olar battery charger - American Radio History

olar battery charger - American Radio History
t__
OLAR BATTERY CHARGER
_.
e
>7
.J
JULY 1966
SF ilìings
Three
.
.I
.
,,Vtrifr
Inductively Controlled
odel Cars
www.americanradiohistory.com
Wireless World
JULY, 1966
RELAYS
KEYSWITCH
I
._
.
:
..
e
e
..
,.
,
e
add versatility
::
to your
circuitry
with the
KEYSWITCH
plug -in unit
-
PULL -OUT
PLUG -IN
SAVE £££ ON MAINTENANCE COSTS
invaluable for sub -circuit fabrication -vibration
proof -completely enclosed- mounted circuit
board -complete with socket and clip-immediate delivery- prices from 15j- to 25;'complete, according to quantity.
always to price
KEYSWITCH
RELAYS
LIMITED
always to specification
CRICKLEWOOD LANE
LONDON
NW2
TELEPHONE:
iJ
WW-001 FOR FURTHER DETAILS.
www.americanradiohistory.com
always on time
GLADSTONE
1152
TELEX: 2627
" Wireless World "
Riffe Electrical Publications Ltd.
Dorset House, Stamford Street,
London, S.E.1
Wireless World
ELECTRONICS, TELEVISION, RADIO, AUDIO
Managing Director:
W. E. MILLER,
M.I.
E. R. E.
Editor -in- chief:
JULY 1966
W. T. COCKING, M.I.E.E.
329
Whither British Television?
330
Remotely Controlled Model Cars
335
Anglo- American Defence Communications
336
Post Office Tower
338
The Diode Transistor Pump
343
A Solar Battery Charger
351
Simple Electronic Organ for the Amateur Constructor
Editor:
H. W. BARNARD
Technical Editor:
T. E. IVALL
ditorial:
.
MILLS
by D. E. O'N. Waddington
by D. Bollen
by T. D. Towers
L. SEXTON, Grad. I.E.R.E.
.
B.
SHORTER,
B.Sc.
Drawing Office:
H. J. COOKE
Production:
D. R.
BRAY
Advertisements:
G. BENTON ROWELL
(Manager)
J. R. EYTON-JONES
© IIHfe Electrical Publications
Ltd., 1966. Permission in writing
from the Editor must first be
optained before letterpress or
illustrations are reproduced from
this journal. Brief extracts or
comments re allowed provided
acknowledgement to the journal
is given.
VOLUME 72 No.
by W. D. Gilmour
357
Transistor
360
New Products Seen at the I.E.A.
369
Recording Level Indicator
373
Computer -" designed " Circuitry
374
Plotting the Root -locus
2
-Metre Converters
by U. L. Rohde
by Murray Ward
by W. Tusting
SHORT ITEMS
346
Automatic Profile- controlled Aerial
347
The U.K. Space Programme
348
Local Radio Plan
356
British Acoustical Society Formed
REGULAR FEATURES
329
Editorial Comment
359
Literature Received
334
The Month's Conferences
and Exhibitions
370
Letters to the Editor
H.F. Predictions
377
Books Received
342
World of Wireless
378
News from Industry
347
380
Personalities
Real and Imaginary
349
by
"Vector"
7
PRICE:3s.
(3rd Monday of preceding month). Telephone: Waterloo 3333 (70 lines).
Telegrams/Telex: Wiworld Iliffepres 25137 London.
Cables: "Ethaworld, London, S.E.1."
Annual Subscriptions: Home £2 6s Od. Overseas: £2 15s Od. Canada and U.S.A. %.00. Second -class
PUBLISHED MONTHLY
FIFTY -SIXTH YEAR
PUBLICATION
OF
mail privileges authorised at New York N.V. BRANCH OFFICES: BIRMINGHAM: 401,
Lynton House, Walsall Road, 22b. Telephone: Birchfield 4838. BRISTOL: 11, Marsh Street, 1.
COVENTRY: 8 -10 Corporation Street, Telephone: Coventry
Bristol 21491,2.
GLASGOW: 123 Hope Street, C.2. Telephone: Central 1265 -6.
MANCHESTER:
25210.
NEW YORK OFFICE U.S.A. 300
260, Deanseate, 3. Telephone: Blackfriar< 4412.
East 42nd Street, New York 10017. Telephone: 867 -3900.
Telephone:
www.americanradiohistory.com
WIRELESS WORLD
56
5
JULY, 1966
new high frequency
odUdoom EDUccauoCUT
transistors for radio and audio
Three new A.F. silicon planar epitaxial transistors have
been designed to meet a wide but important range of
application requirements.
BC 107
for audio driver circuits
The high voltage rating (45V), low bottoming characteristics, and high gain of this transistor make it ideal for
audio driver applications.
BC 108
for amplifier circuits
High gain and the ability to operate down to very low
currents make the BC108 highly suitable for a wide
range of audio applications. It is recommended for firs-.
audio stages of radios and the input and amplifying
stages of record players and other audio equipment.
BC 109
for low noise pre -amplifier
circuits
An outstanding new transistor for use in audio amplifier
input stages where extremely low noise is an absolute
requirement. Low noise has been coupled with very high
gain enabling the designer to meet the most stringent
requirements in tape recorders and high performance
audio amplifiers.
Two high frequency radio transistors, the BF184 and
BF185, offer low cost and first -class performance in
every h.f. stage found in modern a.m. and a.m. /f.m.
radios. They also have closely limited spreads -a
valuable design feature.
BF 184
for high gain i.f. amplifiers
with low circuit costs
Outstanding features are its high gain and the high input
impedance associated with excellent high frequency
characteristics. These, together with high d.c. current
gain, make this transistor particularly suitable for all i.f.
amplifier stages in car radios and both mains and battery
operated a.m. /f.m. receivers.
BF 185
for low noise applications
up to 100 Mc /s
The very low noise figure over a wide range of source
impedance makes the BF185 invaluable in all low noise
applications up to 100Mc /s. It is recommended for the
self -oscillating mixer stage of f.m. receivers and the first
stages of car radios and f.m. receivers. It is also ideal for
portable receivers, in which, due to its narrow current
gain spread, components may be saved and circuit
design simplified.
These transistors are recommended for use in all new equipment designs and Design Engineers can obtain full informaTorrington Place
London, W.C.1
Mullard House
Entertainment Markets Division
tion from Mullard Limited
Telex: 22281
Telephone LANgham 6633
:
:
Oó
Mullard planar
17. C88
WW--090 FOR FURTHER DETAILS.
www.americanradiohistory.com
Wireless World
ELECTRONICS,
TELEVISION,
RADIO,
AUDIO
Whither British Television?
THIS journal has been an ardent supporter of the 405 -line television standard from its
inception in 1936 and campaigned for its retention rather than a change to 625 lines
which we considered gave only a marginal improvement in definition. When, however,
in the light of the 1960 recommendation of the Television Advisory Committee
(endorsed by the Pilkington Committee) the Government decided in 1962 that all
television in the U.K. should eventually be on 625 lines with negative modulation and
f.m. sound, we accepted the inevitable; or did we? Over the past four years there has
been a good deal of correspondence in our pages on the problems of propagation in the
u.h.f. band; on the pros and cons of 625 (50 c /s) and 525 (60 c /s); and on the question
of an international standard.
Now, once again, the question of 405 lines is raised by a correspondent who advocates
the introduction of colour on this standard. The technical arguments have been aired
again and again and despite the Government's decree there is a strong and increasingly
vociferous band of engineers both in the broadcasting organizations and in industry who
consider that there is ample justification in the light of experience with the BBC-2 monochrome u.h.f. service for the retention of the 405 -line standard -and furthermore the
introduction of colour on this standard forthwith.
We must not lose sight of the fact that some of the more vociferous advocates of the
introduction of colour on 405 lines are in the I.T.A. and programme contracting companies and it may therefore be claimed that they have a vested (mot juste) interest.
We would remind readers, however, that in 1959 the B.B.C. asked the Postmaster General for permission to introduce colour on its 405 -line service but was refused.
It may be asked why, if they so desire, the I.T.A. programme contractors do not start
colour transmissions as part of their existing programmes? The answer is, of course,
the P.M.G. has decreed that colour television in the U.K. is to be on 625 lines and as
they are licensed by him to operate they cannot go against his ruling.
It seems that the introduction of colour television in this country is bedevilled by
political, economic and technological intrigue. We are not qualified to speak on the
political and economic climate, but what of the technological? We do not need to go
over all the arguments for and against the various systems (a letter in this issue brings
to light a deficiency in PAL which has not previously been aired) although we wish
it were permissible for us to make public a report prepared by the B.B.C. engineering
department for the T.A.C. in which three systems N.I.R., PAL and N.T.S.C. are
compared. This report is said to be confidential although we know it has been circulated
to some organizations in this country and abroad.
VOL 72 NO 7
JULY
\t
lr
1966
Irrespective of what colour system is finally adopted, whether independently in the
U.K. or in concert with our European neighbours, we in this country have still to decide
when colour will be added to BBC -1 and the I.T.A. services. Are viewers of these programmes to be denied it until some unforeseeable date when these services will be
transferred to the u.h.f. band? If, as seems likely, the P.M.G. stands by his decision
for colour on 625 lines both BBC -1 and I.T.A. should be given the opportunity of
sharing with BBC -2 the time devoted to colour transmissions. Our associate journal
Electrical and Electronic Trader says that such a compromise "has much in its favour
and could well be the initial answer to a difficult problem, if both sides are prepared
to collaborate in making it work."
329
:ELESS WORLD, JULY 1966
www.americanradiohistory.com
Remotely Controlled Model Cars
System uses inductive coupling with al. control signals. One
model is steered by a conventional actuator, the other by differential drive to the rear wheels
THE remote control of small, free, model cars presents
certain differences from the control of other free
models. First, because of the small -diameter wheels
used, a smooth surface on which to run is essential, and
this precludes outdoor running except under very special
conditions; the area available will therefore be measured
in (tens of feet)" at the most, and the maximum distance
over which control will be required will be of the same
order. Secondly, the control accuracy required will be
greater, as running will generally take place on a track
bounded upon both sides, thus reducing the permissible
margin of error. Thirdly, because cornering in such a
small area will be frequent, the control system must be
quick in operation and proportional. Probably because
of these limitations, model cars have received less attention from control enthusiasts than model ships or aircraft. However, the control system to be described is
simple, using only audio frequency techniques, and is
well within the capabilities of the amateur constructor
to assemble and use.
Fig. I. Plan view of cable loop surrounding an area.
coil on the vehicle is mounted vertically.
CONTROL
AMPLIFIER
i
Fig. 2. Pian
of cable delineating
a
track.
The
pick -up
By
W. D. GILMOUR,
B.A., A.M.I.E.E.
W. D. Gilmour graduated at St. John's College, Oxford,
and after wartime service as T /Captain in R.E.M.E. was
employed by St. Dunstan's on guiding devices and
reading machines for blind people. He then went to
Westinghouse Brake & Signal Co. Ltd. where he worked
on semiconductors ; then to Marconi International
Marine Communication Co. Ltd. and since 1954 has
been with EMI Electronics Ltd. on applied research.
Mr. Gilmour, who is 43, and has written for Wireless
World on several occasions, is the author of the Iliffe
book " Electronic Equipment in Industry."
Radio control over these short distances is quite unnecessary. Instead use is made of inductive loop
signalling, in which the area within which the cars are
to run is encircled by a cable carrying the audio frequency control tones. This cable acts as a large single
turn solenoid, thus producing a magnetic field as shown
in Fig. 1. If a coil, mounted on the model as shown,
is introduced into the field, a voltage will be induced in
it from the field, and this may be amplified in a simple
audio frequency amplifier, detected, and used to perform the necessary control functions. If running on a
track is required, the cable may be laid at the edge of
the track, as shown in Fig. 2; this arrangement gives
a very uniform field across the track, whereas in the
single loop arrangement the field is stronger near the
cable.
Each car requires a minimum of two channels, to
control its speed and steering, and a third channel is
often convenient to act as an automatic gain control
channel so that variations in strength of the picked -up
signal as the car moves about the track can be automatically allowed for. Naturally this last channel is
common to all the cars on the track. The number of
cars that can be simultaneously controlled will be set
by the difference in frequency between the channels
necessary to avoid interaction. It has been found that,
using well designed tuned inductance -capacitance drcuits on the cars, a channel spacing of 500 c/s with
frequencies ranging from 800 c/s to 4,000 c/s proves
satisfactory; and thus seven channels are available,
allowing three cars to be simultaneously and independently controlled.
A common audio amplifier for all the control tones
(Fig. 3) feeds signals to the cable. It is important that
this amplifier should not overload when peaks from all
tone generators occur simultaneously in one polarity,
and thus the amplifier must not distort for input signals
seven times the amplitude of a single input signal. The
current in the cable for a single tone should be about
0.5 A, and thus the peak current in the cable will be
about 3.5 A. The inputs to the amplifier (Figs. 8 and 11)
will be dealt with later.
WIRELESS WORLD, JULY 1966
330
www.americanradiohistory.com
In the model car, a
SUPPLY 10
mon amplifier amplifies TONE GENERATORS
a signals picked up from
coupling coil. A suitt
aiole design is shown in
Fig. 4, but almost any
small audio amplifier could
be used. Care must be
FROM
GENERATORS
taken to design the amplifier to work over a large
I,000µ
váriation in supply voltage,
as the model's battery discharges, and steps must
also be taken to eliminate
MAIN GAIN
interference
from the
(PRESET)
motors on the model -this
is most easily achieved by
limiting the high frequency
Fig. 3. Circuit of control
relpponse of the amplifier.
A typical detector is included in Fig. 4. The
series tuned circuit,
and L, filters out the desired
frequency (see table for values), and the transistor Tr 6
acts as an infinite impedance detector, delivering a direct
control voltage at the emitter, which may be amplified
and used to control the desired function. An a.g.c.
arrangement, Tr 4, is also included. If a negative voltage
is applied to the base of this transistor it will conduct,
thus shorting out some of the amplified signal at the
output from the high impedance amplifier, Tr 3. The
emitter follower, Tr 5, delivers a low impedance output
to the filters, thus reducing mutual interaction.
The pick -up coil in Fig. 4 is a short length of
Ferroxcube rod (type A2 or A4 material) wound with
1,500 turns of 48 s.w.g. wire.
A non -reversing drive circuit for the propulsion motor
is shown in Fig. 5. The output from the detector is
amplified in Tr 1 and applied to the emitter follower,
Tr 2, with the motor as load. A drive using a tapped
R11
c
10V
680
24V
°
TO
CABLE
amplifier (common to all tone generators) feediag the cable.
C
TUNED CIRCUIT DATA
In the car receiver L, consists of 2,000 turns of 48
s.w.g. single enamel coated wire wound on a Mullard
Vinkor Type LA 2702 core. The tuning capacitances,
C,,, are then as follows
:
-
Model and function
Value of
Frequency
Mk. I, a.g.c.
Mk. I, steering
Mk. I, speed
..
..
..
..
Mk. 2, full left
Mk. 2, full right ..
..
..
..
..
..
C
(,,F)
(c /s)
..
..
800
1,300
1,800
0.056
..
2,300
2,800
0.0068
0.0046
..
..
0.021
0.011
In the control panel tuned circuits (Fig. 8), similar L
and C values to those in the car can be used.
AMPLIFIER
DETECTORS
A
Rq
-3V
Tr5
R2
MAS20
IE
I0k
FOR VALUES
SEE TABLE
T
O.005µ
-
FSEEWTABLE S
Hi100
PICN-UP
COIL
i'
Trl
Tr2
Tr3
MAS20
MAS20
MAS20
Tr4
MA520
O.05µ
CB
0lµ
O405µ
CI4
aµ
+2V
A.G.
D.C.
STABILIZATION
C.
SIGNAL TO
OTHER DETECTORS
(STEERING 6 PROPULSION)
Fig. 4. Receiver circuit for the cars. D.C. control voltages are taken from the emitters of the steering and propulsion control detectors corresponding to Tr6.
WIRELESS WORLD, JULY 1966
331
www.americanradiohistory.com
the circuit of Fig. 5 controlling the propulsion motor.
The circuit of Fig. 6 controls the steering, driving a
Graupner Bellamatic II actuator, which is equipped
with heavy self-centering springs so that the steering
motor is allowed to stall against these springs for small
angles of steering. At full lock, an integral slipping
clutch prevents damage to the steering motor. The
maximum scale speed of this model is about 100 m.p.h.,
and an adjustment of steering from lock to lock takes
about 0.25 second. Five DEAC type DK450 nickel cadmium rechargeable cells give a running time of about
forty minutes.
RI
2.2k
FROM
DETECTOR
Fig. 5. Circuit of non -reversing drive for cars.
FROM
DETECTOR
0071
OC71
Fig. 6. Reversing drive circuit for cars.
battery to secure reversing is shown in Fig. 6. When
the amplitude of the control tone is low the control line
feeding the common base drive of the complementary
pair of transistors, Tr 3 and Tr 4, will be negative, and
Tr 3 will be conducting. When the control tone is at
maximum, the control line will be positive, and Tr 4
will conduct. At intermediate levels of the control tone
the common emitter point of Tr 3 and
Tr 4 will be intermediate in voltage.
PROPULSION MOTOR
Thus proportional bidirectional con-
Simplified model (Mk. 2). -The steering actuator for the
first model, Mk. 1, is expensive and occupies a fairly
large volume. In the second model each rear wheel is
driven by a separate motor, with its own control tone,
and steering is effected by energizing these differentially.
A third, castor mounted, wheel in the nose automatically
follows the changes of direction resulting from the
differences in speed of the driving wheels. The two
visible front wheels are dummies and do not touch the
ground. This technique allows 1 /32nd scale models to
be easily built.
In the receiver, a.g.c. is not used, but otherwise the
circuit of Fig. 4 is used, together with two driving circuits, Fig. 5, modified as shown in Fig. 9 to provide
rapid deceleration of a de-energized motor. Normally,
when Trl is conducting, its base is slightly positive
relative to its emitter, and Tr2 is accordingly cut off.
When drive is removed, however, the base of Tr2 is
negative with reference to its emitter, which is held positive by the back e.m.f. from the motor. Tr2 is thus
bottomed, and the armature of the motor is effectively
short circuited, and thus rapidly decelerated.
For its power supply the Mk. 2 model uses a nontapped battery of 3.75 V nominal output, and this suits
the 3 V max. requirement for the drive motor. In this
case the +2 V line is earthed and the receiver (Fig. 4)
and driving (Fig. 5) circuits operate between earth and
-3.75 V. The chassis of the car is left floating.
The motors are Graupner type 03/60, geared 60:1,
and give a maximum scale speed of about 60 m.p.h.
The steering is naturally effective only when the vehicle
is in motion, and reversing is not possible because of
the pivoted wheel, but within these limitations the vehicle
handles well. A rear view is shown in Fig. 10. A
BATTERY
trol of the motor is achieved.
BRASS PLATE
PLASTIC PLATE
PT
/
Model with controlled steering (Mk. 1).
-The first model to be described
(Fig. 7) uses two control channels for
steering and speed (non -reversing) respectively, together with a third a.g.c.
channel. The tone generators are
identical, except for the value of the
tuning capacitance (C, in Fig. 8), and
have proved very stable in frequency.
The amplitude of the output is controlled by the potentiometer, RV,. The
frequencies used are 800 c/s for the
a.g.c. tone, 1,300 c/s for the steering
control, and 1,800 c/s for the speed
control.
In the vehicle, which was based on a
1/24th scale model of a sports car,
the receiver of Fig. 4 is used with
LTUNED-CIRCUIT
INDUCTORS
F
E
BUSH
FORK
STEERING
PICKUP COIL
MOTOR
Fig. 7. Model with conventional steering by actuator. The pick -up coil is just ahead cf the
front wheels (right). The wire structure in front of them is a bumper. The large battery is
mounted between the rear wheels.
WIRELESS WORLD, JULY 1966
332
www.americanradiohistory.com
8V
RV!
5k
STEERING
Pb
TO OTHER
I00k
TRANSISTOR BASES
FILTER
TO
X'
TO
AMPLIFIER
WINDINGS E.
VALUES, SEE TABLE
AMPLIFIER
OR
C
Fig. 8. Audio frequency control tone generator.
Fig. 11. Control tone generator for simplified model in Fig. 10.
Capacitors C are about 0.01µF and can be found by experiment, as
can the L and C values of the filter (using a Vinkor core for the
inductor).
AMPLIFIER
OF Fig.
Trl
at 2.3 kc /s, for straight ahead at 2.55 kc /s, and for full
right at 2.8 kc /s. The speed of the model is controlled by
the amplitude of the control tone. The 2.55 kc /s filter
ensures that the transmitted voltage is increased to overcome the reduced response of the model's filters at the
mid -frequency.
The transistors in the models are selected for small
size. Any small transistor will serve in the amplifier,
and one of the new plastic encapsulated types such as
the Texas Instruments 2N3702 would be very suitable.
Tr 6 in Fig. 4, however, must be selected for low leakage
current, and should therefore be a silicon type, such as
a 2N3702 again, or the HT 100 as shown in the circuit
diagram. The transistors in the control panel are normal
types.
5
Fig. 9.
Braking circuit.
The control box contains appropriate equipment to control both models, i.e. three circuits of Fig. 8, one of
Fig. 11, and one of Fig. 3, together with a normal power
battery of three DEAC type
225 DK cells gives a running time
before recharging of about 60
minutes.
For the control tone generator for
the second model, although it would
be quite possible to use two generators of the type shown in Fig. 8, with
the potentiometers RV, ganged
differentially, a simpler scheme is
shown in Fig. 11, where a voltage
controlled oscillator generates a frequency
between
2.3 kc /s
and
2.8 kc /s, which is passed via a filter
tuned to 2.55 kc /s to the common
amplifier. The two channels in the
model are tuned to 2.3 kc /s and 2.8
kc/ respectively. Thus for a full left
turn, the control tone is transmitted
(Left) Rear view of the simplified
model with steering by differential drive to
rear wheels. The battery fits in the circular
holder amidships. In this model the drive
motors are not equipped with spring mounting.
(Right) Front view of the Fig. 7 model.
Fig. JO.
WIRELESS WORLD, JULY 1966
333
www.americanradiohistory.com
supply delivering -24 V at about 1.5 A. A constantcurrent charger is also provided for recharging the
DEAC batteries used on the models; this provides 22 mA
for the 225DK battery used on Mk. 2, and 45 mA for
the DEAC DK450 battery used on Mk. 1. Recharging
times are about 10 -15 hrs from fully discharged, and
therefore a switch is provided so that the electronic
circuits can be switched off if the control unit is needed
for charging only. The accelerator and steering controls
are fitted with knobs that can be distinguished by touch.
Operation. -The two models described have now run
for several tens of hours and have performed satisfactorily. The precision of control is limited by the operator rather than by the vehicles themselves, and it takes
several hours practice to become proficient in handling
the cars.
NOTES ON MECHANICAL CONSTRUCTION
Mk. 1 model. -The general arrangement of the Mk. 1
steering system may be followed from the accompanying
sketch and Figs. 7 and 10. The Graupner Bellamatic
II steering motor is mounted as shown. To the plastic
plate supplied with the mechanism is attached a brass
plate with a fork, which engages with the spring loaded
so that the whole tube can absorb shocks, etc. (As a
three -wheeler suspension is used, springing is not
primarily necessary to hold the wheels in contact with
the ground on rough surfaces.) The Graupner motors
are equipped with integral 60: 1 reduction ratio gearheads,
and the road wheels are fitted directly to the gearhead
output shafts. The front wheel is mounted as shown
below. In the prototype the height of the shaft, and
FREELY ROTATING
SHAFT`
TOP COLLAR
_
PTFE COLLAR
FREE TO SLIDE BETWEEN LIMITING STOPS AA,
SPRINGS REMOVE SHOCKS FROM ACTUATOR
BRASS
STEERING FORK
UPPER
TRACK ROD
POLYTHENE
BUSH
(THREADED)
CHASSIS
'-
NUTS
BOTTOM COLLAR
-v-- BRAZED
hence the clearance given to the dummy front wheels is
adjustable, as is the effective length of the castor arm.
The chassis itself is just a sheet of 12 s.w.g. Dural, bent
up at the sides for stiffness and at the ends to pick up
the mounting points of the model body.
All the mechanical components may be obtained from
RipMax Ltd., 39 Parkway, London, N.W.1 (Tel : Gulliver
1818).
LOWER
-
THE MONTH'S CONFERENCES
AND EXH181TIONS
KING PIN
TRACK ROD
Further details are obtainable from the addresses
BEAM AXLE
KING PIN
BEAM AXLE PIVOT PIN, FIXED TO CHASSIS
WHEEL AXLE
upper track rods, which are joined at their extremities to
the firm lower track rod, supplied with the model. The
whole front axle assembly is mounted on a single central
pivot, so that it can accommodate itself to rough surfaces.
The model is driven on the near side rear wheel only
by a Mighty Midget motor (made by Victory Industries
(Surrey) Ltd., of Guildford, Surrey).
Mk. 2 model. -The two propulsion motors (Graupner
type T03/60) are mounted back to back in a brass tube,
which is mounted onto the chassis by means of springs,
SUSPENSION
CHASSIS
BRASS TUBE
LONDON
Imperial College
4 -6
The Economics of Automated Materials Testing
(Inst. Phys. & Phys. Soc., 47 Belgrave Sq., London, S.W.1)
July
r -' titi: -,.<,-;e
N41
GEARED MOTOR
Ng 2
GEARED MOTOR
Olympia
July 6 -12
Ships' Gear Exhibition
(Brimtex Ltd., 3 Clements Inn, London, W.C.2)
Imperial College
July 11 -14
Applications of Thin Films in Electronic Engineering
(I.E.R.E., 8 -9 Bedford Sq., London, W.C.1)
BRISTOL
The University
Spectroscopy and Automation
(Inst. Phys. & Phys. Soc., 47 Be :grave Sq., London, S.W.1)
GLASGOW
University of Strathclyde
July 7 -8
Chemically Grown Surface Films
(Inst. Phys. & Phys. Soc., 47 Be :grave Sq., London, S.W.1)
LEEDS
The University
July 10 -22
Microwave Techniques (Summer School)
(I.E.E., Savoy Pl., London, W.C.2)
ABROAD
Warsaw
July 3 -8
International Measurement Conference
(IMEKO Secretariat, P.O.B. 457, Budapest 5, Poland)
July 7 -8
July
.OUTPUT SHAFT
in parentheses
8
-24
British Industrial Exhibition
(Industrial & Trade Fairs, 1 -19 New Oxford St., London, W.C.1)
San Francisco
July 11 -13
Electromagnetic Compatibility
(A. Fong, Hewlett- Packard Co., 1501 Page Mill Rd., Palo Alto,
Cal.)
July 11 -14
Seattle
Aerospace Systems
(T. J. Martin, 3811 E. Howell St., Seattle, Wash.)
WIRELESS WORLD, JULY 1966
334
www.americanradiohistory.com
Anglo- American Defence Communications
USING SPACE "JUNK" AND
ARECENT visit to the Signals Research and Development Establishment at Christchurch, near Bournemouth, the site of the first of three ground stations
being built by the U.K. for the U.S. Initial Defence Corn munications Satellite Project, disclosed some interesting
facts regarding the Anglo-American experimental defence
communication project. Although a cloak of security covers
some of the research work being carried out, especially as
regards detailed technical information, one intriguing aspect
discussed was the use of the moon and space " junk " as
passive communication satellites. Space junk is the term
used for the various "dead" objects in orbit round the earth
including for example defunct satellites and the final booster
stages of rockets. There are about 900 in orbit but apparently
only about 200 are suitable for reflection of signals.
Flashing
Paths of signals transmitted in a beam width of 0.25°
from the ground station obey to a very close approximation
the laws of geometric physics. It was disclosed that the
received signal power was normally in the order of pW,
but if the object -which is continually turning-happened
to be in a particular orientation during the reflection of a
signal, a phenomena termed " flashing " occurred. This
causes a mirror -like reflection of the signal and a large increase in the power of the received signal is evident.
Apparently, this phenomena has been investigated by using
specially shaped objects (possibly a spheroid) to give continual flashing but pitting of the surface caused by meteorites
prevents full use of the optical effect.
The ground station
The ground station, designed and constructed by the
Marconi Company in 12 months and believed to have cost
about £1.5M, will work in conjunction with six or eight nonsynchronous active satellites (life expectancy of three to four
years) to be launched from Cape Kennedy later this summer.
However, the station has already proved itself by bouncing
200 message elements a second off the moon or space junk.
lorne idea of the directional accuracy of which the equipment is capable is given by the fact that at an altitude of
about 400 miles it is possible to detect a reflecting surface of
about 0.1m2.
The ground station has been designed to be fully transportable and it takes only about 48 hours to assemble.
Each station comprises a fully steerable dish aerial with a
number of separately mounted containers housing the transmitter, its power supply and cooling system and control and
computer equipment. The aerial system consists of a 40ft
diameter paraboloid, with a focal length of 12 feet and a
surface accuracy of better than 10 thousandths of an inch
within 20ft of the centre, increasing progressively to 25
thousandths of an inch at the edge.
The base of the support structure contains a circular
track, running on six bogie units rigidly attached to the
ground and which allow the complete aerial to be driven in
azimuth. The aerial can be turned in elevation through an
arc of 320° and in azimuth through an arc of 350 °. Cables
NON -SYNCHRONOUS SATELLITES
are run through the two planes of rotation without the use
of slip rings. The transmitter power supply and cooling
system are mounted in two units in the base of the aerial
structure and they rotate in azimuth with the aerial. The
transmitter is a five -stage, cavity klystron, driven by a travelling -wave tube amplifier. The output from the klystron
passes through a waveguide to a rotating coaxial joint in
the elevation mounting, and then to a horn feed assembly
at the centre of the dish aerial. A cassegrain feed system is
used, with a hyperbolic sub-reflector mounted at the focus
of the dish. Energy from the feed horn is then reflected from
this hyperboloid on to the main dish. Four separate wave guide horns in the aerial system are fed into a waveguide
network which produce a signal from the sum of the four
input signals. Two difference signals, are also derived from
pairs of horns in two planes at right angles and the dish is
automatically controlled to direct itself until 0° phase occurs
between the two horns. These three outputs are then
amplified by two -stage cryogenic parametric amplifiers
operating at 20 °K
253 °C) and mounted directly on the
backing structure of the dish aerial together with mixer
stages for the three channels.
Intermediate frequency signals of 70 Mc /s are then routed
by coaxial cable to the control container, where after further
amplification and demodulation, the final communication
output is produced. The outputs of the two difference
channels are used in a servo system to provide automatic
tracking, i.e. the dish locks on to the satellite after the
initial acquisition. Tracking information derived from the
receiving system, is translated into suitable co- ordinates to
control the motors which drive the aerial in these two planes
of rotation as required. The dish together with transmitting,
receiving and control equipment is housed in a double walled, inflated radome approximately 60ft in diameter and
60ft high and is strong enough to support a weight of about
two tons.
The station has been designed to operate as part of a larger
system of satellite links, and a Marconi Myriad computer
performs a major function in ensuring continuity of operation in multiple satellite systems, particularly during the
hand -over of high -speed digital data transmission from one
random -orbit satellite to another. All traffic will pass through
the computer which will insert a variable path delay, depending on the position of the satellite, to produce a constant
total path delay, irrespective of the actual distance of the
satellite from the ground station and it will be possible to
transfer high -speed digital data traffic from one satellite to
another without any break in the transmission.
The computer may also be used to control the acquisition
of new satellites appearing over the horizon. In this role,
orbital prediction data would be supplied for each satellite,
and fed into the computer in the form of a paper tape from
which initial information for directing the dish would be
required. When signals have been received the system would
revert to the auto -tracking mode, using error information
derived from the horns.
In addition to the earth station itself an experimental
system-control and analysis centre has been set up at
S.R.D.E. to co- ordinate the operation of the satellite network.
WIRELESS WORLD, JULY 1966
(-
335
www.americanradiohistory.com
POST OFFICE TOWER
The fact is that the restaurant
and public observation galleries were
an afterthought which necessitated
modifications to the original plan in
order to accommodate them above the
aerial galleries which initially were to
top the tower.
Primarily, of course, the tower was
ROM the publicity given in the lay
level.
to the recent official opening
F press
of " The G.P.O. Tower " it would
appear that it has been built primarily, if not exclusively, to give visitors
a bird's -eye view of London while
enjoying a meal in the slowly revolving restaurant some 500 ft above street
IIIIIIIIIIOIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIi1110 101
MAIN CITY -CITY BOTH WAY LINKS!
FOR 960 OR 1,$00 TELEPHONE CIRCUITS
PER MICROWAVE CARRIER
40_
-
.
MAIN CITY -CITY BOTH -WAY LINKS
FOR 615-LINE COLOUR TELEVISION
O
SPUR LINKS TO OS LINE
TELEVISION BROADCAST TRANSMITTERS
INVERNESS
ABERDEEN
O
I
DUNDEE
DUMBARTON/
/
O
URRI
The new landmark on the London skyline.
On the right of the new 580 ft Tower is the
150 ft lattice mast and paraboloids which it
EDINBURGH
GLASGOW
replaces.
Q SELKIRK
es"
i -LONDONDERRY
NEWCASTL
CAR LI SLE
BELFAST
NTH. YORKS
RKS
EEAyIISSII
LEEDS
Q
MANCHESTER
EAST LINC L
DUBLIN
1
NORW CH
STAFFORD
Q
t
'/ O
PETERBOROUGH
BIRMINGHAM
`
/ MENDLESHAM
AEN PLWYF
e
s.
WEST WALES
V
:::..
\
BRISTOL
CARDIFF
LONDON
i."24:Ty_KENT
HAMPSHIREO_
.
SOUTH
.0
EXETER
CARA
ON H
LLO.
STOCKANDO DORSET
O
H
LL
STH. DEVON
PLYMOUTH
* DVOVER
SUSSEX
..........Q
SOUTHAMPTON
ISLE OF WIGHT
EAST
SUSSEX
UROPE
trunk to ephonc and television traffc the Post Offce plans th's m'crowave
network for 1970. Relay stations at intervals of 25 -30 miles are being built along each route.
To meet expected
conceived as the hub of what will
eventually be a national radio telecommunications network which by
1970 will be able to handle as many as
150,000 simultaneous telephone conversations and up to 40 television
channels.
Below the galleries for the horn and
paraboloid aerials are 15 floors on
which are housed the radio and line
equipment. The floor immediately
below the aerial galleries is used
mainly for routing the waveguides and
cables from the aerials to the equipment floors below on each of which are
radio transmitters and receivers in a
descending order of frequency from
11Gc /s to 2Gc /s. It is interesting
to record that with the advent of
solid -state equipment it has been
necessary to provide space on one of
the floors for 24- and 50 -volt batteries.
The first link to be brought into
service was that to Birmingham which
operates in the 7 Gc /s band and
initially provides 5,400 telephone
channels and three 8 Mc /s television
channels.
In addition to providing the focus
for the country's telephone network
the Tower also houses the London
switching centre for the television network. Although both the B.B.C. and
I.T.A. are responsible for providing
WIRELESS WORLD, JULY 1966
336
www.americanradiohistory.com
and operating their own transmitters
the links between studios and transmitters and between the stations in
each network are provided by the
Post Office. Because the B.B.C. network is essentially national the
amount of switching from region to
region is small, but in the case of
L.T.A., each programme contractor is
free to sell or buy programme material
from any other contractor and therefore the number of changes in inter connecting links required by the
I.T.A. is considerable.
one of the
horns,
supplied by
G.E.C., on the aerial
Fixing
gallery.
Control room of the
Television Switching
Centre in the Tower.
V
4,4sat.t,
woomallagt
11,
,.,:
,
The Post Office, through the net-
work switching centre, provides
circuits for organizations who perform specialized work (such as videotape recording) for the various pro gramme companies. It may not be
realized that there are permanent
cable links from the London switching centre to over forty studios,
centres of entertainment and programme contractors' offices in the
area. The input to and the output
from all these points are routed
through the switching centre. In fact,
the new control room in the London
switching centre is the focal point of
well over 100 incoming and outgoing
vision links, in addition to which there
are 250 sound and 300 control lines.
One little known aspect of the work
of the switching centre is its use in the
educational television scheme introduced by the Greater London Council. It will provide over 1,300 schools
in the G.L.C. area with six educational television channels.
r,,;._.
.
+N
'°
i
a/.-
VC%a1e:
eï..1.°'òi
t5:'
ÁPye
microwave (6 -8
GO)
transmitter and
receiver as supplied for
the Tower.
411
Typical racks of transmitting and receiving
gear on one of the equipment floors (G.E.C.
photograph).
WIRELESS WORLD, JULY 1966
337
www.americanradiohistory.com
The Diode- transistor Pump
PRINCIPLE OF PUMP
OPERATION, APPLICATIONS AND A USEFUL CIRCUIT
By D. E. O'N. WADDINGTON, A.M.I.E.P..E.
The principles of the diode and diode- transistor
pump are outlined. A diode- transistor pump
and limiter circuit using silicon transistors and
developed by the author is given. The circuit
may be used as a discriminator in f.m. receivers
giving very low distortion, and requiring only
10 mV for proper limiting, or as a frequency
meter, the response extending up to around
1 Mcfs.
Other applications of the diode -transistor pump are also discussed.
very small proportion of the supply voltage.) When S,
is moved to position 2, D1 is reverse biased and D2
conducts allowing the charge on C, to be shared between
C2 will now charge up to a voltage
C, and C2.
e= VC,I(C,- +-C2). When S, is returned to position 1,
C, is again charged to VC,. When S, is again moved to
position 2, charge sharing takes place once more but, as
C, already has a charge of VC,C2/(C, +C2), the total
Thus
charge to be shared is VC, +VC,C2i(C,+CD.
the voltage across C2 will now
be:
VC,
C,+-C2
C2
l(1+
C,+ Cg)
A third cycle will produce a voltage of across C2
VC,
r
C2
Ll+
C,+
C2
CI +C2
e
FOR many years the simple diode pump has been
known and used in applications where an untuned
frequency to voltage converter or frequency divider has
been needed. With thermionic valve circuits it has,
generally speaking, proved adequate but with the lower
supply voltages usual with transistor circuits, it no longer
supplies a sufficiently linear output. This has led to a
fuller appreciation of the action of the diode pump and,
hence, to the development of several diode- transistor
pump circuits which out -perform the original circuit.
The diode pump
In its most basic form, the pump principle is as follows
(see Fig. 1). The first half cycle of the input signal causes
the pulse standardizing circuit to produce a pulse containing a fixed amount of energy, e.g. CV. This energy,
or a proportion of it, is then transferred either automatically (or by using the second half cycle of the input
signal to operate a switch) to a reservoir where it is stored.
For use as a frequency-to- voltage converter, a discharge
path is added which drains the reservoir at a rate dependent upon the level of energy stored in it. Thus, the voltage developed across the discharge path will be directly
proportional to the rate at which the pulses are generated,
i.e. the input frequency.
The practical implementation of this principle is very
simple. In the diode pump (see Fig. 2), the pulse standardizer is a diode and capacitor (D1 and C,), the energy
transfer mechanism is a diode D2, the reservoir is a
capacitor (C2) and the discharge path is a resistor (R).
The full action of this circuit may be analysed as follows.
The input square wave is simulated by means of S1.
When the switch is in position 1, D2 is reverse biased
but D1 switches on allowing C, to charge up to the supply
voltage thus storing a charge of VC,. (For the purposes of
simplicity, the effect of the forward voltage drop across
the diodes is neglected. In practice, this voltage is only a
¡
+
C2
of:-
121
C, +C2 )
Hence it may be shown that the voltage across C2 after
n cycles will
en=
be:-
rl
VC,
C, + C2
or
C2
C1
en= V
+ C2
)2-}-(
+C2
C2
-;-(
C1
...-}-
C2
,n
11
C, +C2/
-( C,+
C2
1
C2
[
From this expression it is readily seen that the increment in output voltage, due to each input cycle, is less
than the previous one by an amount proportional to the
charge already present on C2. Thus, the relationship
between the number of input cycles and the output
voltage will be non -linear.
When the resistor, R, is connected across C2, the circuit becomes a frequency to voltage converter. It may be
shown' that in this case
e
=fRC,(V -e)
C, R
or e=1/1f
1
-fCR
a non -linear relationship similar to that obtained above.
However, if e is made a small proportion of V, the equation
PULSE
STANDARDIZER
t
ENERGY
TRANSFER
MECHANISM
RESERVOIR
DISCHARGE
PATH
1
OUTPUT
1
u
I
INPUT
SIGNAL
Fig. I.
Basic pump arrangement.
WIRELESS WORLD, JULY 1966
338
www.americanradiohistory.com
approximates to e =f C1RV. Practical tests have been
made' using this circuit as the frequency discriminator
of an f.m. receiver and the distortion introduced due to
discriminator non -linearity has been claimed to be less
than 0.1%. This was achieved by using an input voltage
of the order of 150 V pk -pk and an output of less than
1 V pk -pk.
Thus, while this circuit was adequate when
used with thermionic valves, where the high voltages may
relatively easily be obtained, with transistors, the output
voltage to retain linearity would generally be very low.
Diode-transistor pump
In order to improve the linearity of the circuit, a simple
modification may be carried out3 (see Fig. 3). The diode
D2 is replaced by a transistor, the base of which is connected to the junction of D2 and C2. This transistor
acts as an emitter follower and causes the voltage at B to
follow the voltage at C fairly closely. This permits C,
to always charge to V+e thus ensuring that, during the
charge sharing process, C2 receives the same amount of
additional charge each cycle and hence the frequencyvoltage characteristic will be linear, i.e. e =f C1RV.
A more advanced form of diode- transistor pump is
shown in Fig. 4. The operation is similar to the simple
diode pump but charge transfer takes place instead of
charge sharing. When S1 is in position 1, D1 conducts
and a charge VC1, is stored in C1. When S1 is in position
2, C1 is discharged through the emitter -base diode of Trl.
The actual charge which will be conducted through this
diode will be C, Vf coulomb /sec (i.e. the current will be
CI Vf amperes) and, as the common base current gain
of a transistor approximates to 1, the collector current will
be C1Vf amperes. Thus, the voltage drop across R
will be C1VfR, the same result as was achieved in the
previous circuit. This circuit has the advantage, however,
that the output circuit is entirely isolated from the input.
It is also interesting to note that, whereas the output of
the simple diode pump circuit is limited to the peak -topeak value of the input signal, the maximum output of the
diode- transistor pump depends mainly upon the supply
voltage to the collector of the transistor. Thus, by using
a supply voltage of 100 V and a transistor having a
VcRO rating in excess of this, it would be possible to
obtain an output of up to 100 V more or less regardless of
the peak-to -peak value of the input signal.
,a)
+I
A
o
o
B
e=V
C
C i+Cz
o
C
t
(b)
I
Cz
[+C+C
Cz
e-VC+C
z
t
e-v
i
Ci+Cz
+C'+Cz+(Ci+Cz
z
Si simulates a rectangular
Fig. 2(a). Simple diode pump circuit.
wave input. (b) Voltage waveforms at points A, B and C in (a).
Ci
e
i
Applications
(a)
NEGATIVE SUPPLY
The uses for this type of circuit are many and varied as the
following examples show.
Frequency discriminator
Recently5'6 the use of pump -type discriminators in
f.m. receivers has aroused a certain amount of interest.
Table 1 gives a comparison between pump and tuned
circuit discriminators. From this it is seen that the pump
has considerable advantages. Probably the only reason
why it has not been generally exploited in commercial
radio receivers is that the components necessary may
prove more expensive, particularly as a low intermediate
frequency is essential for efficient operation.
The circuit shown in Fig. 5 is an example of a practical
discriminator using the diode -transistor pump. The input
amplifier (Tr1) serves the dual purpose of providing a
reasonable input impedance (about lkt2) and of converting the signal into a suitable form to drive the limiter.
As the gain of this stage is relatively unimportant, no
negative feedback has been applied to it. The signal from
A
o
0
b
1
e=V
CI
Ci+Cz
e=z(V
C
i
+C z
e=3
o
(V
C
i
+C z
/
C
(b)
Fig. 3 (a). Diode -tra -Astor pump (original version). DL is replaced
by theemitter follower Trl so that the voltage at B follows that at C.
(b) Voltage waveforms at points A, B and C in (a).
WIRELESS WORLD, JULY 1966
339
www.americanradiohistory.com
TABLE
+Vc
I.- RELATIVE
SUPPL.'
Linearity
Bandwidth
MERITS OF PUMP AND TUNED
DISCRIMINATOR CIRCUITS
Pump Discriminator
Tuned Discriminator
Good linearity is an inherent
Function of the accuracy
with which the circuit has
characteristic of the diodetransistor pump and the introduced distortion may be less
than 0.1 %.
only by switching
speeds, design limits, stray capacity effects, etc. Bandwidths
of up to 50 Mcls have been
achieved with experimental
Limited
circuits.
Fig. 4. A more refined version
of the diode -transistor pump.
been set up.
Generally speaking, pro portional to the centre frequency and is usually fairly
restricted, although large
bandwidths may be obtained at frequencies in the
v.h.f./u.h.f. range.
Output
The output voltage is proporclonal to the intermediate frequency chosen and thus the
modulation signal will be a
proportionate fraction of this
voltage. In practice, this means
that a relatively low i.f. must
be chosen.
Virtually independent of the
Other
No setting up required.
Setting up is a specialised
procedure, usually requiring the services of a skilled
level
Comments
centre frequency.
operator.
oV
f
TrI,Tr2,Tr3
Di, D2
ST 53
BC 1013
BCY 42
BFY 12
PEP S
IS 44
IN 914
ZS 140
SD 10
BAY 31
BAY 38
BAY 41
TI 416
ZT 80
2N
2N
2N
2N
SEE TEXT
706
914
2926
3708
25 512
Fig. 5. Pract'cal discriminator circuit. For use in f.m. receivers, C1,
C2 and C3 may be reduced to 0.11AF, 0.1µF and 0.01µF respectively.
2N2926 transistors (coded red) are obtainable from Amatronix Ltd.
at 306 Selsdon Road, Croydon for Ss. each.
the collector of Trl is fed via C3 to the base of Tr2. Its
amplitude is sufficient to switch Tr2 on and off, and the
diode D1 ensures that the mark to space ratio of the
switching waveform remains approximately 1: 1. Thus a
square wave having an amplitude very nearly equal to the
supply voltage will appear at the collector of Tr2. This
signal is then fed to the diode-transistor pump circuit
which consists of C4, D2, Tr3, C5 and R 7.
In order to assess the linearity of the discriminator,
the charging time constant R 6C4 was varied. The graph,
Fig. 6, shows the sort of variation to be expected. In
practice, it appears that a time constant 1/50 of the
period of one cycle of the highest input frequency to be
encountered, is necessary in order to achieve less than
0.5`), deviation from absolute linearity. However, from
the curves it is possible to calculate' that the distortion
which would be caused by 0.5% non -linearity would be
approximately 0.044% second harmonic, 0.003% third
harmonic and 0.006% fourth harmonic. Unfortunately,
it is almost impossible to prove these figures at present
as f.m. signal generators with distortion of this order are
still a designer's dream. However, even if these figures are
out by a factor of 10, the performance should be good
enough to satisfy the most fastidious listener.
Frequency meter
o
RC=Olpsee
RC=O.2psee
RC=oepsec
40
40
80
120
zoo
160
240
This is an obvious development from the frequency
discriminator. Thus the circuit of Fig. 5 may be adapted,
as shown in Fig. 7, to provide a direct reading frequency
meter. The main change is that the discriminator load
resistor is replaced by a meter linearly calibrated in frequency. The required frequency range is then selected
by switching the value of C4. Tests made using the circuit
over a frequency range from 10 c/s to 1 Mc/s have shown
that an input of the order of 10 mV is all that is necessary
to ensure proper limiting.
The accuracy of this type of frequency meter is limited
by the meter accuracy and the absolute value of the capacitor C4. In practice, a reading accuracy of the order of
±5% on the low frequency ranges and ±2% on the high
frequency ranges should be fairly easy to attain.
FREQUENCY(kcls)
Fig. 6. Effect
of charging time constant
on
linearity.
Frequency divider
The diode -transistor pump is very easy to adapt to
make a frequency divider, see Fig. 8. The action of the
WIRELESS WORLD, JUNE 1966
340
www.americanradiohistory.com
circuit is as follows. Initially, the potential at A is 0 V
with the result that Tr2 and Tr3 are both cut off. When
the input signal is applied, C. acquires a charge each
cycle and the voltage at A builds up as shown in Fig. 8(b).
When this point becomes sufficiently negative to switch
Tr2 on, the collector current of Tr2 switches Tr3 on and
the cumulative effect causes Tr2 and Tr3 to bottom, thus
discharging C2 very rapidly.
When the discharge is complete, the potential at A is
at its initial condition, i.e. 0 V, so that the circuit is ready
for its next operational cycle. It would be possible to
replace the switch (Tr2 and Tr3) with a single unijunction
transistors but, although this would be ideal for low
frequency work, the switching time would be too long for
high frequencies (100 kc /s and above).
From the earlier discussion it will be recalled that the
amplitude of the voltage steps at A will be proportional
to the supply voltage and to the ratio between C2 and C1.
The switch trigger level is set by the voltage at the junction of R1 and R2. Thus, the division ratio may be set by
making the potential at A just exceed the switching level
after the required number of cycles. In order to avoid
spurious switching, the steps should be made as large as
possible and the switching level should be chosen to be
about midway between two step levels. When these
precautions are observed, dividers of this type are very
reliable, both over a range of temperatures and a wide
range of input frequencies.
METER FULL SCALE
C4
IOOc /s
1µF
Ikc /s
0.1µF
lUkc /s
O.OipF
IOOkc /s
0.00lµF
IMc /s
IoopF
+izV
IOWA
I00µ
METER
F =1k
COLLECTOR
FROM
OF Trz
o
Fig. 7. Frequency meter modification. The value of the 5 kSl resistor
(R8) should be adjusted so that f.s.d. is obtained with 120 pA collector
current through Tr3.
INPUT
Trl
Oscilloscope deflection signals
The pump circuit may be used to advantage in applications where it is necessary to produce an oscilloscope,
or x-y recorder, display1° which has an axis proportional
to frequency. Generally speaking, sweep generators
provide an output voltage to perform this function but
there are many swept frequency applications where no
deflection information is available. The block diagram,
Fig. 9, shows a method of connecting the equipment
to use a pump discriminator. (The circuit given in Fig. 5
is still applicable.) If the centre frequency of the swept
signal is too high to permit the efficient working of the
discriminator, a simple heterodyne converter may be
used to convert the input to a suitable frequency.
A rather novel application of the pump circuit is its
use in the construction of a complete linear time -base
system. The block diagram, Fig. 10, shows how this may
be done. A 10 Mc /s crystal controlled oscillator is used
to provide the time standard for the time base. The output
of this oscillator is fed to a frequency divider chain consisting of a series of pump dividers, each dividing by a
factor of 10, thus providing outputs at 10 Mc/s, 1 Mc /s,
100 kc /s and 10 kc /s. One output is selected from this
section by the switch and fed, via a gate, to the final
frequency divider which may be set to divide by approximately 100, 200, 500 or 1,000, thus providing a variable
frequency output. In order to keep the sweep length
constant, the division ratio is set by adjusting the value
of C2 (see Fig. 8). The gate is controlled by a bistable
circuit which is switched on by the synchronizing signal
and off by the fly -back pulse from the divider. The final
frequency divider may be made to " free -run " by disconnecting the reset pulse time, in which case the synchronizing signal will start the time -base running initially
and then have no further effect. This time -base system
will provide a constant length output sweep having a
frequency variable from 10 c/s to 100 kc /s.
Strictly speaking, this is not a linear time-base as the
sweep is actually made up of 100 or more equal steps.
This will only really distort a sine wave presentation when
DIZ
(a)
INPUT
o-WAVEFORM
AT
A
TRIGGER LEVEL
o
WAVEFORM
AT
8
BIAS VOLTAGE
(
b)
Fig. 8(a). Frequency divider circuit using the diode- transistor pump.
(b) Voltage waveforms at points A and B in (a).
OSCILLOSCOPE
EORIPMENT
SWEPT SIGNAL
UNDER
SOURCE
TEST,
r
Y
X
INPUT
INPUT
1
Ifl
1
Fig. 9.
HETERODYNE
CONVERTER
L
"Frequency base
I
" generation for
FREQUENCY
DISCRIMINATOR
oscilloscope displays.
341
WIRELESS WORLD, JULY 1966
www.americanradiohistory.com
generation is that it is a true " time " base and may thus
be used for making precise measurements. By expanding
the sweep width sufficiently to permit individual steps to
be seen and counted, it becomes a very simple matter to
measure both rise time and period.
Io Mc /s
CRYSTAL
OSCILLATOR
Linear time -base generator using
pump frequency divider.
Fig. ID.
REFERENCES
:
OIOMc/s
PUMP DIVIDERS
RAMP GENERATOR
c/s
OUTPUT
:100,200,500
GATE
OR
=10
" The Diode Pump Integrator," J. B. Earnshaw, Electronic
Engineering, January 1956, p.26.
2. " Generation and Detection of Frequency Modulated
Waves," W. S. Seely, C. N. Kimball, A. A. Barco, R.C.A.
1.
4,000
.--0100kc/s
Review, 1942.
Transistor Circuits," P. L. Burton, J. Willis,
Wireless World, March 1958 p. 107.
4. "Electronic Designer's Handbook," T. K. Hemingway,
Chapter 13 (Business Publications Ltd., 1966).
5. "A Simple Transistor F.M. Tuner," J. C. Hopkins, Wireless
World, September 1965, p. 420.
6. " Pulse Counting F.M. Tuner," E. D. Frost, Wireless
World, December 1965, p. 586.
7. " Radio Laboratory Handbook," M. G. Scroggie, p. 302.
(Iliffe 1954, 6th Edn.).
8. "A Micro-electronic Frequency Divider with a Variable
Division Ratio," B. Preston, Electronic Engineering, April, 1965,
p. 240 (Vol. 37).
9. " The Unijunction Transistor," H. R. Henly, Wireless
World, March 1964, p. 110.
o IOkc/s'
3. " Unusual
RESET
=10
LINE
BIST ABLE
JREE RUN
=10
SYNC INPUTS
the input frequency is more than 1/20 of the step rate, i.e.,
the highest usable frequency in this case will be about
500 kc/s, although higher frequency may be displayed if
desired. The advantage of this method of time-base
10. "A Simple Frequency -Voltage Converter," D. E. O'N.
Waddington, Marconi Instrumentation, April 1965, p. 8 (Vol.
10).
H. F. PREDICTIONS -JULY
Mc/s
40
30
20
15
10
B
1
Mc/s
AIRE
BUENOS
30
v/ ANIMM
-
f- -
'111.1.1111T
6
8
5
12
16
20
0
0
4
Mc/s
8
12
16
20
Mc/s
HONG KONG
40
40
30
30
20
20
15
15
k 10
Io
8
8
6
6
5
5
4
4
30
4
8
LOWEST USABLE
H F
10
4
4
OPTIMUM TRAFFIC FREQUENCY
15
6
WITH AURORAL CORRECTION
4
30
20
- ---
s
I
5
MEDIAN STANDARD MUF
40
12
G.M7.
16
20
0
0
4
8
2
16
20
G. M.T.
03
The prediction curves show the
median standard MUF, optimum traffic
frequency and the lowest usable frequency (LUF) for reception in this country. The MUF is, by definition, the
frequency at which communication
should be possible for 50 % of the time,
Satisfactory communication will, of
course, be possible slightly above the
MUF, but only for smaller percentages
of time. The optimum traffic frequency
is usually taken as 85 % of the MUF.
Unlike the standard MUF, the LUF is
closely dependant upon such factors as
transmitter power, aerials, local noise
level and the type of modulation. The
LUF curves shown were drawn by Cable
& Wireless Ltd. for commercial telegraphy and assume the use of transmitter power of several kilowatts and aerials
of the rhombic type.
WIRELESS WORLD, JULY 1966
342
www.americanradiohistory.com
Solar Battery Charger
A
By D.
BOLLEN
The ability of solar cells to convert sunlight directly into useful
electrical energy has been well demonstrated in satellite applications. An advantage of the solar battery is that it allows true,
unattended operation in locations remote from a power supply and,
even moro important, promises an outstanding degree of reliability.
If a solar battery can be arranged, economically, to trickle charge
an accumulator, it will then power apparatus which has a low, longterm current demand, such as electric doorbells, alarm systems,
and clocks. With a NlCd accumulator, the apparatus could function
for many years without any maintenance.
The purpose of this article Is to outline some of the problems
involved, and describe a solution, in the form of a low cost, solar
powered, transistor trickle cha-ger.
THE two main types of solar cell normally encountered
are selenium and silicon; both give a maximum open circuit output voltage of about 0.5 V per cell. Obviously, if an accumulator is to be charged by solar cells
then their output voltage must exceed the accumulator
voltage. Ordinarily, a bank of at least four solar cells,
giving 2 V maximum, is required to charge a 1.2 V NiCd
accumulator, in bright sunlight. A dull day will not
furnish sufficient energy to keep the solar battery potential above that of the accumulator, unless more cells are
wired in series.
The solar battery charger.
light, and ingrained surface dirt. With a four -cell silicon
solar unit, the circuit of Fig. 1 will supply enough energy
to operate a low- current doorbell for nearly one minute
every day of every year.
Sunlight
Fig. 2 gives the rounded curves of average bright sunlight per month, recorded in three places in England over
Simple circuit
The circuit of a simple charger is given in Fig. 1. The
base -emitter junction of a germanium transistor is employed as a diode, to prevent the day's charge leaking
back through the solar battery during hours of dull light
and darkness. It was found that the junction diode
reverse current, under dark conditions with this circuit,
was somewhat less than 2 /IA, too small to be significant.
If a sealed type of NiCd accumulator is charged, such as
the Varta Deac, this will give a predicted working life of
perhaps 15 years. Even after the accumulator has been
replaced, the solar battery will continue to function until
its lens becomes darkened by the action of ultra -violet
Ge TRANSISTOR
Fig. 2. The graph depicts the yearly average over the period 19211950 of hours of bright sunshine. The figures are 1,792 hr for Ventnor,
1,525 hr for Croydon and 967 hr for Manchester.
2V SILICON
I
SELENIUM
BATTERY
OR
I
Fig. I.
24V
Nt
-5mA
IN BRIGHT
SUNLIGHT
Circuit of a simple charter.
Cd
the period 1921 -1950. Ventnor and Manchester may be
take to represent extremes for the whole of the British
Isles. In passing, it is interesting to note that the curves
correspond closely to a sine wave. The reader might
feel inclined to calculate r.m.s. and peak -to -peak sunlight values for his own particular area. Taking the
yearly averages, we can expect from 1,000 to 1,800 hours
WIRELESS WORLD, JULY 1966
343
www.americanradiohistory.com
COMPONENT SUPPLIERS
Varta Deac NiCd G. A. Stanley Palmer Ltd.,
accumulators.
West Molesey Trading
Estate, Island Farm Avenue,
East Molesey.
Assorted selenium B. W. Cursons, 78 Broad St.,
wafers.
Canterbury.
Assorted selenium and International Rectifier Co. (Gt.
Britain) Ltd. (Display
silicon single cells.
Stands), Hurst Green,
Oxted, Surrey.
4-cell silicon
solar Service Trading Co., 47 High
battery.
St., Kingston-on-Thames.
Repanco TT45 trans- Henry's Radio Ltd., 303
Edgware Rd., London W.2.
former
Circuit of the converter charger. Suppliers
components are listed in the panel on this page.
Fig. 3.
of some of the
Therefore, allowing for accumulator inefficiency-a Deac for example needs a charge of 1.4
of bright sun.
times the drain, to maintain equilibrium -to realize a
capacity of 1,000 mAh per year the charge rate should be
at least 1.4 mA per bright sunlight hour. The " dull day "
performance of the charger must also be reasonable, to
offset accumulator leakage and unexpectedly heavy demands during the winter, or in a year of bad weather.
To achieve a good all -round performance, without resorting to large numbers of series connected solar cells, a
converter is, therefore, needed to step up the voltage.
Converter circuit
With the converter circuit of Fig. 3, a useful output can
be obtained from a single selenium cell. With two
selenium cells, the output can be as much as 500 mAh/
year, and two NiCd accumulators may be charged in
series. Using a readily obtainable silicon four -cell solar
battery, a capacity of more than 1,000 mAh /year, from
accumulators of up to 4 V, is possible, and the " dull
day " performance is exceptionally good.
In the circuit of Fig. 3, the d.c. output from the solar
cell is " chopped " by a multivibrator and the resulting
square wave is fed into a step -up transformer, to give
the required higher voltage. With the prototype, oscillation commenced when the supply voltage reached a mere
0.075 V.
Transistors type 0071 are specified for Trl and Tr2,
because of their low cost, but transistors with a higher
switching speed will give an improved performance.
As the dissipation is unusually low, almost any type of
p -n -p transistor will work in this circuit with no danger
of its maximum ratings being exceeded.
The transformer Tl is a standard push -pull driver
giving a voltage multiplication of 4, measured with no
load. The charge current is about 1 /5th the input
current at normal levels.
Once again, a transistor base -emitter junction is used
as a charging diode, and also for rectification. C, is
wired across the solar battery to slightly improve efficiency. If it is thought that an electrolytic might suffer
from a short life, due to slow deterioration, then this
component may be omitted, and no great loss of output
will result. The circuit is not at all critical, but for
longevity, some care is needed with
its construction.
Construction
An etched circuit was employed for
the prototype. It was considered that
this would give greater strength and
durability compared to the " wired
underneath " form of assembly.
Details of the solar battery charger. In the foreground selenium wafers. Left to right: four cell silicon sun battery, converter and box. In the background: 1,000 DKZ Deac cell.
Etched circuit plan and component
placement diagram are given in Fig.
4. During assembly, cleanliness and
good quality soldering are essential.
When all components are finally
mounted, together with the input and
output leads to the etched circuit
panel, the underside of the panel
should be scrubbed with a clean
suede brush, dipped in methylated
spirits, until the surplus flux is
removed and the copper and solder
shines. Finally, the entire assembly
may be coated with polyurethane
varnish, to inhibit corrosion.
WIRELESS WORLD, JULY 1966
344
www.americanradiohistory.com
Vein PI.WOOD
BOX OF
3/
f
27Áo
2
/4
HOLE FOR
OUTPUT LEADS
GLUE FILLET
POLYSTYRENE
PANEL
GLUE
Fig. 5. Details of the housing of the charger. All joints are glued with
epoxy resin and finished with polyurethane varnish.
Fig. 4. Etched circuit and disposition of components.
SELENIUM CELLS ; ACTIVE AREA Isq.in.
SILICON CELLS; ACTIVE AREA 037sq.in.
The original intention was to enclose the circuit panel
and solar cells inside a box with a glass front, but it was
discovered later that a layer of ordinary glass, interposed between the cells and the light, reduced output
by about 25°jß. It was decided, therefore, to dismantle
the solar battery, and waterproof it, so that it could be
mounted on the outside of the box, with its polystyrene
diffuser lens exposed to the weather. The method of
waterproofing depends on the solar cell or battery.
Selenium wafers may be coated with several layers of
polystyrene cement, and can be stuck to a sheet of polystyrene cut from a lunch-box. Some solar batteries have
cases and lenses made of incompatible plastics, and here
the simplest method is to waterproof with an epoxy glue,
ensuring an even layer of glue at all joints. The leads
from the cells must be rearranged, if necessary, and
brought through the mounting panel. Silicon batteries
are delicate and should be dismantled with care. The
diffuser lens is sometimes lightly glued to one cell and
an attempt to separate them could have disastrous results. Details of the plywood box and final assembly
are given in Fig. 5.
Tests and performance
Artificial light tends to give misleading results so tests
must be carried out in daylight. Fig. 6 shows the output
obtained with the original charger, and light-meter readings are included for the photographically minded
reader. The light -meter was set for a film speed of 100
ASA for all measurements. The curves indicate the
amount of charge to be expected under varying light conditions with different accumulators and solar cells. There
is little point in using more than two selenium cells as
their current output would be too low to make a corresponding improvement in converter output. The equivalent input resistance of the converter is around 200 i ?,
being a good match for two selenium or four silicon cells.
`
á
cma
025
0.5
075
CONVERTER
I
I.5
1.25
.75
INPUT (VOLTS)
Fig. 6. Charging characteristics of the converter. An indication of
the light conditions is given by the letters against the vertical lines
on the graph. With the light meter pointed at the sky and set to a film
speed of 100 ASA, the exposure values (E.V.),
number and shutter
speed for various conditions are as follows. (A) E.V.14, f11 at 1/125,
at 1/500, part overcast, sun
bright total overcast. (B) E.V.16,
behind cloud. (C) E.V.18,
at 1/2000, sunlight through thin cloud.
(D) E.V.20 plus (light meter off scale), winter sunlight.
f
fl
fll
Less than four silicon cells may be employed, and their
performance will be slightly better than the same number of selenium. The converter will tolerate a maximum
input of 6 V at about 20 mA, but it is hardly likely that
this level of operation will be required. Although a
single selenium cell will only produce a 100µA charge
WIRELESS WORLD, JULY 1966
345
www.americanradiohistory.com
in bright sunlight, this will be adequate for a low capacity NiCd 1.24 V accumulator, such as the Deac 100DK,
and will offer a 100 mAh /year supply, enough for a small
alarm bell or warning light.
When measuring solar cell and converter outputs, the
readings given by even a good quality testmeter must
be regarded with some suspicion, remembering that a
full-scale deflection on low current ranges might produce
Shunt the testmeter
a voltage drop of perhaps 0.5V.
leads with a high value electrolytic for voltage readings
of the converter's output, this will give a reading slightly
below the peak voltage supplied. A more reliable indication of performance is obtained when an accumulator is
actually placed on charge, under different lighting conditions, and the current fed to it is monitored.
Solar cells and converter should be left in a warm
place for a few hours before sealing, to remove any
moisture. The solar battery panel can then be glued to
the box. Functioning of the multivibrator may later be
checked at any time, by placing the ferrite aerial of a
transistor radio close to the converter output leads, to
pick up the feeble radiation of square wave harmonics.
Installation
Mount the charger, with its solar battery facing south at
an elevation angle of 45 °, preferably high up and away
from obstructions, with no trees in the foreground to
overshadow it. The accumulator, bells, and other equipment, can be placed under cover somewhere near. The
accumulator should be kept dry. Either the side of a
house or a tall pole will make a good mount for the
charger,
THE large,
Applications
The possible uses of the charger are almost too numerous
to list, but a few suggestions are outlined here. In Table
mAh
1 which is based on the calculated output of a 1,000
NiCd cell, coupled to the charger, there is given a
TABLE I
Once
per day
C
I
mA
10mA
IOOmA
IA
10A
2.74 hr
min
16
min
9.8s
1.6
I
s
Once
per week
19
Once
per month
hr
83
1.1
min
6.8
s
hr
8.3 hr
50 min
1.9 hr
11.4 min
5
30
min
s
Once
per year
1,000
100
10
-
hr
hr
hr
48 min
The intermittent discharge rate of a 1000 DKZ cell (obtainable from G. A.
Stanley Palmer Ltd.), when connected to a 4-cell silicon charger, is shown In
the above table. The continuous discharge rate is 10µA. The manufacturer
cell
quotes a maximum admissible discharge current for 10 seconds as ISA at a
potential of 0.2 to 0.3V.
1
range of loads at differing time intervals. This data will
be of assistance to the reader who may want to assess
the possibility of using a solar device for various applications. The figures given in the table are a conservative estimate, based on bright sunlight only. The surplus
charge of a good " dull day " performance will therefore
cover bad years, unexpected loads, and keep the accumulator well topped up. In practice, with a good year, the
capacity could almost be doubled. The short period
discharges, such as 1 A for 9.8 sec /day, indiçáte that the
unit might be used for intermittent switching of power
relays and the triggering of other equipment.
AERIAL PROFILE AUTOMATICALLY CONTROLLED
correct position. The response -up to 25 c /s -is fast enough
steerable paraboloid aerials used for satellite
to
communications and radio astronomy must be designeddiagive the highest possible gains, and this demands bowl
meters of 1,000 or more times the working wavelength.
With such large diameters it is extremely difficult to make
the bowl support structure sufficiently rigid to maintain the
loading
correct reflecting profile when the aerial is subject to steering
forces resulting from wind, water, ice, gravity and
acceleration, and to distortions due to heat, ageing and
structural resonances. A new approach to this problema
being tried by the Royal Radar Establishment is to build
relatively light and compliant structure but give it higheffective rigidity against disturbing forces by applying closed
loop position control to individual elements of the bowl
surface.
At an I.E.E. conference on the design and construction ofa
large steerable aerials, D. W. Burr of R.R.E. described
(actu23 -ft experimental aerial constructed on this principle
ally a scale model of a 170 -ft aerial for use in the X band).
honeycomb
The bowl surface is formed from 54 glass fibre flexible
alupanels and these are mounted on a light and
supminium back frame by 36 hydraulic jacks-each jack
At any
porting the corners of a group of adjacent panels.applied
to
such support point, a disturbing force (e.g. wind)
back frame,
the bowl is transmitted through the jack to the
which yields. The resulting positional error in the reflector
transducer,
is detected by the movement of an optical
to a fixed
mounted on the bowl 1urface at that point, relative
reference light beam, and the resulting error signal -10mV
electroper 0.001 inch displacement -is used to control the
jack
hydraulically the flow of oil into the jack. Thus
bowl
extends or retracts to adjust the distance between the to
its
surface and the frame so that the surface is restored
to deal with turbulent wind forces.
Each of the 36 control loops limits excursions of the bowl
surface relative to the reference light beam to 0.005ín, and
in this way the true paraboloidal shape of the reflector is
maintained. Positional errors due to permanent manufacturing inaccuracies are also automatically corrected.
Each jack can move up to 6in, and when the control
system is energized a 1 -c /s oscillation is applied to the oil
valves so that the jacks and optical transducers search for
the reference light beams.
The 36 reference light beams are formed from a common
light source consisting of an iodine quartz lamp, a set of
lenses and a light beam chopper. This unit is mounted near
the front feed of the aerial and directs a beam on to a pyramid of 36 mirrors mounted in the centre of the bowl. Each
of these mirrors reflects An image of the light source on to
an associated optical transducer on the bowl surface. Thus
the true paraboloidal reference shape is defined by the
various angles of the 36 reflected beams relative to the central incident beam and by the corresponding beam lengths
(mirror -to- transducer). The light chopper breaks the inciare
dent beam at 400 c/s and the transducer amplifiers
tuned to this frequency so that the control loops are made
immune to ambient light.
The total weight of the aerial is 2,0001b. It has been
estimated that a comparable aerial of conventional isconstrucyet
tion would weigh at least 20 tons. Development not
be succomplete but there is confidence that the project willapproach
cessful. Aerial steering will require a special a possibecause of the profile control system, but there is for fine
bility of making use of the profile control facility
adjustment of aerial beam direction.
WIRELESS WORLD, JULY 1966
346
www.americanradiohistory.com
WORLD OF WIRELESS
The U.K. Space Programme
ACCORDING to a report recently published by the National
Industrial Space Committee entitled " The Case for a U.K.
Space Programme," members of the British aerospace industry
are concerned about the lack of a clear policy definition or of
any significant national programme of work in the space
field. It was because of the importance of this subject that
the N.I.S.C. was recently formed by the following associations
-Society of British Aerospace Companies, Electronic Engineering Association, and Telecommunication Engineering and
Manufacturing Association.
The sectors of space technology singled out as being of
particular importance to the U.K. are defence (reconnaissance
and military communications), navigation, meteorological,
commercial communications, domestic applications of communications satellites, rescue surveillance, and scientific research in space. The committee consider that these aspects
of space technology are of immediate importance to Britain
and her economy, and calls for the most urgent and serious
consideration by the Government.
Of particular interest is the recommendation for using
satellites for direct television broadcasting. With the increase
in the size of satellites that is now practicable, improvements
in electrical power supplies and light- weight amplifiers it is
stated that it will be possible within the next few years to
radiate sufficient power to enable sound radio and television
broadcasts to be received directly by small domestic aerials
of similar size to the paraboloids used for terrestial relay of
television programmes.
For 99'! coverage of the U.K., with four television services 64 major stations, 250 minor stations and 1,000 fill -in
stations would be required, costing £120M against £50M
by satellite relay.
The main theme of the note is summed up in the suggestion that international collaboration should be an extension
of the national effort and not a substitute for it. The document outlines possible courses of action to achieve a more
balanced participation in space technology at moderate cost.
The cost of the programme outlined by the committee is
estimated at less than £40M p.a.; tittle more than double
the present gross British expenditure on all space activities.
Another paper of vital interest to those planning trunk
routes was that by G. C. Rider and D. J. Palmer, of Marconi,
who gave data indicating that fading on two tropospheric
scatter links connected end -to -end is virtually independent.
This means that present planning methods are too cautious
since they assume that in the worst conditions one must allow
for both links to be fading simultaneously.
With modern high- capacity systems operating at 6 and
even 11 Gc /s it is certain that a great deal of effort will continue to be deployed on examination of the fine structure of
the troposphere and its influence on propagation.
Computer Designs Computer
A METHOD has been devised by International Computers
and Tabulators Ltd. by which an existing computer is made
to behave like one being designed. In this way, the performance of the new computer can then be assessed and the
system perfected without the use of a new physical layout,
but once the system design is finalized, the existing computer
Tropospheric Propagation
SEVERAL contributors to the recent I.E.E. colloquium on
tropospheric propagation described efforts to correlate transmission loss with meteorological measurements -especially
refractivity of the surface atmosphere which can be obtained
from published data. Although this parameter was shown
to be of little use it was equally clear that radiosonde
measurements tracing the change of refractive index with
height gave good correlation with received field strength.
The work described was exclusively concerned with frequencies above 100 Mc /s at which propagation is controlled
by the atmosphere rather than by properties of the ground.
All the measuring techniques of the meteorologist are relevant
to studies in this field. If any meteorologist had been present,
which was not evident, he would have been surprised to find
a radio engineer such as Dr. D. T. Gjessing, from Norway,
having progressed to the point of using an instrument that
measures changes in atmospheric temperature during a
1 'sec interval.
Mr. M. W. Gough (Marconi) presented an elegant examination of the factors causing a drop out which is sometimes
observed between terminals apparently in line -of- sight.
Microwave circuits of only 25 miles in length have been
found to fail under certain conditions when the terminals are
at widely different heights. Mr. Gough explained how this
was caused by total internal reflection at an atmospheric layer
at a height intermediate between transmitter and receiver.
Ancient and modern techniques are sharply contrasted in this illustration of an operative measuring the contents of a storage tank at a
Shell refinery. Information is reported to the refinery's central control
room by use of a modern two -way radio -telephone but absence of
automatic electronic (or mechanical) level reading equipment (highly
developed during recent years) is emphasized by the dip stick used
by the operative. The radio -telephone is o Lancon I.S. manufactured
by G.E.C. (Electronics) Ltd., and is the first compact lightweight v.h.f.
transmitter- receiver to be issued with an Intrinsic Safety Certificate
by the Ministry of Fuel and Power permitting its use in hazardous
atmospheres.
WIRELESS WORLD, JULY 1966
347
www.americanradiohistory.com
can be used to optimize the layout of the new computer.
One of the advantages of design simulation is the elimination
of manual mistakes. This is of particular significance when
developing a computer comprising several thousand logic
circuit elements and their associated interconnections. In
fact, this complexity has been the major problem encountered
in previous attempts to achieve design simulation with a
computer. Owing to the time period and storage capacity
involved, the part -by -part testing of a simulated model was
necessary. I.C.T. have evolved a method by which the
system to be simulated is described in a new specialized
The new
language known within I.C.T. as SIMBOL.
language offers facilities for simulating complete systems, or
individual parts of systems in greater detail, or both simultaneously. To imitate the operation of a new computer, the
power of a large existing computer is required and a SIMBOL compiler has been developed for the I.C.T. Atlas and
is currently under test.
Local Radio Plan
A MEMORANDUM, issued by the Local Radio Association
of 35 Connaught Square, London, W.2, has been submitted
Postmaster General outlining a plan for the introduction of local radio stations in the U.K. This follows an
earlier recommendation, submitted in January 1965, putting
forward reasons for proposing that local broadcasting be
financed by advertising.
Among the questions discussed are frequency allocations,
the number of stations and field strengths. On the assumption that a population of 50,000 is the smallest community
for which it is economically possible to operate a local radio
station, the Association considers, in the light of technical
advice from the Marconi Company, that there would be a
Assuming that station coverage
need for 276 stations.
extends to a 20 -mile radius it is estimated that 98.5 % of
the population of the U.K. would be covered. The L.R.A.
recommends that all stations should be equipped with both
low-power medium -wave and v.h.f. transmitters.
The medium-wave transmissions would be restricted to
daylight hours and during the day the v.h.f. transmissions
might offer an alternative programme. The memorandum
states that " The introduction of local broadcasting would
necessitate using many of the channels in the band which are
not allocated to the U.K. and hence a depature from present
policy would be involved." It would appear, therefore, that
the scheme could not be introduced until a new European
wavelength plan had been agreed.
Suggested field strengths are: city service 10 mV /m, town
service 4 mV /m, and rural service, 1.5 mV /m. The protection ratio for co- channel stations (the level of the wanted
signal compared with an unwanted signal) may have to be
as low as 20:1 and some degradation at the edges of rural
Transmitter radiated
service areas may be experienced.
powers to give a rural-grade service at a distance of between
20 and 30 miles are between 25 and 200 W depending on the
frequency. On the v.h.f. band necessary field strengths,
according to the C.C.I.R., are city service 3 mV/m, town
A maximum
service 1 mV /m, rural service 0.25 mV /m.
radiated power of about 250 W is envisaged.
to the
An expedition organized by Imperial College is to go to
Malta this year to study underwater communications. Existing systems have been limited to wires which have serious
attendant disadvantages, and the use of ultrasonic carrier
waves which not only limit communications to one way but
also need complex equipment. Both types of system have
been tested at Imperial College with the conclusion that
future research and development will be directed towards a
form of voice frequency link. Investigations will be made of
intelligibility, microphone and transducer sensitivity, power
requirements and threshold and directional measurements. A
film will be made of the expedition and equipment will include two surface ciné cameras and a camera for operating
more than 200ft below the surface.
The Standard Time and Frequency transmissions from
MSF Rugby on 60 kc /s have been extended from the daily
one hour period of 1430 -1530 G.M.T. to 24 hrs a day, but
for an interim period the one hour schedule of A2 modulation pulses consisting of 5 cycles of a 1 kc /s tone for the
seconds pulse and 100 cycles of 1 kc /s tone for the minute
pulses will be retained. The new extension to the service
for the remaining 23 hours of the day will operate with a
modulation of interrupted carrier pulses of 100 ms duration
for the seconds pulses and 500 ms duration at the minute,
the epoch of the pulses in each case will be at the start of
the break in the carrier signal. The estimated decay time
of the pulses is 0.5 ms. A shutdown period for maintenance
will be observed on the first Sunday of every month between
1000 and 1400 hrs.
More programme time for channel 6. -For many years
the frequency band 176 -180 Mc /s which falls in channel 6
has been used in this country for radio astronomy observations and, to enable an important part of this work to be
completed, it was necessary for television stations using this
channel to share it on a time basis with radio astronomers.
This restricted the times of operation of the BBC -1 stations
at Sandale and Moel-y -Parc to the period midday to midnight. The radio astronomical survey has been completed
and the restrictions withdrawn so that the stations now
operate from the start of trade test transmissions at 09.00,
or from the start of programmes if earlier.
Improved I.T.A. Service. -The Independent Television
Authority's new transmitting station at Belmont, Lincolnshire, recently began transmitting a full programme service.
The programmes from this station will be available to nearly
Trade transmissions, however, will be
a million people.
liable to interruption for some time, particularly during the
period of installation and testing of the u.h.f. aerial for
BBC -2 which is mounted on the same mast as the I.T.A.
aerial.
The 1966 edition of the British Standards Yearbook, which
has just been published, contains a complete numerical list
of all the five thousand British Standards and Codes of
Practice, with a brief summary of each one, and a subject
index. The publications of international standards organizations, which were covered in previous editions of the
Yearbook, are now dealt with in a separate Supplement. The
first edition of this Supplement, published last November,
is complete up to June 30th, 1965, and contains a complete
numerical list of all recommendations by the International
Organization for Standardization and similar bodies. The
Yearbook (15s) and the Supplement (7s 6d) are available
from the B.S.I. Sales Office, 2 Park Street, London, W.I.
From America, the Electronic Industries Association
announce that the first national Consumer Electronics Show
will be held during June 25th -28th, 1967, at the New York
Hilton and Americana Hotels. On display in an exhibition
space of over 200,000 square feet will be monochrome and
colour television receivers, record players, tape recorders and
electronic organs; it will be the largest trade show ever held
in New York using hotel facilities. The second annual show
will be held at the same location during the fourth week of
June, 1968.
A one week course of lectures entitled Digital Circuit Techniques is to be held from June 20th-24th at the Borough
Polythechnic, Borough Road, London, S.E.1. The fee is 7gn
and further details are available from the college secretary.
I.E.E. Membership.-The annual report of the Institution
of Electrical Engineers gave the figure of 55,578 for total
membership as at April 1st, 1966. This represents an increase of 1,428 on the figures for the previous year. Corporate members numbered 27,768, i.e., 49.96% of the total.
WIRELESS WORLD, JULY 1966
348
www.americanradiohistory.com
PERSONALITIES
Earl Mountbatten of Burma, chairman
of the National Electronics Research
Council, has been elected a Fellow of the
Royal Society. Lord Mountbatten, who
is this year's president of the British
Computer Society, has been the prime
mover in the Selective Dissemination of
Information project being conducted by
the N.E.R.C. in an endeavour to find a
solution to the problem of information
retrieval.
C. L. G. Fairfield, M.A. (Cantab.),
M.I.E.E., has been appointed managing
director of Submarine Cables Ltd., of
Greenwich, in succession to F. W. H.
Shaw who has retired after more than
50 years' service with the A.E.I. Group
which together with B.I.C.C. are joint
owners of Submarine Cables. Mr. Fairfield, who has been assistant managing
director of the company for the past
year, joined the Telegraph Construction and Maintenance Company (a subsidiary of B.I.C.C.) in 1953 as manager
of the overseas division. Mr. Fairfield,
C. L. G.
Fairfield
who is 54 and is a barrister -at-law, was
with Mullard from 1947 to 1953 first as
assistant to the directors on technical
matters and later as manager of the
valve division.
E. V. D. Glazier, Ph.D., B.Sc.,
M.I.E.E., head of the Physics & Electronics Department at the Royal Radar
Establishment, Malvern, has had the
title of honorary professor in the
Department of Electronic & Electrical
Engineering conferred on him by Birmingham University. Dr. Glazier, who
is 53, was director of scientific research
(electronics and guided weapons) in the
Ministry of Aviation from 1957 to 1963.
He received his early training in electronic and mechanical engineering in
industry and joined the Post Office in
1933. He transferred to the Signals
Research and Development Establishment at Christchurch in 1942 and in
1950 took charge of the research division.
C.
O.
Stanley, C.B.E., LL.D.,
F.C.G.I., chairman of Pye of Cambridge
Ltd., has relinquished his responsibilities as chief executive of the Group and
has tendered his resignation as a
director of the company and its subsidiaries. He has been invited to become
honorary president of the Company
" in recognition of the invaluable services which he has rendered during a
period of over forty years and as a
tribute to his great contribution to the
growth and development of the Group
during his chairmanship." Mr. Stanley,
who became chairman of the board in
1946, joined the organization in 1928
when W. G. Pye & Co., the Cambridge
instrument-making firm, became interested in radio receiver production and
formed Pye Radio Ltd.
Frank B. Duncan, chairman of Ether
Controls and deputy chairman of
British Relay Wireless & Television
Ltd., who joined the board of Pye in
February and a fortnight later was
appointed deputy chairman, has become
chairman and chief executive of the Pye
Group. The executive management of
the group has been reorganized and
there is now a managing director for
each of four divisions. J. R. Brinkley,
M.I.E.R.E., who joined the group in
1948 after six years in the Home Office
Communications Directorate and has
been managing director of Pye Telecommunications Ltd. since 1956, is
appointed managing director of the telecommunications division. R. M. A.
Jones, a director of Pye Ltd., and
managing director of E. K. Cole Ltd.
since 1962, becomes managing director
of the radio and television division.
C. A. W. Harmer, O.B.E., who has been
a director of several of the Pye subsidiaries for some years, is appointed
managing director of the overseas corn panies. Mr. Harmer is this year's
chairman of VASCA (the Electronic
Valve and Semi -Conductor Manufacturers' Association).
The managing
director responsible for the instrumentation and controls division is F. W.
Coulling who is on the board of several
companies including Ether Langham
Thompson Ltd., Electro Methods Ltd.,
and Ether Ltd.
J. Stuart Samson', A.M.I.E.R.E., is
the new chief engineer of ABC Television Ltd. in succession to Howard
Steele, who (as stated in our April issue)
has joined the Independent Television
Authority as chief engineer. Mr. Sansom joined ABC in 1959 as head of
J. Stuart Sansom
maintenance st Teddington and to 1960
was appointed head of development
(electronics). Since 1963, he has been
head of the cqu.pment and installation
group of the Engineering Department.
Prior to joining ABC Mr. Sansom was
for two years with T.W.W. (Television
Wales and the West) as head of maintenance having previously been with
High Definition Films Ltd.' for four
years.
W. D. H. Gregson, D.F.H., M.I.E.E.,
assistant general manager of the Ferranti Group in Scotland, has been
elected chairman of the Industrial Control and Electronics Board of the British
Electrical & Allied Manufacturers'
Association. He succeeds G. S. Lucas,
group director and general manager of
AEI Electronics, Leicester. The new
vice -chairman is L. E. Thompson, B.Sc.,
A.R.C.S., managing director of West-
James H. Bonnett is appointed editor
of Computer Weekly-the new newspaper for the computer world, to be
launched this autumn by Iliffe Electrical Publications. He was previously in-
dustrial and components editor of Electronics Weekly, and was also for 22
years the group European editor for the
Heywood- Temple electronics publications. Prior to taking up journalism Mr.
Bonnett spent ten years as a design engineer in the electronics industry. Tony
J. Higgins is appointed news editor of
Computer Weekly.
WIRELESS WORLD, JULY 1966
www.americanradiohistory.com
W. D. H. Gregson
inghouse Brake and Signal Company.
Mr. Gregson, who is a member of the
Economic
Government Electronics
(" little
Committee
Development
Neddy ") and a member of the Scottish
Economic Planning Council set up by
the Secretary of State for Scotland, received his technical training at Faraday
House. At the beginning of the war he
was with the Admiralty Signals and Research Establishment at Portsmouth,
followed by service with the R.A.F. He
joined Ferranti Ltd. after the war and
has been in his present position since
1960. Mr. Thompson received his technical education at the Royal College of
Science where he took a degree in
physics, In 1931 he joined Westinghouse and in 1944 became assistant
chief engineer and three years later chief
electrical engineer. He has been group
managing director since 1965.
F. T. Christmas and J. Ayres,
M.I.E.E., have been appointed executive directors of Standard Telephones
and Cables Ltd. Mr. Christmas, who is
to be responsible for special projects,
OBITUARY
F. T.
L. E. Thompson
Raymond F. Brown, O.B.E., has had
to resign from the position of chairman
Christmas
covering the co-ordination of major contracts involving several divisions of the
company, and the establishment of new
plants, has for the past five years been
general manager of S.T.C's Transmission Systems Group. He joined the
company in 1928 as á telecommunication engineer in the central laboratories
and four years later went into the transmission division. In 1936 he became
chief production engineer for the Woolwich factory, moving in 1942 to New
Southgate as deputy chief production
engineer. Between 1957 and 1960 he
was company chief production engineer.
Mr. Christmas is succeeded at Basildon
by Mr. Ayres, who becomes executive
director of the Transmission Systems
Group. He recently joined S.T.C. -as
a special assistant to the managing direc-
and managing director of Racal Electronics Ltd. on his appointment by Her
Majesty's Government as Head of
He is succeeded by
Defencè Sales.
E. T. Harrison who joined the company in 1951 as secretary and chief
accountant at the age of 25 and since
1961 has been deputy managing director.
Mr. Brown, who is 45, started his career
as an engineering apprentice with
Redifon in 1934. He left that company
in 1948 and joined the Communications
Division of Plessey as sales manager.
He founded Racal with G. Calder Cunningham in 1950.
E. R. Laithwaite, Ph.D., M.Sc., Professor of Heavy Electrical Engineering
at Imperial College, has received the
Royal Society's S. G. Brown Award
and Medal for 1966 for his work on
the linear induction motor. Professor
Laithwaite, after war -time service in
the R.A.F., graduated at Manchester
University in 1949, obtaining his M.Sc.
in 1950 following a year's full -time postgraduate research on electronic computer design, and his Ph.D. in 1956 for a
thesis on induction motor development.
He became a lecturer at Manchester in
1953 and a senior lecturer in 1956, and
has occupied his present Chair in London since 1964.
J. H. Mitchell, Ph.D., B.Sc., M.I.E.E.,
F.Inst.P., who has been for some years
technical director of research with
Ericsson Telephones Ltd., and also,
since 1960, chairman of Associated
Transistors Ltd., has joined the board
of Astaron -Bird Ltd., of Poole, Dorset,
with special responsibilities for radar
and telecommunications. Dr. Mitchell,
who joined Ericsson in 1947, was at the
Bawdsey radar research station in 1936
and later at the Royal Aircraft Establishment, Farnborough, where in 1939 he
took charge of radio aids to navigation.
For his " contribution to the development of radar installations " he received
in 1952 an award from the Royal Commission on Awards to Inventors.
J. Ayres
tor-from G.E.C. (Telecommunications)
Ltd., Coventry, where he had been
managing director since 1963. From
1953 to 1962 he was managing director
of Simms Motor Units.
In the note in our last issue (p. 281)
on appointments in the S.T.C. Components Group the position to which
G. Thornton has been promoted was
inadvertently omitted. He has become
product sales manager of the Capacitor
Division.
Vice -Admiral James W. S. Dorling,
C.B., M.I.E.E., M.I.E.R.E., director of
the Radio Industry Council from its
formation in 1946 until 1958, died on
May 12th at the age of 77. He entered
the Navy as a cadet in 1904 and throughout his naval career was a signals
specialist. During the First World War
he was for a time assistant to the Fleet
Wireless Officer, Grand Fleet, and in
1919 became Fleet Wireless Officer,
Atlantic Fleet. He was later successively director of the Admiralty Signal
Dept. and Captain of H.M. Signal
School, Portsmouth. Admiral Dorling
was responsible for the setting up in
1932 of the Royal Naval Wireless
Auxiliary Reserve. Among the positions
Admiral Dorling held during the second
war was that of deputy controller, Admiralty, Bath, and head of the British
Admiralty Supply Mission in the U.S.A.
Air Commodore W. E. G. Mann,
C.B., C.B.E., M.I.E.E., director- general
of navigational services in the Ministry
of Transport and Civil Aviation from
1950 to 1959, died on May 4th aged 67.
During the early part of the war he was
Chief Signals Officer, R.A.F., Middle
East, and on retiring from the R.A.F.
in 1945 joined the Ministry as Senior
Signals Officer and U.K. representative,
Middle East. The Times obituary recalls that it was while in Egypt immediately prior to the war that " he
evolved the first truly mobile radio
communications unit on which designs
were based all the R.A.F. mobile signals and radar units which performed
such vital work with the Tactical Air
Forces." When Air Comdre. Mann retired from the Civil Service in 1959 he
became Middle East representative of
the Decca Navigator Company.
Horace Freeman, who died on May
14th aged 84, organized the first allBritish Wireless Exhibition which was
held at the Royal Horticultural Society's
Hall, Westminster, in 1922. He retired
in 1960 after nearly 40 years in radio
and electrical advertising. He was elected in 1960 the first and only honorary
vice -president of the Radio Society of
Great Britain with which he had been
associated for 35 years.
WIRELESS WORLD, JULY 1966
www.americanradiohistory.com
Transistor Electronic Organ -3
Simple Electronic Organ for the
TOWERS*,
Amateur Constructor
By T. D.
THE first two articles in this series examined electronic
organs generally (May 1966 issue) and their use of
semiconductors (June 1966.) The present article
sen out the basic design of a simple instrument capable
of being built by an amateur with a moderate knowledge
of electronics, carpentry and music.
The transistor organ to be described is a single -manual,
divider-type instrument with twelve " speaking " stops,
and vibrato provision. It does not include some features
conventionally standard in some commercial instruments,
such as pedals, a second manual, sustain, reverberation
and percussion. The reason for the restricted specification is that anyone setting about building it has a better
chance of finishing it than if he tries something more
ambitious. (The " project-mortality " rate is probably
higher in electronic organ building than in any other
branch of amateur electronics!) However, the design is
such that the extra facilties mentioned can be covered by
a separate article later if the demand arises (and the
Editor agrees!).
COMPLETE ORGAN SYSTEM
Fig. 1 gives a block schematic of the various parts of the
design, and is largely self -explanatory. In it you can
easily follow the course of the electrical signals from the
tone generators to the loudspeaker.
Tone generator assembly. -This comprises a set of
twelve tone -" boards ", a vibrato oscillator, and a stabilized d.c. supply.
The set of twelve tone boards are the source of tone
signals. There is a separate board for each note of the
MC, B. Each board gives
chromatic scale, C, C#,
six octave- related square wave outputs in the standard
equal- tempered scale, taken off from an LC master oscillator and five synchronized divide-by -two circuits. For
example, the "C" board gives C6 (2093.0 c/s), C5
(1046.5 c /s), C4 (523.3 c /s), C3 (261.6 c /s =" Middle C "),
C2 (130.8c /s), and CI (65.4c/s).
The vibrato oscillator provides a sinusoidal output at
around 6c /s which can be switched in at will to
frequency-modulate the set of twelve tone -board master
oscillators simultaneously and thus provide a vibrato
effect in the sound output from the organ.
The regulated supply ensures that the master oscillators in the tone boards do not vary in frequency with
mains voltage variation. This helps to prevent the organ
going out-of -tune.
...
Keyboard switching.-From the tone generator, the
square wave signals pass via isolating or " anti-robbing "
resistors (not shown in Fig. 1) to the keyboard switches.
Each playing key actuates a 5 -pole change -over switch.
When the key is at rest, the tone generator outputs are
Newmarket Transistors Ltd.
M.B.E.
shunted to earth. When the key is depressed, the generator outputs are fed into five busbars, one for each of
the pitches-16 ft, 8 ft, 4 ft, 2J ft, and 2 ft. (For the newcomer, 8 ft is " unison " or normal pitch., 16 ft is an octave
lower, 4 ft an octave higher, 2}ft an octave-plus -a -fifth
higher, and 2 ft two octaves higher than the key being
played.)
The output from the five busbars are each amplified at
the keyboard by identical busbar pre- amplifiers, and the
amplified outputs are fed onwards on five busbars to the
tone shaper.
Tone shaper. -In the tone shaper you will see from Fig.
1 that the five busbars feed directly into tone shaping
filters. There are twelve of these filters -two 16 ft, five
8 ft, three 4 ft, one 2; ft and one 2 ft. Individually, each
provides a different sound or " stop " in the organ output,
when a playing key is depressed, and they can be used
together to give various combinations of stop sounds.
The switches controlling the stop filters are mounted
on the separate stop panel, which is usually sited handily
immediately above the keyboard. In its off position, each
stop switch shunts to earth the square wave signal being
handled by the corresponding filter and there is no output
from that filter. When the stop is "pulled ", i.e. switched
in by depressing the stop tab, the switch open - circuits
TONE SHAPER
TONE GENERATOR
BUSBAR MIXER
PREAMPLIFIER
(OSCIBL
TOR) 'CA
PPLY)
y_ - -1
_
STOP SELECTED
TONE-SHAPED
4I
OUTPUTS
SET OF TWELVE
(C
TO
B)
TONE SHAPING'
TONE-GENERATOR BOARDS
(EACH WITH SIX SQUARE -WAVE
OUTPUTS FROM
MASTER OSCILLATOR
PLUS FIVE DIVIDERS)
FILTERS
STOP
SWITCH
SQUARE
WAVE
LEADS
INPUTS
I
4
STOP PANEL
(STOP CONTROL SWITCHES)
--
72 SQUARE
WAVE OUTPUTS!
I
S
L-
---
KEYBOARD
(KEYSWITÇHES)
1<TTI1(PLAYÌNG
SPEAKER
AMPLIFIER
- - -J
- --
KEYS)J1111
I
Fig. 1.
System block diagram
OUTPUT
BUSBAR
!!
---J
SWELL
PEDAL
D.C. SUPPLY
-J
I
IOW
BUSBARS
(MAINS -DRIVEN,
UNSTABILIZED)
of home-constructor's simple single-
manual. bolyphonic, divider -type transistor electronic organ.
351
WIRELESS WORLD, JULY 1966
www.americanradiohistory.com
12 V d.c. supply at 0.5 A to the 9 V stabilized supply in the
tone generator section. Any power amplifier used should
incorporate its own d.c. supply, and the on /off switches
of the power amplifier and the organ d.c. supply can be
ganged together for convenience of operation.
and ceases to shunt the input signal to earth. This allows
the signal to pass through the filter where it is appropriately shaped. The outputs of all twelve filters are
permanently commoned into a single mixer preamplifier.
Swell pedal, amplifier, and d.c. supply. -The output from
the mixer pre-amplifier passes to the swell pedal, which
in this design is a potentiometer volume control operated
by a foot pedal. From the swell pedal, the signal is
transferred to an amplifier feeding a loudspeaker. Any
standard commercial amplifier can be used, but it should
be noted that the " voicing " of the stops (i.e. the design
of the tone filter circuits to get the desired timbre or tone
colour of sound) depends quite critically on the frequency
response characteristics of this amplifier and the loudspeaker system it drives. The design given, therefore,
gives details of the 10 W, 3 -ohm power amplifier that
was specifically used in developing the instrument.
The d.c. supply, shown as a separate section of Fig. 1,
is a mains -driven unit designed to give an an unregulated
GENERATING SQUARE WAVE TONES
Going into more detail on the individual parts of the
system, Fig. 2 sets out full circuits of the items grouped
in the tone generator assembly.
Tone board details.-One tone board circuit is shown in
detail at Fig. 2(c), the others being identical except for the
LC tank components in the master oscillator.
The master oscillator (Tr1, Tr2, Tr3) is a "long tailed- pair" oscillator of my own design which gives a
sinusoid across the LC tank circuit and a square wave
final output from the collectors of Tr2, Tr3.
The oscillator tank circuit inductance, L, uses an
adjustable ferrite high -Q pot core (LA2517, MM484 or
T
r
75
-9V°
9v STABILIZED
(ON
STABILIZED
400TnA
3"x3"PLATE)
-12/I5V
UNSTABILIZED
# N KT 403
12k
561,
VIBRATO
o
JSPEED SWITCH
Sµ
(EXTERNAL)
I.5µ
250µ 10V
250µ
NKT213
IOV
I0k
SET
VIBRATO
LEVEL
ON /OFF
470
SWITCH
(EXTERNAL)
1,000µ
12V
3.3k
9V
ADJUST
ZENER
I.Sk
FOR LOW
IOA
DISTORTION
o+
(b)
(a) VIBRATO DRIVE
L_
L
9V STAB.
D.C. SUPPLY
SIX SQUARE WAVE OUTPUTS
VIBRATO
BUS
(FOR
L,C,C
TEXT)
SEE
Q4
Q
6
Q32
ó9V
STABILIZED
4.7k
4.7k
LcT
Trl
30mA
Ik
k
TO STRING OF FOUR
FURTHER DIVIDERS
IDENTICAL WITH FIRST
270k
Tr2
Ok
0.0Iµ
(ALL TRANSISTORS GERMANIUM
0.0Iµ
A
F
pnp
MEDIUM GAIN EG NKT272
DIODES GE.JUNCTION OR
PC)
0.0lµ
COMMON EMITTER LINE -3V
ZSOµ 10V
.- I00µ
10k
Ik
I,
6V
OSCILLATOR
t
Ist.DIVIDER
(c) SINGLE TONE BOARD (ONE
L_
OF
%
o+
7'L
TWELVE)
VIBRATO
LINES
(d) ELEVEN OTHER SIMILAR TONE BOARDS
-i
oscillator and driver, (b) 9 V, 400 mA stabilized d.c. supply, each of
Fig. 2. Circuits grouped in tone generator assembly, including (a) vibrato
and five dividers, together giving square wave outputs at six
an
oscillator
with
each
boards,
tone
the
twelve
and
(d),
(c)
which is ancillary to
pitches.
octave- related
352
WIRELESS WORLD, JULY 1966
`
.,D..
similar). The tank capacitors,
C and C', are polystyrene
with a negative temperature
coefficient of around 120
p.p.m. / °C which offsets the
cotemperature
positive
efficient of the pot core inductor. The main capacitor
C is 0.035 p.F and the
" padder " C', is selected between 0.001 and 0.007µF to
set up the frequency during
assembly so that the pot core
adjuster slug is in the middle
of its travel. The inductance
values are selected approximately as in Table 1 for the
The
different oscillators.
string of five dividers on each
tone board are identical with
the bistable multivibrator,
Tr4, Tr5, which is a conventional " diode -steered " circuit.
.,A..
TONE BOARD
gm=mg
r7
D2
D
D4
D3
A19
D5
I
TONE BOARD
A29 A39 A4.
A.
A69
A5
A4
D6
4101mOilmilowor
Vr11mwENO=1 r
D5
D4
D3
D2
iiiÍ
ttttt
Whom
1
DI
Dok
TREBLE D
SWITCH
,>
D2
D3
A4
D4
EARTH LINE
D5
16ft
BUSBAR
Sit
BUSBAR
Oft
TABLE
Note
C6
C#6
06
0#6
E6
F6
F#6
G6
G#6
A6
A#6
B6
BUSBAR
I
Freq. (c/s)
Inductance
2093.0
2217.5
2349.3
2489.0
2637.0
2793.8
2960.0
3136.0
3322.4
3520.0
3729.3
3951.0
160
ft
22/3
BUSBAR
(mH)
145
130
115
100
90
80
70
63
56
50
45
Vibrato circuit details. -The
vibrato drive circuit at Fig.
2(a) is an RC feedback oscillator with a buffer driver
stage which feeds a sine -wave
signal into the vibrato busbar
to provide vibrato drive to all
twelve tone boards together.
The oscillator has provision
for varying the output amplitude level by a preset " setlevel " 10kí variable resistance, and is switched for fast/
stow vibrato speed, and
vibrato output on /off. (The
two control switches for these
functions will be mounted
separately away from the tone
generator assembly with the
stop switches on the stop
panel.) The drive from the
common vibrato bus to the
individual tone board oscillators is controlled by a separate
preset 10 kit on each board
feeding into the top end of the
110-2 emitter resistor in the
master oscillator.
The 9 V stabilized power
supply at Fig. 2(b) is a conventional emitter -follower cir-
2ft
7.5k
BUSBAR
22k
Square -wave signal paths from tone generator boards via 100 k isolating resistors to key
switches and thence to five separate busbars for different pitches. The distribution system is illustrated
for the middle and treble D keys. (The 7.5 k and 22 k resistors in the busbars are equalizing networks
inserted at intervals along the bars to attenuate notes progressively towards the low-frequency end
to prevent bass notes from overpowering treble. For details of networks for a complete single busbar
Fig. 3.
see Fig. S.)
Key contacts connection chart; with black note connections omitted for clarity. Observe
"doubling back" of pitches in brackets at top of 2 ft, 23 ft and 4 ft ranks and bottom of 16 ft rank; six
outputs in octaves from each generator board; five change -over switches per playing key; standard
61 note, 5 octave organ manual keyboard, C to C.
Fig. 4.
GENERATOR BOARDS
CI
TO
TOO
CS
C6
Di
EI
TO
TO
F1
TO
D6
E6
F6
(TWELVE)
Fr
GI
G°
A1
TO
TO
TO
TO
TO
TO
F¿
G6
G6
A6
A6
B6
B1
KEYBOARD
2
tt
C3 D3 E3 F3
22/3 'It G2 A2
Ott
8
ft
G3 A3 B3
C3 02 E3
Fj
C1
DI El F1
E1
F1
A3 B3 C4
as C5
D5 E5 F5
04 E4 F4 G4 A4 B4 C5
G1 A1 B1 C2 D2 E2 F2 G2 A2 B2
G
At Bi) CI
I
3
F1
DI
GI
A1
B1
05
A5 B5 C6 D6 E6 F6 G6 A6 86 (C6 D6 E6 F6 G6 A6 66 Co)
D5 ES
F5 G5 AS B5 C6 D6 E6 F4 G6 As B6 (Co cis E6
C3 03 E3 F3 G3 A3 63 C4 04 E4 F4 G4 A4 64 C5 D5 E5
C2
' ii ' u
ï
C1 DI E1
119.
E4 F4 G4 A4
G6)
C2 02 E2 F2 G2 A2 62 C3 D3 E3 F3 C3 A3 By C4 04 E4 F4 G4 A4 84 CS DS E5 F5 G5 A5 B5 C6 D6 E6 F6 G6 A6 66 (C6)
16tt (CI
KEY
E34
5
F1 BI
6
8
A1 B1
10
12
02
E2 F2 G
FS G5
A5 BS C6
B2 C3 D3 E3 F3 G3 A3 B3 C4 04. E4 F4 04 A4
i i i Il i i i i i
as C5
i1i
C2 02 E2 F2 02 A2 62 C3 03 E3 F3 G3 A3 63 C4 04 E4 F4 04 A 64 C5 D5 E5 F5 G5 As 85 C6
13
15
17
18
20 22 24
2.5
MIDDLE
C
27
29 30 32 34 36 37 39
41
42 44 46 48 49
SI
53 54 56 58 60
61
(261-6c/s)
353
WIRELESS WORLD, JULY 1966
www.americanradiohistory.com
OUTPUTS FROM KEY SWITCHES
27
=9V
STABILIZED
C1-
F2-
E2
KEYS
E3
-1
F3
-E4
F4-
F5-
E5
KEYS
KEYS
KEYS
C6
0
KEYS
\I 0
\-4-1
15k
BUSBAR
.8k
n6
FROM BUSBAR
p
TO
PREAMPLIFIER
10k
(b)
(a)
.5. Bus equalization and amplification provision; identical for each of five busbars(I6ft, 8 ft, 4ft, 24 ft, 2 ft); (a) attenuation network
elements inserted between notes E and F in every octave except lowest; (b) bus pre -amplifier with output level setting provided by preset
potentiometer in transistor collector circuit.
Fig.
IM
I00k
220k
I00k
220k
8ft
470k
16ft
DIAPASON
left
o005p
470k
OooSN
47k
8ft
47k
001p
STOPPED
I
8ft
H
DIAPASON
DIAPASON
I00k
IOOk
8ft
8ft
100k
0O02µ
47k
8ft
FLUTE
I00k
150k
0'oosp
8
loo
100k
dit
4ft
STRING
FLUTE
4ft
4ft
REED
STRING
47k
00111
200k
22/3ít
Fig. 6. Stop (tone filter) circuits; square wave input
o-002p
tt
REED
270k
DIAPASON
IOOk
4ft
I00k
0'005/1
from left, out of corresponding bus pre-amplifier
5k
output
resistance;
shaped outputs commoned
to right into single bus
mixer -amplifier input of.
approximately 10k resistance with same circuit as
bus pre -amplifier, Fig. 5 (b).
IOOk
2ft
DIAPASON
354
WIRELESS WORLD, JULY 198e
www.americanradiohistory.com
cuit deriving its unstabilized input from a separate 12 V
d.c. power supply unit which is described later (Fig. 8).
SELECTING SQUARE WAVE TONES BY
PLAYING KEY SWITCHING
The square wave output signals pass from the tone
generator board outputs via 100kí2 isolating resistances to
the key switches and thence to five separate common
busbars for different pitches. This is illustrated in detail
in Fig. 3 for two typical notes an octave apart. The
key switches are shown in the " off " position, short circuiting the generator outputs to earth. When actuated,
they feed the square waves to the appropriate busbars.
The purpose of the 7.5 kk, 22 kn resistor networks in the
busbars is explained below.
The full routeing of the square -wave signals through
the complete key switch assemblies is set out in simplified chart form omitting black notes in Fig. 4 for use
when it comes to wiring up the keyboard.
-It
Busbar equalization.
is necessary to insert progressive attenuating networks in the busbars following the
key switches to scale down the lower frequency notes
which would otherwise tend to overpower higher notes.
In each busbar, breaks are made between the notes E
and F (except in the lowest octave), and attenuator pads
inserted. The provisions are the same in each busbar
and the arrangement for one busbar is given in Fig. 5(a).
The completed electronic organ showing cabinet construction and
mounting of keyboard. Further constructional details will be given
in the final article.
Conventionally the break is usually made between the
notes B and C, but in this design between E and F.
This is because, if it is decided to use a split- keyboard
with the notes from middle F upwards on one set of
busbars, and from middle E downwards on another (to
be able to vary the balance of output between the two
halves), this is possible without much rewiring of the
busbars.
as a collector load enables the output level of each
busbar to be set up independently or " scaled." The
output from each pre -amplifier busbar is connected to
the corresponding tone filter circuits.
SHAPING BUSBAR SQUARE WAVE OUTPUTS IN
STOP FILTER CIRCUITS
Busbar pre-amplifiers. -From the busbar output (D)
in Fig. 5(a) the signals, still square wave, are amplified
in the busbar pre -amplifier (one for each busbar) shown
in Fig. 5(b). The preset 5 kn potentiometer which serves
The filter circuits that shape the square -wave signals
into complex waveforms to simulate various " voices "
in the organ are set out in detail in Fig. 6. As noted
-24V. IA
2,O00µ
68k
60VIA
50V
64p
25V
18V
á'8k
A.C.900mA
00000000QÇ
NKT279A
SWELL PEDAL
VOLUME CONTROL
°
150mA
2,000p
50V
100k
11
32p
ON/OFF
10V
FROM BUSBAR
MLXER AMPLIFIER
OUTPUT
NKT213
0075
25p
64V
22k
L
240V
IOk
120
LOG
52p
A.C.
T4V
F=
Fig. 7.
10
W power amplifier; mains driven;
Sw
47
ON Sin x
oE
3in MIN. HEAT SINK
3 ohm load, 100 mV
input sensitivity;
10
k input resistance; swell pedal volume control at input
WIRELESS WORLD, JULY 1966
355
www.americanradiohistory.com
earlier there are two 16 ft, five 8 ft, three 4 ft, one 2; ft,
and one 2 ft filters. The square wave output busbars
from which the filters collect their inputs are indicated
on the left hand. All the outputs, on the right -hand side,
are fed in parallel into a single mixer pre -amplifier with
the same circuit as Fig. 5(b).
The filter circuits given are fairly typical. Diapasons
have no great accentuation of any frequencies, strings
use high frequency accentuation, reeds use " formant "
tuned -circuit accentuation of a specific band of medium high frequencies, while flutes feature attenuation of
higher frequencies. These are, of course, brutal generalizations, but if you examine each filter in turn you will
see that they conform generally to this. Tone filters are
the ideal area for amateur experimentation. By varying
component values around the values given, you can
modify the tone colours steadily to suit your own " aural
palate."
The stop switches controlling the filters operate by shortand open -circuiting the middle of the filter to earth. This
arrangement has several advantages. Cheap single-pole
switches can be used. The large resistances between
the stop switch and the input to the following amplifier
reduce the key -clicks which arise when the stop switch
is operated. Finally, the actuating of a stop switch does
not substantially change the loading on the busbars at
the inputs or the amplifier at the output.
AMPLIFYING MIXED COMPLEX TONE SIGNALS
AFTER TONE FILTERS
From the output of the mixer pre -amplifier, the complex signals are fed via a screened coaxial cable to the
swell control. This is a 10142 logarithmic potentiometer
Fig. 8. 12V, 500 mA d.c. unstabilized supply for oscillators, dividers
and bus pre -amplifiers and mixer -amplifiers.
controlled by the swell or expression pedal as shown at
the left of Fig. 7.
The remainder of Fig. 7 is the power amplifier proper.
This is a typical transformerless, Class B push -pull circuit, which is able to provide 10 W into a 3-ohm load,
such as four or five 15 -ohm speakers in parallel, or four
3-ohm speakers in series parallel.
POWER SUPPLIES
7 has its own power supply, but
Fig. 8 gives the detailed circuit of the 12 V, 0.5 A, power
supply used to provide the unregulated d.c. for the 9 V
voltage regulator described earlier (Fig. 2(b)) which
supplies the tone generator boards. The busbar preamplifiers and mixer amplifier also derive their d.c. from
this same 9 V regulated supply.
The amplifier in Fig.
BUILDING THE ORGAN
is proposed to give practical information on how to assemble the various circuit sections
described above.
In the next article, it
British Acoustical Society Formed
THE British Acoustical Society has been founded to provide
persons having acoustical and allied interests with a single
society to promote and disseminate knowledge of acoustics
in all its aspects -scientific, technological and medical. The
need for such a society has been apparent for some time
now and in December 1963 the Royal Society set up an
ad hoc committee to explore the possibilities.
The committee, headed by Sir Gordon Sutherland (chairman of the British National Committee for Physics), reported
in March 1965 and a provisional council was established to
launch the society. The chairman of the society is Dr. A. J.
King (Manchester University); vice -chairman is Professor
E. J. Richards (Southampton University) and the joint hon.
secretaries are Dr. P. Lord (Royal College of Advanced
Technology, Salford) and Dr. R. W. B. Stephens (Imperial
College, London).
Members of the council are : W. A. Allen (Associated
Architects and Consultants), Professor R. E. D. Bishop (University College, London), D. E. Broadbent (Applied Psychology Research Unit, M.R.C.), Professor W. Burns (Charing
Cross Hospital Medical School), C. T. Chapman (Derritron
Ultrasonics Ltd.), Professor C. Cherry (Imperial College),
H. Creighton (consultant), P. E. Doak (Southampton University), F. B. Greatrex (Rolls -Royce Ltd.), Dr. P. Grootenhuis (Imperial College), Professor G. M. Lilley (Southampton
University), Dr. H. D. Parbrook (Liverpool University), A. T.
Pickles (Building Research Station), Dr. L. H. A. Pilkington
(Pilkington Brothers Ltd.), Dr. D. W. Robinson (National
Physical Laboratory), T. Somerville (B.B.C. Research Dept.),
Dr. W. Taylor (St. Andrews University), Professor D. G.
Tucker (Birmingham University), Dr. J. Dunstead (Admiralty
Underwater Weapons Establishment), and F. E. Williams
(G.P.O.).
The proposed constitution and rules have been issued
and are subject to an Extraordinary General Meeting to be
held in October this year.
Applications for membership
should be made to Dr. R. W. B. Stephens, Department of
Physics, Imperial College, Prince Consort Road, London,
S.W.7.
The inaugural meeting was held in May at Imperial College
when papers on the subject of aircraft noise were presented
by invited speakers from Britain, the U.S.A., Switzerland and
the Netherlands. In July there will be a four-day meeting
(12th-15th) on " Numerical methods for vibrational
problems " and a two -day meeting (20th -21st) on " Impulsive
noise and sonic bangs," both being held at the Institute of
Sound and Vibration, Southampton. On the 19th there is a
meeting on " Non -linear underwater acoustics " at the Dept.
of Electronics & Electrical Engineering, Birmingham
University.
" The
Field- effect Transistor at Y.H.F."
WE have been asked by Texas Instruments Ltd., of Bedford,
to point out that they were not in any way associated with the
circuit design work described in U. L. Rohde's article " The
Field -effect Transistor at V.H.F. " (January 1966 issue) and therefore regret that they cannot deal with readers' enquiries about the
article. The f.e.t. types referred to in the article were erroneously
described as insulated gate m.o.s. transistors, whereas in fact they
are junction gate f.e.ts (see subsequent correspondence, April 1966
issue, p. 197).
WIRELESS WORLD, JULY 1966
356
www.americanradiohistory.com
Transistor 2 -Metre Converters
F.E.T. design with valve -like performance and a low -cost
design using germanium epitaxial planar transistors
By U. L. ROHDE
TWO years ago the author published a transistor circuit
for a 144 -Mc /s converter in a German journal' and in
this design the low noise figure of 1.8 dB was obtained
at the expense of image rejection and cross modulation.
Now that field-effect transistors are available which are
even better than some of the valves used in this type of
application, it has been possible to produce a new converter with a superior large- signal performance to that
of valve designs. The author has also developed a lowcost converter using germanium transistors, and the circuit of this will be given in the latter part of the article.
F.E.T. circuit.-Fig. 1 is the circuit of the converter using
field-effect transistors. It can be seen that the input circuit
has a junction f.e.t. with a series resonance circuit between
gate and source electrodes. The 2N3823/T1S34 junction
f.e.t. made by Texas Instruments has following two -port
parameters.
Input resistance
..
Input capacitance..
Transconductance
..
2 kû
6 pF
6 milli -mhos
1.5 pF
..
Feedback capacitance
Output resistance
10 kIl
Output capacitance
21 pF
The capacitive voltage divider across the secondary of
the input transformer, together with the output circuit of
the transistor, transforms the input admittance so that
power matching with simultaneous noise minimum is
obtained at the input. As a result of the much smaller
increase of noise from the noise minimum with the same
noise factor at the band edges, a considerably wider trans+12V
10p
2N3823/TIS34
V---
60íl
6
82
/
°
35p
Converter circuit
Fig. I.
using field effect transistors.
2N3823/T1S34
8p
T
DA73
T500p
+12V
Ik
A.C.
G:
WV
5k
--
GM290/TIXM07
3 -15p
500p
.20k
T
Construction of converter
using field -effect transistors.
mission bandwidth is obtained. The bandwidth, to the
-3 dB points, is 3 Mc /s.
In order to obtain the highest possible image rejection,
a high -Q bandpass filter circuit is used between the
r.f. amplifier and mixer at 144 Mc /s. The lower part of
the r.f. amplifier section in Fig. 1 is concerned with the
correction of waveform deformation. A small fraction
of the signal power is taken from the output of the r.f.
amplifier and fed to a diode circuit, which is assumed
to have a quadratic charactertistic. The voltage developed
by this circuit, between points A and B, can be considered
as a variable -voltage battery. From Fig. 2 (a) it can be
-
357
WIRELESS WORLD, JULY 1966
www.americanradiohistory.com
(a)
The oscillator uses a crystal operating at 116 Mc /s, but
a modified version for 172 Mc /s is currently being examined. This will avoid the image coming in the 88 to
108 Mc /s broadcasting band with an i.f. of 30 Mc /s. The
oscillator has a distortion of 1 %, which is substantially
(b)
better than in the earlier version of the converter.
Measurements :
Power gain of pre -amplifier
Power gain of mixer ..
..
Image rejection ..
Gain control ..
Noise figure
Input voltage for 1% intermodulation
Wipe -out effect for 3 dB loss of gain
.
.
..
..
15 dB
80 dB
35 dB
2 dB
800 mV
300 mV
Cross-modulation of 1 % will be generated between one
signal of 2 mV, unmodulated, and another of 300 mV,
40 % modulated, spaced in frequency by 500 kc /s.
i
Illustrating the " double deformation " process : (a)
deformation caused by square -law characteristic of r.f. amplifier;
(b) cancellation of distortion by "variable battery" effect.
Fig.
10 dB
1.
Low-cost converter.-Using germanium epitaxial planar
transistors type 'TIXMO5 /7 (made by Texas Instruments), a spot -noise figure of 2.2 dB can be achieved
with the circuit shown in Fig. 3. For the mixer a silicon
epitaxial planar transistor type TI408 by the same manufacturer has been chosen, giving 10 dB better cross modulation than a germanium type.
The converter uses a rather sophisticated filter design
providing high image rejection and low harmonic mixing.
The transistors, because of the relatively high operating
emitter current of 2.5 mA, offer a more linear characteristic than is commonly obtained. The bandpass filter
has relatively high losses, but these are offset by the
high gain of the input amplifier. The crystal- controlled
oscillator uses a 5th overtone crystal providing a stability
of better than ± 50 c/s for large changes of temperature
and supply voltage.
From the input, the r.f. signal is applied to a bandpass
filter, which suppresses unwanted image frequencies and
harmonic mixing. This filter,
is a parallel arrangement of two series resonance circuits, one resonating
at 97 Mc /s and the other at 203 Mc /s, and the values
of the capacitors have been chosen so that a parallel
resonance occurs at 145 Mc /s.
seen that the square -law characteristic of the r.f. amplifier
transistor would normally cause a deformation of the input
waveform. Owing to the 90° phase shift within the transistor, however, the effect of the " variable battery " is to
cancel this distortion (which is predominantly 2nd order)
as shown in (b). Because of this " double -deformation "
all mixing products resulting from unwanted stations are
cancelled and the mixer receives a substantially distortion free signal.
The signal from the bandpass filter is applied to the
gate electrode of the additive mixer, which also uses a
junction f.e.t. Since the junction f.e.t., because of its
structure, has an exact square-law mixing characteristic,
under no circumstances can ambiguous signals appear,
not even with overloading at the input. Thus the
problem of overloading by nearly 100 -watt amateur transmitters in the band is solved. The mixing transconductance has been found to be 3.3 milli -mhos. Since the gate
electrode is "hot" with respect to i.f., the output resistance will be increased (at an oscillator voltage of 1.8 Wu).
F
5.6k
Ip
5.6k
60f1
3p
3p
II
60p
FI
TIXMO5
o
500p
8p
i
¡
k
500p
1
5p
T17P
Fig. 3. Converter using germanium transistors. (F, - -=F2= series resonance at 97 Mc /s
and at 203 Mc /s with parallel resonance at 145 Mc /s. F3 =series resonance at
19 Mc /s with Q = 10).
Mixing amplification is then identical to that of straightforward amplification. Because the oscillator employs
a crystal (using the 3rd overtone) no bridge arrangement
is necessary to isolate the input voltage and the oscillator
signal. Since the i.f. bandpass circuit requires a 2 Mc /s
bandwidth, it must be damped at both ends, and this is
done by matching in the secondary circuit of the transformer.
358
WIRELESS WORLD, JULY 1966
www.americanradiohistory.com
In the amplifier use is made of the well -known cascade
arrangement, which is very rarely used with transistors,
since this configuration allows high gain with the best
possible stability. In fact, it was found extremely difficult
to make this stage oscillate, and only with heavy external
feedback could this be achieved. A second bandpass
filter, using three 145 -Mc/s circuits, one as already
described, is used to suppress unwanted heterodyned
signals. The amplified signal is then applied to the
silicon transistor mixer, which, because of the high baseemitter voltage of the transistor and the high oscillator
voltage of about 700 mV, offers an improvement
of
10 dB in linearity. The mixer output
is tuned to 28 -30
Mc /s.
The oscillator requires zero phase shift between the
emitter and collector currents. This is provided by the
input tuned circuit consisting of a 10-pF capacitor
and
a variable inductor. In the feedback loop
the 5th overtone crystal (made by Quarzkeramik of W. Germany)
operates at 116 Mc /s. Several crystals made by
companies were tested, but they failed to give the other
stability of ± 50 c/s over the temperature range high
to +85 °C provided by this type. No frequency -10°
greater than 50 c/s can be made in the tuned change
for after detuning by 50 c/s the oscillator rapidlycircuit,
stops
working. A sampling oscilloscope (Tektronix)
was used
to control the harmonics of the oscillator,
spectrum analyser adaptor for the 'scope a and with the
distortion of
0.5
was measured. This was about the same
as with
the f.e.t. circuit.
Measurements :
Power gain
28dB,
Spot noise figure
2.2dB
Image rejection..
90dB
Suppression of harmonic mixing
65 dB
Grossmodulation will start with an input
signal of
. .
100 mV into 6052
Intermodulation will start with one signal of 5 mV,
and another of 70 mV. Fig. 4 shows the spot noise figure
1
f
= 200Mc/s
RG
= 7sf1
,I
TIXM05
6
4
VC9=-4V
111111j7= -IOV
2
2
3
4
EMITTER CURRENT
Fig. 4.
7
5
(mÑ
Spct noise figure vs. collector current of TIXMOS transistor.
of the TIXMO5 transistor at 200 Mc /s, the value at
150 Mc /s being 1.9 dB.
REFERENCES
-m converter with transistors, by U. L. Rohde. Funk technik, vol. 11, 1964, pp. 412 -414.
2. Design of v.h.f. cascaded stages with u.h. f. silicon
transistors and u.h.f. m.o.s. f.e.ts, by U. L. Rohde. Elektronische
Rundshau, vol. 11, pp. 633 -640.
3. The field -effect transistor at v.h.f., by U. L. Rohde.
Wireless Wcrld, January, 1956, pp. 2 -6.
1.
2
Literature Received
"Microcircuits Planar Selector " gives brief
of the
wide range of microcircuits currently availabledetails
from SGSFairchild Ltd., 23 Storefield Way, Ruislip, Middlesex.
The
range is divided into three groups; military, professional,
and industrial.
WW 338 for further details
" Designers' Digest 4," issued by the Standard Telephones
and Cables Ltd., Components Group, Footscray,
is
intended as a quick 47 -page reference to componentsKent,
manufactured and marketed by the Group.
WW 319 for
further details
WW 342 for
" Solid -state video switching and mixing equipment "
(B /VSE issue 3) has been received from the Broadcast
and
Recording Equipment Division of E.M.I. Electronics Ltd.
This 19 -page publication is concerned with E.M.I. solidstate equipment designed to replace electromechanical
switching and mixing equipment.
ww
3411
Four application reports produced by Brookdeal
tronics Ltd., Myron Place, Lewisham, S.E.13, describeElec(1)
optical spectrometer measurements with the phase sensitive
detector; (2) measurement of Hall effect using a phase sensitive detector; (3) measurement of the edge shift in gallium
arsenide using a phase sensitive dectector; and (4) the a.c.
operation of thermistor bridges.
further details
further details
"Loctite Single Components Structural Adhesives is a
general description 4 -page leaflet No. S155 covering " most
of the Loctite self- curing resins. Full technical data
any
one of six new adhesives is available from Douglas onKane
(Sealants) Ltd., Swallowfields, Welwyn Garden City, Herts.
WW 343 for further details
for further details
WW 301 for
" Induction Brazing," a new technical data
on low temperature silver brazing, explains why high-sheet
frequency
duction can provide the most rapid form of heating and inparticularly useful in the brazing of large runs of ferrousis
components of relatively simple shape. Uniform and reproducible results can be obtained at high speed and heating
cycles can be controlled by automatic timing devices.
This
data sheet 110;143 is one of a series on low- temperature
silver brazing issued by Johnson, Matthey -& Co. Ltd., 73
-83
Hatton Garden, London, E.C.1.
"Silicon & Germanium Transistor Selection."-Considerations that must be made when designing equipment around
either silicon or germanium devices are discussed, together
with advice on making the choice between the two types
in the 31 -page booklet by Newmarket Transistors Ltd.,
Esening Road, Newmarket, Suffolk. Tables are included,
indicating the more commonly used silicon and germanium
devices both industrially and professionally.
WW 344 for further details
WIRELESS WORLD, JULY 1966
359
www.americanradiohistory.com
Selective Null Detector
A tunable null detector for the range
25c /s to 100kc /s is introduced by
PRODUCTS
VIEM
"ivL
o
and equipBECAUSE of the vastness of the range of instruments, components
&
Electronics
Instruments,
recent
the
at
exhibitors
ment shown by the 800
of
Automation Exhibition held at Olympia we have not attempted a review
the Show but have chosen the following items for our New Products Section.
S.T.C.
The selective detector, type
26 16 -A, has a sensitivity of 1µV for
full -scale deflection. It can be used as
a bridge balance detector and as a low -
noise pre -amplifier for oscilloscopes and
electronic voltmeters.
Either a linear or a logarithmic response can be obtained from the output
amplifier by operating a switch, thus
simplifying the search for a null point
by widening the scale in the logarithmic
condition. Provision -is made for using
external filters to extend the scope of
INSTRUMENTS
Digital Gaussmeter
A Hall -effect probe with linear flux
response, good thermal stability and
temperature compensation is one of the
main features of the F. W. Bell (American) Model 660 gaussmeter just introduced into the U.K. by Livingston
Laboratories Ltd., North Watford, Herts.
Furthermore, in the four higher ranges
(0 -100 G to 0 -100 kG) the instrument
gives its read -out in digital form on
three manually operated dials. On the
two lowest ranges (0 -1 G and 0 -10 G)
the indication is presented in conventional analogue form on a meter.
In the digital mode of operation the
three dials are operated to give a null balance with the signal from the Hall
probe, and this is shown by a centre zero indication on the meter. Further
resolution is obtained by using the meter
scale for interpolation between digits on the last
(least -significant) dial. Analogue measurements 0.1 at
G full scale can be made
using the 0 -100 G digital
range.
The temperature dependence of probe sensitivity
(compensated, digital mode
of reading,
percentage
-20 °C to +65 °C) is
0.005 %, per °C nominal.
-
WW
301
TRANSISTOR AND DIODE TESTER
A transistor and diode tester, type TT
537, is announced by Avo. Provision is
made for measurement of transistor hFF
up to 1500 at a frequency of 1 kc /s and
the measurement of leakage current down
to 1µA. Collector currents up to 1 A
can be measured and base currents from
100 nA to 50 mA.
Both forward and reverse characteristics of diodes can be measured. Reverse
characteristic up to 1000 V can be measured (at 200 µA) and to 100 V at 3 mA.
(Avo Ltd., Vauxhall Bridge Road, London, S.W.1.)
WW 302 for further details
360
for further details
the instrument, which is suitable for
field use since it operates from two 6 V
internal dry batteries.
The instrument has an input impedance of 25 -k11-1 M!' depending on
gain control setting and output impedance is about 300 SI in series with 5 pF
Maximum output is about 1 V r.m.s.
ww
303
for further details
DIGITAL TACHOMETER
Speeds of rotating parts are indicated
digitally on a four-numeral indicator in
digital tachometer introduced by
a
Smiths Industries, Industrial Instrument Division, Wembley, Middx. The
instrument operates from pulses received
from, say, an inductive pick -up, and the
standard instrument is calibrated on the
basis of 60 pulses /rev. (though other
ratios can be provided for). The pulse
rate is sampled at intervals of either
I or 5 sec, as selected by a switch on
the front panel, and the speed indication, in r.p.m., is held steady during
these intervals. Any input pulse frequency up to 100 kc /s can be accepted
and the pulse amplitude can be anything from 200 mV peak to 175 V peak.
The minimum input pulse width permissible is 5 µsec.
WW 304 for further details
WIRELESS WORLD, JULY 1966
PRE -SET COUNTER -TIMER
HIGH- OUTPUT PRESSURE TRANSDUCERS
counter -timer (model 805R)
using silicon transistors, is offered by
Racal Instruments Ltd., Dukes Ride,
Crowthorne, Berks.
It is intended primarily for industrial
application. The digital timebase is
variable between 1 ms and 100 s, and the
five -digit readout will display such units
as gallons per hour, or revolutions per
second, etc. The instrument may be
used as a frequency meter, ratiometer,
chronometer for time interval measurement or as a delay generator (1 ,s100 s). Optional features available include serial output for a digital printer,
remote display and supplementary
switching modules.
The crystal standard has a stability
of ± 1 part in 10" long and short term.
Sensitivity is 100 mV and input impedance is either 10 or 100 kf2.
Pressure transducers with integral
solid -state d.c. amplifiers giving a full range output of 1 V d.c. (at up to 1 mA)
have been introduced by Intersonde
Ltd., The Forum, High Street, Edgware,
Middx. Available for liquid or gas
pressure ranges from 0 -100 p.s.i.g. to
0- 10,000 p.s.i.g., these devices, because
A pre -set
WW 305 for
The tubular beryllium-copper pressure sensitive element has bonded to
it resistance strain gauges wired into a
four -arm bridge circuit; which is energized from a 24 V d.c. supply (a separate unit available from the
company). Applied pressure
unbalances the bridge, and the
unbalance signal is amplified
in the d.c. amplifier
Ferrand ZLD2S integrated circuit. Static or varying pressures can be measured, and
the operational temperature
range of the transducers is
-40 °C to +120 °C. Electrical connections are flying
leads or Plessey plugs and
sockets. (Transducer seen on
left of picture.)
-a
further details
Transistor scope with
Plug -in Units
One of a new range of instruments from
Cossor Instruments, of Harlow, is the
CDU110 general purpose oscilloscope,
priced at about £330. The timebase
sweep range extends from 0.2 %s cm-' to
0.5 s cm-' and a 5 x magnification control extends the fastest sweep speed to
40 ns cm-'. Single sweep and sweep
delay facilities are provided.
The vertical deflection system incorporates a 180 ns signal delay so that
the leading edge of the sweep triggering
waveform can be observed. Three plug in y amplifiers are available: a single
trace unit (CAM 110) with a sensitivity
of 5 mV cm-' and a bandwidth of
20 Mc /s; a double trace unit (CAM
111); and a differential unit (CAM 112)
with a sensitivity of 500íV cm-t, a 1
Mc /s bandwidth and a common mcde
rejection of 10,CC0 to 1.
WW 306 t:r further details
of their large outputs, give- signals which
can be directly indicated on 0 -1 mA
meters calibrated in p.s.i. or other units.
WW 307 for
lurther details
Logic Trainer
training unit for demonstration.
experiment and teaching of modern control and computer techniques was
shown by Allen West Automation Ltd.,
of Brighton, and utilizes the industrial
range of static switching modules produced by the parent company Allen
West & Co. Ltd. The equipment comprises a cabinet, containing a power supply
unit, with an inclined
A logic
with jumper leads, which have colour
coded plugs, into socket positions on the
faces of the modules. The input module
has switches which give a " 1 " and
"0" output; presence of a 1 output is
indicated by an energized lamp. Logic
levels are 0 (false) between 0 and 1.1 V,
1 (true) between 8.2 and 12.9 V.
face
to which
10
modules covering up
to 60 logic elements
can be fitted. Functions of modules currently available are
AND,
OR,
NOT,
NAND,
FLIP -FLOP, INPUT, DISPLAY, STEPPING UNIT
a n d
MISCELLANEOUS
COMPONENTS.
Th e
logic elements
the
within
module are
arranged in groups of
6 or 4 and are of the
resistor -transistor logic
type except for AND
and NAND which are diode -transistor
logic. Power supplies are made to the
elements via a connector mounted at
the bottom of the module; fitting the
module in position makes the connections. The positions of the modules
on the cabinet are interchangeable.
Interconnection of logic signals is made
WIRELESS WORLD, JULY 1966
The power unit built in the cabinet
operates from 230 -250 V a.c. and supplies the ten modules with a stabilized
operating voltage of ± 12 V. The power
unit will also supply an extension
cabinet which can be used to increase
the number of logic elements to 120.
WW 308 for further details
361
www.americanradiohistory.com
Low -cost Digital Voltmeter
Only one switch, the range selector, is
involved for most measurements on the
LM 1450 digital voltmeter, manufactured by Solartron, of Farnborough.
Range maxima are from 20 mV to
1,000 V, with a resolution of 10 V on
the lowest range. Accuracy is ±0.05 ,,
of reading ±0.05% of range for a
+5% variation of mains voltage. Input
resistance is 100 M12 on ranges less
than 2 V and 10 M!2 otherwise. Internal
calibration is p.cvided by a low
temperature coefficient Zener diode.
Common mode rejection at 50 c/s can
be as high as 140 dB with a filter which
has a 60 dB rejection from 50 -120 c /s.
Three modes of operation are commanded by front panel pushbuttons:
an auto mode, when the instrument
samples at 50 readings per second; a
slow mode, providing repetitive sampling at approximately one reading every
two seconds and single shot sampling.
In operation the LM 1450 converts
the unknown voltage into a proportional time interval, this time interval
then being measured and displayed
using conventional counter circuitry
with an internal crystal clock. The
initial conversion depends upon charging a capacitor to a voltage derived from
the input signal and then discharging
the capacitor with a constant defined
current. The time taken is then a direct
measure of the unknown voltage.
WW 309 for
further details
DOUBLE -BEAM OSCILLOSCOPE
The D56 special purpose oscilloscope,
manufactured by Telequipment Ltd.,
Chase Road, London N.14, is designed for applications involving twin
timebase techniques. Two independent
timebases are provided, each variable
between 0.5 us cm-' and 5s cm-' The
B timebase can be used to delay the
A timebase, allowing a selected portion
of the lower trace to be expanded and
shown on the upper trace.
Two marker facilities are provided.
One, a television line marker, uses a
0.5 V pulse from the A timebase which
may be fed to a video monitor for line
identification of video waveforms. An
internal 10 Mc /s generator produces
marker pips on the upper beam for rise
time measurement.
The two identical y amplifiers have
bandwidths of 15 Mc /s and sensitivities
of 10 mV cm-1. Price is about £300.
.
WW 311 for further details
0.1 per cent Universal Bridge
One of the most recent instruments to
be announced by Marconi Instruments,
of St. Albans, is the TF 1313A general
purpose impedance bridge with a measurement accuracy of ± 0.1%.
At balance, the L, C or R value is
read from two concentric dials of the
coarse and fine balance controls. The
coarse control is 110- position switch and
the reading from this is added to that
from the continuously variable fine control. There are eight resistance ranges
with maxima from 1112 to 110 M12 and
permitting measurement down to
0.00312. Inductance and capacitance
ranges cover 0.1,uH to 110 H and 0.1 pF
to 110µF with lowest range maxima of
110 ;4H and 110 pF respectively. On
the resistance ranges, the bridge is energized with d.c. and a photoelectric
chopper is used to provide a.c. for the
amplifier preceding the balance indicator. The internal a.c. bridge source
operates at 1 or 10 k /cs, but measurement can be made at any frequency
with the use of an external source. An
output is provided for use with an external balance indicator. A polarizing
voltage of up to 350 V is provided for
measurement of electrolytic capacitors.
The Q range extends to 310 and D
( =1 /Q) range to 0.0005. Accuracy on
these ranges is about ± 5
of reading
and ± 5% of reading plus respectively
f.s.d.
3
f
WW 310 ti.r further details
INFORMATION SERVICE FOR PROFESSIONAL READERS
To expedite requests for further information on products appearing in the editorial and
advertisement pages of Wireless World each month, a sheet of reader service cards is
included in this issue. The cards will be found between advertisement pages 16 and 19.
We invite professional readers to make use of these cards for all inquiries dealing
with specific products. Many editorial items and all advertisements are coded with a number, prefixed by WW, and it is then necessary only to enter the number on the card.
Postage is free in the U.K. but cards must be stamped if posted overseas. This service
will enable professional readers to obtain the additional information they require quickly
and easily.
362
Portable Weighing
Equipment
Maximum loads between 200 lb and
2,090 tons can be measured by a new
portable weighing equipment consisting
of a load cell and an electronic indicator
unit measuring only 8 in x 6 in x 10 in.
Made by Ekco Electronics Ltd., South end-on -Sea, Essex, the equipment uses
either a single load -cell or a flat weighing platform containing a number of
small cells. Weight is sensed by the
resistance change of the strain gauge
elements in the load cell and indicated
by a null -balance method using a ten turn 3 -digit dial and a centre -zero meter.
Accuracy is claimed to be +0.25% of
full load and the temperature coefficient
0.0002% of full load per °C.
WW 312 for
further details
WIRELESS WORLD, JULY 1966
www.americanradiohistory.com
COMPONENTS AND MATERIALS
J-band "Agile" Magnetron
A 1-band coaxial pulse magnetron, the
L -4500, capable of high -speed random
tuning in the range 15.5 -17.5 Gc /s, and
rapid dithering about a given frequency
to improve radar resolution by reduction of clutter and glint, has been developed by the Litton Industries Electron Tube Division of California. It
is also capable of precise electronic settini to any fixed frequency within the
range.
The unique feature of this
magnetron is the electro- magnetic actuator that works on the same principle as
a loudspeaker, except that it employs a
servo feedback. The tuner follows any
signal -sine, triangular, or square wave
-and provides an output voltage for
monitoring the frequency to an accuracy
of 0.3 X, under all environmental conditions. There is also a low voltage
cooker magnetron, the Microtron, which
is a permanent magnet, air cooled tube,
operating at 3.6 kV. Power output is
1 kW at 2,450 Mc /s.
Warm -up time is
3
seconds.
Available in the U.K. from
High Speed Counting
Module
The E.N.M. high -speed electro -mechanical counting module with electrical
read -out, is described as the Series 444
UNIDEC. A single decade counter, in
modular design, it can be arranged in
groups as a high-speed counter with any
number of digits and it is fitted with
internal contacts to enable groups to be
driven in either serial or parallel form.
These contact: will permit remote resetting to zero, and provide an electrical
read -out. The count frequency is 60 c /s,
with minimum pulse length of 8.5 ms at
109r, below rated voltage.
Maximum
pulse length is continuous duty at 25 ° C
ambient and 10
above rated voltage.
Nominal count voltage 12 V d.c. or 24 V
d.c.; coil resistance (nominal) 52 or 210
onms. The reset pulse length is 200 ms.
The figure size is 0.14 in high by 0.09 in
wide and is black on a white background. The front panel is 1.68 in high
Litton Precision Products, 503 Uxbridge
Road, Hayes, Middlesex.
WW 313 for further details
Tantalum Film Resistors
Sample arrays of 12 fixed resistor elements having a common substrate are
now available from A.E.I. These sample
arrays (TFC 4/1 and 4/5) are typical in
every aspect of those elements used in
tantalum film hybrid circuits.
Since
these sample resistors are not internally
connected, any combination of the 12
elements with each other, or with any
of borosilicate glass, the contacts are
gold /chrome evaporated films, the whole
array being inside a moulded polystyrene package 2.4 x 1.3 X 0.28 in. The
elements consist of two each of the following values 10052, 1k, 5k, 10k, 25k,
50k, while the recommended power
ratings at 20 °C ambient are 100, 100,
100, 100, 200 and 500 mW respectively.
The tolerance on nominal value for
TFC 4/1 = ±1";, and for TFC 4/5=
r 5 ",.. Temperature coefficient of resistance less than ± 50 p.p.m./ °C. From
Associated Electrical Industries, Re-
by 0.25 in wide. A printing wheel is
available in addition, for printed readout. Overall length is 4.216 in, and
body height 1.125 M. English Numbering Machines Ltd., Queensway, Enfield,
Middlesex.
ww
316
for further details
:
search Laboratory, West Road, Temple
Fields, Harlow, Essex.
ww
311
for further details
Solid -state Gas Igniter
Brush -Clevite Mark 1
gas
igniter employs a patented method of
mechanically distorting two piezoelectric
ceramic elements to produce a series of
sparks. Appliances such as fires, cookers,
heaters and lighting equipment can have
their gas fuel ignited by this solid-state
device. It does not require batteries or
power supply. The energy produced by
the Mark 1 will ignite the following
gases :
acetylene, butane, hydrogen,
methane, pentane, petrol, propane and
natural gas. Brush -Clevite Co. Ltd.,
Hy the, Southampton.
"l'his
ether device or component may be made
by the use of external connections. The
leads which are gold- plated nickel tapes
may be soldered or welded. The resistor elements are etched and anodised
sputtered tantalum films on a substrate
WW 315 for
further details
SILICON PHOTO SENSOR
The MSP /3 and MSP /6 are composite silicon photo -electric sensors that
can operate relays directly in optical
links that utilise gallium arsenide light
sources. With a peak spectral response
of 9500 .C, they have an output current
intensity of 100 mA, and a maximum
voltage rating of 25 V. These sensors
can also be used in photo -electric
systems employing visible light sources.
They are 25.4 mm long and have a diameter of 6.4 mm. Sensitive area is 2 mm
diameter. The MSP /3 has a frequency
response of
dB at 10 kc /s, and a dark
current of 0.003 mA at 25 °C. The
MSP /6 has a frequency response of
dB at 3 kc /s, and a dark current of
0.02 mA at 50 °C.
Manufactured by
-3
-3
M.C.P. Electronics Ltd., Station Wharf
Works, Alperton, Middlesex.
ww
317
for further details
WIRELESS WORLD, JULY 1966
363
4
www.americanradiohistory.com
+
Magnetic Materials
Thumbwheel Rotary
Switch
Among the magnetic materials shown
by Standard Telephones and Cables
Ltd. was a range of high permeability
ferrite ring cores which although
designed especially for pulse and wide band transformers are suitable for other
applications. The cores are available
in two grades (SA503 and SA601) of
high stability manganese -zinc ferrites
with low losses. The main differences
in the properties of these materials are
listed in the following table.
Property
SA503
SA01
Initial permeability
2250±20°ó
3500±20%
3750
5800
0.14
0.12
Maximum perme-
ability
Coercivity Bmax =
4000 (Oe)
Temperature coefficient of permeability from 20'C to
70 C
per0`C)
0.21-0.1
Typical loss factor
('
at 300 kc/s
10
0.35±0.35
12
The proportion, of the cores ensure
maximum inductance per turn and the
range of dimensions is such that for
a given value of A,, or Ar /I,.. he
NEW GRADES
designer always has a choice of two
sizes. Twelve core sizes ranging from
an o.d. of 4 mm and i.d. of 2 mm to
an o.d. of 24 mm and i.d. of 12 mm are
available. Cores are normally available
with a tolerance of ± 25% on the
specific inductance value. This tolerance
allows for variations of material permeability and core size. Closer tolerances
are available. Cores are finished with
two coats of stoved enamel so that self fluxing wire can be wound directly on
the core without danger of insulation
breakdown.
WW 318 for further details
A thumbwheel or edge control rotary
switch by Nucleonic Accessories Ltd. of
Lee Green, Mirfield, Yorks., is produced
in a wide variety of output codes, that
include decimal, binary coded decimal
(with and without complementary codes)
or zero and parity contacts. The switch
has 10 positions and is available with or
without detent. Rotation may be continuous (360 °) or stops may be provided
to limit the rotation to any position
from any two adjacent ones (36 °) up to
the full ten positions (324°). The
printed circuit is gold plated, and a life
expectancy of at least 1 million revolutions is claimed for it. Maximum current rating is 0.5 A, with an insulation
resistance greater than 1000 MO. Width
0.5 in, height 2.3 in (overall), and fixing
is by two studs. Connections may be
made by direct soldering or edge conModels available : -Type A
nectors.
1 -pole 10 -way break before make; Type
B 10- position for binary coded decimal.
Also available is a mains version (250 V
a.c. S A; 110 V a.c. 10 A) with micro switch for instrument on /off control
operated by a thumbwheel.
ww
321
for further details
OF INDUSTRIAL LAMINATES
Bakelite Limited. London, S.W.1, have
new
paper -based and
introduced
Grade
epoxide -glass based laminates.
copper -clad
" EW
54943 natural "
laminate is paper based and can be
guillotined and punched at room
heat
temperatures, thus obviating
This
grade,
problems.
shrinkage
intended to comply with the relevant
British Standards, has all the characteristics of the American National
Electrical Manufacturer's Association
(NEMA) standard for XXXPC laminates and is suitable for many applications in the electronics industry. The
Bakelite range of epoxide glass fabric
laminates has been extended with new
grades which have fire retardant and
cold punching properties combined with
excellent mechanical and electrical
characteristics. " DH 138 Natural," in
standard form and "DH 90 /E
Natural," the copper -clad version, are
intended to comply with relevant
British Standards and with the fire
retarding requirements of NEMA FR4.
The copper -clad version employs an
adhesive system specially formulated to
provide additional resistance to electroplating conditions and translucency of
the substrate is such as to allow the
circuit pattern to be visible from the
opposite side of the board in daylight.
WW 319 for
further details
MINIATURE MERCURY CELL
Claimed to be the world's smallest primary dry battery the Mallory RM -212 is
0.22 in wide, 0.13 in high and weighs
0.3 gram. This new mercury oxide battery has been specifically designed for a
new type of hearing aid that fits inside
the ear duct. Its size, and stability of
power output will make it suitable as a
power source for " radio pills " and the
diagnostic probe techniques now being
developed. The capacity of the RM -212
Mallory
is rated at 0.5 mA 16 mAh.
Batteries Ltd., Gatwick Road, Crawley,
Sussex.
ww
320
for further details
Diagram Preparation
Until recently Chart -Pak pre -printed
tapes and symbols have been imported
from America but now a British Corn pany Chart -Pak Ltd., Weybridge,
Surrey, has been formed to manufacture and market the product in the
U.K. Basically, Chart -Pak is the use of
pre -printed self- adhesive tapes and symbols to prepare diagrams but one
interesting aspect is its use with respect
to printed circuits. If a prototype circuit
is required the tape- available in widths
from 1/64 in to 2 in and cut to a slitting
accuracy of ±0.002 in -can be placed
directly on the copper clad sheet and
immersed in the acid bath to protect
the copper during circuit formation.
WW 322 for further details
WIRELESS WORLD, JULY 1966
364
www.americanradiohistory.com
Printed Circuit Production
STANDEE TYPE RESISTOR
The Vitrohm UBT standee -type carbon
composition resistor has been developed
for printed circuits and conforms to
DEF- 5115 -1, and MIL -Rll -E specifications. The resistance range available is
from 22 t to 22 Mil, power rating 0.3 W
at 70 ° C derated to no load at 130 ° C.
Tolerance ± 10 ¡;, ± 5 % and a 250 V
maximum rated continuous voltage. Body
dimensions 8.5 x 5 X 3.5 mm. The sole
U.K. distributors of this Danish product
are Dubilier Condenser Co., Ducon
Works, Victoria Road, North Acton,
London.
WW
323 for further
details
Ceramic i.f. Filters
These filters by Brush -Clevite are now
available for use in military and commercial equipment. Designed for the
i.f. stages of superheterodyne radio receivers, they possess advantages over the
i.f. stages in common use. Fixed -tuned,
they require no alignment, and possess a
centre frequency stability of within
for 5 years, and within 0.2
+ 0.2
from -40 ° C to + 85 ° C (their operating temperature range). Five standard
models are available, TL1OD9 -20A,
TL16D9 -32A, TL20D9 -38A, TL30D957A, and TL40D9 -72A; the minimum
bandwidths at 27° C at 6dB are 10kc /s,
16 kc /s, 20 kc /s, 30 kc /s, 40 kc /s respectively, while the maximum bandwidths
at 27° C /60 dB are 20 kc /s, 32 kc /s,
38 kc /s, 57 kc /s and 72 kc /s respectively.
These five models are available with
centre frequencies of 455 kc /s and
500 kc/s. Input and output impedances
at 27 ° C are 2 kilohms for a 10 kc /s
bandwidth and 1 kilohm for a bandwidth
of 16 kc /s and above. Stop band rejection is 50 dB and the tolerance on the
centre frequency at 27° C is ± 3 kc /s, although filters can be supplied with a
tolerance of ± 2 kc /s at 27 ° C.
Brush -Clevite Co., Ltd., Hythe,
Southampton.
WW 324 for further details
On the stand run jointly by Kodak and
MPC of Milan was shown the first commercially available fully automatic production unit for printed circuits. Three
separate units are used: the MPC UNI 63
for cleaning the copper laminate; the
MPC Photomat E for printing; the MPC
UNI 100 for etching and rinsing the
final printed circuit. Other versions are
available but only the three types mentioned will be discussed. In addition,
the units have other uses besides those
briefly mentioned. For example, the
MPC UNI 63 can be used for cleaning
and deoxidizing copper -clad material
before photoprinting, but it can also be
used for deoxidizing of printed circuits
after the application of solder resist with
subsequent fluxing. Material for the
printed circuits is placed on a flat conveyor and subjected to a spray of cleaning or deoxidizing liquid which
is
automatically
maintained at the
desired temperature. The material
is then burnished
and cleaned.
After this stage,
the material, for
example, copper
laminate board,
is passed to the
printing unit. The
MPC Photomat
E has six radial
arms which rotate
through 360° and
Atter the completion of each stage,
the arms rise, 'rotate, stop and then
descend for the next process. The
first stage entails manual loading
and subsequent unloading (at the end
of the cycle) of the boards which are
prepunched to enable them to be clipped
on the arm. During the succeeding five
automatic stages, the boards are (2)
coated with a photosensitive resist (3)
dried (4) printed on one side (double
sided printing is possible with a larger
model) (5) first stage developed (6)
second stage developed.
Etching and rinsing of the boards can
be done by the MPC UNI 100. The
boards are placed on a horizontal conveyor, then pass into an etching section,
where an etchant is sprayed on to them.
Two rinsing sections are provided.
WW 325 for further details
.
go through a six s t ag e
process.
Colour Tubes
19 -in Mazda rectangular shadow mask tube has been introduced and pilot
production of this and the 25 in size has
started at the Thorn -AEI factory at
Brimsdown, North London. These 90°
deflection tubes have a neck diameter
of 36.5 mm and a
heater rating of
6.3 V. The three
electrostatic
focusing
anodes,
connected in parallel, run at about
5 kV and the final
anode operates at
A
red phosphor is used resulting in a
higher overall brilliance without desaturation of reds and giving the unlit
screen a less yellow appearance.
WW
326
for further
details
25 kV.
The grey tinted
glass screens a re
and
integral
metal suspension
lugs. Rare earth
aluminized
have
WIRELESS WORLD, JULY 1966
365
www.americanradiohistory.com
COMMUNICATIONS EQUIPMENT
Point -to -point Tx /Rx
Designed to meet the British Post Office
specification for point -to-point corn munications equipment the Cossor
CC -RTR4A transmitter- receiver operates in the 450-470 Mc /s band allocated
to fixed and mobile radio services. The
complete equipment, including the
power unit, is mounted in a 19 in
chassis, 9 in deep, and has a 31 in front
panel. Silicon planar transistors are
used
throughout, and there
are versions for
d.c.
operation
or 24 V)
(12
and a.c. mains
90 -120 V or 180240 V (50 or 60
A front
c /s).
panel meter can
be switched to
the
monitor
transmitter operates on one crystal controlled frequency between 450 and
470 Mc /s and has a frequency stability
of +5 parts in 106 from -10° to
+40 °C.
The receiver, which is also crystal
controlled, has a signal /noise ratio of
42 dB for ±5 kc /s deviation in the range
300-3,400 c/s for an r.f. input of 5µV
p.d. and 1 kc /s modulation.
Pocket Transmitterreceiver
Code -named MITRE (Miniature Individual Transmitter -Receiver Equipment) this pocket set covering the
68 -100 Mc /s or 145 -174 Mc /s bands
provides four f.m. channels with a
25 kc/s separation. Produced by Rank
Bush Murphy at Welwyn Garden City,
Herts, the set has been designed
specifically to slip into a uniform pocket.
It is a little over 1 in thick, is 5; in
high and 31 in wide including the
clip -on rechargeable sealed nickel cadmium battery. The miniature loudspeaker- microphone can be clipped on
to a lapel enabling the wearer readily to
hear incoming calls. The set is muted
until an in- coming signal is received.
transmitter
power output,
the receiver r.f.
input level or for testing selected points
in the equipment.
The transmitter employs a crystal
oscillator of approx. 4 Mc /s and the
second harmonic is selected by the
tuned circuit in the collector of the
oscillator transistor. Part of the tuning
capacitance of the tuned circuit is
formed by a variable capacitance diode.
the capacitance of which is varied
according to the amplitude of the modulation signal applied across it from the
modulation pre- amplifier. The 2 -W
This is the first point -to -point equipment to be produced by Cossor Communications Co., Ltd., of Elizabeth Way,
Harlow, Essex. It is designed to provide one or two speech channels with
or without supervisory control tones.
WW
ww
329 for further details
327 for further details
H.F. Communications Receiver
Continuously tunable over the frequency range 600 kc /s to 32 Mc /s in
eleven 3 Mc /s bands, the R7020
receiver, recently introduced by C. & N.
(Electrical) Ltd., of The Green, Gosport, Hants, employs a 6 ft scale
" folded " in 7 -in bands. Detector. outputs are provided for a.m., p.m.., s.s.b.,
f.s.k. and c.w. Sensitivity on a.m. with
30'
modulation at
1
20 dB S +N /N ratio.
kc /s is 6
%V
for
On c.w. it is
3 µV and on p.m. (deviated 1 kc /s at
1 kc /s)
it is 4 p.V.
The receiver
employs a 100 kc/s crystal controlled 1st
local oscillator and frequency stability is
quoted as within 3 kc/s from -10° to
+40°C.
The R7020 has a
built-in 2; in
monitor speaker and
an " S " meter. The
receiver has an integral voltage regulator
and will operate on
input voltages of between 12 and 50 V
d.c.
WW
328 for
details
further
366
H.F. Pack Set
Weighing only 18 lb, complete with
battery and carrying harness, the Racal
" Squadcal " pack
set transmitter receiver has an output of 5 W p.e.p. and
operates in the band 2 -7 Mc /s. It can
be operated on any of 29 crystal -conttolled channels which are pre -selected.
The channel crystals can readily be
changed in the field. Speech communication over a distance of up to 12 miles
is possible with a whip aerial.
WW 330 for further details
WIRELESS WORLD, JULY 1966
www.americanradiohistory.com
Communications Equipment
I
continued
ACOUSTIC TRAFFIC SENSOR
An echo sounding principle is used
M.F. /H.F. Receiver
A short -term frequency stability of
better than 5 c/s and over a long term
better than ± 30 c/s is claimed for the
Plessey PR155 communications receiver
which covers 60 kc /s to 30.1 Mc /s. It
is continuously tunable over this range
in thirty 1 Mc /s bands. The linear
scale provides 70 inches per megacycle giving a frequency setting resolution of better than 100 c /s. The incoming
signal is fed to one of eight
sub-octave bandpass filters
-
other modes. Typical sensitivity figures
quoted by the manufacturers are:
s.s.b., measured at 15 Mc /s with 3 kc /s
filter,
carrier -on /carrier -off
ratio
>20 dB for 1µV; a.m. (measured at
150 kc /s with 6 kc /s filter) 41tV for 10 dB
s +n /n ratio; and c.w. (as for s.s.b. but
with 300 c/s filter) carrier-on /carrier-off
ratio 26 dB for 1 f,V.
WW
331 for further fetails
which are automatically
selected by the " megacycles " switch. An on -off
calibration switch injects
100 kc /s crystal markers
for the adjustment of the
zero setting cursor on the
tuning scale. A reception
mode switch provides for
u.s.b., I.s.b., c.w. and a.m.
with two spare positions for
INFORMATION SYSTEMS
Magnetic Thin -film Store
store with the
short read /write cycle time of 0.33 µsec
and access time of 0.2 /esec is the latest
A digital information
in a series of magnetic thin -film stores
being developed by E.M.I. Electronics
Ltd., Hayes, Middx. It stores 128
words, each of 16 bits. The
storage elements of one plane
are formed by evaporation of
80/20
nickel -iron
alloy
through a mask on to a glass
substrate, and this is done
in the presence of a magnetic
field so that the elements
have a preferred direction of
magnetization, thereby providing two stable magnetic
states. A matrix of printed
wires, formed by two identical two -layer printed circuits,
wraps the magnetic elements,
providing address drive, digit
drive and sense conductors.
A number of such storage
planes are stacked and connected together by printed
circuits to form a complete
storage module.
WW
WIRELESS WORLD, JULY 1966
332 for further details
in an ultrasonic traffic sensor introduced
by The Marconi Company Ltd., of
Chelmsford, Essex, for counting or
measuring the speed of vehicles on
roads.
The advantage of the system
over conventional switch pads in the
road is claimed to be lower capital cost
and lower installation cost (since no
work on the road is required).
The unit is mounted above the road
on a bridge or standard, and transmits
2 msec
bursts of ultrasonic energy
downwards so that they are reflected
from the tops of vehicles passing beneath. The return times of the echoes
are measured. In the absence of vehicles
the echo return time corresponds to the
distance of the road surface, and this is
used as a reference. When a vehicle
enters the acoustic beam the echo
return time is shorter than the reference
value because of the shorter path length;
this indicates the presence of a vehicle
and is used to operate a counter or other
recording device. By means of an electronic gating circuit the device can discriminate between vehicles of different
heights (different path lengths), signifying cars, lorries, buses, etc., so that
traffic analyses may be made. Speed
may be measured by the use of two
spaced sensors and a common signal
processing unit.
The ultrasonic pulses are transmitted
and received at a p.r.f. of 20 -25 per
second by barium titanate transducers
working into beam forming horns, which
give fan -shaped beams. Beam width is
variable from 10'to 60 °.
ww
333 for further details
PROCESS CONTROL COMPUTER
Latest addition to the Elliott- Automation range of ARCH process control
digital computers is the ARCH 102.
This new silicon diode /transistor computer is suited to direct digital control
duties, but it has greater flexibility in
application than the earlier ARCH 101
fixed -programme machine and can work
in conjunction with other computers
and accept programme alterations.
The ARCH 102 is a binary, parallel
(synchronous) machine with a word
length of 13 bits, a speed of approximately 40,000 operations per second, an
add time (with access) of 23.41as, and a
multiply time of 76.5/ +s. A small store
version (4096 words) and a large store
version (8192 words) are available.
Inputs may be scanned at either 10 or
128 points per second.
Elliott -Automation Ltd., 34 Portland
Place, London, W.I.
WW 337 for further i.toll.
367
VERSATILE C.R.T. DISPLAY SYSTEM
The distinguishing feature of a c.r,t.
computer -output display system just
introduced by Ferranti is that it is not
restricted to a standard list of symbols.
New symbols may be created as required
by " drawing " them on the screen with
the aid of a joystick control, after which
they are stored in the associated computer. Types of display mode available
include letters and numerals (two sizes)
with random or tabular positioning,
graphical symbols as in circuit or mimic
diagrams, straight lines of any length
and angle starting at any required
point, circles of any radius and any
centre position, and graphs. The equipment does not use the
raster bright -up principle but forms all
characters
out
of
straight -line sections
and circles, from positional information supplied by the associated
computer. Lines are
specified by their resolved x and y corn ponents expressed in
numerical terms.
Circles, once their
radius and position
have been determined,
are generated on the
screen by the Lissajous
figure technique.
The equipment comprises a c.r.t. display
console, a control panel for generation
or modification of spatial information,
a teleprinter keyboard for writing
letters and numerals, and a logic unit
for format control, character extraction
and selection of display mode (this unit
includes the hardware for line and circle
drawing). Up to 12 displays can be
driven from the such logic unit. Digital
data referring to the display need not be
kept in a particular part of the associated
computer store but can be dispersed
according to programming requirements.
Ferranti Ltd., Hollinwood, Lancs.
ww
335
for further details
" Touchwire "
Manual
Input to Computers
A new manual method for giving in-
structions to computer -controlled or data
processing systems has been introduced
by The Plessey Company Ltd., of Ilford,
Essex. Originally developed for air
traffic control, the equipment comprises
an array of 32 short wire electrodes set in
a transparent screen mounted in front of
a c.r.t. display unit.
Against each of
these " touch wires " the c.r.t. displays
a code group of characters signifying an
available control decision.
The wires are electrically sampled in
turn by a scanning system controlled
by a binary clock. When the operator
touches one of the wires his hand capacitance unbalances an a.c. energized
circuit, and this is detected by a phase
detector. The output of the phase detector passes into a logic circuit which stops
the scanning system, erases the corresponding control code sequence from the
c.r.t. screen and sends a binary number
to the computer identifying which wire
has been touched. When the computer
has recorded this number it sends a
re -set pulse to the logic circuit which
Automatic Alarm Unit
Intended for use at unmanned stations,
such as remote oil storage tanks,
the Parkinson Cowan Ltd. " Dialarm "
automatically rings a pre- arranged telephone number or numbers when an
out -of- limits condition is detected in
the plant (e.g. high or low level in the
storage tank) and gives a 12- second recorded verbal message to this effect to
the person who answers the phone.
Four alarm conditions can be accommodated, the appropriate messages being recorded on four magnetic discs in
the unit. An alarm is initiated by the
closing of a pair of normally open
contacts. The Dialarm is allocated a
telephone number like an ordinary subscriber and can be interrogated by ringing that number; if at that time the plant
is operating correctly an " all conditions
normal " message is sent. The equipment uses solid-state circuits and can be
operated from the mains, from a primary
battery or from a secondary battery with
a trickle charger.
Parkinson Cowan Ltd., Stretford,
Manchester, Lanes.
ww
33o
for further details
re- starts the scanning system. This
causes the next control decision (in some
pre- arranged sequence) to be displayed
on the c.r.t. If the operator touches the
wires in a sequence which is illogical for
the system concerned, a warning indication is displayed on the c.r.t. The
number of control sequences which can
be performed is limited only by the
work capacity of the associated computer, and can be varied by modifying
the computer programme.
The advantages of the equipment over
keyboard input devices are claimed to
be that it is simpler to operate, has no
moving parts to wear and requires very
little operator training.
WW 337 for further details
368
WIRELESS WORLD, JULY 1966
www.americanradiohistory.com
Recording Level Indicator
AN AUDIBLE WARNING
DEVICE
By MURRAY WARD, B.sc.
THIS circuit was designed primarily to give aural indication to a blind person of the recording level on
a tape recorder. When the recording level is set too
high an audio tone is given out. Other possible applications are as an aid to the amateur ciné enthusiast during
the recording of sound on film, or a simple industrial
level -warning device.
Circuit
The circuit is shown in Fig.
1. An audio signal taken
from the monitor socket of the tape recorder is fed to
an emitter follower, Trl, which acts as a buffer stage,
reducing the loading at the monitor point and prevent mg any of the audio tone produced by the indicator feedmg back into the recorder.
Transistors Tr2 and Tr3 are cross coupled in a monostable multivibrator circuit with Tr3 normally bottomed
and Tr2 cut off.
Tr4 and Tr5 form an astable multivibrator, with the
emitter of Tr5 directly coupled to the audio output stage
Tr6. In the quiescent condition Tr5 is non -conducting
due to its base being kept near earth potential by a
connection via Dl from the collector of the bottomed Tr3.
Operation
When an audio signal is fed to the input stage Trl, a
proportion of the output voltage, determined by the
pre -set potentiometer RV1 is applied to the base of Tr3.
This voltage switches Tr3 off and triggers the monostable
multivibrator to its unstable state in which Tr2 is bottomed and Tr3 cut off.
The collector voltage of Tr3 is now negative and
reverse biases D1, thus allowing the astable multivibrator to operate and produce an audio tone from the
loudspeaker.
After a short time, C2 in the monostable multivibrator
EMITTER
FOLLOWER
The illustration shows the mounting of the components in a 2 oz
tobacco tin. The on -off positions of the switch are marked in Braille.
discharges and the circuit reverts to its stable state,
causing the astable multivibrator to become non- operative, thus cutting off the tone in the loudspeaker.
Further audio peaks from the tape recorder cause the
sequence to be repeated, the pre-set potentiometer RV1
being adjusted so that the indicator works when the meter
or magic eye on the tape recorder shows maximum
recording level.
Power supply
6.3 V a.c. from the tape recorder is fed to a bridge rectifier, D2 to D5, and smoothing circuit to give the necesR12
22
MONOSTABLE
ASTABLE
MULTIVIBRATOR
MU LTIVIBRAT OR
AUDIO
OUTPUT
-6
.} 63V
A.C.
R
220k
INPUT
0__i
FROM
TAPE
RECORDER
C1
0'00Iµ
R2
4.7k
RV'
5k
LINEAR
Di
N
Fig. I.
Circuit diagram of the recording level indicator.
WIRELESS WORLD, JULY 1966
369
www.americanradiohistory.com
sary power. A small battery would be equally suitable.
Quiescent current is about 2 mA, rising to 12 mA when
operating.
Construction
The components were mounted on a piece of Veroboard
and the author housed the unit in a 2 oz tobacco tin. An
illustration of the finished unit is shown on the previous
page.
The recovery time of the monostable multivibrator
(and hence the length of output tone) is determined by
the time constant C2R5. The frequency of the tone
depends on the time constants in the multivibrator cross coupling networks, i.e. C6R9 and C7R10.
Other component values are not critical. Practically
any small -signal p -n -p transistors may be used for Trl
to Tr5, e.g. 0071. Tr6 is a low power p -n-p output
transistor; an 0081 was used in the original unit. By
connecting a 7552 loudspeaker directly in the emitter lead
of Tr6, no output transformer is needed.
Acknowledgement
I am indebted to Mr. B. E. A. Vigers for his guidance
during the construction of this device.
LETTERS TO THE EDITOR
The Editor does not necessarily endorse opinions expressed by his correspondents
The Future of Television
in the
U.K.
be obvious by now to all interested parties that
this country's 625 -line u.h.f. television service is a flop.
There is evidence that many viewers with properly
installed receivers get a poorer picture on 625 lines
(u.h.f.) than they do on 405 lines from v.h.f. stations
of BBC-1 and I.T.A.; and TAM ratings for BBC -2 vary
from unmeasurable to an audience of about 10% of
the possible.
Yet it is this same 625 -line u.h.f. service which is
expected to revive the flagging receiver industry in some
magical way by the addition of colour. We are asking
to repeat the long slow drag which marked the first
years of colour in the U.S.A.
If we look back a few years, all the high hopes of
the Pilkington Committee and others about the technical
future of British television can be seen now to be so much
hot air. For the facts are that
(1) the viewer considers the improvement in picture
quality with 625 -line u.h.f. to be marginal, and
(2) that four years after the Pilkington report, no
plan has yet emerged to change -over the existing
405 line services to 625 lines.
I think it will be accepted by all but the colour blind
that the addition of colour to a monochrome television
service is a far more worthwhile improvement than the
addition of 50 %, more scanning lines. Therefore it
makes commercial, political and engineering sense to
add colour to our national (98% coverage) 405 -line services rather than to the almost moribund 625 -line BBC -2
service. If this is done, a case can be made for putting
BBC-2 back to 405 lines, which is easier than might be
thought, or it can be left as a monochrome 625 -line
service, with no detriment to any one. Several
B.R.E.M.A. members admit privately that they would
far rather have colour on 405 lines than on both standards, and they are becoming painfully aware of the
small market for a receiver that has to show 405 -line
monochrome and 625 -line monochrome and colour.
Recently, a B.R.E.M.A. and T.A.C. spokesman gave
it as his opinion that it would be disastrous for the
country and the industry if colour were transmitted on
405 lines for BBC -1 and I.T.A. This has been the opinion
of the T.A.C. for some years, but one cannot help thinking that if asked to reconsider and in the light of subse-
IT will
quent developments reverse their decision, they would
earn the gratitude of the industry and the public for
the magnanimity of such a gesture.
The cost to the country of introducing colour on the
existing 405 -line service would be small, as it involves
only some relatively minor engineering work on the
transmitters and inter -city links, whereas the cost of
introducing one 625 -line national coverage u.h.f. service
is now estimated to be about £130M for the 2,000, or
more, transmitters which are now known to be necessary.
It only remains to look briefly at the reasons given
for the change to 625 lines. The first is that of parity
with Europe; but the T.A.C. fixed different sound -vision
carrier spacings. The second reason was that by going
to 625 lines, we could exchange programmes with the
Continent but the B.B.C. has nullified that point by
introducing the line store standards converter. The
only major remaining point is that of making the export
of receivers easier. However, since the Continental
market has different standards of performance for d.c.
maintenance, sound quality and scanning linearity, it
is understood export receivers are developed in different
laboratories to receivers for the home market. Last year
about 3-4 "., of the U.K. receiver production was
exported, which is hardly reason to change our entire
system.
I look forward to at least two competing colour programmes next year, and I am sure that your readers
do too.
Richmond, Surrey.
MICHAEL COX
Interference ?
I THINK it advisable to draw attention to a shortcoming
of the PAL colour television system which does not
appear to have been considered carefully, if at all.
It is a fundamental characteristic of the PAL system
that while the phase of the B-Y colour vector is held constant that of the R -Y vector is switched 180° on alternate lines of the same field. To a first approximation
this is mathematically analogous to having two sub carriers whose frequencies are separated by exactly one
half the line scanning frequency.
It has been necessary in the past to allocate to cochannel transmitters carrier frequencies whose difference
is a multiple of one third of the line frequency. By this
PAL Co- channel
370
WIRELESS WORLD, JULY 1966
www.americanradiohistory.com
means the visibility of co- channel interference is reduced
some 14 dB.
The presence of two colour subcarriers separated by
one half line and the need to allocate co- channel luminance carriers separated by one third line are incompatible
and it can be shown that the visibility of interference
will be increased by at least 8 dB compared to monochrome or N.T.S.C.
If PAL is adopted in this country, as the P.M.G. has
intimated, are we going to be troubled by co- channel
interference? There can be little doubt of this. If this
is so are we right in adopting a largely untried system for
the sake of partial European unity?
It is being argued that it is too late to reconsider
systems. Is it not remarkable that after years of indecision it now becomes imperative to settle promptly for
what may prove to be an inferior system?
Morden, Surrey.
D. SMART
Electronic Organs
IN the present series of articles on electronic organs Mr.
Towers suggests in the May issue that random mistunings can add to the true organ effect. I would like to
suggest that permanent maladjustment of the " bearing "
(or more scientifically, the relationship between the
notes of the scale and their reference) is most unmusical.
The point that Mr. Towers should make, surely, is that
the " chorus " effect experienced when listening to a
pipe organ consists of random changes in frequency due
to minute variations in wind pressure and low " Q "
factor of the organ pipe, the free phase effect and tone
of individual pipes, and the different characteristic starting transients of each pipe sounded in each register or
stop.
Clearly, if we are going to design an electronic instrument to meet the above requirements we are letting ourselves get involved with considerable financial and technical complications.
In a divider organ there are only 12 basic generators
and an attempt to mistune say 3 or 4 of these would
mean that all the divided frequencies from these generators would be out of tune and this would be painfully
obvious even for small errors.
To summarize, if we are unfortunate enough to have
separate generators for each note and indeed for every
note on all stops, minute variations in frequency could
be tolerated and would enhance the sound of the instrument especially if random changes in frequency could
be introduced (not to be confused with vibrato, where
considerable frequency change is produced in a cyclic
manner). The divider organ is a reasonably cheap and
simple instrument, and very useful because of this. Consequently, the divider organ must be treated as a single
scale organ and no attempt should be made (by accident
or design) to alter the tempered scale to produce pseudo
chorus effects.
Axminster, Devon.
B. W. DANIELS
-
The author replies:
With my " purist's " hat on, I agree completely with
Mr. Daniels. His exposition of the scientific aspect of
the problem reflects well the views widely expressed in
the literature on the subject. There can be no doubt
that the " lifeless, dull, dead, hooty, tubby quantity "
(Peabody, 1936) of the electronic organ is due to the
use of too few and too non -disonant partials, its lack of
pitch fringe and its lack of temporal (or agogic) fringe.
Accepting this, when I put my " practitioner's " hat
on, I find that part of the charm of pitch fringe and
agogic fringe (which both stem from the aesthetic principle of uncertainty) can be achieved simply by deliberate random mistuning as I indicated. My own proof
of this in the practice tests carried out with a crystal
controlled divider organ I built which held equal temper
to 0.01% with a corresponding coldness of sound. This
my critic friends found less sympathetic than a conven-
tional LC-oscillator -controlled divider organ with imperfect tuning.
Cambridge.
T. D. TOWERS
Matrix Algebra
ON reading Mr. Hina's letter in the April issue of Wireless
World we too were " shaken to fragments." So much so
we have had to read and re -read the letter to discover the
cleverly concealed criticism amid the myriads of irrelevant data such as the state of Stevenage and Hatfield
libraries, the poor perspiring, aspiring H.N.C. candidates,
their exam trials, etc., ad nauseam.
Like Mr. Hina we have discovered the power of the
matrix in circuit analysis, particularly phase shift networks and we have constantly referred to last year's excellent articles by Mr. Olsen.
The crux of the criticism is that he requires a derivation
of the matrices for the shunt impedance and the ideal
transformer. We indicate how by similar methods we reapplied Mr. Olsen's argument.
With reference to the articles in the March and April
issues of last year and using the same nomenclature, the
defining equations of the A- matrix are
v1= a11v2 -a12i2
it =a 2 -a22í2
v
Therefore
=
a
v
which is for the output open- circuited.
v2 }i
v
a, 2= {_'
}
i2 v2 =0
which is for the output short- circuited.
{il
which is for the output open- circuited.
a21= v2 }i2 -0
a22 ={
72L,-
which is for the output short circuited.
By applying these equations to the four terminal networks
shown and remembering Ohm's Law and the voltage
ratio for the ideal transformer we obtain these matrices:il
z
ll
12
"----Avvw-41-
12
1V2
1v2
vlI
VI
n:I
SERIES IMPEDANCE
IDEAL TRANSFORMER
all
al2
azl
a22
I
[o
Fan
atz
La21
a22
n
o
D
I
SHUNT
[all
a12
a21
a22
IMPEDANCE
Ol
Y
Difficulty may be experienced on Mr. Hina's part in
explaining the a22 element in the ideal transformer case
but it will be remembered from elementary transformer
theory that the current ratios are inversely proportional
to the turns ratio with the output short- circuited.
In deducing the above we make no claim to be latter-
WIRELESS WORLD, JULY 1966
371
www.americanradiohistory.com
day Sylvanus P. Thompsons nor for providing a method
of passing H.N.C. " standing on one hand." A full
exposition in two short articles is expecting too much and
we hold little faith in the comprehendability of a full
treatise in three eloquent pages.
Finally may we point out a trivial mistake in Table I.
In converting from y- parameters to z- parameters the z21
has been ',tinted as z12.
Framwellgate Moor,
NEVILLE HOGARTH,
Durham.
CHARLES LEWIS
Automatic Car Parking Light
READERS of the May 1966 issue who try to use an
as suggested by Mr. S. K. Chawla as
a phototransistor may be disappointed by the results.
Some while ago, Mullard stopped putting a translucent
jelly in the 0071, and substituted an opaque type. The
solution is to use an OCP71, although a cadmium sulphide cell is likely to work out cheaper. The ORP12
would suit. Should the OCP71 be chosen, better results
will be obtained by connecting the base to the emitter
with a 4.7 k!2 resistor. This improves the ratio of light
to dark impedance of the transistor.
JOHN WEBSTER
Stroud, Glos.
0071 minus paint
Amplifier Noise Level
I READ with interest Mr. Driscoll's letter (June issue)
regarding signal-to -noise ratio (or to be more academic
signal+noise -to-noise ratio) in transistor amplifiers.
First, with regard to maximum settings of volume
and tone controls this would in itself result in clipping
of transient peaks even if the average modulation were
low. In addition the effect of the tone controls would
be nullified by clipping and considerable distortion
result by overload of the earlier stages of the preamplifier due to poor positioning of these controls in
several popular transistor amplifiers.
Secondly, regarding specifying a minimum acceptable
noise power delivered to the loudspeaker, surely this is
in itself meaningless considering the widely varying
efficiencies of loudspeakers currently available. In any
event the manufacturer has already stated the noise
power delivered by his amplifier, since this is always
(or should be) measured with volume controls at maximum and expressed as signal -to -noise ratio in decibels
with reference to maximum power output. The very
definition of the decibel (10 log P, /P,) means that it is
a measure of the ratio of two powers. Hence, a 30 -watt
amplifier with a 70 dB signal -to -noise ratio delivers a
noise power of 30 x 10- watts or 3 microwatts to the
loudspeaker system.
C. ARTUS
SGS-Fairchild Ltd.,
Applications Laboratory,
Ruislip, Middx.
Collision Avoidance Systems
HAVE read with interest Flight Officer Perry's letter
in your May issue, and would agree with him that an
active bearing measuring scheme would be most satisfactory, provided that sufficient range and accuracy
could be achieved at a sufficiently low weight penalty.
Unfortunately, however, this is not the case at the present state of the art. This is not the place to go into
detailed calculations, but a radar such as he envisages
I
would be likely to weigh at least 50 lb, and to require
an exposed radome, which would increase drag, in order
to obtain an all-round look. My system, on the other
hand (April issue), uses equipment that will have to be
carried and maintained by law in the near future, and
the additional weight penalty attached need not be more
than, say, 10 lb. Although, therefore, my system is less
attractive operationally than that of Flt. Officer Perry,
I feel that it presents at present a more useful solution
to the problem of avoiding collisions in areas without
ground surveillance. Later, when more is known of the
behaviour of active laser systems, a scheme such as Fit.
Officer Perry's may well have considerable merit.
W. D. GILMOUR
EMI Electronics Ltd.,
Wells, Somerset.
Pick -up Arms
I HAVE followed with interest the correspondence subsequent to Mr. Bickerstaffe's articles on Unipivot pick -up
arms, and I should like to add that I have never experienced any trouble with either rumble or torsional modes
of vibration. My problem, to the exclusion of nearly
all others, has been the h.f. resonance of the arm. The
present model, my third, is excellent in all respects bar
this one, and although I have managed to damp the
resonance considerably by filling the tube with a soft
silicone potting- compound and rubber -mounting the
counterweight the resonance still shows up on extreme
orchestral peaks as a tendency for the sound to " break
up." This admittedly only occurs on one or two records,
but is, as any hi -fi man knows to his cost (! ), disturbing. I was surprised therefore, to see almost no mention
of how to avoid this trouble in the aforementioned
article, and I would suggest that anyone building Mr.
Bickerstaffe's arm should try filling the tube with various
substances to see which is the most effective -two suggestions are Rawlplugs, or tight-fitting p.v.c. sleeving,
but I would be intrigued to hear of any other methods
used successfully.
ANTHONY H. KING
Langford, Beds.
Tailpiece !
THE more observant of your readers will have no doubt
noticed that I.T.A. has been carrying out colour television tests for the past two years or so during the natural
breaks occupied by advertising.
Tests on P.A.L. have been going on for some time and
the enriching experience gained has now led to tests on
Colour Harmonic Unit Modulation (C.H.U.M.) which
is similar to P.A.L. in as much as a long tailed pair is
used to produce a vector resultant which wags either
side of the B -Y Leg. This system is German and is
attributed to Hr. Dr. Ing. D. A. Chshundt whose original
idea of the quadruped colour vector diagram gave the
lead (after much vetting) to the British System -Black
Unit Line by Line (B.U.L.L.).
Dogged by the inability to produce a colour bar, the
B.U.L.L. system which suffered with degraded whites,
has been superseded by the British Entirely Alternating
Gamma Line Experiment (B.E.A.G.L.E.). This experimental system has had some small success in the aviation world and has bred some interesting colour displays
for flight simulators.
Apart from this, British Television, both monochrome
and colour, has been lying doggo for sufficiently long to
be considered thoroughly up the (Barking) Creek.
J. J. BELASCO
Backwell, Glos.
WIRELESS WORLD, JULY 1966
372
www.americanradiohistory.com
Computer room at the new premises of
Racal Research Ltd. at Tewkesbury, Glos.
Corn puter-
"designed"
Circuitry
WHEN Sir Walter Cawood, chief scientist in the Ministry
of Aviation, officially opened the new premises at
Tewkesbury, Glos., for Racal Research Ltd., the highlight was the use of a computer in the design of electronic
equipment. Racal established this company a few months
ago as a central research organization for the group under
the direction of Eric Wolfendale. Design problems from the
subsidiary companies throughout the world will be sent to
this new establishment for investigation.
The computer, an Elliott 4120, is being used for trial
designs and visitors were invited to specify requirements
(bandwidth, voltage gain, gain stability, source resistance and
ambient temperature) for a wideband r.f. amplifier The
basic circuit, shown below, had been prepared and the computer programme stored on punched tape. The parameters
prescribed by Wireless World were fed into the computer,
followed by the first part of the programme, from which
the computer then calculated the values of the bias resistors,
the decoupling capacitors and the load and emitter resistors
to give the gain and the gain stability. As the total component list has to be printed out as input data for the next
part of the programme, the remaining capacitors, also, have
to be given values, hence the series capacitors are given large
salues of 1,000MF arid the shunt capacitors very small values
of 0.1 pF. The values of these capacitors do then not affect
the calculation of the mid -band gain.
The second part of the programme calculates the values
of the series and shunt capacitors which determine the bandwidth of the amplifier. In each case, they are adjusted to
be within 10",, of the actual value required to give the bandwidth in the specification. The complete component list is
then printed and is reproduced below. The nearest preferred
values are used except for the components which determine
the gain, gain stability and frequeriîy response. These are
printed out to greater accuracy to enable 1
components
to be used, where the bandwidth is requited to be specified
accurately and /or the gain stability required is very small,
Tho third part of the programme then calculates the gain
frequency response of the designed amplifier and shows where
the amplifier differs from the bandwidth called for.
Initially the computer is using papel tape input, but with
the aid of a random access backing store a much more
sophisticated programme could be prepared.
The specification prescribed by the guest of honour on
his arrival was programmed, the amplifier made and graphs
prepared showing the measured performance against the
calculated frequency response before he left the establishment three hours later with the equipment and the data.
AMPLIFIER DESIGN.
SUPPLY VOLTAMO =120.
ANDIENT TEMPERATURE FROM
40
BANDNIDTM FROM
VOLTAGE GAIN.
50URCC
1000.2
z
27.0
3
0.2
h
303.0
5
4.1
G
7
8
9
10
11
12
13
14
15
16
17
750
600.0
020.0
10.0
5.0
938.0
57.7
390.0
12.0
OIMS
nMls
KlLrnuaS
Y.I
17 kc /s to 17.351 Mc /s at which frequency the gain was
- 13.55dB.
C
1
L^I i1:5
E111:45
o111.0
eons
1:I
LEI c:5
Ln11,15
011115
011115
CIII;S
1:1
CROF.:+:.05
CROFARADS
IIICROFARADS
PICOFARAOS
P1C0FACAOS
111
I:1CR0FARA05
2062.5
0.330
MICROFARA05
P100040005
PICOFARA05
MICROFANADS
0.022
3.900
.1ICROFARACS
IfICOCFARADS
C
TRA05257OG5
TYPE BFY
WIRELESS WORLD, JULY 1966
www.americanradiohistory.com
;i:FREQUENC
FEP.CEIIT
ßE1P0115E.0;TIIEAMP11FIER
' ::CKILOCYCLIS
011115
0.560
4.1
1
IZ6ACYCLES
00
:'iFREQUENC,
014.5
16.0
C
CEr.CEES CENT1:1'.,,C
L
Cm ''S
I.ILOMI%
2660.5
0.560
circuit and (right) the print -out of the list of components.
Inset is part of the frequency response calculated over the range
70
KI LOIC<5
0.047
0.390
2.200
The basic
30
cnMS
3.9
250.0
115.0
150.0
TO
STCEILITY.
Co:,:
CEGISTAUCE.
1
0
".IIOCYCLES TO
70
13
16
19
21
24
27
20
30
FALL OF GAIN FROM
BAND IN DECIBELS
12.74
10.56
8.55
7.15
6.05
5.17
4.45
1.87
P:IO
The Root -locus Technique
2- Plotting
the Root -locus
cerned with the right -hand half, for this is the region of
instability. It is usually permissible to omit most of this
part therefore. The root -locus plot is always symmetrical
about the real (horizontal) axis. Unless there are complex
poles and /or zeros, it is permissible to suppress the lower
half.
For the present we shall use the full diagram, but in
practice where it is practical to employ little more than
a quadrant it is as well to do so, since it enables a bigger
scale plot to be made on a piece of paper of given size.
Having prepared the paper the following steps are
taken in turn as far as they are applicable to the given
case or as far as it is necessary to take them to obtain the
required information.
Step 1. Plot the values of the poles and zeros on the
diagram. Indicate a pole by a cross on the point, and a zero
by a circle around it. If there are two or more equal poles
or zeros, the convention is to indicate this by a vertical row
of crosses or circles, one for each pole or zero, placed symmetrically about the actual point.
Fig. 1 shows the diagram appropriate to the example
with the three poles plotted on it.
Step 2. Draw that part of the root -locus which exists on
the real axis. The locus exists on the real axis at every
point which has to its right an odd number of poles and
zeros, counting them together. Any complex poles or zeros
Continuing this series of articles, the
main steps in plotting the root -locus are
now described in detail and illustrated
by an example
THE necessary form for the equation and the
meanings of the terms root, pole and zero have
At this stage, therefore, we know
been explained.
the values of the poles and zeros with which we are
concerned. We are using as an example an equation
which represents the performance of a three -stage RC
amplifier having time constants of 1, 2.5 and 1Oµsec.
This equation is
0
=
GoH0K
1
By W. TUSTING
-f-(p+ 1)(p ±0.4)(p +0.1)
K = 1/25.
In the general case
with
a sheet of graph paper is required
with horizontal and vertical scales crossing in the middle
at the zero of both. Positive is upwards and to the right,
negative downwards and to the left. It is essential that
bcth scales be the same. In practice, we are rarely con-
are ignored.
In the example of Fig.
1,
there are no zeros and there
16)
+ -I.0
-0.8
06
- 0.4
-02
,
X
1.4
12
1.0
X
04
06
04
0.2
P
X
0
0.2
0-4
0 6
0.8
I.0
Fig. I . For a root -locus plot the real and
imaginary axes must be scaled the
same. The first step is to plot the values
of the poles and zeros. Here there are
three real poles indicated by crosses.
02
0.4
0.6
-08
--1.0
WIRELESS WORLD, JULY 1966
374
www.americanradiohistory.com
are poles at -0.1, -0.4 and -1. To the right of -0.1
there are neither poles nor zeros and so the locus does not
exist on the real axis. At all points between -0.1 and
-0.4, there is an odd number of poles and zeros (viz
one pole) to the right. The locus exists between 0.1
and -0.4 and is indicated by drawing a line between
these two poles, as in Fig. 2.
Between -0.4 and -1, all points have an even number
of poles and zeros on the right (viz, two poles). Consequently, the locus does not exist between these poles.
All points to the left of -1 have an odd number of poles
and zeros (viz, three poles) to their right; hence, the
locus exists at all points on the real axis to the left of - 1.
This is indicated by drawing a line coincident with the
real axis to the left from -1 to the limit of the figure;
in reality it goes to infinity. This is again shown in Fig. 2.
-
Step 3. Compute the value of
-
EP
EZ
Pa= P
where EP and
are respectively the sums of the
-Z
P
Z
values of the poles and zeros and
and
are respectively
the numbers of poles and zeros. Plot the value of pa on the
diagram.
In the example, there are poles at -0.1, -0.4 and -1
and no zeros. Hence, LP = -0.1 -0.4 -1 = --1.5,
3Z = 0,
= 3,
= 0. Consequently, pa = -1.5/3
_ -0.5, and this is plotted in Fig. 2.
Complex poles and zeros, if any, are included. As
they always occur in conjugate pairs their imaginary
parts always cancel out and so pa is always real.
P
Z
Step 4. Compute angles from the relation
9 =
180
P-Z
(1
+
2n)
for n = 0,
1, 2, 3, etc. Draw lines at these angles from the
real axis from the point pa found in Step 3.
In the example
= 3, so e = 60 (1 + 2n). For
P-Z
=
=
= 1, 9 = 180 °; and for n = 2,
Higher values of n only give angles
repetitive of these. An angle of 180° is given by a line on
the real axis to the left of pa. The angles of +60° are
given by lines drawn from pa at +60° to the real axis
on the right of pa.
These lines are asymptotes to which the root -locus
tends when Golfo tends to infinity. They are shown dotted
in Fig. 2.
n
e =
0, 9
300°
60 °; for n
_ -60 °.
Step 5. Determine the breakaway point(s) on the real
axis. A breakaway point pb is one at which for a certain
value of Golfo two real roots become equal, so that for a
further change of G0110 they become complex. It is necessary
to find a point by guessing a likely value and then checking it,
repeating until it is found with the required accuracy.
A breakaway point on the real axis is always at some
point which lies on that part of the locus which exists on the
real axis and which has been drawn in Step 1. It usually
lies between two poles or two zeros. An even number of
breakaway points can exist between a pole and a zero
(which may be at infinity).
Having guessed a likely value for a breakaway pb
(a) measure the distances between this point and all the
poles and zeros on the real axis, and assign negative
signs to the pole distances and positive to the zero
distances
(b) take the reciprocals of all these distances
(c) add all the reciprocals which refer to poles and zeros
to the left of pb
(d) add all the reciprocals which refer to poles and zeros
to the right of pb
(e) subtract the right sum from the left sum. If the result
is zero the guessed point is indeed the right point. If
it is not zero the guessed point is wrong, so try again
(f) if there are complex poles and /or zeros, further quantities
must be added to (c) and /or (d)
(g) measure the distances AC and BC, Fig. 3, and evaluate
the quantity 2 BC /(AC)2. Assign a negative sign if it
0.8
0-6
0.4
Fg. 2. Here the diagram is shown with
the part of the locus which lies on the
real axis drawn in. The origin of the
asymptotes pa and the asymptotes themselves are also shown. These are lines
to which the locus tends as G0H0 tends
to infinity. The breakaway point pb is
also plotted.
0.2
+
X
1!4
12
Ó
08
Ott. 0.4
Pa
-X
/i40'.2
'': Pb
o
0.2
0-4
0.6
0.8
WIRELESS WORLD, JULY 1966
www.americanradiohistory.com
0Z
04
1;6
08
I!0
would be halfway between them, at -0.25. As the third
pole is relatively far, away and to the left, the breakaway
point will be to the right of -0.25. A first guess for
A;(
pb is
-0.24.
The pole distances are 1 -0.24=0.76, 0.4 -0.24 =0.16
and 0.24 -0.1 =0.14. Assigning minus signs because they
-
-(
-
B
This diagram shows a pair of complex poles with a
trial breakaway point at C. As explained in the text the
contribution of these poles (or zeros) is determined from a
measurement of AC and &C.
Fig. 3.
for a pair of complex poles and a positive if it is
for a pair of complex zeros. If the poles or zeros lie
is
to the left of a vertical through Pb (= C), add the
quantity to (c); if they lie to the right add it to (d).
It is much less difficult than it would appear to determine pb, but nevertheless doing so can be rather tedious.
Fortunately, in many practical cases it is not necessary
determine it with great accuracy.
.n the example, the locus exists on the real axis between
poles at -0.1 and -0.4 so the breakaway point must lie
between them. If it were not for the third pole at -1, it
to
1.317
are pole distances and taking reciprocals gives us
and -6.25 on the left of pb and -7.15 on the right. The
sums on the left and right are -7.567 and -7.15. The
-7.15)
difference left sum minus right sum is -7.567
= 0.417. Thus the breakaway point is not at -0.24.
Now try -0.23. The distances are 0.77, 0.17 and
0.13, giving for the reciprocals -1.3 -5.88=7.18 on the
left and -7.68 on the right. The difference is -7.18+
7.68=0.5. Thus the breakaway point is not at -0.23.
However, the difference found is only slightly greater
in this second trial than in the first but is of opposite
sign. Therefore, the true point is approximately halfway
between our two trials, or at -0.235. Probably -0.236
would be nearer and we should try this if we were going
to make a third check. However, when we look at the
scale of our graph we can only plot clearly in steps of
about 0.005, and we take pb= -0.235 as being quite
accurate enough.
Having determined pb, we plot it in Fig. 2.
Step 6. Sketch in freehand the complex part of the
root-locus, guided by the knowledge that it passes vertically
through pb and tends towards the asymptotes.
In the example, a freehand sketch of the locus will probably pass through the imaginary axis at co = 0.75 -0.8.
This may be near enough for a first rough estimation of
the critical gain. However, it may be necessary to find
the crossing point accurately.
Step 7. To determine whether or not any point is on the
root-locus. Draw straight lines from the point to all the
the poles and zeros. Measure all the angles between these
In the complete root -locus diagram the curved part of the locus is
drawn freehand through pb, guided by
the asymptotes. The only point which
must be known at all accurately is the
This
crossing point of the jm axis.
can be checked by drawing lines
(dashed) to all the poles (and zeros)
and measuring their angles from the
horizontal. If the point is correct they
sum to 180 °. The lengths of these
lines enable the critical gain to be
computed.
Fig. 4.
WIRELESS WDRI.D, JULY 1966
www.americanradiohistory.com
and a horizontal to the left through the point and assign
negative signs to angles to zeros and positive to angles to
pules. Add all the tingles with due regard to sign. If the
result is an odd multiple of 180 °, the point is on the root locus.
A word of warning is needed on. the application of
Step 7. It sometimes happens that one tries to find a
point on a locus in a region of the diagram where in fact
the locus does not exist at all. Instead of the sum of the
angles being widely different from 180 °, it may be quite
near it, 190° perhaps. A few further trials may give one,
say 184 °, and there is a temptation to say that this is near
enough. This temptation must be firmly resisted unless
it is certain that the locus actually does pass in that
region. This certainty demands that another nearby
point be found for which the angles sum to less than 180°.
If all the trial points lead to sums of the angles which are
above or below 180 °, even if they are quite near to it,
there is a strong suspicion that the locus in fact may be
nowhere near these points.
In Fig. 4 is shot that complete root -locus passing
through the imagin y axis at (0=0.74. The dash lines
are the ones joining this point to the poles and also the
horizontal to the left through this point. We measure
the angles with a protractor and find that they are -36 °,
-62° and -83 °. The sum is -181 °. In this case we
may consider this as being near enough. There is no
doubt that the locus passes in this region, for it must
cross the imaginary axis between zero and the asymptote.
Step 8. To find the value of GoHOK corresponding to a
given point on the locus, measure the distance of the point
from every pole and zero. Form the product of all the
distances to poles and a second product of all the distances
to zeros. The value of GoH0K is the first product divided
by the second product. As K is known, the open -loop gain
GoH0 can be obtained at once.
In Fig. 4, the relevant distances are the lengths of
the lines drawn from the point at which the locus crosses
the zero axis to the poles. They are 1.245, 0.84 and 0.745.
Their product is 0.78. Now K=1/25, so G0110=0.78 x 25
=19.5.
With w =0.74, f= 0.74/6.28 = 0.1175 Mc /s.
Thus, the amplifier will start to oscillate at 117.5 kc /s if
the open -loop gain exceeds 19.5 times.
'
BOOKS RECEIVED
Principles of Aerial Design; by H. Page. In this book the
author has written at a level which is a compromise between
the two normal approaches, i.e. either a complex mathematical treament or a purely descriptive treatment. Where
approximations have been used to avoid mathematical complication the reasons have been given and the significance of
the approximations indicated. The contents cover the laws
governing electric and magnetic phenomena, Maxwell's
electromagnetic wave equations and their application, theoretical design principles and practical forms of aerials. A sixpage appendix details vector rotation and formulae. Pp. 172;
Illustrations 102. Price 50s. Iliffe Books Ltd., Dorset House,
Stamford Street, London, S.E.1.
Guide to Radio Technique Vol 1; Fundamentals, Valves,
Semi -conductors, by E. Julander. This is a book intended
to provide a basis for the enthusiast or student who seriously
wishes to improve his theoretical knowledge of radio rather
than make a continued practical approach. The opening
chapters deal with the fundamentals of electricity, complex
numbers and transmission units. A complete chapter deals
with the construction, theory and operation of thermionic
valves; this is then followed by a similar chapter on semiconductors in which a comparison is made with valve circuits.
The final chapter deals with generation, propagation and
modulation of electro- magnetic waves. A 14 -page appendix
contains useful data presented in the form of tables and
monograms. The book is issued as part of the Philips Technical Library and is a translation of the original Swedish
ed;tion. Pp. 238; Figs. 214. Price 37s 6d. Distributed by
Macmillan & Co. Ltd., Little Essex St., London, W.C.2.
Servicing Electronic Organs, by C. R. Pittman & E. J. Oliver.
The book is written as a practical guide to the theory and
operation of electronic organs for the serviceman requiring
basic knowledge of this application of electronic techniques.
Details are given of the basic circuits encountered and
include illustrations and information reproduced by courtesy
of ten American organ manufacturers. Originally published
in America in 1962. Pp. 191; Figs. 112. Price 30s. W.
Foulsham & Co. Ltd., Yeovil Road, Slough, Bucks.
Handbook of Electronic Circuits, edited by R. Feinberg.
Intended for the experimenter and laboratory worker, the
book contains brief descriptions and circuit diagrams of a
wide range of practical circuits which can be used in the
overall design of equipment. About 75 circuits are given
some of which use valves. Applications include power supplies, waveform generators, amplifiers, frequency converters,
counters, waveform modifiers, clamps and electromechanical
control. Pp. 195; Price 50s. Chapman & Hall Ltd., 11 New
Fetter Lane, London, E.C.4.
Building and Using Sound Mixers, by R. E. Steele. Essentially of a practical nature, the book contains complete
constructional details, parts lists and layout drawings for a
number of the author's designs for sound mixers. However,
the practical aspect is well supported by explanatory text
on mixing technicalities and mixer design. Pp. 152; Figs. 124.
Price 30s. Focal Press, Ltd., 31 Fitzroy Square, London,
W.I.
Materials Used in Semiconductor Devices, edited by C. A.
Hogarth. The book is divided into eight sections, the first
being a brief introduction describing the applications of materials. The remaining sections deal with germanium, silicon,
selenium, lead sulphide -selenide and telluride, indium, antimonide, bismuth telluride and antimonides of cadmium and
zinc. Extensive cross references are included and a 23 -page
author and subject index is given at the end of the book.
Pp. 243; Figs. 101. Price 5gn. John Wiley & Sons Ltd.,
Glen House, Stag Place, London, S.W.1.
Transistor Receivers and Amplifiers, by F. G. Rayer. Treatment used in the book is essentially of a practical nature and
is intended to provide guidance for the home constructor
interested in the use of transistors. Basic contents include
general descriptions of operating characteristics of semiconductors, aerials, and r.f. amplifiers, superhet circuitry and
power supplies. Additional chapters describe printed
circuits, test equipment and fault finding. Pp. 164; Figs.
129. Price 30s. Focal Press Ltd., 31 Fitzroy Square, London,
W.I.
WIRELESS WORLD, JULY 1966
377
E
www.americanradiohistory.com
NEWS FROM INDUSTRY
The Flight Simulator Division of
Redifon Ltd., has received an order
worth £1.5 M from the Ministry of
Aviation for three Phantom digital flight
simulators. Two simulators are for the
Royal Air Force and one is for the Royal
Navy. Each simulator has a digital
computer for the operational flight
trainer and a second computer for the
tactical simulation. Each simulator is
provided with a Redifon three axes
motion system and a colour visual
system, allowing the pilot to make his
visual approaches and landings at his
base. Each simulator also includes a
64 sq ft terrain model, permitting the
pilot to gain experience and training in
the ground attack mode. The tactical
section fully simulates the weapons
system, and the aircraft operational
radars using the Redifon land mass
multi- colour system. Western Airlines
of Los Angeles, U.S.A., have ordered
three simulators to the value of $ 1.5 M.
Believed to be the first time that a
complete town has been wired to relay
radio and television, the new town of
East Kilbride is to have all 13,500
houses connected to a wired relay
system. British Relay will lay 300 miles
of wire, erect a receiving aerial, and
establish a central control station from
which all television and radio programmes will be fed into the network.
East Kilbride council said the system
would ensure satisfactory reception of
BBC -2 and eventually colour transmissions, and also subscription television if
it is developed in Scotland.
£300,000 contract has been
awarded by the Costa Rican government to G.E.C. (Telecommumcations)
Ltd., who are to supply a microwave
communications system. This will link
all the principal towns of Costa Rica
together by a ten station network, from
the northern town of Liberia to the
Panamanian border in the south. It will
be installed, commissioned and initially
maintained by G.E.C. engineers, who
will subsequently hand over the system
to Costa Rican engineers, trained
locally, and at Coventry. The transistor
microwave radio equipment is constructed on the card assembly shelf
principle. Design of the circuitry permits the dropping and inserting of
circuits at intermediate stations without
demodulation of the through traffic.
A
-It
is regretted
M.R. Supplies Ltd.
that the address of this company was
incorrect in their advertisement last
month. The correct address is 68 New
Oxford St., London, W.C.1.
Following Decca's acquisition of Set point Ltd., Decca Radar Ltd. will introduce a range of products in the field of
marine automation equipment. Setpoint
Ltd. will develop for Decca Radar a
range of equipment designed to meet
the requirements for monitoring and
control of main engines and auxiliaries
in large motor ships and refrigerated
cargo vessels.
A.E.I., G.E.C. and Plessey have
formed a satellite communications consortium, to build and sell complete
communications satellite ground terminals. It is estimated that the market
for civil terminals is worth £100 M, and
joint engineering teams are working on
designs to meet the requirements of this
international market.
Claimed to be the largest micro -electronics factory outside the U.S.A., the
new Marconi micro- electronics plant
will occupy a 61- acre site at Witham,
Essex. Granted final planning approval, this £1 M project should be fully
operational in 1967, and will house the
Division's research, development, production and commercial departments.
Domestic Receiver Supplies. -Manufacturers' disposals of television sets to
the home trade during the first quarter
of this year, r^ording to B.R.E.M.A's
figures, were 409,000 compared with
470,000 for 1965. Radio receivers delivered were 382,000 against 444,000 in
1965 and 489,000 in 1964.
The Switch and Control Company, of
Vale Road, Watford, manufacturers of
heavy current multi -bank and multi position switches for industrial control,
has joined the Livingston Group.
Manufacturing facilities have been
transferred to the Livingston factory on
the Greycaines Estate, North Watford.
The aerial structure for the satellite earth
station on Ascension Island in the South Atlantic to be owned and operated by Cable and
Wireless Ltd. has been set up for tests on the
Marconi radar test site at Rivenhall, near
Witham, Essex.
The
aerial structure
is
undergoing nearly two months of overall
testing before being shipped to Ascension,
and COMSAT has agreed ¡to loop tests being
conducted with the Early Bird satellite.
Design studies have been completed
by English- Electric- Leo -Marconi for a
multiple access machine computer system (4-75) which can be used by up to
200 people simultaneously. It can transmit data for processing via a terminal
point in the user's office. The paging
technique used, permits the addressing
unit to select the pages of the programmes most frequently employed by
subscribers, ensuring that they are always immediately available. The least
used pages of each programme will be
held in the computer's backing store, but
will be obtainable in one sixtieth of a
second. The system has been designed
for universities, research centres, Gov-
ernment departments and commerce.
The contract for a complete 625 -line
television station, to be installed in the
Songkla Province of Thailand, has been
awarded by the Thai government to
Pye TVT Ltd., of Cambridge. It is
the second Pye station to be ordered
for Thailand and will include two
5 kW transmitters operating in parallel
with a high gain aerial system which
will give an e.r.p. in excess of 100 kW.
A reciprocal agreement for the interchange of microwave technology has
been concluded between Microwave &
Electronic Systems Ltd. (M.E.S.L.) of
Newbridge, Midlothian, Scotland, and
Sanders Associates, of Nashua, New
Hampshire, U.S.A. Under the agreement, M.E.S.L. will manufacture exclusively in the U.K. the entire range of
microwave products of Sanders Associates for a period of seven years.
DISA. -Dansk Industri Syndikat A/S
of Denmark, have now opened a United
Kingdom branch at 116, College Road,
Harrow, Middlesex (Tel: 01 -427 9263)
where stocks of equipment, accessories,
and spare parts will be held. This
equipment provides for the measurement of flow, turbulence, vibration,
strain, torsional vibration and pressure.
DISA also handle Lyrec tape recorders
for correlation measurements and freTechnical
transformations.
quency
bulletin DISA Information (No. 3) is
now available. The manager of the U.K.
branch is Mr. A. L. Cussens.
WIRELESS WORLD, JULY 1966
378
www.americanradiohistory.com
The annual report of BSR Ltd.
of
during 1965 78.6
their production was exported compared
records that
THE HOUSE OF BULGIN
with 65.3 % the year before. Exports
to the U.S.A. totalled 55 % in 1965 and
38% the previous year. The company's profit after taxation was
£1.378 M in 1965 and £ 1.404 M the
year before.
Racal Electronics Ltd. -An increase
of some 20 % in the company's profit
before taxation is reported for 1965/6
compared with the previous year. The
figures are £730,745 and £610,641.
The turnover of the A.E.G: Telefunken Group for 1965, amounted to
The increase over the
DM 4,135 M.
previous year is about 9 °i,.
Exports
rose to DM 884 M, an increase of 2
compared with 1964.
Belling -Lee Ltd., have sold 74
of
their share capital to the industrial
bolding company, Ada (Halifax) Ltd.,
which is a subsidiary of Philips Electronics and Associated Industries.
Directors and management of the company remain unchanged.
SGS- Fairchild Ltd. is moving its
administrative headquarters and applications laboratory from Ruislip, Middx,
ro a new office block to be named
Planar House at Aylesbury, Bucks. The
present premises at Ruislip will be
devoted entirely to the production of
monolithic semiconductor devices.
A.B. Metal Products Ltd. of 119/127
Marylebone Rd., London, N.W.1, are
handling the range of quartz crystal
products produced by CTS Corporatiop
of Elkhart, Indiana. Mr. R. P. Caton
is in charge of this section.
Fantovox receivers, manufactured in
Japan by Novel Dempa Co., are being
marketed in the U.K., by E.R.
(Factors) Ltd. of 374 -378 Harrow
Road, London, W.9. (Tel.: CUNningham 0361.)
AT YOUR SERVICE
For details of our NEW PRODUCTS
you missed at the IEA
send for our TECHNICAL LEAFLET
List No.
List No. K.512
K.521
List No. K.523
List No. M.C.L.22 List No. M.C.L.14 List No. M.C.L.64
List No. B.
I
to
B.3
List No. B.6 or
List No. S.800/L
List No. 5.800
List No. 5.800/2L
Nine new Battery Holders are
offered, either panel, chassis -lug
or base mounting and accepting
from one to three cells of various
sizes. Send for Battery Holder
Data Sheets.
switch which can be fitted with
various subsidiary operators. Also
available in a high specification
range. Send for leaflet 1501 /C.
Another new Kit! The Bulgin
Security Alarm System. Pre-
the Pioneer Electronic Corporation of
Tokyo are now Swisstone Ltd., of 26
Leigh Place, Cobham, Surrey (Tel.:
Cobham 2853). Previously, Pioneer were
represented by C. E. Hammond.
Among the recipients of The Queen's
Award to Industry was Multicore Solders Ltd. whose name was misquoted in
the list in last month's issue (p. 324).
A new and unique range of six
Pilot Lampholders accepting the
S.6/8 cartridge cap lamps, and
fitted with various fixing brackets.
Send for S.6/8 Lampholder
Data Sheet.
Sub -miniature, mains rated Micro
Switch range. Based on a S.P.C.O.
Pioneer. -The U.K. distributors for
Lansing. -Ad. Auriema Ltd., of Impectron House, 125 Gunnersbury Lane,
London W.3 (Tel.: ACOrn 8765) are
the agents for James B. Lansing Sound
Inc. of Los Angeles, some of whose
equipment was referred to in the article
" Audio 66" in last month's issue.
B.7
Three knobs from fifteen new
models. Left, medium version of
six, ultra modern styling. Centre,
tall thin knob with decor cap;
Right, miniature collet fixing
model. Send for leaflet 1500/C.
built, versatile, easy to
install,
adaptable
and
very
economic; a unique development.
Basic Kit contains: two latching
switches, one pressure pad, key
switch control box, battery magazine, 4in. Underdome bell, and
all wire, screws, etc. Send for
cision
List No. KIT.5
A. F. BULGIN & CO. LTD.,
Bye Pass Rd., Barking, Essex.
Tel: RlPpleway 5588 (12 lines)
leaflet 1523/C.
MANUFACTURERS AND SUPPLIERS OF RADIO
AND ELECTRONIC COMPONENTS TO
ADMIRALTY
MINISTRY OF WORKS
WAR OFFICE
IMINISTRY OF AVIATION
AIR MINISTRY
MINISTRY OF SUPPLY
HOME OFFICE
RESEARCH ESTABLISHMENTS
CROWN AGENTS
U.K.A.E.A.
B.B.G.
G.P.O.
N.P.L
0.5.1.
R.
WW-101 FOR FURTHER DETAILS..
WIRELESS WORLD, JULY 1966
379
www.americanradiohistory.com
Red
4,4
lr
fwel4.6,,,,
Vtett2A
Report on Terran Electronics
FROM an unusually unreliable source come these extracts
from a repart issued by a delegation of Martian scientists
who visited Earth in order to study our electronics
industry.
The year which has passed since the first Martian exploration of the planet Terra has seen a considerable change in the
attitude of the human inhabitants. It will be remembered
that our original landing produced an intense fear -reaction
which increased to manic proportions when it was found that
Martians were impervious to attack by any Terran weapons.
This mass hysteria slowly subsided when it became clear
that our intentions were peaceful, giving way in due course to
a somewhat grudging co- operation. It is, however, evident
from our telepathy traffic that underlying the veneer of cordiality is a feeling of intense chagrin that, in the event, Ter rans were the visited not the visitors. This is understandable
seeing that, following their moon -landing success, plans were
well advanced for the construction of a crude form of space
vehicle with which it was hoped to navigate to Mars. What is
less understandable is that even now that the Terrans know
that we can read their innermost thoughts they still continue
to use their acoustic powers in an attempt to conceal the
truth.
It may be asked why our delegation limited its studies to
electronics. The answer is that we had no option because
Terran technology has progressed no further than the harnessing of the electron, and even in this the degree of knowledge is shallow and the techniques rudimentary. It soon
became clear to us that even such long out -moded Martian
concepts as fifth -dimensional zeta radiation were not only
completely unknown but were likely to remain forever beyond
the Terran power of comprehension.
Terrans, as the first exploration party has pointed out, are
essentially animus -type computers with animal attachments
and as such are somewhat reminiscent in structure to the
lower forms of life on our own planet. Their computer element is primarily a survival mechanism and is therefore ill adapted to scientific research except on a primitive scale. A
built -in defence mechanism operates to discount the existence
of physical laws which cannot be detected by their senses.
senses.
To say that our mission was to study the Terran electronics industry is in itself a statement demanding explanation, in the absence of any Martian counterpart. It appears
that some two centuries ago, some external influence-as yet
unidentified-caused a degenerate mass mutation in the animus computers and impelled them to herd together to expend
their energies in buildings called manufactories. A collection
of these manufactories of similar character formed an
industry.
The electronics industry is a lineal descendant of this socalled Industrial Revolution; its general mode of operation is
similar to most others, and will not be given in detail here.
One small facet of the system may be mentioned however, as
illustrative of the serious nature of the degenerate mutation.
At a given time of the day the operatives of a manufactory
leave their homes and converge in a mass upon the industrial
building. So primitive are the forms of transport and so
congested the routes that the journey may take an hour to
complete (a conservative average figure). As, later in the day
a reverse mass migration takes place, two hours per day per
Terran are taken up. Thus, in
a
manufactory which employs
5,000 operatives -and some employ far more -there occurs
a wastage of at least 50,000 human -hours in every week. Yet
this continues unquestioned because the Terran animus
computer (or brain as it is termed) has been conditioned to
regard it as inevitable.
The control structure of a Terran electronics manufactory
is similar to that of most industries. Metaphorically it is in
the form of a pyramid with a master -computer or brain at the
apex and a succession of levels of lesser computers beneath
him in a series -parallel arrangement Each computer at the
level nearest the apex is in direct Contact with the master
computer (or chairman as he is called) and each computer is
in control of several on the level below. Each of these lower
level brains is in turn controlling a number on a still lower
level, and so on down to the base of the pyramid.
As is to be expected, the broader the base lengths of the
pyramid in relation to the perpendicular, the less efficient the
organization becomes. In spite of this the trend is toward the
amalgamation of manufactories into even more unwieldy
structures. The explanation of this can only lie once again
in the degenerate mutation and the damage it has caused to
the Terran computer.
Each independent manufacturer is in fierce competition
with many others. This means, among other things, that every
organization which markets a similar type of device, carries
out its private research and development on it. Every device
thus tends to be researched and developed perhaps a hundred
or more times over, with a total expenditure of about 600
years. It might be thought that this terrifying wastage of
effort is an argument in favour of amalgamation, because of
the reduction in competition. Our investigations show that
this is not the case -again, because of inadequacy of control.
It is not uncommon to find, in the largest categories of manufactory, that similar devices are being developed by separate
teams in adjacent laboratories, each being ignorant of the
others' activities.
A detailed account of the present state of the art in electronics is contained in the Appendices to this report. It can
be summarized by saying that it is rudimentary; as an example, the most advanced technology yet known consists of the
laborious processing of components and metal connections
on to, or within, glass or semiconductor platforms. A corn monly used term for an electronic equipment is hardware,
which in itself is indicative of its primitive nature.
It is no part of the Scientific Delegation's function to
decide what action should be taken to assist the Terrans in
their plight. We could, of course, in our specialist sphere,
advance their knowledge of electronics or even, to a limited
degree, enlighten them as to other, and far better, means of
communication. But this is not the real issue. The crux of
the matter is that Terran industry and commerce (of which
electronics is a small part) is so crazily constituted that it
periodically promotes global warfare. Our recommendation
to the Martian Tribunal, therefore, is that we be authorized
to correct the damage which the Terran animus -computer
sustained two centuries ago, as it is all too evident that their
own scientists, having the defect implanted in their own computer elements, can never hope to effect the repair.
380
WIRELESS WORLD, JULY 1966
www.americanradiohistory.com
Wireless World
JULY, 1966
iii
IIMIT
More TrN
YN
SOLDERING EQUIPMENT
TOTAL PERFORMANCE
ILLUSTRATED
ILLUSTRATED
L64 IN PROTECTIVE SHIELD L700
R300 BENCH STRIPPER
For
full information
and sales apply direct to
HEAD OFFICE SALES & SERVICE
ADCOLA PRODUCTS LTD
ADCOLA-HOUSE, GRUDEN ROAD,
LONDON, S.W.4.
Telephones: Macaulay 0291;3
Telegrams: SOLJO!NT, LONDON, S.W_4
AUSTRALIAN ASSOCIATES: ADCOLA PRODUCTS PTY LTD., 673 WHITEHORSE ROAD, MONT ALBERT, MELBOURNE
AGENTS IN ALL LEADING COUNTRIES
www.americanradiohistory.com
Wireless 'òr
JULY, 1966
Why do leading
electronic manufacturers use
aErsin
Solde
I
Because they realise that their reputation
can rest upon the quality of the solder
they use. For utmost reliability they use
Ersin Multicore, the only solder containing
the purest tin and lead, plus 5 cores of
extra -active, non -corrosive Ersin flux.
Whatever the application
the speedy
soldering of miniature components or the
ndividual production of large units -there
is an Ersin Multicore Solder which is
exactly right for the job.
-
Standard Telephones & Cables Ltd. The
manufacture of Automatic Telephone Exchange
Equipment involves many million soldered
joints. The accessibility of many of the wires is
restricted after they have been terminated, and it
is therefore essential that only solder of proved
reliability be used. Consequently Ersin Multicore
Solder, 60/40 alloy, has for many years been
preferred by Telephone Manufacturers in Britain
and overseas. It is approved by the British G.P.O.
and many foreign Telephone Authorities, and is
shown here being used by Standard Telephones
and Cables Limited in the manufacture of a
G.P.O. Line Finder Rack.
Ferguson
Radio
Corporation
Ltd.
The
critical characteristics of U.H.F. tuners in
television sets demand the highest quality of
soldering. Ersin Multicore Savbit alloy, containing
live cores of extra fast Ersin 366 flux, is seen here
being used by Ferguson Radio Corporation Ltd.
H. J. Leak & Co. Ltd. Ersin Multicore Savbit
alloy is shown in use in the wiring construction
of the Leak Integrated Stereo 30 Transistor Amplifier. In order to ensure the utmost reliábility, the
Leak High Fidelity equipment has been made with
Ersin Multicore Solder for more than 20 years.
reliability
ngineers and technicians are inuited to write on their Company's letter
heading for the booklet ' MODERN SOLDERS" containing data on
melting points, alloys, etc.
MULTICORE SOLDERS LIMITED, MAYLANDS AVENUE, HEMEL HEMPSTEAD,
3636. TELEX: 62363
WW-002 FOR FURTHER DETAILS.
www.americanradiohistory.com
M.26
Was this manual useful for you? yes no
Thank you for your participation!

* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project

Download PDF

advertisement