cores A can also be tested. start-stop oscillator (SN 7413N) with a frequency of about 3 kHz
triggers the two flip-flops (SN 74107N), which control the two tests (short and open circuit).
First, the cable is tested for short-circuit,
the decoder SN 741 59N is unblocked and
switches one core to a low potential and thus effects the test for open circuit. The
SN 74450N amplifies the signals coming
from the test specimen and drives the two flip-flops (SN 7472N). If a fault is detected, the corresponding flip-flop is set by the clock signal and the fault is indicated with a luminiscent diode (TIL 210). The
oscillator is thus stopped, and the number of the defective core can be read off on the TIL 311 display. This device works in the hexadecimal code, i.e. the figures to 9 are shown in the conventional way, while the numbers 10 to 15 are denoted by the
letters A to F (Figure 26.15). This has the
advantage, that only a one-digit display is necessary to represent 16 states and the complex and thus expensive code conversion
into the BCD code is eliminated.
By actuating the "Start" button, the fault indication is cleared and the oscillator is started. The SN 7493 counter is advanced by one and selects the next core to be tested through the demultiplexer (SN 74154N andSN74159N).
Optocouplers are used as potential-free switches. They ensure that, under no circumstances is the switching element
m m
·
1
2'
3
4
,,
· ·
5
6
7
mm
aa
..
· «
·
8
9
10
u
· >* ·
a*
12
13
14
15
Figure 26.15 Display format of the TIL 311
433
(phototransistor) affected by the control part (luminescent diode). Also, they have the advantage, that they can be driven
directly from TTL circuits.
The demultiplexer SN 74159 is used as the
switch which connects the corresponding cable core during continuity testing to ground potential. It must be noted, however, that
every output of this module draws a leakage current, in the "Off" condition, of 50 fiA max. The sum of these currents (maximum 800 HA) could, in some circumstances, falsify the measurement and cause the amplifier (SN 75450N) to operate. Therefore
this current is diverted to the supply line through resistors. Diodes in the individual leads ensure reliable decoupling.
434
Index
Absolute refractive index, 96 Absorption:
basic lattice, 130 basic laws of, 88-90 extrinsic, 130 intrinsic, 130 and transmission spectra, 90-95 Absorption factor, 85 A.C. Electroluminescence, 8
Accommodation, 59 Actinic values, 77-81
of IR luminescence radiation from GaAs
diodes, 145-151 Adaption, 59 Amplifiers, driving of: with phototransistors,
photodiodes, and photocells, 280-285 with three-pole phototransistors, 293-
294 Analog indication of digital values, 355-357
Analog measuring instruments with LED
indication, 357-360 Apostilb (asb) unit of luminance, 26 Area-dependent sensitivity for junction
photodetectors, 144-145 Artificial temperature radiators, 104-105 Atomic models, 5-8 Attenuation, basic laws of, 88-90 Avalanche gain, 316 Avalanche photodiodes:
dark current of, 153-154 quantum yield and photocurrent
amplification of, 126-129
Band models, 5-8
Basic lattice absorption, 130 Betaluminescence, 8 Bioluminescence, 8 Black bodies, laws of radiation from, 23, 43-
56 defined, 45-46 emittivity, 50 Kirchhoff's law, 51 non-black bodies, 46 Planck's law, 46-47 radiation isotherms, 51 reduced law of radiation, 51-56 Stefan- Boltzmann law, 47-48
Black bodies, laws of radiation from (Cont.): temperature and, 45-46 Wien displacement law, 48-50
Calculation in radiation physics and optics,
principles of, 19-32 radiant energy-luminous energy, 21-22
radiant flux-luminous flux, 21-22 units related to radiation source, 22-27
radiance-luminance, 24
radiant efficiency-luminous efficiency,
26-27
radiant emittance-luminous emittance,
22-23
radiant intensity-luminous intensity, 23-24
units of luminance, 26-27
units related to the receiver, 27-32 irradiance-illuminance, 27-28
irradiation-light exposure, 28
reflection and irradiance, 29 spectral radiation, 29-32
Candela, 69, 80
Case reflections, 170-173
Cathodoluminescence, 8
Cavity radiator, 45 cd/ft2 unit of luminance, 26 cd/in2 unit of luminance, 26
CDI (Collector Diffusion Isolation) process,
photodiodes by the, 134
Celsius (°C), 45-46
Chemiluminescence, 8 Chromatic radiation, 137-139
Clouds as radiation source, 104
Color
temperature
T f,
46
Conduction band, 6-7
Cone vision, 59, 60
Contrast ratio of very small photocell
junctions, 181-184
Conversion constants C and C,
determination of, 62-64
Couplers (see Optoelectronic couplers)
Coupling characteristics of very small photocell junctions, 176-179
"Cross-eyed" optoelectronic component,
173
Crystal luminescence, 9
435
Dark current of junction photodetectors, 151-154
avalanche photodiodes and, 153-154 Data transmission with optocouplers (see
Optoelectronic couplers, data transmission with) Daylight vision, 59 D.C. voltage amplifier with chopper, 429-431 Detectivity, 156 Diffuse reflection, 86-87 Digital rotary knob, 387 Digital voltmeters, optocouplers and, 425-429 analog section, 426-427 digital section, 428-429 voltage converter, 429 Diode testers, 352 Diodes (see specific types of diodes) Direct couplers, 249-250 Direct semiconductors, 9-11 Directional reflection, 86-87 Discharge through gases, 105-107 Discharge lamps, 107 Discharge tubes, 107-109 Dispersion, 96
Display units (see Numeric and alphanumeric display units)
Distribution temperature Tv , 46
Dualism of waves and particles, 4 Dynamic data, 186-191
Electrical parameters of luminescence diodes, 226-228
Electroluminescence, A.C., 8 Electromagnetic radiation spectrum, 3 Electron transitions and semiconductors, 10-
11
Electronic flash units, 341-347 circuit of, 345 principles of, 343-344 Si phototransistors for automatic, 346-347 types of exposure control, 344-345
Emitter and receiver parameters, 159-204 dynamic data, 186-191 half-power and half-value points, 184-186 optical tolerance of optoelectronics components, 167-175 internal case reflections and wafer geometry, 170-173 tolerance levels, 173-175 wafer centering, lens quality, distance from lens to wafer, refractive index of
epoxy resin, and shape of dome and
case, 168-170 optoelectronic components and, 161-167 reliability of optoelectronic semiconductor
components, 191-204
Emitter and receiver parameters (Cont.): very small photocell junctions, 176-184 contrast ratio, 181-184 coupling characteristics, 176-179 transmission ratios, 179-181
Emittivity, 50 Energy bands, 6-7 Escape energy, 13-14 Evaluation of radiation, general and
photometric, 57-81 actinic value, 77-81
of Planck and luminescence radiation with the eye as a photodetector, 80-
81
of radiation spectrum for photodetectors and for the eye, 7780
calculation of the photometric radiation
K equivalent for different radiation
spectra, 69-74 for luminescence radiation from light-
emitting diodes (LED's), 70-74 for Planck radiator with the
temperature laid down for the
definition of the candela, 69 for standard light source A, 69-70 conversion of radiometric units into
photometric, photopic units, 74-77 for the luminescence radiation of light-
emitting diodes, 76-77 for Planck radiation, 74-76
human eye as a light receiver, 59
optical sensitivities, 59-60 radiation spectra, 60-68
conversion constants C and C, 62-64
photometric radiation equivalent, 6468
photopic sensitivity of the eye, 67-68 External photo-effect, 13-14 Extrinsic absorption, 130
Filament lamps, 104-105 simple couplers with, 298-301
5 x 7-point matrix display units, 374-380 Flash discharge tubes in photography, 111-
112 Fluorescence, 9 Fluorescent lamps, 1 10 Foot-Lambert (fL) unit of luminance, 26 Frequency limit, 188-189
GaAs diodes: actinic value of IR luminescence radiation
from, 145-151 "low-cost," 168-170
436
GaAs diodes (Cont.):
parameters of, 159-229 silicon-doped, 113 zinc -doped, 114, 115
GaAsP diodes, 114, 115 GaP diodes, 113-114
Grey emitter, 46 open fire and, 104
Junction photodetectors, spectral sensitivity of (Cont.):
possibilities for shifting the, 134-136
Schottky barrier PIN photodiodes, 132
temperature and, 134 two-pole, detector circuits for, 276-288 Junction photo-effect, 13, 16-18
Half-power points, 164 half-value points and, 184-186
High-pressure mercury-vapor lamps, 109110
Hot gases as radiators, 105
Human eye as a light receiver, 59, 67-68
actinic value of a radiation spectrum for, 77-80
optical sensitivities, 59-60 Planck and luminescence radiation and,
80-81 radiation spectra and, 60-62
Illuminance, 27-28 reflection and, 29
Indirect semiconductors, 9-10 Injection luminescence, 9-13 Internal photo-effect, 13-16
Intrinsic absorption, 130
IR detector sensitivity, 154-158 (See also Junction photodetectors)
IR luminescence radiation from GaAs diodes,
145-151 Irradiance-illuminance, 27-28
reflection and, 29 Irradiation-light exposure, 28
Junction photodetectors, 123-158 actinic value of IR luminescence radiation
from GaAs diodes for silicon, 145-151
amplifying, 139-144 area-dependent sensitivity parameters
for, 144-145 dark current of, 151-154
avalanche photodiodes, 153-154 defined, 121 IR detector sensitivity, 154-158 non-amplifying: for chromatic radiation,
137-139 for monochromatic radiation, 136-137
Q quantum yield, of, 125-129
avalanche photodiodes and, 126-129 spectral sensitivity of, 130-136
photodiodes by the CDI process, 134
planar-diffused silicon photodiodes, 132
Kelvin (K), 45-46 Kepler's rule, 73-74 Kirchhoff 's law of radiation, 51
Lambert radiator, 36-38 radiation and receiving characteristics and, 162-165
Lambert (L) unit of luminance, 26 Lambert's cosine law, 37-38 Lens quality, 168-170 Lenses, optocouplers with, 254-259 Light-emitting diodes (LED's), 9-13
circuits with, 349-360 analog indication of digital values, 355357 analog measuring instruments with, 357360 diode tester, 352 large format seven-segment display unit, 355 logic tester, 352-354 polarity and voltage tester, 354-355 simple indicators, 351
luminescence radiation from, 70-74, 7677
Light exposure, 28 Light exposure switch for photographic
enlargers, 413-418 construction of, 415-416 practical circuit of, 418 timing system of, 416-417 Light measurement with Si phototransistors
in electronic flash units, 341-347 Line tester, 431-434
practical circuit of, 433-434 testing principle, 431-432 Logic circuits, 269-271 numerical display units with integrated,
368-371 with phototransistors, 302-304 Logic testers, 352-354 Loss of power for luminescence diodes, 217--
218 in metal cans: with heat sinks, 217
without heat sinks, 217-218 in plastic packages, 217 Low-pressure mercury-vapor lamps, 109
437
Luminance, 24 units of, 25, 26
Luminescence diodes, 113-115, 205-229 electrical parameters, 226-228
GaP diodes, 113-114
measurement of: irradiance, 242-244 radiation, 240-245 total radiant power, 245
modulated transmitters with, 309-324 pulse-, 320-324 simplest circuits, 311-312 sine-wave, 312-320
operation of: from constant-current sources, 265-269
with direct current, 263-271 drive with logic circuits, 269-271 through series resistances, 265 pulse operation, 228-229 quantum efficiency, 207-213 quantum yield, 1 14 radiant efficiency, 223-224 radiant power, 220-223
silicon-doped GaAs diodes, 1 13
spectral radiant efficiency, 224-226 thermal calculations, 213-220
basic principles, 213-215 loss of power calculations, 217-218
maximum permissible forward current,
218-220 thermal resistance, 215-217 Luminescence phenomena and radiation, 5-
13
Luminescence radiation, 105-110 actinic value of, 80-81 light-emitting diodes and, 70-74, 76-77
Luminous efficiency, 26-27 Luminous emittance, 22-23 Luminous energy, 21-22 Luminous flux, 21-22 Luminous intensity, 23-24
Matter {see Optical radiation and matter, interaction between)
Measurement units:
for illuminance, 28 for irradiance, 27 for irradiation, 28 for luminance, 25, 26 spectral, 30 Metal vapor lamps, 109-110 fluorescent, 110 mercury-vapor, 109-110 sodium, 109 Mixed-light lamps, 110-111
Mixed radiators, 46
Mixed reflection, 86-87 Modulated transmitters with luminescence
diodes, 309-324 pulse-, 320-324 simplest circuits for, 311-312 sine-wave, 312-320
with bias voltage sources, 313-318 with constant-current sources, 318-320 Monochromatic radiation, 136-137 Monochromator, 242
Moon as radiation source, 104
Multivibrators, control of, 285-288
Natural radiation sources, 103-104 Night vision, 59, 60 Noise equivalent input power, 155 Non-black bodies, 46 Numeric and alphanumeric display units,
361-380 5 x 7-point matrix, 374-380 monolithic, 371-374 multiplex operations of, 365-368 numeric, with integrated logic, 368-371 seven-segment, 363-365
Open fire as temperature radiator, 104
Optical radiation, physics of, 1-18 light and, 3-5 basic definitions, 3 dualism of waves and particles, 4 quantum nature of radiation, 3-4 spectrum of, 3 wavelength and propagation speed, 45
luminescence phenomena and, 5-13 atomic and band models, 5-8
fluorescence, 9
luminescence, 8-9 phosphorescence, 9 in semiconductors, 9-13 photoelectric effect, 13-18 external, 13-14 internal, 14-16 junction photo-effect, 16-18 (See also Calculation in radiation physics and optics, principles of) Optical radiation and matter, interaction between, 83-99 absorption, transmission, and reflection
factors, 85
absorption and transmission spectra, 9095
basic laws of absorption, attenuation, and scatter, 88-90
438
1
Optical radiation and matter, interaction
Photocells:
between (Cont.):
direction-dependent, 381-387
reflection of radiation, 86-88
counter, 384-385
spectral reflectivity, 87-88
digital rotary knob, 387
refraction, 96-99
optoelectronic couplers, 385-386
scatter, 86 spectral transmission and spectral
absorption factors, 85-86
principle of operation, 383 driving amplifiers with, 280-285
phototransistor TIL 81 and, 288-290
Optical sensitivities, 59-60
circuits with, 328-334
radiation spectra and, 60-62
(See also Photocell junctions)
Optical tolerances of optoelectronic
Photoconductors, 120-121
components, 167-175
Photocurrent amplification of avalanche
Optoelectronic components:
photodiodes, 126-129
optical tolerances of, 167-175 radiation and receiving characteristics of,
161-167
Photocurrent sensitivity of Si phototransistors, 335-339
Photodarlington circuits, 276-277
Optoelectronic couplers, 247-261 data transmission with, 399-411
common-mode suppression of, 410-41
Photodetector circuits, 273-307
to drive TTL integrated circuits, 304-307
logical circuits with phototransistors, 302-
construction and characteristics of, 401-
304
404
for modulated radiation, 325-334
current transfer ratio and, 401-402 duplex operation with, 409-410
with phototransistors, 327
with the TIL 81 as photodiode and as
dynamic performance and, 403
photocell, 328-334
improvement of switching performance,
principle of operation, 275
406-408
simple couplers with filament lamps and
interference on transmission links, 401 mechanical construction and, 402-403
two-pole phototransistors, 298-301 with three-pole phototransistors, 288-298
photodiode operation and, 403-404,
driving of amplifiers, 293-294
408-409
operating modes, 288-293
phototransistor operation and, 404
phototrigger and photomultivibrators,
simple transmission links and, 404-405
294-298
direct couplers, 249-250
with two-pole junction photodetectors,
direction-dependent, 385-386
276-288
examples of, 252-253
direct relay control with
with lenses, 254-259
phototransistors, 276
with modulated optical radiation, 260-261
multivibrators with phototransistors,
with non-stationary source emission, 252 reflected, 250-251
driving of, 285-288 photodarlington circuits, 276-277
as switches for analog signals, 419-434 D.C. voltage amplifier with chopper, 429-431 digital voltmeter and, 425-429
thyristor and triac control with phototransistors, 277-279
transistor and operational amplifiers with phototransistors, photodiodes, and
line tester, 431-434 phototransistor, 421-424 semiconductor switches and potential
photocells, 280-285 Photodetectors, 117-121
actinic value of radiation spectrum for, 77-
isolation, 421
80
with unmodulated optical radiation, 259-260 ambient and/or background radiation and,
Optoelectronic rangefinder, 389-397 phase management and, 391-394 practical circuit of, 394-397
103
to drive TTL integrating circuits, 304-307
with external photoeffect, 1 19- 120 the eye as, 80-81
with internal photoeffect, 120-121
Particle wave, 9-10 Phosphorescence, 9
photoconductors, 120-121 photoresistors, 120-121
Photocell junctions, very small, 176-184
(See also Junction photodetectors)
439
Photodiodes:
circuit with the TIL 81 as, 328-334
driving amplifiers with, 280-285 optocouplers and, 403-404, 408-409
phototransistor TIL 81 and, 288-290
(See also Junction photodiodes) Photoelectric effect, 13-18 Photoelectric threshold value, 13-14 Photographic enlargers, light exposure switch
for, 413-418 Photoluminescence, 8 Photometric evaluation of radiation (see
Evaluation of radiation, general and photometric) Photometry, law of, 36-37 Photomultivibrator circuits, 294-298 Photon, 4 Photopic sensitivity of the eye, 67-68 Photopic units, conversion of radiometric units into photometric, 74-77 Photoresistors, 120-121
Phototransistors: circuits with, 327
logic, 302-304 control of multivibrators with, 285-288 direct relay control with, 276 driving of amplifiers with, 280-285 optocouplers and, 404
dynamic performance, 423-424 steady state performance, 421-423 switches and, 421-424 Si: light measurement with, 341-347 photocurrent sensitivity of, 335-339 three-pole, 288-298 thyristor and triac control with, 277-279 TIL 81, 288-293 circuit with, 328-334 Phototrigger circuit, 294-298 Planar-diffused silicon photodiodes, 132-133 Planck's law of radiation, 46-47 actinic value and, 80-81 conversion of radiometric to photometric
units and, 74-76
K photometric radiation equivalent and, 69
Point radiator, 38-40 inverse square law of, 40-41
Polarity and voltage tester, 354-355 Propagation speed and wavelength, 4-5 Proportional radiation, 53-55 Pulse operation of luminescence diodes, 228-
229 Pulse transmitters, 320-324
Quanta, 4
Quantum efficiency, 207-213
Quantum yield:
Q: of junction photodetectors, 125-129 photocurrent amplification of avalanche photodiodes, 126-129
radiation sources and, 114
Radiance-luminance, 24 Radiant efficiency, 26-27, 223-224
spectral, 224-226 Radiant emittance, 22-23 Radiant energy, 21-22 Radiant flux, 21-22 Radiant intensity, 23-24 Radiant power, 220-223
measurement of, 235-237 of standard light A, 237-238 total, of a luminescence diode, 245 Radiation, laws of, 33-41
Lambert radiator, 36-38 Lambert's cosine law, 37-38 law of photometry, 36-37
small surface radiators and surface receivers, 38-41
inverse square law, 40-41 solid angle, 35-36 (See also Black bodies, laws of radiation
from) Radiation isotherms, 51 Radiation measurements, 231-245
color temperature of standard light A, 233235
general considerations, 233
A irradiance of a standard light radiation
with Si photodetectors, 238-240 luminescence diode radiation with Si photo-
detectors, 240-244 general measurement problems, 240-242 irradiance, 242-244 relative spectral sensitivity with a
monochromator, 242
A radiant power: of standard light radiation
with the thermopile, 237-238 with thermal photodetectors, 235-237 total, of a luminescence diode, 245 Radiation sources, 101-115 artificial temperature radiators, 104-105 filament lamps, 104-105 open fire, 104 flash-discharge tubes in photography, 1 1 1-
112 luminescence radiators, 105-110
discharge through gases, 105-107 discharge lamps, 107 discharge tubes, 107-109 fluorescent lamps, 110
440
Radiation sources, luminescence radiators (Cont.):
hot gases, 105 mercury-vapor lamps, 109-110 metal vapor lamps, 109-110 sodium vapor lamps, 109
xenon lamps, 109 luminescent diodes, 113-115
GaP diodes, 113-114
quantum yield and, 114 silicon-doped GaAs diodes, 1 13 zinc-doped GaAs and GaAsP diodes,
114, 115 mixed-light lamps, 1 10- 111 natural radiation sources, 103-104
clouds, 104
the moon, 104 the sun, 103-104
Radiation spectrum: actinic value of, 77-80 photometric evaluation of, 60-62
Radioluminescence, 8 Radiometric units, conversion of, into
photometric, photopic units, 74-77
Ratio temperature, Tr , 46
Receiver parameters (see Emitter and receiver parameters)
Reflected optoelectronic couplers, 250-251 Reflection of radiation, 86-88
law of, 86-87 Reflection factor, 85 Reflectivity, spectral, 87-88 Refraction, 96-99 Refractive index of epoxy resin, 168-170 Reliability of optoelectronic semiconductor
components, 191-204
Scatter, 86 basic laws of, 88-90
Schottky barrier PIN photodiodes, 132
Selective radiators, 46 Semiconductor switches and potential
isolation, 421 Semiconductors and luminescence, 9-13
Sensitivity:
of IR detectors, 154-158 of Si phototransistors, 335-339 spectral, 130-136, 242 Seven-segment display units, 363-365
Si photodetectors: irradiance: of a luminescence diode radiation and, 242-244
A of a standard light and, 238-240
luminescence diode radiation, 240-244 Si photodiodes, 415-416
Si phototransistors, 335-339
SI unit (luminance), 26
Silicon-doped GaAs diodes, 113
Simple couplers, 298-301
Sine-wave modulated transmitters, 312-
320
Small surface radiators, 38-40 inverse square law of, 40-41
Sodium vapor lamps, 109
Solid angle, 35-36
Specific detectivity, 156 Spectral absorption factor, 85-86
Spectral bands, 86
Spectral emittivity, 50
Spectral photosensitivity, 13 Spectral radiant efficiency, 224-226 Spectral radiation-distribution units, 29-
32
Spectral
radiation
temperature
T s,
46
Spectral reflectivity, 87-88
Spectral sensitivity: of junction photodetectors, 130-136
measurement of relative, 242
Spectral transmission factor, 85-86
Standard light A, 31-32
color temperature of, 233-235
defined, 80
irradiance of, 238-240
K photometric radiation equivalent for,
69-70
radiant power of, 237-238
Stefan-Boltzmann law of radiation, 47^48
Stilb (sb) unit of luminance, 26
Sun as radiation source, 103- 104 Surface receivers, 38-40
Temperature and spectral sensitivity of photodiodes, 134
Temperature radiators, 45 artificial, 104-105
Thermal calculations, 213-220 Thermal photodetectors, 235-237 Thermoluminescence, 8 Thermopile, 236-238 Three-pole phototransistors, 288-298 Thyristors, 277-279 Tolerance levels of optoelectronic
components, 173-175 Transmission, absorption spectra and, 90-
95 Transmission factor, 85 Transmission links, 401 Transmission ratios of very small photocell
junctions, 179-181 Triacs, 277-279
441
Triboluminescence, 9
TTL integrating circuits, 304-307
Two-pole photodetectors: detector circuits for, 276-288 simple couplers with filament lamps, 298301
Valency band, 6-7 Very-high-pressure mercury-vapor lamps,
109-110 Very small photocell junctions, 176-184 Visible radiation, 3
Voltage converter, 429
Wafer centering, 168-170 Wafer geometry, 170-173 Wavelength and propagation speed, 4-5 Wien displacement law, 48-50
Xenon lamps, 109
Zinc-doped GaAs diodes, 114, 115
442
Selected Titles in McGraw-Hill's TEXAS INSTRUMENTS ELECTRONICS SERIES
DIGITAL INTEGRATED-CIRCU1T,
OPERATIONAL-AMPLIFIER, AND
OPTOELECTRONIC CIRCUIT DESIGNS
Prepared by TEXAS INSTRUMENTS INCORPORATED
Aimed at the circuit designer, this book covers a wide range of important topics in the digital, operations I -amplifier, and optoelectronic areas. The emphasis throughout is on practical applications to help working designers save time and trouble when creating circuit designs from a wide array of semiconductor devices. The first section describes Schottky transistortransistor logic and related subjects. The second section deals with applications and ends with a concentrated chapter on stereo amplifiers. The final section illuminates the theory and applications of optoelectronics with practical considerations of these new semiconductor devices.
DESIGNING WITH TTL INTEGRATED CIRCUITS
Prepared By the IC APPLICATIONS STAFF OF TEXAS INSTRUMENTS INCORPORATED Edited By ROBERT L. MORRIS and JOHN R. MILLER
322 pages 7% x 9%, 330 illustrations
Designed as a ready reference for all electronics engineers, computer designers, systems analysts, and professionals involved in logic planning and
design, this is the first book devoted exclusively to the transistor- transistor
logic family of integrated circuits. It familiarizes the reader with the entire
TTL family, covering not only design philosophy, economics, basic descrip*
tions. and electrical performance of the devices, but many practical appli-
cations of the circuits in digital systems. It will give even the uninitiate a
clear picture of how virtually any logical function may be implemented
TTL using
integrated circuits.
ELECTRONIC POWER CONTROL AND DIGITAL TECHNIQUES
Prepared by TEXAS INSTRUMENTS INCORPORATED
Unlike other available works, this volume is not arranged by device types. Rather, it is divided into two broad application areas: power control and digital processing techniques. In the first section, rectifiers, high-voltage power transistors, programmable unijunction transistors, touch control,
and the control of light through sound are examined. Section Two defines
and groups microprocessors and microcomputers, as well as peripheral
A devices and circuits. wealth of circuit designs that can be used as they
stand or adapted for specific purposes are included.
MOS, SPECIAL-PURPOSE BIPOLAR INTEGRATED CIRCUIT
AND R-F POWER TRANSISTORS
Prepared by TEXAS INSTRUMENTS INCORPORATED
A wide range of semiconductor devices are covered in this volume, all
grouped in broad device-type sections. The chapters are divided into four
--MOS parts
integrated circuits, special -purpose bipolar ICs, field effect
transistors, and radio-frequency power transistors. The first section covers
important areas of memory, terminals, calculators, and interconnecting or
interfacing. Next, the book examines a broad selection of special bipolar ICs. The volume then turns to major areas in which the FF.T is used, and finally explains the use of power transistors in tuned and single 'Sideband
amplifiers.
FIELD EFFECT TRANSISTORS
By LEONCE J, SEVIN
%%X ISO pages,
9%, 137 illustrations
In this concise volume you can get all the practical data you need on the theory, characterisation, and application of field-effect transistors. Its presentation of physical theory h based on Maxwell's equations applied to the motion of charged particles in a semiconductor. From this theory the book develops and uses a lumped linear model or equivalent circuit to describe the interaction of the device and its electrical environment. The book examines the physical behavior of the field-effect transistor, and then explains
in detail the electrical characteristics of field effects in circuit applications.
It also covers the development of the FET as a circuit element in low-level
linear, non-linear, and power circuits.
SILICON SEMICONDUCTOR TECHNOLOGY
By W. R. RUNYAN, (Semiconductor Research and
Development Laboratory, Texas Instruments Incorporated)
288 pages, 7% x 9%, 278 illustrations
Presenting the first extensive coverage of silicon from the semiconductor standpoint, this comprehensive volume fully explains the use of silicon in transistors and integrated circuits. It brings together information on silicon manufacturing, casting processes, crystal growth and orientation, doping procedures, diffusion, electrical and optical properties,, and metallurgy. Hundreds of illustrations, diagrams, and tabular data help clarify such essential information as measuring diffusion coefficients, silicon breaking strengths, melting silicon in molds, distribution of impurities during zone melting, silicon tetrachloride and trich lorosi lane processes, and shape and position of mel ted-solid interface.
SEMICONDUCTOR MEASUREMENTS AND INSTRUMENTATION
By W, R. RUNYAN, Manager of Materials Processing, Special
Products Department, Texas Instruments Incorporated
As anyone involved with semiconductor measurement, analysis or defect detection knows, keeping pace with the latest advances in the field is a formidable task. With the aid of this unique guide, however, you can be brought up to date immediately. The only reference volume devoted solely to semiconductor measurements, it covers the latest information available on the subject, including techniques never published before. Because of the wide variety of techniques listed, the advantages and disadvantages of each measurement can be quickly assessed, and the reader may be stimulated to experiment with new aproaehes.
SOLID-STATE ELECTRONICS
A Basic Course for Engineers and Technicians
By ROBERT G. HIBBERD, Semiconductor-Companents
Division, Texas Instruments Incorporated
164 pages, 7 x 10, 90 illustrations
This unique work presents the principles of semiconductors in an unusual way. It starts with a description of semiconductors and their properties, the p-n junction and the junction transistors, and the characteristics of transistors and basic transistor amplifier circuits. It then describes the manufacture of transistors and other semiconductor materials and devices. The book also discusses the whole family of semiconductor devices and examines the
applications of integrated circuits. Down to earth in presentation, it can be
used by non-technical readers to obtain a working familiarity with the subject.
CIRCUIT DESIGN FOR AUDIO, AM/FM, AND TV
Prepared By THE ENGINEERING STAFF OF TEXAS INSTRUMENTS INCORPORATED
7% 352 pages,
x 9%, 145 illustrations
Stressing time- and cost-cutting approaches, this practical volume is packed
with proven new procedures lor solving typical problems in audio,
TV AM/FM, and
circuit design. With examples and procedures illustrating
the latest available transistor devices, it covers such important topics as
design of IF strips, neutralized and unneutralized amplifiers, IF amplifier
AM/FM FM designs for
and
IF amplifier circuit applications, specific de-
TV UHF VHF sign examples for each major
receiver system,
and
tuners,
TV sync separators, vertical oscillators, video amplifier systems.
automatic
gain control, and more.
McGRAW-HILL BOOK COMPANY
Serving the Need for Knowledge
1221 Avenue of the Americas, New York, N.Y. 10020
At your bookstore, or write Suite 26-1, McGraw-Hill,
0-07-063755-5
TEXAS INSTRUMENTS INCORPORATED
621, 36 OPT SHORT LOAN
t ,11
Digitized by the Internet Archive Recoded by Lura PDF v2.28