date electronics for lab and leisure

dod ©am

60

april 1980

UK.55p.

U.S.A Can. $1.75

in mpg

improve

the ignition system

read about

cars in the Eighties

* recommended

contents

elektor april 1980 — UK 03

page 4-10

There are many transistor/
thyristor ignition systems
being produced commer-
cially but they all seem
to suffer from one dis-
advantage or another. In
this article we describe a
system which combines
the most significant
advantages of all the
others, including the con-
ventional type. This
recipe, mixed with a few
of our own ideas, results
in the Elektor transistor
ignition system — a spark
for all seasons?

page 4-18

An extremely annoying
aspect of poor wind-
screen wipers is that they
are fine — until you have
to use them. We unfortu-
nately are unable to do
anything about that but
we can help them cope
with varying weather
conditions. How? By
designing an intelligent
wiper delay circuit!

about this issue

selektor

electronics in the car of the eighties

battery protection

Forgetting to turn off the headlights need no longer be a
motorist's nightmare. The circuit described here is designed
to monitor the battery voltage and switch off the lights
automatically in all kinds of motor vehicles.

transistor ignition

a dip-stick probe

There are all kinds of indicators and warning lights in modern
cars, but an oil temperature indicator is rarely included. The
easiest way to add this type of indicator to an existing car is
to mount a temperature sensor in the dip-stick.

intelligent wiper delay

active car aerial

Aerial amplifiers are often used for domestic receivers — in
some cases, even, where they do more harm than good. In
cars they are not nearly as common, surprisingly, since that
is one place where they can often be extremely useful! This
'car issue’ seems the ideal place to present our ideas on the
subject.

Stop thief! (B.H.J. Bennink)

There are all sorts of systems for protecting cars, but the one
described here is unusual: it is deception, rather than protec-
tion.

4-01

4-01

4-03

4-08

4-10

4-16

4-18

4-24

4-28

page 4-34

Do you waste fuel by
poor driving? For most
drivers (like us) it's all
guesswork. The only
way to know what's
really happening as you
drive is to use an instant
fuel consumption meter.
A unit, in other words,
that gives instant
indication of your
present fuel consumption
in miles per gallon.

ejektor

thyristor burst ignition system

battery voltage indicator

As described here, only a few components are needed to
obtain an 'optical' indication of the battery condition: a
single lamp that changes colour from green to red as the
battery goes into the danger area.

free protection!

aide-de-camp

It is probably true to say that every touring caravan uses a
1 2 volt power supply for something, even if only for lighting.
With this in mind, many caravan holiday sites are fitting
mains power point facilities. This article describes a con-
verter which allows the caravan's existing 12 volt system to
be supplied from the mains connection point.

4-29

4-30

4-31

4-32

As this month's cover
illustrates, this issue is
devoted to cars. More
specifically, electronics in
cars. We have a feeling
that at least some of the
ideas presented here will
find their way into the
cars of the future. But
why wait? They can all
be built and installed
now!

fuel consumption meter

old hat?

market

missing link

advertiser's index

4-34
4-43
4-47
UK 16
UK 26

selektor

elektor april 1980 — 4-01

Automative electrics in the
'Eighties'

As we enter the 1980's the motorists of
Europe — often led by over-optimistic
motoring journalists — are waiting im-
patiently for the great breakthrough in
battery technology which will come in
the next few years, will sweep away the
world's dependence on oil with all its
political implications, and will herald
the age of the electric passenger car.
Unfortunately, as experts from many
areas point out, the prospect is not
quite so exciting as that.

Constant development

The first fact to remember is that as in
most technologies, 'breakthroughs'
seldom come in dramatic form and all at
once. Although a hundred years old, the
conventional lead-acid battery which we
all use in our cars has been the subject
of progressive breakthroughs during
most of its life. These developments
have accelerated during the past two
decades, and although the battery you
buy today may look similar to those on
the market twenty years ago, in practice
it is different in almost every detail. Its
container, its seperators, the design and
configuration of its vents, the design of
its plates and the composition of the
alloys used to make them, have all
dramatically changed — mainly in the
interests of improved life and reliability,
and above all to obtain the greater
power for less weight and size, on which
modern car designers insist.

Lead acid has by no means yet reached
the end of the road, and there is plenty
of room for creative technical develop-
ment, enabling the designer to emphasise
whatever performance characteristics
are required by the motoring public at
the time — currently, low maintenance,
which will be prominent in the Motor
Shows of 1980 and 1981 and which will
be dealt with in more detail later in this
article.

The electric car

But such developments to this conven-
tional arrangement of internal-combus-
tion engine plus lead-acid battery for
starting, lighting and other services will
not satisfy the eager motorist who
wants to see an all-electric car, to reduce
the world's dependence on politically-
unstable and increasingly expensive oil
supplies.

There is a great deal of historical logic in
this. Electric cars competed very success-
fully with petrol-engined designs in the
early days of motoring — they even held

the world's land speed record, and
'electric broughams' were the carriages
of Royalty and the aristocracy.

Problems of power supply

Professor Bird went back to the familiar
petrol hose on the garage forecourt, as a
base from which to begin his study, and
some frightening figures emerged.

Les us examine what happens in terms
of energy and power when a petrol tank
is filled. If it is assumed that it takes one
minute to fill a ten-gallon tank and that
one gallon of petrol contains 45 kWh of
energy, when the total energy transfer
which occurs during that one minute at
the petrol pumps is 450 kWh, and the

rate of transfer of this energy — i.e. the
power flowing through the filler hose —
is 27 MW. Thus filling a petrol tank on a
garage forecourt can be equated to
taking the entire output of a 27 MW
power station for one minute and it can
be said that a motorway forecourt with
20 petrol pumps represents the energy
equivalent of a 550 MW power station.
This is clearly a significant calculation
for electric transport and contains a
host of engineering implications. It is
valuable to look at this energy equival-
ence from a different viewpoint. If a
vehicle is plugged into the domestic
13 A mains for 12h overnight, a maxi-
mum energy transfer of 37 kWh occurs,
which can be equated to the energy
content of less than one gallon of petrol.

The implications are obvious. To obtain
an electric vehicle which has similar
characteristics to the i.c. -engined vehicles
is extraordinarily difficult, and the use
of electricity for transport purposes
could be expected to require a radically
different engineering approach in order
to achieve a satisfactory performance.

The industrialist's view

The industrial view, expressed by
Mr. Burnet, is equally forthright, though
he naturally saw the problem from the
practical point of view of the Motor
Industry and its customers.

The effect of the energy crisis is likely
radically change the design of cars.

which in turn will have an important
influence on battery design. However,
with an electric car or truck costing
approximately twice the price of its
internal combustion equivalent and with
the internal combustion engine having
such advantages of convenience, flexi-
bility and economy for ordinary
motoring, the public will continue to
use it in most circumstances. Many
battery manufacturers are deeply
involved in development work on electric
vehicles with the belief that the perform-
ance of electric vehicle batteries for
commercial trucks can be increased by
50%. This, plus the necessity to reduce
urban pollution and the need to save
energy, may well cause electric vehicles
to capture 10% of the 1 - 5 ton road
transport market by the end of this
century.

Unfortunately, traffic conditions, per-
formance and other requirements have
all changed since and electric power for
private motoring is no longer viable —
nor does it seem likely in the foresee-
able future.

A number of prominent designs have
appeared in the past ten years, some of

4-02 — elektor april 1980

selektor

which are still in the early stages —
ranging from an experimental car by
Fiat, with front wheel drive and room
for two passengers and 20 kg of luggage,
to a much more ambitious town car
now being developed in America, with
thermostatically controlled heaters to
facilitate fast charging of a specially
designed lead-acid battery. But they are
all restricted to a cruising speed of
about 50 mph and a range of about
50 miles, which although acceptable in a
town delivery vehicle is of little use to
the ordinary business or family motorist.

Two points of view

The present situation of the electric
vehicle has recently been analysed in
depth by two leading authorities,
one, a scientist — Prof. B.M. Bird,
M.Sc. Chairman of the Power Division
of Britain's Institution of Electrical
Engineers; the other Mr. David Burnet,
Chairman of Chloride European
Automotive Battery Operation. It is not
always that Science and Industry agree,
but on this subject they both homed in
on to exactly the same viewpoint.

The academic view was that in spite of a
substantial research effort, it is unlikely
that the lead acid storage battery will be
substantially replaced by advanced
technology (such as sodium sulphur)
much before the end of the 20th
century. The industrial one was that
most people in the motor industry, as
well as motorists, are likely to see the
internal combustion engine's dominance

continue throughout their working and
motoring lives.

The reasons for this unanimity between
two sides of the fence, which do not
always see eye to eye with each other,
was clearly spelt out by both.
Remembering that electric delivery
vehicles have already proved to have
great practical advantages in safety,
silence and flexibility and are already in
extensive use, points the way to signifi-
cant developments.

Granted that new battery technologies
are not going to transform road transport
overnight, then what is the energy crisis
likely to do to the Motor Industry?
Petrol at £ 2 a gallon in a Mini is no
more expensive than £ 1 a gallon in a
Rover 3.5. It has taken us seven years to
get from 30 pence a gallon petrol to a
pound a gallon yet it seems reasonable
to believe that it is not going to take us
too long before we get to £ 2 a gallon in
the 1980s. In the next decade we are
going to see a very rapid trend towards
smaller and much lighter cars with fewer
accessories in order to improve fuel
economy so that buyers have to make as
few compromises as possible on vehicle
size.

Chloride Automotive Batteries Limited.

(538 S)

Spectacular boost in
semiconductor sales

According to a 323 page report published
in two volumes by Frost & Sullivan Inc.
there will be a terrific increase in sales
of semiconductor memories on the
European market. The turnover during
1978-1982 will increase by 230%
(from 210 million dollars in 1979 to
494 million in 1982) and production
is expected to rise from 2.88 billion
bits in 1978 to 17.7 billion in 1982.
These figures correspond to Elektor's
predictions on 1C memories in this
year's January issue. The enormous
increase in sales over the next five years
will be remarkable and the general
trend will be for bits to drop in price.
Now it will be possible for 1C memories
to be adapted on an even larger scale.
As we mentioned once before, this will
cause the rate of development, in audio
technology especially, to accelerate.

Table

Expected developments on the West
European market in semiconductor
memory sales per sector (turnover in
millions of dollars) :

1979 1980 1985

1990

Computers

238

298

650

1136

Industry

13

18

54

113

Private

7

10

30

109

Enterprise

Others

4

6

24

62

CWS'Rl&l

TMWiMbi

W|x»t

3SS?

CHLORIDE

“ttw E

MBS:

Until 1985 production in dynamic
and static RAM's will soar and at the
same time, the introduction of double
pole memories will be more wide
spread. CCD's and bubble memories,
however, will only gradually take on
more importance. The F & S report
once again confirms the fact that
American companies completely
dominate the semiconductor memory
market. Intel Corporation takes the lead
by owning 25% of the shares. In Europe,
Texas Instruments comes second with
18%, while Mostek, having 13.3%, has
rapidly gained third place. As far as the
Japanese are concerned, only Nippon
Electric Corporation (NEC) have
managed to reach the top ten.

For further information, contact:

Frost & Sullivan Limited,

104-112 Marylebone Lane,

London W1M 5FU.

(542 S)

electronics in the car of the eighties

elektor april 1 980 — 4-03

Electronics in the engine

Saving fuel is undoubtedly the most
significant factor in today's motoring
and is almost certain to remain so in the
foreseeable future. A flexible and
'intelligent' control system will enable
the engine to remain fully efficient
while catering for the many and various
motoring conditions which occur. One
example of this is an electronic ignition
control system of the type that is
currently gaining more widespread use
(mainly in the performance car market).
After the initial developments such as
the transistor ignition and fuel injection
systems under mechanical or analogue-
electronic control, an important step
forward was recently taken by Bosch
designers with their introduction of

dedronics In die ear

of die eighties

The recent progress in electronic technology has been extremely rapid
but, however, the motor industry has been rather slow on the uptake.
At a time when awareness of the energy crisis is ever increasing,
automobile designers are, at last, turning to microelectronics in an
effort to increase the efficiency of motoring in many aspects — not the
least of which is engine control.

1

engine control unit

Figure 1. Block diagram of the engine control unit. The microcomputer assembles data from
the various sensors to calculate the exact requirements of the fuel injection and ignition systems.

digital motor electronics or 'motronics'.
It consists of an integrated digital con-
trol system governing both the ignition
and fuel injection. At the heart of this
system lies — what else? — a microcom-
puter.

Figure 1 gives the block diagram of an
engine control unit which will be familiar
to the electronics designer. A row
of sensors, pulse generators, pulse
shapers or analogue-to-digital converters,
enables the microcomputer to gather
the necessary information about the
state of the car at any instant in time.
Information on the position of the
crankshaft, manifold depression, and
the carburettor butterfly is fed to the
microcomputer.

According to the assembled data,
sampled at a rate of up to 400 times a
second, the requirements of the ignition
and fuel injection systems are calcu-
lated. Data concerning the correct
moment of ignition and fuel metering
are read by the computer from its
memory. This has the following advan-
tage: the computer's built-in program
determines the type of control required.
Thus, the circuit can be adapted for
different levels of engine power and
different types of fuel (methanol, for
instance).

Figure 2 shows the ignition character-
istics stored in memory at the time of
ignition as compared with a 'normal'
range of characteristics found in a tran-
sistor ignition system (figure 2b). How
the various adaptations affect the
engine is noticeable in the amount of
fuel consumed.

Figure 3 gives an overall picture of the

4-04 — elektor april 1980

electronics in the car of the eighties

entire system. The highly compact con-
trol unit shown in photo 1 has been in
production since the summer of 1979.
The components and mechanics in-
volved have to comply with tough en-
vironmental requirements, before they
can be considered to be reliable. They
must be able to withstand an impact of
100 G and have a temperature range
from — 40°C to +1 40°C.

Microcomputer motronics could also be
introduced into a 'drive control' system.
This would involve the introduction of
a central electronic unit to monitor the
entire driving system and would eventu-
ally lead to further improvements in
fuel consumption and driving standards.
What will be important to further
development depends not so much on
electronics, which has already made
sufficient progress, as on sensors land
control units which will have to drop
in price and rise in efficiency.

Electronics for driving safety

Electronic aids to brake systems were
an important development during the
last decade. Work started on an Anti-
lock Brake System (ABS) during the
sixties as part of a co-operation between
Daimler-Benz and AEG-Telefunken. The
system, which originally consisted of
analogue electronics, was converted to
digital processing at Bosch.

Photo 1. The heart of the Bosch Motronic
engine control system.

Time

80095 2b

Figure 2. Comparison of the ignition characteristics stored in the Bosch engine unit (2a) and
those of a conventional transistor system (2b).

Figure 3. Overall view of the entire engine control unit system.

electronics in the car of the eighties

elektor april 1980 — 4-05

4

wheel

speed

$<]-

i

digital

signal

processing

4 ,

current

regulator

4 ^

power pressure

stage solenoid

^ 4 ,

brake

pedal

Figure 4. Block diagram of an anti-lock braking system.

5

Photo 2. The transmitter and receiver aerials of this radar controlled proximity detector can be
clearly seen at the top of this picture.

Figure 4 illustrates the principle behind
the operation and gives the layout of the
controls. Sensors attached to every hub
detect the number of wheel revolutions.
When the vehicle is braked violently and
the wheels risk being locked, strobing
electronics recognise this and operate a
hydraulic brake pressure modulator,
thereby reducing the brake pressure to a
safe level. This works independently for
each wheel, so that the best possible
brake effect may be obtained. This
system gives excellent results as test
reports have shown that braking time is
improved and the danger of skidding is
significantly reduced. The security of
the system itself has also been taken
care of. When the engine is started, a
test circuit checks the system and all its
components are constantly monitored
while driving. If an error is detected, the
ABS switches over to the normal
braking system and an indication is
given to the driver.

Mobile radar

Another important safety aspect
being developed is the remote control
warning system using radar for distance
calculation (see photos 2 and 3). In
several countries wide scale tests are
being carried out on a number of
vehicles at the very high frequency of
35 GHz. Pulses of 300 mW are radiated
and a reflection from a vehicle in front
is then detected by the instrument with
the aid of a separate receiving aerial.
The interval between the transmitted
and reflected pulses will determine the
distance between the car and the ob-
stacle ahead. This information enables
the correct driving speed to be calcu-
lated and this will be indicated on a
screen. It is a proximity detector
incorporating a radar system which ful-
fils this task with the aid of a micro-
computer.

From a technical point of view, there
are still a few problems to be solved
with regard to frequencies at which the
system will not work — especially in
narrow bends for instance. There are
also psychological problems. Either the
driver may become too familiar with the
alarm (and not let it 'sink in'!) or he
may find it so disturbing that he will
switch it off.

Tyre pressure warning

This early warning system (see figure 5)
informs the driver if a loss of pressure
in any tyre is detected. While the wheel
is turning, the tyre pressure data is
conveyed to the chassis via transmit
and receive coils. The system operates
at a resonant frequency when the tyre
pressure is normal. The receiver func-

4-06 — elektor april 1980

electronics in the car of the eighties

tions as a grid dip meter and therefore
any change in tyre pressure results in a
change at the output of the receiver.

Systems monitor

Of great importance to safe driving, is
the automatic systems monitor (see
photos 4... 7). Firstly, the on-board
computer relieves the driver of this task
and secondly, it can give fair warning of
errors in the system and of any danger-
ous situations which may occur. Fur-
thermore, long-term requirements can
be indicated, for instance, the need to
renew the brake linings or the spark
plugs. The warning or other data may be
given either via a visual display or in an
accoustic manner, that is, a different
sound for each specific warning. Visual
indications may now be provided in
'bright' alphanumerics.

The future of the on-baord computer is
still open to speculation. Some believe
that having such a mathematician in the
car will provide more amusement than
safety. However, we leave that aspect
open to debate.

Power-ring

The development of digital electronics
has led to something new in car elec-
tronics: a ring circuit which will replace
the many individual cables. This incor-
porates a control system for the binary
coded addresses and commands. At the
same time, the load functions may be
checked and reported back to the con-
trol multiplex centre and to the com-
puter.

The dashboard switches will no longer
be connected directly to the electrical
components (lights etc.), but to the
multiplex centre. Every accessory will
have its own multiplex slave station
which receives addresses and commands
from the centre, acknowledges them
and then converts them into actions, for
example, lights are switched 'on' with a
relay. The advantage of such a system is
that the same hardware and multiplex
centre may be employed in all vehicles.
Adaptations may be made by means of
individual programming (software). In
the long run, fibre-optic cables will also
be applied.

Electronics for easy driving

Electronics also serve to make driving
more comfortable. A recent example
of this is the newly developed seat
positioning memory system now in
production in America. Pre-programmed
seat positions can be selected at the
push of a button for a number of differ-
ent drivers.

Heating and ventilation may of course
be electronically controlled. Even elec-

80095 5

Figure 5. Tyre pressure early warning system. Pressure loss in any tyre is detected via the
transmit and receive coils.

electronics in the car of the eighties

elektor april 1980 — 4-07

Photo 4a and 4b. The results of a design study by Ford into the extensive use of electronics
in motor vehicles. The car. Probe 1, does actually exist and apparently Ford even have a driver
for it.

tronic anti-theft systems may be intro-
duced. Bosch has come up with a very
simple design consisting of a gradient
detector and a small computer with a
memory. When the driver leaves the
vehicle, the state it is in at that moment
is stored as data. When something
occurs to alter it, for instance when
someone tries to remove the wheels,
the alarm goes off. Hard luck for the
would-be car thief!

In the light of all the examples given
above, it is highly unlikely that the car
will become any cheaper to run, con-
sidering the expensive sensors and the
mechanical and hydraulic components
involved in electronically controlled
systems. On the contrary, the high
requirements with regard to quality will
make driving more expensive, if any-
thing. Electronics is therefore only
worthwhile in the car where it saves
energy and thereby counteracts rising
fuel costs. M

Photo 5. This instrument panel is interesting in that it is a complete liquid crystal display, and
what is more, is almost ready for production. The top right hand information is a reminder to
change the brake pads.

Photo 6 and 7. Two illustrations of current design thinking where electronics play a significant
role in instrument panels.

4-08 - elektor april 1980

battery protection

The dashboard of an up-to-date car
displays such an array of push-buttons
and switches that the driver sometimes
feels he is controlling an earthbound
rocket instead of a workaday auto-
mobile. Small wonder that is difficult to
remember to switch them all off!
Unfortunately, many of them regulate
battery-run equipment and, especially as
far as the headlights are concerned, a
moment's forgetfulness could mean a
long walk home on a cold night. (For
people living in Yorkshire this might be
quite a problem!) However, a solution
has now been found in the form of an
'intelligent' battery protection circuit.
This particular circuit monitors the
battery voltage level, and when this
drops below a certain safety limit it

battery.

proMm

Don't let your battery leave you in the dark

Forgetting to turn off the headlights need no longer be a motorist's
nightmare. The circuit described here is designed to monitor the battery
voltage and switch off the lights automatically in all kinds of motor
vehicles. Since driving has become as essential to everyday life as it is
expensive, the introduction of such a circuit into your car will no
doubt be an 'intelligent' solution.

causes the headlights to be turned off
automatically. Thus, where the human
brain fails, a simple mechanism takes
over.

So far it has been considered for use in
the car, but it could also apply to other
vehicles, including motor bikes, motor
boats and caravans, with endless possi-
bilities. All it needs is a little imagination
and wiring. The introduction of a
battery control system into a caravan,
for instance, would enable the television
set to switch off automatically, as soon
as, say, an electric kettle is plugged in.
This is just one example — nevertheless,
it illustrates how much energy could be
saved. The system seems likely to
become an essential requirement in the
near future!

Let us now consider the circuit and its
layout in figure 1. At first sight it may
give the impression of being highly
complicated. In actual fact this isn't
the case at all, as it operates quite
simply. Once the ignition has been
switched off (and the lights are still on),
the battery voltage is monitored, so that
as soon as it enters the danger zone
— after the lights have been on for some
time — they are turned off. When the
engine is running, however, the circuit
is inoperative.

The key at the top of figure 1 marks the
ignition switch. As long as it is turned
on, relay Re is powered via D5.
Depending on whether the headlights
are on full or dipped, IC1 is connected
to the supply through D1 or D2, respect-
ively. As long as the battery is reasonably
charged, the output of IC1 will be at
full supply voltage; relay Re is powered
via D3, even when the ignition is turned
off. If the battery voltage drops below
par, however, the output of IC1 will
switch from 12 V to earth. This causes
the relay to drop out and the lights are
then turned off.

IC1 is a standard voltage regulator,
containing a reference voltage source
and an opamp. In this circuit, it is
'misused' — charmingly. The reference
voltage at pin 6 is fed to the inverting
input of the opamp. A control voltage
is derived from the positive supply
(the battery!) by means of a voltage
divider: R3, PI and R4. This voltage is
applied to the non-inverting input of
I C 1 . When the control voltage is higher
than the reference voltage, the output
of the opamp will be 12 V and Re will
pull in. However, when the battery volt-
age drops too far, the voltage at the
wiper of PI will also decrease, with the
result that the non-inverting input will
now be less than the reference voltage.
This causes the output of the opamps to
swing down to earth and the relay will
drop out — switching off the lights.
Obviously, the battery voltage will then
rise again, so that the opamp might be
expected to switch back, but it doesn't.
R5 is included to pull the non-inverting
input down even further when the
battery voltage drops. Theoreticians call
this effect 'hysteresis'. Furthermore,
once the relay drops out IC1 is no
longer connected to the supply. The
only way the relay can pull in again is
when power is applied through the
ignition switch and D5.

PI presets the battery voltage at which
the lights must be switched off. There
will be enough energy left to start the
engine a few times (20 . . . 50% of full
capacity), when the voltage is some-
where between 11.5 and 12 V. The
exact level will of course depend on the
quality of the car battery.

The battery voltage (during discharge) is
plotted in the graph given in figure 2.
This particular plot is valid for a 44 Ah
accumulator, but other types are similar.
As can be seen, 1 1 .5 V corresponds to a
battery charge of around 20%. This will

battery protection

elektor april 1980 — 4-09

have to rise above 12 V, or 50%, for it 2
to be 'normal'.

A printed circuit board is given in
figure 3. At first sight, the position of
the mounting holes may seem peculiar
— but it so happens that they corre-
spond to a fairly commonly available
plastic box. H

discharged condition

charged

Parts list

Resistors:

R1,R6 = 10 k
R2 = 3k3
R3 = 5k6
R4 = 1 k
R5 = 1 M

Capacitors:

Cl = 100 m/25 V

Semiconductors:

D1 . . . D5 = 1N4001

Miscellaneous:

Re = relay 12 V, maximum 100 mA,
heavy-duty contacts.

Figure 2. The discharge curve of a standard
battery.

4-10 — elektor april 1980

transistor ignition

transistor ignition

the best of both worlds

breaker points is controlled by the
distributor cam. When the points are
opened, the current through the
primary winding (PW) of the coil is
interrupted, thereby causing a large
voltage to be induced in the secondary
winding (SW) .

When the contact breaker points close,
the current through the primary of the
coil will increase exponentially to a
value Ip in the time ti, as shewn in the
graph of figure 2. The value Ip is deter-
mined by the resistance of the primary
winding and the battery voltage. The
spark energy released by the points
opening after time ti will be at a
maximum.

The coil has been designed so that the
dwell angle (which dictates the time
that the points are closed) can be
adjusted allowing the maximum
permitted engine revolutions to be
reached and the coil fully loaded (just!).
We are now faced with a problem. If the
dwell angle drops, the primary voltage
will no longer be able to reach its
maximum value at high engine speeds,
which will mean a decrease in high
voltage and, therefore, spark energy.
This alone is reason enough to design an
electronic ignition system which is
independent of dwell angle and engine
speed .

Although far from ideal, it still makes a
lot of sense to retain the conventional
contact breaker points. Their major
disadvantage is the erosion of the correct
gap due to the high current passing
through them. This problem is drastically
reduced with the use of electronic
systems.

This does not render them entirely
satisfactory, since mechanical wear and
degradation of spring tension (causing
contact bounce) will remain the same.
(A point to note here — yes, the Elektor
circuit does take care of bounce prob-
lems). However, these are minor prob-
lems when the finding/constructing/
fitting of optical or inductive alternatives
are considered. In spite of the above
drawbacks, the main advantage of
retaining the conventional points is that
the system can be reverted back to
normal at any desired time — at the
'flick of a switch' even!

There is another problem to consider.
Any increase in the resistance of the
primary circuit of the coil has a detri-
mental effect on the HT and therefore
spark energy. This means any electronic
switch in the primary circuit must have
a low resistance. Theoretically speaking,
this should be no problem. In practice.

Photo 1.

PrototyP e of the Elektor transistor ignition.

An apparent advantage of the
energy crisis is that it has caused a
great deal of energy to be
channelled into improving old
systems (to save energy of course).
We believe that we have done just
that with the outdated car ignition
system. Yes, there are many
transistor/thyristor systems being
produced commercially but they
all seem to suffer from one
disadvantage or another. In this
article we describe an ignition
system which combines the most
significant advantages of all the
others, including the conventional
system. This recipe, mixed with a
few of our own ideas, results in
the Elektor transistor ignition
system — a spark for all seasons?

It would appear, at first sight, that of
the most common types of electronic
ignition a thyristor based version would
present the best possible solution. This
may be true when only the simpler
circuits are considered but not when
major disadvantages in general are taken
into account, for instance, the spark
duration offered by the thyristor is far
from ideal.

In the Elektor electronic ignition system
there were a number of parameters to
be considered:

1. The spark energy must be indepen-
dent of both the engine speed and
the contact breaker dwell angle.

2. Contact breaker 'bounce' must not
affect operation.

3. The HT voltage must be at least
equal to that of a thyristor ignition
system when a standard coil is used.

4. No current may pass through the coil
when the engine is stationary and the
contact breaker points are closed.

The importance of these requirements
will be obvious to most of our readers,
although the finer points may not be so
well known.

Figure 1 shows the conventional system
which has been with us since the year
'dot'. The opening of the contact

transistor ignition elektor april 1980 — 4-11

Figure 3. The Elektor transistor ignition consists of a Schmitt trigger, a monoflop, an oscillator
Figure 2. The curve in the graph shows that <T 1 > and an output stage (T2). The control current for the output stage is obtained by

time is needed for current in the primary coil rectifying the monoflop output pulses. This ensures that a current can only pass through the

winding to rise to a maximum. coil if the engine is running.

however, technological limits have to be about this later on. breaker points will cause a considerable

taken into account; the switching The ability to switch rapidly has the amount of current to flow through the

transistor will have to be protected from added advantage that the HT rises so coil if the ignition is not turned off.

excessively high voltage. Protection quickly that the chances of a flash-over This chance situation will allow the

measures, unfortunately, also lead to in the distributor are diminished. This is temperature of the coil to rise beyond a

undesirable attenuation. In the Elektor because a flash-over can only occur in permissible level. The electronic ignition

transistor ignition care has been taken the event of air ionisation and that must therefore be designed in such a

to ensure that attenuation occurs only needs quite a bit of time. The sooner way that current can only flow through

whenever the voltage limit of the the peak HT is reached, the less chance the coil if the engine is running. How all

switching transistor is approached. of ionisation. these requirements have been included

High voltages are not only dependent on Not only are the high voltage and spark in the Elektor design will be seen at a

the primary resistance of the coil, but energy of importance to ignition, but so later stage,

also on the amount of primary current is the spark duration. The high voltage

and the speed at which this is interrup- ensures the creation of a spark. The

ted. If higher voltages than normal are spark energy (current) determines the Block diagram

to be achieved with the use of a transis- spark temperature. It must be suf- From the block diagram (figure 3) it
tor ignition, this could be realised by ficiently high to ignite the compressed will be seen that the capacitor, C, across
increasing the voltage to the primary gas mixture. Furthermore, the spark the contact breaker points in the
winding. This would certainly require a must be maintained for some time, conventional ignition is also used in an
re-designed coil with a lower primary ensuring ignition. This is because electronic system. In spite of the
resistance (R p ) and consequent re- ignition is not helped by the unequal reduction in current which flows
duction in primary self-inductance (Lp). petrol to air ratio (the modern engine is through the closed contact breaker by
The latter takes place because the designed to run efficiently on a lot of way of resistor R, some ionisation
charge time, which is determined by the air to a little petrol, for obvious reasons), would take place between the contact
Lp/Rp ratio, has to remain the same, and by fluctuations in the gas mixture, points when these are opened. The
This unnecessary expense can be spared. In order to ensure a regular ignition, the capacitor prevents this and so allows the
if the current is switched by an electronic spark will have to last at least one ignition timing to be more exact,
means which is faster than the mechan- millisecond. Any longer would be an The capacitor and resistor combination
ical method. This not only calls for a unnecessary waste of energy. will result in a slower rise time on the

fast switching transistor, but also for a In the conventional ignition system, a leading edge of the trigger pulse, but
clever master oscillator. More details stationary engine with closed contact this problem is overcome by using a

transistor ignition

4-12 — elektor april 1980

Figure 4. The circuit diagram is of necessity more complicated than a simple single transistor ignition as it complies with every major
requirement of today's electronic system.

Schmitt trigger. This also ensures that
slight variations in voltage across the
contact breakers do not lead to spurious
ignitions; only when the points are
definitely opened and closed, is a trigger
pulse produced. Following the Schmitt
trigger is a non-retriggerable monostable
multivibrator. This prevents repetitive
ignition during the monoflop time delay
by producing a trigger pulse with a
width which is independent of the
opening angle (and the dwell angle). The
monoflop time delay has been calculated
so that a maximum of approximately
6000 revolutions per minute in a four
cylinder engine still allows for enough
time to load the coil, ensuring a suf-
ficiently high HT voltage.

The output pulses from the monoflop
are fed to a master oscillator and a
rectifier. The master oscillator causes
the coil current to be interrupted at
intervals (by switching the current to
earth before the output stage). The
output of the rectifier supplies the coil
primary current via the output stage
(T2). This is only possible if trigger
pulses are produced (in other words:
when the engine is running).

Circuit diagram

A glance at figure 4 will show that there
is a great deal more to this type of
electronic ignition than the conven-
tional system. How does it work and
what sort of technical aspects have been
incorporated to meet the set parameters?
It seems reasonable to start at the very
beginning, which is in this case, at the
far left of the circuit diagram.

The purpose of the contact breaker
(dotted area) now becomes a simple
pulse shaper, as opposed to a power
switch. Resistor R1 has been added in
order to limit the current through the

contact breaker to approximately 250
milliamps. On the one hand, this current
is sufficiently small to control wear and
tear caused by ionisation between the
contact breakers, and on the other, it is
large enough to burn away any dirt on
the contacts.

The trigger pulses produced by the
points are cleaned up by the Schmitt
trigger formed by transistors T1 and T2,
allowing a more precise ignition timing.
The Schmitt trigger is designed so that
the switching levels will be approxi-
mately 2.2 and 7.4 volts with a battery
voltage of 12 V. The trigger pulses
therefore start when the voltage across
the contact breaker rises to at least
7.4 V and end when the level drops
below 2.2 V. The difference between
these switching levels is called hysteresis.
It prevents any residual hum voltage
across the points from influencing the
system.

The trigger is designed to react quickly
to positive going transitions (the contact
breaker opening), but to have a delayed
reaction to negative slopes (the contact
breaker closing). This is achieved by the
combination of R2, R3, Cl and D1,
where D1 provides the shorter and
longer time constants, respectively. The
shorter time constant is required to
trigger the ignition 'on' time and the
longer time constant to suppress contact
'bounce'.

The Schmitt trigger is followed by a
non-retriggerable monoflop (T3 and T4).
This produces a pulse with a fixed
period, independent of the dwell angle.
When inoperative, these transistors are
turned off until the moment the Schmitt
trigger produces a positive pulse when
they will both start to conduct. The
positive going voltage at the collector of
T4 is fed back to the base of T3 via C5
and R10. The monoflop now remains in
this state until C5 is charged through R9.

Parts list

Resistors:

R1 =47 n/5W
R2 = 82 n
R3 = 22 k
R4 = 39 k
R5,R7,R1 1 = 3k3
R6.R12 = 47 k
R8 = 2k2
R9 = 15 k
R1 0 = 6k8
R1 3,R1 5 = 100 n
R14.R18 = 220 n
R16 = 1M5
R17 = 470n
R1 9 = 1 S1/2W
R20 = 8 £72/25 W
R21 ,R22 = 0£21 8/2 W

Capacitors:

Cl = 100 n
C2.C6 = 68 p
C3 = 33 p
C4 = 470 m/25 V
C5 = 47 n
C7 = 220 n
C8 = 470 n

Semi-conductors:

D1 . . . D6 = 1N4148

D7,D8 = 1 N 5406

D9 . . . D18 = zener 200 V/

400 mW

T1 ,T3 = BC 547B
T2,T4 = BC557B
T5 = BD 131
T6 = BC 516
T7 = BD 242A
T8,T9 = BU 208 A

Miscellaneous:
heatsink for T8 and T9
(TO-3 size)

heatsink for T5 and T7
(TO-66 size)

flexible oil resistant wire 2.5,mm !
suitable die-cast box

in 1111

elektor april 1980 — 4-13

transistor ignition

V V V V w

1 n s *r Va

b| n

b

ii

u

b

B

B

'Pg |M

-i U -i

ulm

-i

-1

-i

0

10

U1

u

IU

4-14 — elektor april 1980

transistor ignition

6

terminal and connector
insulating plate
plastic T03 isolator
solder tag

4 BA nut and bolt
locking washer

pin and connector
printed circuit board
metal case

Ficure 6 This detailed drawing shows how to mount the connectors onto the case. The insulating plate prevents thelterminals from touching the
case The plastic isolator is in fact meant to be used when mounting TO-3 style transistors but it may equally well be used here.

With the given values for C5 and R9, the
monoflop period will be approximately
two milliseconds. After that time, any
further energy pumped into the coil
would be completely converted into
heat. The monoflop output pulses are
rectified by diodes D5 and D6, which
(via T6 and T7) only provide transistors
T8 and T9 with base drive current when
the engine is running. These transistors
switch off more rapidly if the charge
carriers are removed as fast as possible
from the base. This happens more effec-
tively when the base drive current is
short circuited to earth rather than inter-
rupted. All the charge carriers that are
still moving due to base/collector capaci-
tance during the blockage do not then
arrive at the emitter, and this results in
a very fast switch off. Switching the
base drive current to earth is carried out
by T5. The coil current must be
switched off quickly to achieve high
ignition voltages, and for this reason, T8
and T9 are high speed switching types.
With a view to safety and reliability,
transistors T8 and T9 are protected
from extreme voltages by means of
zener diodes (in case of flash-over
between secondary and primary
windings, for instance).

It is possible for the primary coil voltage
to go negative. The switching transistors
T8 and T9 do not respond favourably to
this and therefore have to be protected
by means of diodes D7 and D8.

Why are two switching transistors used?
The BU 208A is relatively cheap and
complies with the technical require-

Photo 2. Resistors R1 and R20 are special
power types which may be attached to the
chassis with aid of nuts and bolts.

ments. Unfortunately, however, it can
only just withstand the considerable coil
current. Since the ignition forms a vital
part of the car, it is better to run no
risks and that is why there are two
switching transistors in parallel.

Construction and setting up

Mounting the components onto the
printed circuit board (figure 5) should
not cause any problems. The power
resistors R1 and R20 dissipate a fair
amount of power and are therefore
chassis mounted. Transistors T5, T7, T8
and T9 need to be mounted on separate
heatsinks. Do not use mica insulation
washers between the cooling surface and
the transistor, as this would hinder the

heat transfer. Cooling electronic com-
ponents under the bonnet is a problem
as it is, especially if your car is in a
traffic jam on a warm summer's day.
Special attention should therefore be
paid to cooling the transistors. As a last
resort, a special heat conducting silicon
paste may be used.

Because they are 'live', care should be
taken to ensure that the heat sinks are
not electronically connected to any
other components or the metal case.

It should be remembered, that if one of
the switching transistors T8 or T9 were
to become loose and move, it is possible
for the emitter to touch the heatsink
and therefore the collector. This will
most decidedly stop the engine. It is
advisable to sleeve both the emitter and
base pins and ensure that the transistors
are fitted securely.

The ignition system should be housed in
a sealed metal case, a die cast metal box
would probably be the most suitable.
The six connections to the outside
world can be made as shown in figure 6,
although a minor difference to the
illustration may be preferred. The pin
and connector on the printed circuit
board could be left out and the con-
necting wire soldered directly to the
board. The tag could then be placed on
the other side of the 4BA nut (to the
right of the illustration) and held in
place by a further nut, which will also
double as a locking nut for the whole
assembly. The tag could still be removed
if required without loosening the
terminal connector.

transistor ignition

elektor april 1980 — 4-15

Figure 7. This drawing should help avoid any problems while constructing the electronic ignition. The arrows indicate how to convert it back
to the mechanical system, should anything go wrong.

The connecting wire used should be
2.5 mm 2 (at least), flexible and oil
proof cable (obtainable at any car
accessory dealers). Mounting the two
power resistors requires a little more
attention. In order to prevent the solder
connections from loosening because of
heat development, connecting wire
needs to be passed through the solder
eyes of these resistors and then wrapped
firmly before being soldered. The power
resistors are a special type (see photo 2)
and can easily be mounted onto the
chassis using nuts and bolts.

The drawing in figure 7 clearly shows
the external connections. Should any-
thing go wrong as far as the electronics
is concerned, it would of course be
advantageous if the system could revert
to the conventional ignition without
any difficulty. For this purpose a 'loose'
double connection clamp is attached to
the case. By connecting the leads from
the contact breaker and the coil to this
point, the ignition operates once more
in the conventional manner. It would,
however, be advisable to disconnect the
supply to the electronic section also.
Arrows in the drawing show which
modifications to carry out in order to
'switch off' the electronic system. An
alternative, of course, is to include two
switches. One (double pole) switch to
break the connections from the contact
breaker and coil to the unit and join the
two together, and one (single pole) to
creak the positive (or negative) supply
connection. This has the added advantage
mat the second switch can be used as an

anti-theft device. If the system is
switched to the transistor ignition and
the power to the unit disconnected, the
engine will not run.

As mentioned before, the capacitor
across the contact breaker is still used.
However, its value must be 0.1 /iF. Once
the unit has been assembled, it can be
tested. The ignition timing will remain
unchanged. The points gap is now no
longer critical, but it is advisable to set
it as recommended by the manufacturer.
Otherwise, mechanical problems could
arise.

To make it easier to set the ignition
timing, an LED in series with an 820 £2/
0.5 Watt resistor (shown dotted in
figure 4) may be placed across R1. Note
that the LED must be connected the
right way round. When the contact
breaker points close, the LED lights (in
contrast to a control bulb placed across
the points). It would be far better,
however, to have the ignition timing set
dynamically with the aid of a strobo-
scope. External connections to the coil
and distributor are as numbered in
figures la and 7.

Having completed the construction,
fitted the unit into the vehicle and
tested the system, it is then highly
advisable to take one somewhat strange
step further. Remove the unit from the
vehicle and take it back to the
workshop — for a very good reason. As
previously mentioned, the under bonnet
environment is extremely hostile and
for the sake of reliability a protective
measure is called for. The printed circuit

board should be removed and given two
or three coats of clear polyeurathane
laquer. This will prevent damp from
affecting the circuit. When this is dry,
the whole unit should then be reas-
sembled paying particular attention to
all fixing points, nuts and bolts etc.
When satisfied, the unit can then be
refitted into the vehicle. It should of
course operate perfectly.

This last step may seem a lot of extra
effort but take note of the following.
Murphy's Law states that if the ignition
of any vehicle should fail, it will do so
late on a Sunday evening, miles from
anywhere, during the worst rainstorm in
ten years.

Some cars are equipped with a resistor
in series with the coil primary winding
which is bypassed during operation of
the starter motor. This must not be
removed, as overheating of the coil
would result. Apart from the number 1
connection to the coil (which, in this
case, should be fitted to the other side
of the resistor) all connections remain
the same. This ensures that the by-pass
resistor system is maintained.

If a rev counter is fitted to the vehicle,
it should be connected to the coil
(connection number 1) when this elec-
tronic ignition system is used. N

4-16 — elektor april 1980

a dip-stick probe

The existing temperature meter in most
cars measures the water temperature.
Since the water is passed through the
radiator, with various thermostatic
controls to keep the temperature in the
optimum range, it cannot be expected
to give a reliable check on the oil
temperature. Any indication is likely to
be 'too little and too late'.

electrical wiring!— the dip-stick seems
the only possibility.

What is required is a replacement
dip-stick with a thermistor (NTC
resistor) built in. VDO supply two
versions, which can be adjusted to fit
almost any car; it seems likely that
other manufacturers will soon be
supplying similar probes. We would not

stick probe

There are all kinds of indicators
and warning lights in modern cars,
but an oil temperature indicator is
rarely included. One might
conclude that the oil temperature
is not important . . . however, one
would be wrong.

When oil is overheated, it tends to
go 'thin' and in extreme cases it
may even oxidise. Either way, it is
no longer a good lubricant.
Keeping an eye on the water
temperature may give some
warning, of course, but it doesn't
tell the whole story — to play it
safe, the oil temperature itself
should be measured.

The easiest way to add this type
of indicator to an existing car is to
mount a temperature sensor in the
dip-stick.

So how do you measure the oil tempera-
ture? Not from the outside, that's for
sure: the oil is deep inside the engine.
Some way must be found to insert a
temperature sensor into the oil — in
the sump, for instance. Only the most
courageous do-it-yourself mechanics are
likely to attempt drilling a suitable hole
through the engine casing, so it seems a
more attractive proposition to look for
existing holes. Apart from the plug
underneath the sump —garage mechanics
tend to manhandle that with heavy
spanners, so it is not a good place for

advise anyone to try making their own,
unless they are very experienced: there's
not much room inside an engine, and it
gets hot and vibrates. If any bits fall off
the dip-stick, the results may be very
expensive.

A VDO dip-stick is shown in photo 1 . In
an extensive series of drawings, the
manufacturer shows how to fit it in a
large number of different cars.

The circuit

As can be seen in figure 1, the bulk of
the circuit is contained in a single

Photo 1. A dip-stick with built-in temperature sensor (made by VDO).

1

R1

fi

A1 . . . A4 = LM 324 = IC1

D3 = yellow or red
D4 = green
D5,D6 = red

< 2 >

80102

U 0 ut

Figure 1. The complete circuit of the oil temperatures indicator. The bulk of the circuit — the four opamps — are contained in a single 1C.

a dip-stick probe

elektor april 1980 — 4-17

integrated circuit: the LM 324; this 1C
contains the four opamps A1 . . . A4.
The first of these, A1, is used in a
voltage regulator circuit that includes
D1 and T1. The reference voltage from
D1 is applied to the non-inverting input
of the opamp; the latter controls the
base current of T1 in such a way that
the voltage at the inverting input of A1
(R2/R3 junction) is kept almost identi-
cal to the reference voltage. With the
voltage across R2 held constant at 5.6 V,
the total voltage across R3 and R2 is
equally constant at 8 V. This voltage is
used to drive the actual measuring
circuit.

The dip-stick thermistor (o.- 'NTC') is
connected in a so-called bridge circuit:
R5, the NTC, R6, R7 and PI. Opamp
A2, with resistors R8 and R9, is a buffer
stage with a gain of xl. This means that
the difference between the voltage at
the R5/NTC junction and the voltage at
the R6/R7 junction appears at the
output of A2. As the oil temperature
rises, the resistance of the NTC will
drop; the voltage at the non-inverting
input of A2 will decrease, so that the
output from this opamp will also fall.

In principle, a voltmeter can be connec-

ted to the output of A2 and calibrated
in degrees Centigrade or Fahrenheit.
The actual scale obtained depends on
three factors: the type of NTC used
(and the value of R5); the values of R8
and R9; and the setting of preset PI.
With a given NTC, R8 and R9 determine
the temperature range that can be
measured: with the values given
(R8=R9 = 47k) 1 V output swing

corresponds to approximately 10°C
temperature change (*» 18°F), so that a
total range of some 120°C (21 6° F ) can
be measured. The actual scale limits are
set by means of PI — in the example
given, 0 ... 120°C (32 ... 248°F) might
be chosen.

A different scale can be obtained by
modifying the values of R8 and/or R9,
and readjusting PI. For a Fahrenheit
scale, for instance, R8 = 56 k and
R9 = 100 k; this gives an output from
A2 of approximately 1 V per 10°F.
However, a thermometer using an NTC
is unlikely to be highly accurate over
the full temperature range; furthermore,
the only oil-temperature indication
needed in a car is on the lines of 'on the
low side— just right— too high'. For
these reasons, a simple optical indi-

Figure 2. Printed circuit board and component layout.

Parts list:

Capacitors:

Resistors:

Cl = 47 p/10 V

R1.R12 = 1 k

C2 = 470 p/25 V

R2 = 10 k

R3 = 4k7

Semiconductors:

R4 = 18 n

A1 ,A2,A3,A4 = IC1 = LM 324

R5 = 1 80 n (see text)

D1 = 5V6, 400 mW zener diode

R6 = 4k7

D2 = 1 N4001

R7 = 220 n

D3 = LED (yellow or red)

R8.R9 = 47 k

D4 = LED (green)

R10 = 3k3

D5,D6 = LED (red)

R 1 1 = 3k9

T1 = BC 107B, BC 547B or equ.

R1 3,R1 5 = 470 fl

R14 = 390 «

Miscellaneous:

R16 = 10 n

Dip-stick with built-in thermistor;

PI = 2k2 preset potentiometer

for example, VDO type 5310024.

cation was chosen for this circuit. In
figure 1, D3 . . . D6 are LEDs (Light
Emitting Diodes) that light up to
indicate the oil temperature.

Once the circuit has been calibrated
correctly (more on this later), the
indication circuit works as follows. At
low oil temperatures (below 80°C or
176°F), the output voltage from A2 is
higher than the reference voltage at the
junction of RIO and R 1 1 . The outputs
of both A3 and A4 will be 'low'
(virtually 0 V), so only D3 will light. As
the temperature rises above 80°C, the
output of A2 drops below approxi-
mately 4.8 V. At this point, the output
of A3 swings 'high' (to full supply
voltage); D3 goes out and D4 lights.
This (green) LED indicates that the oil
temperature is in the optimum range. If
the temperature rises even higher,
however — above 120°C, or 250°F —
the output from A2 will drop below the
second switching point (approximately
1 V). The output of A4 will then also
swing high, so that D4 goes out and D5
and D6 both light. These two (red)
LEDs therefore spell 'Danger'!

Construction and calibration

The complete circuit is mounted on the
printed circuit board given in figure 2. It
shouldn't be too difficult to find a small
plastic box to fit it. According to taste,
the whole lot can be mounted 'in sight'
or, alternatively, the box canbe mounted
behind the dashboard and the connec-
tions to the LEDs brought out to some
suitable point.

Calibration is not too difficult. With the
component values given, the changeover
points from one LED to the next should
be at 80°C and 120°C (176°F and
250°F, respectively). The simplest way
to calibrate the unit, therefore, is to dip
the sensor into a pan of water at 80°C
and adjust PI until the indication just
switches over from D3 to D4. The upper
indication will then correspond to
120°C (it is difficult to heat water that
far . . . ). As a further check, the output
of A2 can be measured when the sensor
is dipped in boiling water — it should
then be approximately 3 V. Absolute
accuracy is not essential for this type of
indicator, of course.

By way of conclusion, a word about
the sensor itself. If the VDO unit is used,
the value given for R5 (18012) is
correct. For other sensors, however, it
may have to be modified. As a rule of
thumb, R5 should be between one-half
and two-thirds of the resistance of the
NTC when the latter is immersed in
boiling water. As an example, say that
the thermistor value is measured as
100 12 in boiling water; a good value for
R5 is 56 12 in this case. This, by the way,
should also be taken as the minimum
permissible: only NTCs with a value
between 10012 and 10 k at boiling
temperature can be used in this circuit
(note that this will normally have to be
measured: the resistance of an NTC is
usually specified at room temperature) .M

4-18 — elektor april 1980

intelligent wiper delay

An extremely annoying aspect of
poor windscreen wipers is that
they are fine — until you have to
use them. We unfortunately are
unable to do anything about that
but we can help them cope with
varying weather conditions.

There have been many circuits
published relating to wiper delays
but never one quite like this. '»

It is intelligent, it only needs to be
told once what is required of it.

It will then carry out your orders
until you change them, which you
can at any time, instantly. It's so
good that we (being male
chauvinist pigs) are currently
converting one into a secretary.

wiper delay

single push-button control and continuously variable

The main disadvantage of conventional
delay circuits is that the correct setting
is often awkward to find. Usually, it is
a devil of a job, when using a pot, to
find the right setting for keeping the
screen clear without the wipers having
to make too many wipes. Once the
correct mode has eventually been
found, more often than not, the rain
suddenly starts pelting down or dwindles
to a quiet drizzle, or a car passes and
floods the entire windscreen with
buckets of water ... All in all, such a
circuit is nothing to be proud of! With
Elektor's intelligent windscreen wiper
circuit, however, things are different.
It is so easy to operate that the number
of wipes per minute can be changed in
a fraction of a second. Pushing a button
twice is all that is needed to effect a
change. It will 'remember' how much
time passed between the two button
presses and will maintain this interval
for as long as is required.

Such a simple operation is due to the
fact that neither components nor efforts
were spared during its design. The cir-
cuit itself is fairly complicated, but has
been simplified to a reasonable level.
Only four leads have to be connected to
the existing windscreen wiper wiring.
Under the heading 'construction' some
reference will be made to the standard-
isation of the terminal coding in cars.
Since this is not generally accepted as
yet, a method will also be described of
how to find the right terminals easily by
using a test lamp. In short, the construc-
tion and setting up should not pose any
problems. The exact working of the
circuit will be a little more difficult to
understand. It therefore seems appropri-

intelligent wiper delay

elektor april 1980 — 4-19

figure 1. The simplified block diagram of the windscreen wiper delay circuit that includes a
digital memory.

ate to start by explaining this, before we
go any further.

The block diagram

At the heart of any delay circuit there
must always be a memory. This may be
of the simple analogue type, an RC net-
work, for instance, where the RC-time
determines the delay duration. If the
delay is to be regulated as required, the
memory will have to be slightly more
complicated. Then a pot or a step switch
will have to be included, so that the RC
time constant may be adjusted.

More up-to-date technology was used in
designing the delay circuit in this
project. There are of course memory
IC's which can store digital information
and quite a large amount of memory is
available in an 1C of this nature. Infor-
mation may be stored in the form of a
'0' or a 'V. The information to be stored
is fed into the data input. Further-
more, there are ten lines through which
information may be fed to the 1C to
determine the selected memory lo-
cation. In the 1C used here there are
1024 memory locations, or addresses.
If these are numbered from 1 to 1024,
they may then be selected by setting
the number on the ten address lines in
a digital (binary) code. The code will
be given by a counter which can count
to 1024. If we allow the counter to
count pulses from a clock-generator, a
time factor is introduced into the
memory. Should the clock-generator,
for example, produce five pulses per
second, five memory addresses per
second will be selected in succession.
The memory also has a read/write con-
trol input, thereby allowing information
to be written into or read out of
memory. Previously stored information
appears at the data output during the
read cycle. By utilising this method of
operation, a useful delay circuit can be
realised.

The block diagram of the circuit is
shown in a simplified form in figure 1 .
Any contact bounce in the wiper con-
trol switch is eliminated by the pulse
shaper. The output of the pulse shaper
controls an electronic switch which
determines the read/write mode of the
memory section. Also included is a
clock-generator which produces a fixed
number of pulses per unit of time.
Apart from this, there is a reset-section
and a small block to control the wind-
screen wiper motor. The most import-
ant part of the system is the block
marked counter and memory.

When the wiper switch is operated, the
wipers wipe (once), the counter is reset,
and a number of memory addresses are
filled with 'noughts'. When the wiper
switch is operated a second time, a 'one'
is entered into memory and the counter
again reset. The memory is now placed
nto the read mode and scanned until
the '1' is found. This causes the wipers
to operate and the counter to be reset.
As the memory is read at the same

speed that information was entered the
time delay between successive wipes will
be the same as that between the two
initial operations of the wiper switch.

All this sounds simple enough, but in
practice quite a few obstacles will have
to be overcome. For instance, it is
highly important that everything hap-
pens at the right time and in the right
order. Usually all important events
start the moment the clock-signal
changes in level. However, when the
memory is controlled, the events have
to occur in a fixed order. Thus, of
course, the address first has to be
chosen before the information is written
into it. Trying to do this simultaneously
would probably cause the information
to end up in the wrong memory lo-
cation. Instead of a single clock-signal,
therefore, several will be needed. These
will change in level at different
moments, but will have the same fre-
quency. That is why only one clock-
generator is used, but its output is fed
to a sequencer before it controls various
IC's. Each of the outputs from the
sequencer go high in turn for the
duration of one clock pulse.

Only four of the ten available output
signals are used in the delay circuit. In
figure 2 an attempt has been made to
illustrate how the various important
parts of the circuit function.

The circuit

The complete circuit diagram of the
'intelligent' wiper is shown in figure 3.
The actual switching of the windscreen
wiper motor is carried out by a relay.
When the relay is activated, it will
operate the wipers. But first it is better
to discuss the relay control. The clock-
generator is formed by the circuit

around N1. As soon as the supply volt-
age is applied, the frequency of oscil-
lation, determined by the values of
C2/R3, will be fairly high (a few kilo
Hertz). When Cl is sufficiently charged,
D1 will conduct, holding the voltage
at the junction of C2/R3 high. This
makes the oscillator run at a much
lower frequency (approximately 25 Hz)
because of the greater time constant of
R4/C3.

The output of the clock-generator is
fed to 1C 1 , a 4017, which serves as the
sequencer mentioned previously. The
output at Q1 is high during the second
clock period, Q5 during the sixth, etc.
The memory is contained in a single
2102, IC2, which is connected directly
to the binary counter, IC3, via its
address lines (A0 . . . A9). Two flip-flops
are contained in IC4. They operate as
follows. When switch SI is operated the
output of N3 will go low. This '0' is fed
to the D input (pin 5) of the first flip-
flop. During the positive transition of
the next clock pulse (Q5 of I C 1 ) this '0'
will appear at the Q output (pin 1) and
will be inverted at the Q output (pin 2) .
The flip-flop in the second half of IC4
reads the 'old' condition of the first
flip-flop during the first clock-pulse and
remains unchanged. It is not until the
second clock pulse that the second flip-
flop passes on the new data from the
D input to the Q output (inverted of
course).

During the second clock pulse however,
the first flip-flop does not react (at least
not unless SI is operated again). The
signal change at the D input of the first
flip-flop does not arrive at the 0 output
of the second flip-flop until after two
clock pulses. The remaining 1C, IC5, is
a monostable multivibrator which is
triggered by a negative going edge. When

4-20 — elektor april 1980

intelligent wiper delay

figure 2. An extensive block diagram of the wiper circuit.

the level at its input (pin 5) goes low,
its output (pin 6) will go high for a
period of time determined by the values
of RIO and C5. When this output is
high, transistor T2 is turned on and the
relay is energised.

That just about covers the main com-
ponents of the circuit, but now we go
into greater detail.

Normally, when the windscreen wipers
are inoperative, the wiper of SI will be
connected to ground. The input of N3
will therefore be low and its output will
be high, when the supply voltage is
applied. As soon as the supply is 'on'
the clock-generator starts to oscillate.
The information at the D input of the
first flip-flop then arrives at its output
as soon as Q5 of IC1 goes high. The data
input of IC2 will be low. At the next
pulse from the Q5 output of IC1 the
Q output of the second flip-flop will go
high. The output of N2 is only low
when both inputs are high, therefore,
IC2 will be put in the 'write' mode and
a 'nought' written into the memory when
the Q1 output of IC1 goes high. Let us
suppose that the output of N6 is high
immediately after switch-on. The out-
put condition of N5 is then determined
exclusively by the signal from Q7 of
I Cl . This clock signal is applied to the
counter (IC3) via N5 and so success-
ively incrementing the address lines of
IC2. This causes a 'nought' to be written

into each memory location. After a
maximum of 1 024 clock pulses the Q1 1
output of IC3 goes high. The output of
N6 will then be low thereby inhibiting
the clock pulses to IC3. The circuit is
now ready for use and the LED (D2)
will go out. All this happens in about
one second flat, because the clock
frequency is high immediately after
power-up. If the wiper switch is oper-
ated and held closed, the D input of the
first flip-flop will be taken low. At the
following positive transition from Q5
of IC1 the data input of IC2 will go
high. The memory will still be in the
'write' mode as the Q output of the
second flip-flop is still high. The counter
receives a reset pulse, via N7 and N8,
when the Q output of the first flip-flop
goes high as Q1 1 of IC3 is also high at
that moment. This means that a '1' is
written into the first memory location
as soon as the Q1 output of IC1 goes
high. A little later, Q3 of IC1 will go
high and as the memory is now in the
read mode, both inputs of N4 will go
high providing a trigger pulse for IC5.
The output of IC5 goes high turning on
T2 and activating the relay. At the
following pulse from Q5 of IC1 the
memory is put in the 'read' mode via
the second half of IC4 and N2. The '1'
in the first address location is read once
more and the wipers move again. At
the same sime, however, IC3 is once

again reset. In this way the circuit
continues to read the first address.
The interval timing is started as soon as
the driver releases SI . This is because
the information at the D input of IC2
then changes to 'O'. A little later, the
memory 1C is switched to 'write' by
means of the second half of IC4 and
'noughts' are written into memory. The
'1' in the first memory location is over-
written with a 'O'. When the memory is
read only 'noughts' appear at the output
of IC2 and the windscreen wiper stops.
IC3 is no longer reset.

After the required interval has passed,
SI is again operated. Now the infor-
mation at the D input changes to 'V,
while IC2 is still in the 'write' mode.
After this '1' has been written (while Q1
of IC1 is high), the same 'V is read
(while Q3 is high) . This causes the wind-
screen wipers to operate but soon after,
IC3 is reset by way of N7. With the aid
of Q5 the R/W input is switched to
'read' via the second flip-flop. All the
'noughts' written in the memory are
now read in succession until the '1' is
read again. Then IC3 is reset, so that
the cycle starts at the beginning.

The monoflop time of IC5 can be
selected in such a way that every time
IC5 is triggered the wipers make a single
stroke. By enlarging the value of capaci-
tor C5, two strokes can be made instead
of one.

intelligent wiper delay

elektor april 1980 — 4-21

The LED D2 lights whenever the Q1 1
output of IC3 is low. This is the case if
the supply voltage has been available for
long enough and SI is not closed.
D2 will therefore indicate when the
circuit is ready for use. If the wind-
screen is to be continually wiped while
D2 is still lit, this can be done quite
simply by presetting a very short
interval. SI must then be switched very
briefly.

Finally D3 lights when the relay is
under power. Now it will be clear, even
without the LED, when the relay is
activated, but the LED will indicate a
j defect or, at least, show whether the
circuit is working properly.

Construction

The printed circuit board and com-
ponent layout for the intelligent delay is
shown in figure 4. Mounting the com-
ponents onto the printed circuit board
should not cause any problems, as long
as 1C sockets are used. A good idea is to

insert the entire circuit into a plastic
case. Secure terminals for outside con-
nections will also be necessary. At least
eight connections are required: one for
the 12 V line, one for D3, one for D2,
three for the relay, one for the SI
connection and finally an earth connec-
tion. A suitable case is the OKW type
number 90-30-087. It can of course be
mounted at the bottom of the dash-
board for instance, or it can be fitted in
a less visible place. The LEDs can either
be mounted on the case or the dash-
board.

The relay must be of a type that is
reliable at 12 volts and must possess two
sets of contacts. One of these (88 and
88a in figure 5) must be a normally
open pair and the other (87, 87a) a nor-
mally closed pair. The contact current
rating of the relay should be at least 5 A
(preferably 10 A). It is also important
that the coil current must be no more
than 400 mA as this is the safe I imit for
transistor T2 (BC 517).

Installation

Without any prior knowledge of wiper
circuits, the installation of the delay
system can present something of a
problem. This is no cause for alarm,
since the great majority of wiper circuits
are not very sophisticated and can be
readily understood.

Let's start with figure 5. The problem
part here is switch 'H' because this is
almost certainly out of sight, tucked
away somewhere inside the wiper motor
mechanism. What is more, there are as
many different types of switch as there
are wiper motor manufacturers. The
first question to answer is — do you
have one in your car? A very simple
check verifies this. When the wipers
are switched off, do the blades park
themselves neatly at the bottom of the
windscreen? If the answer is yes, you
can sit back with a sigh of relief. If the
answer is no, then changing the wiper
motor for a later model might be a
solution for you as the fitting of this

figure 3. The circuit diagram of the 'intelligent' windscreen wiper deley. The relay must be able to switch at least 5 amps.

4-22 — elektor april 1980

intelligent wiper delay

® ©

figure 4. The printed circuit board and component overlay for the wiper deley circuit.

Resistors:

R1 ,R7,R9,R1 2 = 10 k
R2,R1 3 = 1 M
R3,R4,R5 = 100 k
R6 = 82 k
R8.R11 = 390 n
R1 0 = 4M7
ri 4 = ion

Capacitors:

Cl = 22 p/1 0 V tantalum
C2 = 1 00 p
C3 = 68 n

C4,C7 ,C8,C1 1 ,C1 2 = 100 n
C5 = 220 n

C6 = 1 m/ 10 V tantalum

C9 = 470 m/25 V

CIO = 10 m/ 10 V tantalum

Semiconductors:

D1 = DUS
D2,D3 = LED
D4.D5 = 1N4001
T1 = BC 547

T2 = BC 51 7 (Darlington)

IC1 = 4017

IC2 = 2102 (L) 1024 x 1 memory

IC3 = 4040

IC4 = 401 3

IC5 = 4098, 4528

IC6 = N1 . . . N4 = 4093

IC7 = N5 . . . N8 = 401 1

IC8 = 78L05

Miscellaneous:

SI = double pole switch
Re = relay 1 2 V (e.g. Bosch)
case OKW type 90-30-087

intelligent wiper delay

elektor april 1980 -4-23

figure 5. What the windscreen wiper wiring will look like in your car. The dotted area contains
the relay contacts of the delay circuit.

Instructions for use

Even an 'intelligent wiper delay' unit must be told what to do. Fortunately, it is
quick on the uptake. More specifically, it is designed to work 'as you would expect'.
Let's say you're driving along, and it starts to rain slightly. At a certain point, you
switch on the wipers to dear the windscreen — than you switch off again when it's
dear. The fact that the wiper switch actually controls an 'intelligent' wiper doesn't
make any difference.

Some time later, it becomes time for another wipe. So you switch on the wipers.
However, you don't switch them off again! The circuit does. From then on, It
will give one wipe at the same interval, until you switch it off again.

This is great, but, every driver knows rain is anything but consistent. No problem!

• The rain stops. In this unlikely event, switch off the wipers. What else did you
expect?!

• The rain increases. You're windscreen needs wiping, but the automatic system
doesn't realise this. So you switch the wipers off and on again. The 'screen is

wiped, and the shorter interval is remembered. Until further orders, this interval
will be maintained between wipes.

• The rain decreases. So what do you do? You switch off the wipers — just after
they've cleared the windscreen. When it's time for a new wipe, switch on again —

and leave them on! They'll do the job, at the new interval, from then on.

Intelligent? Yessir! You operate the wiper switch in the normal way, whenever the
amount of rain changes. But you only do it once — from then on, the unit does the
job automatically.

switch is not a practical proposition.

The other switch in figure 5, Sa/Sb, is
in fact the car wiper switch (standard
fitting on most cars!). The circuit shown
is of a typical two-speed wiper system.
The numbers indicated on the motor
and switch wiring (15, 31, 53 etc.) are
the European DIN standard desig-
nations, this means you may or may not
find tern on your car. However, for the
sake of simplification it is convenient to
use them here.

The question now of course is, how do
you identify these wires on your wiper
motor? For this a small 12 volt bulb
with two connecting leads is required.
One lead is connected to the chassis
(or the battery's negative terminal) and
the other is used to scan the terminals
of the motor. There is no need to dis-
connect any of the wiring for these
tests. With the wipers switched off, the
bulb will light on only one of the leads
to the wiper motor. This will be 53a on
the drawing. If, both the ignition and
wiper switch are now turned 'on' (to
low speed if there are two positions)
a further lead will be found to light
the bulb and this will be 53. Lead
number 31 b can now be easily found as
on this one the bulb will flash on and
off.

The dashboard switch can now be put
in the second position (fast speed) and
53b will 'light the light'. The lead which
never produces any reaction at all is
number 31, or the negative connection
of the system.

At least 90% of cars will be covered by
this simple testing method. In some cars
however, the negative supply is switched
(positive earth systems). The above test
procedure remains the same, but the
positive and negative (leads 15 and 31
in figure 5) will have changed places.
Having now identified the wiring, only
two points need to be explained to
clarify the complete standard wiper
circuit.

As mentioned earlier, switch 'H' is built
into the wiper motor and is in fact
switched off (to position '0' in figure 5)
by the motor when the wiper blades
reach the edge of the screen. This is
known as the 'park' position and
switch 'H', the 'self-park' switch. There
now remains only one more detail of
the circuit to be explained. Switch 'H'
has two purposes in life. One is that
described above, and the other is to
provide a 'short' across the wiper motor
when it eventually switches off. The
reason for this is that due to the inertia
of the mechanical system, it may over-
run the park position, a 'short' across
the motor acts as a brake and prevents
this. It is now a simple matter to con-
nect the relay to the wiper system by
following the circuit in figure 5. Only
one lead needs to be cut, and this is
labelled 31 and 53e as shown. Please do
not ask us why DIN standards require
two numbers for one cable, we have
quite enough illogical answers to this
already.

active car aerial

Better reception
with less noise

'fringe area' can be anywhere between
20 and 50 miles from the transmitter.
This means that most people are faced
with poor reception more often than
not! AM on the other hand (long and
medium waves, in particular) has a
much wider range. You don't need an
aerial amplifier for Droitwich! However,
if you like to listen to Continental
transmitters, or if you intend to drive
to the Continent, an aerial amplifier
may be a useful investment.

Having made the decision to install an
aerial amplifier, there is still a difficult
choice to make: what type? To be
honest, most commercial amplifiers
supplied for cars are not really worth
writing home about. They often contain
a FET, used as a source follower, and a
wide-band amplifier. The idea is to
cover the whole tuning range of the
receiver in one go. A very laudable
endeavour - it simplifies operation in
that you don't have to switch amplifiers
each time you change to another
waveband — but it also entails some

Whether or not to install an aerial
amplifier is always a difficult choice.
It is often doubtful whether the
reception quality will improve suf-
ficiently to justify the expense. A poor
receiver in a fringe reception area is
virtually a lost cause — aerial amplifiers
don't work miracles! On the other hand,
if reception is already good enough
there is no point in investing in further
improvement.

Aerial amplifiers are intended for the
'normal' case between these extremes,
where reception is quite good normally,
but fades out in fringe areas. Signals
that are on the border-line of being
acceptable can be boosted satisfactorily.
All this is valid for both VHF-FM and
AM reception. But there is a difference.
VHF-FM transmitters have a fairly
small 'service area', and trying to
receive them in a moving vehicle with a
normal car aerial doesn't help matters.
Depending on the transmitter power,
height, amount of housing around
and the sensitivity of the receiver, the

Aerial amplifiers are often used
for domestic receivers — in some
cases, even, where they do more
harm than good. In cars they are
not nearly as common, surpri-
singly, since that is one place
where they can often be
extremely useful! The few aerial
amplifiers installed in cars are
usually the standard commercial
type: 'wide-band', they say, but
not much good for VHF-FM.

This 'car issue' seems the ideal
place to present our ideas on the
subject. With a proven circuit, of
course: an aerial amplifier that can
be switched for AM or FM
reception. Improved performance
on all wavelengths — that's what
an aerial amplifier should provide!

elektor april 1980 -4-25

active car aerial

heavy penalties. Reception on medium
and long waves may be improved at
the expense of poorer VHF-FM per-
formance. At the very least, it would
seem reasonable to add a switch to by-
pass the amplifier when listening on the
FM band. And if you're switching
anyway, why not do the job properly
and include a second amplifier for
VHF-FM?

When we decided to design an active
car aerial for this issue, we felt that
good performance was more important
than avoiding a switch. All effort went
into designing two good amplifiers —
one for long, medium and short waves
and the other for VHF-FM — and the
fact that this calls for one additional
switch beside the radio was considered
acceptable.

Block diagram

Do you really have to switch between
the two amplifiers? Surely it must be
possible to include filters that separate
the low frequencies (AM) from the high
(VHF-FM), feed each to its own ampli-
fier, and then mix them again before

Figure 1. Block diagram of the active car aerial. The amplifier consists of two sections: one for
AM (long, medium and short waves) and the other for VHF-FM.

2

Figure 2. The complete circuit. The two amplifiers are at the left (AM at the top and FM at the bottom); the right-hand section is the supply
smoothing and switching circuit.

4-26 — elektor april 1980

active car aerial

Figure 3. Two printed circuit boards are used. The smaller of these contains the switch and
supply smoothing; it is mounted beside the car radio, so that the switch is within easy reach.
The larger board, containing the amplifiers, is mounted in a screened box and as close to the
aerial as possible.

passing them to the radio? Agreed:
that is a possibility. But it is not the 3
best!

The reason for this is the effect of the
car aerial. For AM it is relatively short,
and so the signals picked up are quite
weak. To obtain good reception, the
input impedance of the amplifier
should be as high as possible. However,
you then run into difficulties with
parasitic capacitances at the input — the
more capacitance, the worse things get.
Connecting a complete FM aerial
amplifier to the same input is just asking
for trouble — even if you use the best
of filters! So we come back to the
switch.

A block diagram of the complete system
is given in figure 1. It may come as a
surprise to discover that the input is
switched by means of a relay, but
there's a good reason for this. An aerial
amplifier should always be mounted as
close to the aerial as possible. After all,
the idea is to boost weak signals, and
there is no point in making matters
worse by first passing them down a
cable that makes them even weaker.

That sort of thing doesn't improve
the final signal-to-noise ratio. However,
if the amplifier is mounted near the
aerial, what do you do with the input
switch? Obviously, you can't run long
cables to the dashboard — that's even
worse than mounting the amplifier
there in the first place. The switch
must be close to the aerial, therefore.

The only solution that combines dash-
board control with switching at the
aerial is to use a relay. It is arranged to
pull in when power is applied to the
VHF-FM aerial amplifier.

A double-pole two-way switch is used
to select either the AM or the FM
amplifier. One pole of this switch
selects the output from the correct
amplifier, and passes it to the aerial
input of the car radio. The other pole
switches the power supply to the
selected amplifier — and to the relay, if
FM is chosen. To avoid making the
wiring more complicated than neces-
sary, the same cable is used for both the
power to an amplifier and the output
signal from it. This means that two coax
cables run from the switch (mounted
beside the radio) to the amplifiers. At
each end of both cables, a coil is used
to isolate the signal from the supply —
its inductance is a high impedance for
the aerial signals, and virtually non-
existent for the DC supply. The coax
cables are shown as dotted lines in the
block diagram.

One last detail remains to be discussed
in the diagram: the zener diode between
the aerial and supply common. This is
obviously not intended for DC voltage
regulation . . . The reason for including
it is to protect the input stages of the
amplifiers from high voltages. These
could be picked up by the aerial as a
result of static charge or even lightning
flashes. Actually, as we will see, the

'zener' symbol is not quite accurate: at
this point, both positive and negative
spikes are limited to approximately
2.5 V.

The circuit

The complete circuit of the active car
aerial is given in figure 2. The input
protection circuit (the 'zener diode' in
the block diagram) consists of eight
diodes in all: two groups of four, each
clamping the input at approximately
+ or — 2.5 V. A 'normal' zener diode
wouldn't work at this point — at the
very least, you would need two in
series, back-to-back. Even then, the
capacitive load on the aerial would be
higher than with the four diodes shown.
The upper part of the circuit is the AM
amplifier. As mentioned earlier, a
high input impedance is required; for
this reason, the first stage is a FET
source follower (T2). This is followed
by a single-transistor amplifier stage
(T3); the gain is determined by the
value of R9. The total bandwidth of the
amplifier is such that a significant im-
provement in signal-to-noise ratio can be
expected at frequencies up to 18 MHz.
Since most receivers don't go above
about 6 MHz, even on the short-wave
band (6 MHz corresponds to 50 m), this

bandwidth is more than adequate.

The VHF-FM amplifier, shown in the
lower half of the circuit, is a straight-
forward design. The BFT66 used here
is a very low-noise transistor, ideally
suited for this type of application. For
FM signals, the length of a car aerial
is approximately equal to a quarter
wavelength. The input is tuned by a
resonant circuit, consisting of LI and
C6. The latter is adjusted by tuning in
to a weak transmitter in the centre
of the band, and trimming C6 for
minimum noise in the received signal.

A resistor, RIO, is shown in series with
the relay. This can be replaced by a wire
link if a 12 V relay is used; however,
if a relay with a lower coil rating is
easier to obtain, a suitable value for RIO
can be selected so that the relay can
run off the 12 V supply.

L8, L9, C9 and CIO are included to
smooth the supply to the circuit. In
particular, the nasty spikes caused by
the ignition system!

Construction

A printed circuit board and component
layout for the active car aerial are given
in figure 3. To be more precise: two
boards. The first is mounted as close as
possible to the aerial: it contains the

active car aerial

elektor april 1980 — 4-27

Parts list

Resistors:

R1,R7 = 10k
R2,R6 = 1 k
R3 = 22 k
R4 = 100 k
R5 = 220 k
R8 = 39 k
R9 = 270 O.

RIO = see text
R 1 1 = 56 n

Capacitors:

C1,C3,C5,C7 = 1 n
C2,C4 = 100 n
C6 = 4 . . . 27 p
C8 = 4n7

C9,C10 = 1 m (tantalum)

Semiconductors:

D1 . . . D8 = 1N4148
T 1 = BFT66
T2 = BF256C

Inductors:

LI = 3.5 turns on VHF coil
former, Toko type SI 8.
L2,L3,L4,L8,L9 = 1 mH
L5,L6 = 10 mH
L7= 1 pH

Sundries:

Re = 12V relay, changeover
contact (e.g. Siemens V23027-
A0002-A101 )

SI = two-pole two-way switch

two amplifiers and the relay. The smaller
board is intended for mounting the
switch and the associated coils and
capacitors; it can be placed near the
radio. The two boards are connected
by means of two coax cables, as
mentioned earlier.

For the VHF-FM signal connection,
normal 'aerial coax' (50 . . . 75 J2) can
be used. However, it is not a good idea
to use this type of low-impedance
cable for the AM connection. Its capaci-
tance would appear in parallel with the
input circuit of the receiver, and it may
well be so high that it cannot be com-
pensated for by the trimmer on the

radio. (Some kind of 'aerial trimmer' is
normally included in car radios.) One
solution is to use the existing coax feed
from the car aerial for the AM signal
connection. However, this is not always
as easy as it may appear: the core is
often very thin, so that the slightest
pull is sufficient to break it. Further-
more, some of these cables have built-in
coupling capacitors. If they can't be
'got at', it is not possible to use the
same cable for the power supply; in
that case, a separate supply connection
will be required between the R8/C2
junction and the positive supply (across
CIO).

The power supply is connected to the
smaller of the two boards. Since this
will be mounted beside the car radio, it
is logical to derive the positive supply
from there. This has the advantage that
radio and aerial amplifier both run off
the same fuse; furthermore, they may
both be switched off together by the
ignition switch, if that's the way the
radio was originally wired.

The larger of the two boards — with the
amplifiers — should be mounted in a
screened box. We always use 'raw'
printed circuit boards, since the copper
makes for good high-frequency
screening. M

4-28 — elektor april 1980

stop thief!

stop

thief!

Who wants to steal
a car with
engine trouble?

Protecting one's property is a
popular hobby — particularly when
that property is attractive to
others. There are all sorts of
systems for protecting cars, but the
one described here is unusual: it is
deception, rather than protection.
It doesn't make it impossible to
steal the car (for that matter, no
system does), but it makes it very
unattractive: who wants to steal a
car with an engine that stalls every
few yards?

B.H.J. Bennink

Even the most effective of theft preven-
tion systems normally suffer from the
drawback that it is immediately apparent
to a thief that some kind of protective
device is built in. If he . is sufficiently
courageous, persistent and experienced,
he can put the device out of action and
make off with the car. If he's a pro-
fessional thief, it's 'Goodbye, car!'; if
he's joy-riding, you may just be lucky . . .
but normally the vehicle ends up
severely crumpled or burned-out. An
alarm system that sounds the car horn
even seems to be attractive to certain
types of 'joy-rider'. Amazing, but true.
All in all, no theft prevention system
can be guaranteed to keep thieves out
of your car. Once in, the chances that
they can get it moving are quite high.
The system described here should give

— no spark — no go. One dead engine!
It can be started again — no problem —
and it will run as smoothly as before.
Until the relay pulls in again, that is.

The circuit is given in figure 1. It is put
into operation by operating switch SI.
A 'secret' switch, of course. Bear in
mind that the best place to hide a
switch is very often the most obvious:
right in the middle of the dashboard,
say. As long as you don't lable it 'theft
protection'. Anyway, to get to the
circuit: the timer 1C (a 555) is used as a
multivibrator. As soon as power is
applied, via the ignition switch and SI,
it starts to produce a squarewave out-
put at about 0.2 Hz. A period time of
5 seconds, in other words. After bridg-
ing the ignition switch (that's how they
do it), the thief can start the engine

Figure 1. Only six components — cheap ones, at that! — are needed for a quite effective
anti-theft device.

adequate protection against joy-riders.
It should even discourage most 'pro-
fessional' thieves — all except the type
who is prepared to steal a furniture
removal van first and then run your car
into it! At the same time, this system
has the advantage that it does the job on
its own, without giving silent or raucous
alarm signals to the owner or other
passers-by. There's no need to chase
after your property in the middle of the
night, wearing only pyjamas and a
dressing gown: you can rest assured that
it will never end up far away. It is highly
unlikely that the would-be thief will
drive more than a few hundred yards.
What is the basic principle behind such
an effective system? Simple! The engine
is about as reliable as that of a twenty-
year old car with water in the petrol
tank!

In practice

The wire from the positive side of the
ignition coil to the 12 V supply from
the ignition switch is cut, and passed
through the break contact of a relay. As
long as the relay is not energised, power
is supplied to the ignition coil and the
engine runs smoothly. However, when
the relay pulls in it breaks the con-
nection. No power to the ignition coil

without any problems. However, after
five seconds the relay pulls in. The
ignition coil is cut off, and the engine
stalls. After a few seconds of frustated
fiddling, the engine will fire again (the
relay has dropped out!), but the feeling
of achievement is doomed. Five seconds
later, the engine will again stall. To
sum it up: the engine will run, so
apparently there is no theft prevention
circuit in the car, but it conks out at
short notice. Very frustrating for any
thief. His best bet is to leave the car
where it is and try someone else's.
Always hoping that owner doesn't read
Elektor as well.

The circuit can be modified, according
to personal taste. The period time of the
555 (corresponding to the time that the
engine will run) is determined by R1
and Cl. Too short is suspicious, too
long corresponds to a longer walk to
your car next morning.

A printed circuit board and component
layout for this unique anti-theft device
are given in figure 2. There is room on
the board for a miniature relay; a larger
type can obviously be mounted 'off-
board'. It is a good idea to look for a
relatively 'silent' type, or else mount it
in a sound-proof box. A loud Click as
the engine cuts out would give the game
away! M

stop thief

elektor april 1980 — 4-29

Parts list:

Resistors:

R1 = 82 k

Capacitors:

Cl = 47 p/16 V

Semiconductors:

IC1 = NE555 or equ.

D1 = 1N4148

Miscellaneous:

SI = single-pole changeover switch
Re = 12 V/100 mA relay with
heavy-duty break contact.

L® ®l

Figure 2. Printed circuit board and com-
ponent layout. There is room for a miniature
relay, but a larger type can be mounted off-
board. 'Sound-proofing' is important!

Electronic ignition systems fall into two keep the spark 'on'. Since a relatively low

categories: inductive and capacitive. The voltage is required to maintain the spark, once

inductivetypeisrelativelysimple.However.it the gas mixture between the spark plug's

cannot produce a spark voltage as high as that electrodes has been ionised, the DC/DC

of a capacitive system unless certain steps are converter can be controlled to reduce the

taken, such as in the case of the transistor voltage during the burst,

ignition system described elsewhere in this The principle of operation is illustrated in the

issue. An inductive ignition system has the block diagram of figure 1 . The signal produced

advantage that the spark can easily achieve a by the (mechanical or electronic) contact

burning period of one to two milliseconds. breaker is 'shaped' and 'debounced' by a

This must be reached, as otherwise com- Schmitt trigger. The output of the Schmitt

bustion would only be partial and this would trigger is fed to a monostable multivibrator

reduce engine power and at the same time which produces a pulse with a duration of

pollute both the engine and the environment. 1 to 2 milliseconds. This pulse controls the

A capacitive ignition system can produce a output amplitude of the DC/DC converter

highly powerful spark, which, unfortunately, and also triggers a gated astable multivibrator

does not usually last long enough (unless the (AMV). The AMV is an oscillator which starts

steps mentioned above are taken). up the moment the level of the control input

A 'proper' thyristor system is comprised of a changes and it continues to oscillate until the

transistor circuit, usually quite simple, and a input level returns to normal. The end result

high voltage section with a thyristor to of all this is that for 1 to 2 ms the gate of the

control the coil voltage. A capacitor having a thyristor is fed with a series of 'ignition'

voltage of approximately 400 V across it is pulses thereby maintaining the flow of

discharged through the primary of the indue- current through the coil and so prolonging the
tion coil by the thyristor. The simple transis- spark.

tor circuit then maintains the spark. An ignition system based on this principle

Alternatively, a burst of pulses can be applied would reap the benefits of both inductive and
to the thyristor. The interval between two capacitive types without having any of their

successive pulse groups can be calculated to drawbacks. H

4-30 — elektor april 1980

battery voltage indicator

It can be rather a nuisance when a
car battery goes 'dead'. This can
happen if the lights are left on
after leaving the car, but it can
also be the result of a fault in the
electrical system — where the
battery is not charged properly.
An 'early warning' system that
gives an continuous indication of
the battery condition is a useful
accessory. As described here, only
a few components are needed to
obtain an 'optical' indication: a
single lamp that changes colour
from green to red as the battery
goes into the danger area.

The circuit in figure 1 uses a two-colour
LED as an indicator. Below a certain
threshold (12 V, for instance) the LED
lights up red to indicate that the battery
is low. When the battery is fully charged,
the LED will be green. The colour
changes gradually from red to green as
the voltage increases.

The heart of the circuit is an LM 10 (C)
(National). This 1C contains a reference
voltage source of 200 mV connected
internally to the non-inverting input of
A1. The circuit round A1 isa differential
amplifier; the gain depends on the
value of resistor R3. The upper threshold

of the battery voltage is set by PI — at
14 V, for instance (voltage across C2).
At voltages at or above this threshold,
the output of A1 is 0 V. As the battery
voltage drops, the output of A1 will
increase until it reaches the positive
supply voltage. The value selected for
resistor R3 determines the range of
battery voltages over which the output
of A1 switches from 0 V to completely
positive. With the value of R3 given
here, it will be equal to the supply
voltage when the battery voltage has
dropped by approximately 2 volts (in
other words at 12 V). With the voltage

indicator

in the green or in the red?

REFERENCE

BALANCE OUTPUT FEEDBACK

LM 10(C)

REFERENCE

FEEDBACK

REFERENCE

OUTPUT

Figure 1. The car battery voltage indicator circuit. The opamps A1 and A2 and the 200 mV
reference voltage source are contained in one 1C: the LM 10(C).

battery voltage indicator

elektor april 1980 — 4-31

free protection!

between 12 and 14 V, the output of
A1 will assume corresponding values.
A1 controls an oscillator circuit around
A2. As Al's output alters, the duty-
cycle of the squarewave output from
the oscillator circuit will also change.
This causes the two-colour LED (D2) to
change colour. At both thresholds (12
and 14 V) the output voltage produced
by A1 will be at a fixed level (12 or
0 V, respectively). In that case, the
oscillator will stop and the LED will
light up red or green according to the
battery condition.

In figure 2, the battery voltage is plotted

as a function of the charge condition.
When the engine is running, the battery
will be recharged according to the
upper curve; the LED will light up
green, when it is 70% charged. What will
happen during discharge is also evident
from the graph.

The printed circuit (figure 3) for the car
battery voltage indicator can be fitted
quite simply behind the dashboard. It is
also possible to use two separate LEDs
instead of one two-colour type — con-
siderably reducing the construction cost.

H

free

protection

r

2

Parts list

12 V44 Ah 210 A battery
Temperature: +27° C (80° F)

voltage during charge
(charging current:

4.4 A)

voltage during discharge I 5
(discharge current: 2.2 A) 5

0 20 40 60 80 100 %

fully charge fully

discharged condition charged

Resistors:

R1 = 1 M
R2= 10k
R3= 3M9
R4,R5,R6 = 100 k
R7= 100 n
R8,R9= 390 n

ri 0 = ion

PI = 10 k preset

Capacitors:

Cl = lOn
C2= 100 m/25 V

Semiconductors:

A1 ,A2 = LM 10(0

Figure 2. The battery voltage as a function D1 = 1N4001

of the charge condition, during charge (upper D2 = two-colour LED (or two

plot) and discharge (lower plot). separate LEDs)

Figure 3. The printed circuit board and component layout.

Elsewhere in this issue, theft protection
circuits are included. They're good. But
no protection can possibly be complete.
Fortunately, most would-be car thieves
are amateurs. Even the simplest pro-
tection scares them off! For instance:
a clearly visible warning: This car is
electronically protected against theft'.
Whether or not some device is actually
built in, a warning like this is liable to
give them second thoughts.

With this in mind, Elektor readers can
take advantage of our FREE protection
sticker. Mount it in a clearly visible
position, and the chances of your car
being stolen are reduced to less than
one-tenth!

Subscribers should find the sticker in
this issue — provided it didn't drop out
in the post . . . Those who buy their
copies at a news stand may have also
received a sticker — we certainly supplied
plenty of them!

If you didn't get yours, you can send up
for one. While stocks last, we are
supplying them free of charge. Simply
send a stamped, addressed envelope to
Elektor, Canterbury, with request 'Anti-
theft sticker'.

Please, readers, don't ask for more than
you need! Stocks are limited, and we
operate on a first-come-first-served
basis. Remember, other readers also
have cars!

4-32 — elektor april 1980

aide-de-camp

'2N3055 1

INVERTING INPUT i

COMP

80118 1

It is probably true to say that
every touring caravan uses a 12
volt power supply for something,
even if only for lighting. This is
often taken from the towing
vehicle's battery via the van's

Most caravanners are aware that they
can connect a mains power supply to
their van and many assume that equip-
ping it with 240 volt light fittings,
power points etc. is a difficult and
costly business. This is of course true
and, fortunately, unnecessary. For a
very reasonable cost, about £5, it is
possible to build a mains to 12 volt
converter with a current capability of
2 Amps, sufficient to cater for all the
lighting needs of most touring caravans.

Shower power

For normal lighting purposes a 2 Amp
power supply will be adequate. However,
more and more electrical appliances
are becoming available for use in
caravans. High on the popularity list is
the electric water pump and this may
require a power supply capable of
delivering up to 5 Amps (especially if
a shower is fitted). Figure 4 shows the
circuit for a very simple design using a

aide-de-camp

mains power supply for caravans

lighting connector. This can be
less than ideal if extra electrical
facilities are fitted into the
caravan and starting the vehicle in
the mornings could be a problem.
With this in mind, many caravan
holiday sites are fitting mains
power point facilities. This article
describes a converter which allows
the caravan's existing 12 volt
system to be supplied from the
mains connection point.

2 Amps to see with

As can be seen in figure 1, the circuit
is very simple. The 1C used is the 723
precision voltage regulator with current
limiting set (by R3) to 2 Amps. The
preset PI should be adjusted to give
12 volts at the output. It is important
that the power transistor (T1) is
mounted on an adequate heatsink of the
type shown in photo 1.

It would be convenient to use a 'potted'
bridge rectifier with a hole through the
centre allowing for a more secure
fitting. The completed converter can be
assembled in any suitable metal case
(not forgetting the fuse) and this can be
placed near the mains input connection
cable. It may be more convenient to
mount the switches remotely on a
separate panel.

It should be remembered that a caravan
on the move is subject to severe
vibration and therefore considerable
efforts must be made to ensure that
components are securely fitted, particu-
larly the transformer, where locking
nuts would be desirable.

high power regulator, I C 1 . This device,
in common with the other 78 series
regulators, features thermal overload
and short circuit protection. A sub-
stantial heatsink will be required for
this regulator together with a potted
7.5 — 10 Amp bridge rectifier B1. Good
quality components should be used
throughout. All components should be
secured as mentioned previously.

Regulations

It should be noted at this point that the
regulations regarding the connection
of mains to caravans are fairly extensive,
not to say confusing. There are of
course two sets, the I.E.C. regulations
and the U.K. regulations. The reader
is advised to obtain a copy of these
from his local caravan dealer.

One thing should be made absolutely
clear, the incidence of fire in caravan
accidents is very high with a large
percentage citing electrical faults as a
reason. Be warned - if you are not sure
of what you are doing, ask a pro-
fessional.

CURRENT LIMIT ( ]2

CURRENT SENSED 3
INVERTING INPUTS 4
NON — INVERTING INPUTC 5
VrefT 6

CURRENT

LIMIT

FREQUENCY

CURRENT SENSE rj ’0 COMPENSATION

Trl = 15 V/2.5 A

Figure 1. Complete circuit diagram of the 12 volt converter. All the components inside the dotted line are mounted on the printed circuit board

aide-de-camp

elektor april 1980 — 4-33

Q0BBB18 Q

Asi

. EPS7Q43h

Figure 2. Printed circuit board for the 12 volt/2 amp converter (EPS number 7043b).
Figure 3. Component overlay for the printed circuit board.

a means of overcoming these problems.
As can be seen, the circuit is connected
between the neutral and earth leads at
some convenient point inside the
caravan. Ingenious constructors may
find a method of fitting it inside the
circuit breaker housing: however, that
is a matter of personal preference.

If the mains inlet cable is connected
correctly the LED (D1) will remain
unlit. If, on the other hand, the cable
is connected wrongly, the neutral lead
will become 'live' and the LED will
glow, thereby indicating an 'unsafe'
condition. The current through the LED
is limited by the reactance of capacitor
Cl . The working voltage of this capacitor
should be at least twice that of the
mains voltage. Diode D2 is included to
protect the LED from excessive reverse
voltages. If polarised connectors are
used, but the LED indicates a fault,
something will have to be rewired. This
can be dealt with however, by incor-
porating the changeover switch S2 to
cross over the live and neutral input
leads to the caravan, as illustrated in
figure 5. Regardless of whether the
mains inlet cable is wrongly connected
at some point, the LED will always
indicate whether or not the changeover
switch is in the correct position. If the

Parts list for figure 1.

Resistors:

R1 = 2k2
R2 = 3k3
R3 = 0.33 n
PI = 1 k preset

Capacitors:

Cl = 220 m/25 V
C2 = 1 n
C3 = 4 m 7/25 V

Semiconductors:

ICIg 723
T1 = 2N3055

B1 = 5 Amp 500 V potted bridge
rectifier

Miscellaneous:

Trl = 15 V/2.5 A transformer
F 1 = 250 mA fuse

51 = double pole two-way switch

52 = single pole two-way switch

Parts list for figure 4

Capacitors:

Cl = 220 m/25 V
C2 = 10 n
C3 = 10 m/25 V

Semiconductors:

IC1 = 781-112
B1 = 7.5-10 Amp 500 V
potted bridge rectifier

Miscellaneous:

Trl = 15 V/6A transformer
Frl = 500 mA fuse

51 = double pole two-way switch

52 = single pole two-way switch

A safety checker

Care must always be taken when con-
necting mains to anything — especially
caravans. It is essential that the mains be
connected correctly, LIVE to LIVE,
NEUTRAL to NEUTRAL, EARTH to
EARTH - to avoid ASHES to ASHES
and DUST to DUST! If polarised plugs

and sockets are used there would appear
to be no problem. However, it may arise
that the actual site connection point is
wrongly wired. Where reversible mains
inlet connectors are used (such as the
Continental types) a 'safe' connection
cannot always be ensured.

The simple circuit shown in figure 5 is

LED is on, switch over S2. Since mains
connections are uncertain it is essential
that the isolating switch SI is a double
pole type. M

12 V from car
battery

^ S2 J* O

78H12

3

C2

i

0 “

\220\i

lOn (

25V

lA

12 V 5 A max
to lights etc.

1 SI = Mains isolating switch
I S2 = Change over switch

Figure 4. Circuit diagram of a 5 A converter.

Figure 5. Simple circuit to ensure correct
connection of the mains inlet cable
— especially useful on the Continent.

4-34 — elektor april 1980

fuel consumption meter

fud

consumption

meter

'Enjoy the ride? Forget it, nowadays.
Agreed, you may enjoy the ride; but if
you have to pay for the petrol, after-
wards . . . you start to calculate in miles
per pound.

It's not at all clear how scarce oil really
is. Governments, oil companies and
'independent investigators' all seem to
have different ideas. However, two
things are obvious: the price of petrol is
not likely to go down — on the contrary,
'a pound a gallon' may soon seem
ridiculously cheap. Furthermore,
economic driving is becoming ever more
important, whether to save pennies or
to save oil. By making the car itself
more efficient and economical (with a

how to get more miles per gallon!

good electronic ignition system, for
instance), by improving your driving
style or simply by driving less. Although
we do not wish to underrate the pro-
ducts of modern technology in this field,
we feel that driving style is perhaps one
of the most important factors. This is
where a fuel consumption meter comes
in. Without this aid the difference
between third and fourth gear, the
effect of coasting or riding the brakes
and the fuel required to pick up speed
sedately or more rapidly remains
purely guesswork. If you want to cut
the petrol bill, there's nothing like a
meter that drops from 35 mpg down to
10 mpg as you tread on the accelerator!

Two transducers.

By now, a large number of regular
Elektor readers will be reaching for
their pens to remind us that we have
published a fuel consumption meter

: igure 1. The two main transducers for the fuel consumption meter. The speed transducer is
nounted in the drive cable to the speedometer; the flow sensor in the petrol pipe between the
jump and the carburettor.

Cold start, choke 30 mph, 70 mph, zoo-o-o-om or purrrr — how much
fuel does it take? Does your car burn fuel at a terrific rate? Do you
waste fuel by poor driving? For most drivers (like us) it's all guesswork.
Car manufacturers prefer to specify the fuel consumption of their
models at a constant 70 mph - or at a lower speed. Under those
circumstances, virtually any car is 'economical'. As car owners know:
unrealistically so . . . The only way to know what's really happening
as you drive is to use an instant fuel consumption meter. A unit, in
other words, that gives instant indication of your present fuel
consumption in miles per gallon. An instrument like the one described

here!

before. True enough. We haven't forgot-
ten. That was in September 1976.
Nearly four years ago. Not that that, in
itself, is sufficient reason: there are
several points in favour of a new design.
First off, there is still considerable
interest in the earlier design as several
letters and telephone calls show.
Furthermore, at the time of the original
publication a fuel consumption meter
was something of a novelty. By now,
the increasing cost of petrol has made
it more important than rev counters or
speedometers - to name two examples.
Another reason is that, in our opinion,
the original design has lost some of its
'charm'. The pointer instrument is
rather old-fashioned in this 'digital' era.

fuel consumption meter

elektor april 1980 — 4-35

and it wasn't a completely home-grown
Elektor design.

Two sensors are needed for measuring
'miles per gallon': a flow sensor that
measures the flow of petrol to the
engine (in 'gallons per hour', say);
and some kind of speed sensor that
gives an output which corresponds to
'miles per hour'. In the 'old' design we
used a pick-up coil, glued to the back of
the existing speedometer. Even though
it does work it is not as elegant as a
'speed transducer' designed for the job.
An American manufacturer, Space-
Kom*, supplies both flow sensors and
speed transducers (see figure 1).

The 'active' components in the flow
sensor are a lamp, a phototransistor and
a small 'paddle wheel'. As petrol flows
through the unit the wheel starts to
rotate between the lamp and photo-
transistor. The latter converts the
flashing light into a series of electrical
pulses, with a frequency that is directly
proportional to the rate of petrol flow.
The complete unit is mounted in the
petrol feed between the fuel pump and
the carburettor, so that it gives almost
instant indication of the petrol used at
any given moment. It is factory-cali-
brated at 13.3 gallons per hour (16.0
American gallons per hour); at that
flow rate it produces an output of
14.22 Hz. Not that 13.3 gallons per
hour is realistic for many European cars
— even half that is really burning it
up — and in practice output frequencies
between approximately 0.5 Hz and
5 Hz may occur.

The speed transducer (or 'tachogene-
rator') is similar to a dynamo. It is
mounted in the cable to the existing
speedometer, and produces a DC
output voltage that is directly pro-
portional to the speed of the car, 5.6 V
corresponding to 70 mph. Obviously,
since all cars are different, there are
several types of speed transducer and
an even more extensive range of con-
nectors and adapters (see Table 1). In
some cases the unit is mounted at the
back of the speedometer; in others, it is
fitted on the gearbox.

Miles per gallon

Given a good transducer that measures
the amount of fuel that is flowing into
the engine, it is no problem to build a
gallons-per-hour meter. The pulses from
the flow sensor are simply fed to a
counter, and the result of each count is
displayed on some kind of digital or
analogue meter. By adjusting the timing
for the counter (the 'gate frequency'),
any transducer can be calibrated to
display gallons (or litres) per hour. As
we said, no problem.

This system may be good enough for
motor boats, but in cars we are less
interested in the number of gallons

*lmported into Europe by Medel B. V.,
Roden, The Netherlands.

f2 miles per hour

display : — = miles per gallon

fl gallons per hour

Figure 2. The basic principle of the fuel consumption meter. The 'speed' information (miles per
hour) is divided, in the counter, by the fuel flow information (gallons per hour). The result
appears in the display: miles per gallon.

3

Figure 3. By extending the principle, the number of ranges can be increased: not only miles per
gallon, but also gallons per hour, litres per 100 kilometers and revs per minute.

that we are using per hour. The true
clue to economic driving is an indication
in miles per gallon. The basic principle
of this type of measurement is given in
figure 2. The thing to realise is that a
counter, in effect, divides the 'clock'
frequency by the 'gate' frequency - it
counts the number of clock pulses that
occur within one period of the gating
signal. By applying a clock signal
that corresponds to the speed in miles
per hour (f 2 ) and a gate signal from the
flow sensor (gallons per hour, f i ) , a
final display is obtained in miles-per-
hour divided by gallons-per-hour: miles-
per-gallon, in other words. Perfect!

With the transducers used here, the
gallons-per-hour information (f i ) can be
taken direct from the flow sensor. The
same is not true, unfortunately, for the
miles-per-hour information (f 2 ): the
speed transducer produces a DC voltage
- not a frequency — that varies with
speed. However, if this voltage is used to

control a VCO (Voltage Controlled
Oscillator) the desired signal is obtained:
a frequency that is directly proportional
to speed. This signal is fed to the clock
input of the counter.

So much for miles-per-gallon only -
we'll come to the circuit later.

Extended block diagram

Sometimes it's a pity to do things the
easy way, when with a bit more effort
you can make them more compli-
cated ... A block diagram of an ex-
tended version of the fuel consumption
meter is given in figure 3. The basic
principles are the same as those discussed
above; however, to make the unit more
flexible some additional 'blocks' have
been added.

The complete unit can not only measure
miles-per-gallon (or kilometers-per-litre,
if required). It can also be switched to

4-36 — elektor april 1980

fuel consumption meter

read gallons-per-hour , or even litres-per-
100 kilometers — the 'upside down'
German way of looking at things. As
if all this wasn't enough, the fact that
a counter and display are already
available tempted us to add a rev
counter position: engine revolutions-
per-minute. Since all the main com-
ponents are already available, only a
minor extension is required for this: in
particular, a lead from the contact
breaker points and a simple 'pulse
shaper' circuit (shown in the block
diagram as 'transducer 3).

Three seven-segment displays are used,
with the decimal point between the
second and third digit.

As can be seen, the complete block
diagram given in figure 3 is rather more
extensive than the basic version (figure
2). There are quite a few additional
sections . . . First off, the 'switch
block'. This is simply a multi-way
switch, that selects two of the available
frequencies (f] . ,.f 7 ) for use as
clock and gate ('latch/reset') signals
for the counter.

The transducers have already been
discussed: '1' is the flow sensor, '2' is
the speed transducer and '3' is for the
rev counter function. The voltage-to-
frequency (u/f) converter between
transducer 2 and the switch block is
actually the VCO described in figure 2.
In this extended version, two output
lines (f 2 and f 3 ) are shown from this
section. With good reason: the infor-
mation from transducer 2 is required for
both miles-per-gallon and I itres-per- 1 00
kilometers, in the former case it is
used as clock-signal, in the latter as
latch/reset. Different ranges of output
frequency are required for these two
applications, so the u/f converter is
switched (by means of SI) between two
frequencies, f 2 and f 3 .

The same applies for the oscillator: as
required, it provides one of two fre-
quencies: f 6 or f 7 . This oscillator was
not included in figure 2: it is added for
measuring gallons-per-hour and revo-
lutions-per-minute. On these two ranges,
only one transducer is used: either the
flow sensor or the contact breaker with
its pulse shaper. The signal from the
transducer is used as clock input for the
counter; a fixed frequency from the
oscillator serves as latch/reset signal. By
adjusting the oscillator frequency as
required for each application, the
display can be calibrated to read gallons-
per-hour or revs-per-minute.

Two further sections are added: f/u
and u/f between transducer 1 and the
switch block. In figure 2, the signal
from this transducer went straight to
the counter. In one position of the
switch in figure 3, it still does: signal f! .

However, when measuring gallons-per-
hour or litres-per-100 kilometers the
signal from transducer 1 must be used
as clock signal, instead of as latch/reset.
For this application, the frequency f]
(0.5 ... 1 0 Hz) is too low — each count
cycle would have to take anything from

4

revs

per

minute

osc.

clock

revs

per

minute

counter

8

8

8

latch

reset

SI in position 4

80096

Figure 4. These four block diagrams show the measuring principle in the four positions of the
selector switch.

10 seconds up to almost a minute,
depending on the desired scale. This
problem can be avoided by converting
the varying frequency fj into a varying
DC voltage, and then using this voltage
to control a VCO (u/f). The output
from this VCO (f 4 ) can be set to any
desired range by selecting the necessary
conversion factors for the f/u and u/f
sections. In effect, the original fre-
quency f 3 is multiplied to produce a
suitable clock signal (f 4 ) for the
counter.

So much for the various sections in the
block diagram. Now it's time to see
how they are used in practice.

Four in one

It will be clear by now that several

of the sections in the block diagram
are only required for one or two of the
possible measuring ranges. This, by
the way, provides a clue to the possi-
bilities of cutting the component cost
when constructing the unit. If certain
ranges are not required, the correspond-
ing components can be omitted! As an
extreme (but eminently practical)
example, for measuring miles-per-gallon
only one u/f converter and the counter
are required. (More on this later.)

This is illustrated in figure 4. The four
block diagrams give the measuring
circuits in the four positions of SI.
The switch unit itself is not shown: the
various signals are taken direct to the
counter. That helps to make things a
lot clearer.

Let's start with figure 4c. This is the
situation when the switch is set to

fuel consumption meter

elektor april 1980 — 4-37

5

Figure 5. The complete circuit for the extended version, capable of measuring on all four ranges. As explained, it can be simplified if only 'miles
per gallon' is required; see figure 7!

position 3, for miles-per-gallon. Com-
paring this diagram with the one given
in figure 2, we discover that it is basi-
cally identical. The DC voltage that
corresponds to 'speed' is fed to a u/f
converter; the output signal (f 3 ) goes to
the clock input of the counter. The gate
(latch/reset) signal is the output from
the flow sensor (f x ) . In this third
position of the switch, therefore, clock
frequency divided by latch/reset fre-
quency corresponds to miles-per-hour
divided by gallons-per-hour. The result,
miles-per-gallon, appears on the display.
Now for figure 4b: position 2 of the
main selector switch. In this case, the
output from the flow sensor must be
used as the clock signal for the counter.
As explained earlier, this involves
multiplying the output frequency from
the sensor by means of an f/u and a

80096 6

Figure 6. The Exar integrated circuit type
XR 4151 that is used for the voltage-to-
frequency and frequency-to-voltage
conversions.

u/f converter, connected in series. The
result, f 4 , goes to the counter. The
latch/reset signal for this measurement
is derived from a fixed-frequency
oscillator. This frequency (f 6 ) is ad-
justed to give a display that is calibrated
in gallons-per-hour.

Figure 4a — position 1 of the selector
switch — is added for the Germans . . .
If they want litres-per-100 kilometers,
they can have them! As in the previous
case, the output from the flow sensor
is frequency multiplied to provide the
clock signal for the counter (f 4 ). The
output from the speed transducer goes
through its voltage-to-frequency con-
verter to provide the latch/reset signal
for the counter. In this case, the u/f
converter is set to give a much lower
range of output frequencies (f 2 ) than
in figure 4c.

Finally, position 4 of the selector
switch: the 'rev counter'. The cor-
responding block diagram is given in
figure 4d. The pulses from the contact
breaker are fed to the clock input of
the counter. We don't intend to count
'sparks-per-second', of course: we are
interested in revolutions per minute.
The two are related, obviously: for a
four cylinder four-stroke engine, there
are two pulses from the contact breaker
for every revolution of the engine. The
display is calibrated in rpm by ad-
justing the output frequency from the
oscillator (f 7 ) accordingly. As is
common with this type of indicator,
the display is actually calibrated in
'rpm x 100': when it reads '42.5', say,
this corresponds to 4250 revolutions
per minute.

The circuit

Figure 5 is the complete circuit of the

most extensive version of the fuel
consumption meter/rev counter. Quite
a jungle, at first sight! The best ap-
proach seems to be to locate the
various sections of the main block
diagram — filling in the blocks one
by one, so to speak. First the easy ones.
A good place to start is at the end:

Counter and display

A single integrated circuit (IC1, type
74C928) does nearly all the work in
this section. Not only does it contain
a complete counter: it also has all the
necessary display drive circuitry on
board — memory, BCD-to-seven-segment
decoder/driver and display multiplexer.
Three BC141s select the three display
digits. As is common practice nowadays,
so-called multiplex technique is used:
one digit is selected at a time (by T1,
T2 or T3) and, simultaneously, the cor-
responding drive to its seven segments is
applied via resistors R34 . . . R40. The
three displays light up in sequence, so

rapidly that it appears to the eye as
if they are all on all the time.

The decimal point after the centre digit
is permanently on, via R41.

Latch/reset signals

To obtain a correct count and display,
the counter must first be reset. Then it
is allowed to count for a specified
period, after which the final count must
be transferred into a memory by means
of a 'latch' pulse. This is immediately
followed by a further reset pulse, in
readiness for the next count cycle.
Another way of looking at this is to say
that each count is preceded (and
followed) by a latch pulse and a reset
pulse in quick succession. In the block
diagrams, 'latch/reset' was considered as
a single signal; it will now be apparent
that this is not strictly true. The
'latch/reset' signal must be split into
separate latch and reset signals that
occur in quick succession.

fuel consumption meter

elektor april 1980 — 4-39

This is done by gates N4 . . . N9. The
combined latch/reset signal selected by
SI is fed to N7 and N8. The latter is
followed by an RC network (C15/R24)
and a diode to clamp the negative-
going spikes. Two inverters, N4 and N5,
clean up the resultant pulse for use as
the 'latch' signal. A similar circuit is
used to produce the reset pulse. As
before, the Schmitt-trigger gate (N7)
is followed by an RC network; the
short, negative-going pulses are 'cleaned
up' by N9 and fed to the 'reset' input of
the counter.

The oscillator

There is no future in going to great
lengths to design an oscillator circuit:
you may as well simply use an inte-
grated circuit designed for the job. An
obvious choice is one of the well-
known 'recipes' using a 555 timer — IC5
in the circuit. Since the oscillator must
be able to provide two different fixed
frequencies (f 6 and f 7 , see figures 3 and
4), pin 7 of the 1C is brought out to
Sic. This switch is used to select one of
two different resistor chains: R25,
P2 and P3 for f 6 , and R26, P4 and P5
for f 7 . The adjustment of these presets
will be discussed later, under 'cali-
bration procedure'.

The transducers

Transducer 1, the flow sensor, is at the
upper left in the circuit. The 'gallons-
per-hour' pulses are taken from the
collector of the phototransistor, pol-
ished up by N1 . . . N3, and fed to
position 3 of Sla. That takes care of
the first signal, fj .

Rather more effort is required to derive
the second 'fuel flow' signal, f 4 , from
the output of N3. Via Cl, the pulses
arrive at the output of a frequency-to-
voltage converter, IC2; the output of
this 1C (pin 1 ) goes to a voltage-to-
frequency converter, IC3. Its output,
at pin 3, is the desired frequency-
multiplied signal (f 4 ); it goes to pos-
itions, 1 and 2 of Sib. Obviously, this
description of the circuit around IC2
and IC3 is very brief — but we will give
a much more extensive explanation
further on.

The speed transducer ('2' in the block
diagrams) is shown at the lower left in
figure 5. As explained, this is a DC
generator that produces a voltage that is
proportional to speed. This voltage is
passed — via an RC network for inter-
ference suppression — to pin 7 of IC4.
The 1C takes care of the voltage-to-
frequency conversion (u/f). Its output,
from pin 3, can be used in two different
ways. In position 3 of SI the signal is
passed directly through Sib to the clock
input of the counter (f 3 ); in position 1
short negative pulses, derived from the
original signal by means of an RC
network and N6, are fed through Sla
to the latch/reset circuit. This signal,
f 2 , must be at a much lower frequency
than f 3 . For this reason, the timing
network of the u/f converter is switched
by means of Sid. In position 1, C18and

R28 are connected to pin 5 of IC4; in
position 3, Cl 9, R27 and P6 are used.
The third 'transducer', for the rev
counter, is actually nothing more than
a pulse shaper connected to the contact
breaker. It is shown at the lower right
in the circuit. As can be seen, the signal
from the contact breaker drives a
transistor, T4; the latter is followed by
a monostable multivibrator (N10, Nil)
that provides a short pulse atTTL logic
level each time the points open. This
output signal, f 5 , goes to position 4 of
Sib. It should be noted that, when
using an electronic ignition system,
it is usually better to connect the 'con-
tact breaker' side of the ignition coil to
the input (R30) — in other words, in that
case it is not connected to the contact
breaker itself.

Frequency-to-voltage converter (IC2)

The frequency-to-voltage and voltage-
to-frequency converters use an EXAR
integrated circuit, the XR4151. Three
of these are needed for the complete
circuit: IC2, IC3 and IC4. We have never
really discussed this 1C in detail, but
there's no time like the present.

Figure 6 is a simplified block diagram
of the interior circuit. For frequency-
to-voltage conversion, it is used as
follows:

The signal to be converted is applied to
one input of the comparator and a
DC reference voltage is connected to
the other (pins 6 and 7). As the input
signal swings up and down around the
reference level, the output of the
comparator switches to and fro between
positive supply and 0 V. Each time
it swings positive, it triggers the
one-shot (monostable multivibrator);
the latter produces a short pulse at the
'logic output' and closes the 'switch'
for a brief, constant period. When the
switch is closed, a constant current
from the current source is passed to
output pin 1, where it can be used to
charge an external capacitor. It will
be obvious that when the input fre-
quency is higher the switch will close
more often; the external capacitor
will therefore charge to a higher voltage.
An output voltage that is proportional
to frequency: frequency-to-voltage con-
version, in other words! The current
source can be controlled by a resistor
connected to pin 2; adjusting the
current in this way alters the range of
output voltages obtained.

In the main circuit (figure 5), this
principle is used for IC2. The pulses
from the flow sensor are passed via Cl
to pin 6 of the 1C. The other compara-
tor input is connected to a reference
voltage, determined by R6 and R5.
The RC network C2/R8 determines the
period time of the one-shot; this, in
turn, corresponds to the time that the
switch in the 1C is closed. In effect,
therefore, the RC network determines
the output voltage range. C5 is the
'external capacitor', it is charged via
pin 1 when the 'switch' is closed, and

Parts list

Resistors:

R1,R2,R23,R24,R28= 22 k
R3,R1 2,R 13.R19 = 100 k
R4,R5,R6,R15,R30,R31,
R32 = 10 k

R7,R8,R16,R21 = 4k7

R9 = 15 k

RIO = 5k6

R11 = 150 k

R14 = 220 k

R17 = 6k8

R18 = 68 k

R20 = 3k9

R22 = 33 k

R25 = 120 k

R26 = 3M3

R27,R33 = 1 k

R29 = 12 k

R34 . . . R41 = 18 «

PI = 10 k preset
P2 = 500 k preset
P3 = 100 k preset
P4,P5 = 1 M preset
P6 = 2k5 preset

Capacitors:

Cl ,C6,C10,C16 = l/u/35 V
tantalum

C2.C3.C 1 8,C22,C25,

C26 = 10 p/35 V tantalum
C4,C1 1 ,C1 2,C13,C20,

C24 = 100 n MKM
C5 = 22 m/25 V
C7,C1 7 = 10 n MKM
C8 = 22 n

C9 = 4m7/35 V tantalum
C14 = 68 n MKM
C15 = 3n3 MKM
C19 = 330 n MKM
C21 = 680 p cer.

C23 = 100 m/16 V

Semiconductors:

D1 =4.7/400 mW

D2 = DUS

D3 = 1N4001

D4 = 15 V/400 mW

T1 . . . T3 = BC 141

T4= BC 107B, BC 547B orequ.

IC1 = 74C928

IC2.IC3 = XR4151

IC4 = XR4151 or LM 331 N

IC5 = 555

IC6 = 4049

IC7 = 4093

IC8 = 7400

IC9 = 7805 or LM340 T5
Dpi ... Dp3 = seven-segment
common-cathode LED display,
e.g. MAN 4640A

Sundries:

SI = 4-pole, 4-position switch
flow sensor and speed transducer:
see text.

*© rao-jjo -‘GUO^O oo >o EO

4-40 — elektor april 1980

fuel consumption meter

oiuiu

|R33 |0

M

99M0]

Figure 8. Printed circuit board layout for the extended version.

Figure 9. Component layout for the p.c. board.

fuel consumption meter

elektor april 1980 — 4-41

50 Hz

5 Hz

Figure 10. This little circuit can be used for calibrating the meter.

discharges through R11 between pulses.
Resistor R9 sets the current supplied by 10
the current source. The various com-
ponent values (for C2/R8, C5 and R9
in particular) are chosen so that the
output voltage varies between 0 and
8 volts over the full range of input
frequencies.

Voltage-to-frequency converters (IC3, IC4)

Two of these are required in the fuel
consumption meter. The first as part
of the frequency conversion circuit
for the signal from the flow sensor
(IC3), the second is used to convert the
signal from the speed transducer (IC4).

The same 1C is used as in the frequency-
to-voltage converter, so we can again
refer to figure 6. In this case, it is used
as follows:

The input voltage is applied to pin 7.

When this voltage is higher than that at
pin 6, the comparator triggers the one-
shot. During the one-shot period, the
current source supplies current through
the switch to pin 1 ; at the same time,
the 'logic output' is at 0 V. After the
one-shot period, the switch opens and
the logic output swings up to positive
supply. As before, the current pulses
from pin 1 are used to charge an exter-
nal capacitor (C6 or CIO in figure 5).

This capacitor is also connected to
pin 6, so that each time the switch
closes the voltage at pin 6 increases by
a constant amount. In practice, the
circuit is arranged so that the voltage
across the capacitor can rise above the
input voltage within one step; the
switch then opens, and the capacitor
starts to discharge through R14 or R18.

Recommended For Use On

Speedometer
Cable Fitting

Volkwagen Beetle

18 x 1,5 MM

Porsche (some models only)

Metric

Volkswagen Dasher, Gulf

18 x 1,5 MM

Audi-BMW-Fiat-Mercedes-

18 x 1,5 MM

Opel-Volvo

Metric

Fiat-Volvo

16 x 1,5 MM

Mazda-Toyota-Datsun-Colt

22 x 1,5 MM

Luv and Courier pickups

Metric

Mazda-Toyota 1973 and later
Rover (some models only)

Snap-On

All American cars except Ford

7/8-18

Products & American Motors

SAE

All American cars 1968 and

5/8-18

older — cars with cruise

SAE

control option

16 x 1,5 MM

Datsun — Most American made
trucks (if in doubt check

THD fitting)

Metric

Ford Products 1 969 and

Snap-On

newer

S-100

American Motors 1972 and

Snap-On

newer

S-400

When the voltage at pin 6 falls below
the input voltage at pin 7, the one-shot
is again triggered: the capacitor is
charged, and then starts to discharge
again.

It will be obvious that if the input
voltage is relatively high, the capacitor
will discharge quite rapidly to this level
and so the one-shot will be triggered
quite often. A low input voltage, on
the other hand, will result in a relatively
infrequent charge cycle. Effectively,
therefore, the control pulses to the
switch must occur increasingly often as
the input voltage rises. As the switch
opens and closes, pulses are also appear-
ing at the 'logic output'. The frequency
of this output signal is also pro-
portional to the input voltage, of
course. In other words, an input voltage
at pin 7 is converted into a frequency
at pin 3. Just what we want.

Speedometer

G-18

Transmission

Transmission

G-1 8
G-18

Speedometer

Transmission

G-16

G-22

Speedometer

G-9

Transmission

G-7

Speedometer

G-8

Speedometer

G-9

Speedometer

G-1 (

The output frequency range can be
modified by changing the values in
the network connected to pin 5. After
all, this RC network determines the
one-shot period. As we have seen, this
is actually done for IC4: one of two
different networks is selected by means
of Sid, two produce either of the two
desired output frequency ranges (f 2 or
f 3 ). P6 is included in one of these
networks for calibration purposes.

The output frequency can also be
modified by altering the control resistor
at pin 2: this determines the current
output from the current source. A fixed
resistor is used for IC3, but IC4 requires
calibration. This is why PI is included.

The main function switch (SI)

SI is simply a switch with four wafers
and four positions. By now it should
be clear which measuring ranges can be
selected. Which signals are switched
through what wafer has also been
mentioned at various points. However,
a brief summary may prove useful —
especially when 'pruning' the circuit to
give only the ranges that are required
for a particular application.

SI a selects the latch/reset signal for the
counter. As required, one of the four
signals is used: f 2 , f6 , fi or f 7 .

Sib passes the clock signal to the
counter: f 4 for positions 1 and 2, f 3
for position 3 and f 5 for position 4.

The oscillator is switched by Sic: fre-
quency f 6 in position 2 and f 7 in
position 4. It is not used in positions 1
and 3.

Finally, Sid is used to switch the RC
networks for the voltage-to-frequency
converter (IC4): frequency range f 2
in position 1 and f 3 in position 3. This
converter is not used in positions 2
and 4.

And then, of course . . .

. . . the power supply. This is the last
section in the circuit diagram that has
not been discussed. For that matter,
there is very little to say. It is quite
obvious from the circuit that we didn't
see any point in trying to be 'different'.
Why bother?

Two supply voltages are required: 12 V

Table 1. Speedtransducer selection table, based on manufacturer's data.

Table 1.

Speedtransducers

Speedtransducer Speedtransducer
Mounting Model No.

4-42 — elektor april 1980

fuel consumption meter

and 5 V. The 12 V supply is readily
available — it is simply derived from
the car electrical system. The second
voltage is stabilised by means of a
regulator 1C, IC9. That is the easiest and
cheapest way of doing the job, and
perfectly adequate.

Miles -per-gallon only

One of the main objectives when
building a fuel consumption meter is
to save money. With this in mind, it
will often be preferable to build a
simple unit that measures miles-per-
gallon only — doing away with all the
other ranges. The result is shown in
figure 7. Compare this to figure 5!

The operating principles are the same
— with the minor exception of the
circuit that produces the latch and reset
pulses ( N 1 . . . N4). If it is to be moun-
ted on the board given in figure 8, it is
possible to connect the output of N3 in
figure 5 to the inputs of N7 and N8.
This will do the same job as N1 . . . N4
in figure 7, but it costs two ICs instead
of one.

A board for the simplified version will
be given next month.

Construction

The complete extended version — in-
cluding the displays -can be mounted
on the printed circuit board given in
figures 8 and 9. The only components
not included on the board are the
transducers and SI. All connections
to these are clearly indicated, so the
wiring should not present any problems.
As indicated by a dotted line on the

component layout, the display section
of the board can be cut off and moun-
ted at right-angles to the rest. This
makes for a 'flat package' construction.
The only component that may be
inclined to run hot is the voltage regu-
lator 1C for the 5 V supply. It must be
mounted on a heatsink. Another point
to watch is the number of wire links on
the board. It is a well-known fact that
these are easily forgotten — but they are
just as essential as the other components!
The connections to and installation of
the transducers require some further
comments. The connection from trans-
ducer 3 to the ignition coil is no
problem. However, it should be noted
that some electronic ignition systems
provide a special output for rev counters.
If available, this should be used; if it is
not provided, the best connection will
usually be to the ignition coil — not to
the contact breaker.

Only one type of flow sensor is avail-
able, suitable for most cars. It can not
be used in cars that have a fuel overflow
feed back from the carburettor to the
petrol tank. In practice, this means
that cars with high-pressure fuel injec-
tion systems are ruled out. It cannot be
used on diesels, either. In all other cases,
it can be mounted in the petrol line
between the pump and the carburettor.
The speed transducer is more compli-
cated. This is mounted in the (mecha-
nical) drive cable to the existing speedo-
meter. Since car manufacturers seem
reluctant to standardise these cables,
several versions of the speed transducer
exist with different connections. Table
1 gives a general idea of the possibilities.
It would seem a good idea, when
ordering, to state exactly in what make

and model of car it is to be installed.
Obviously, retailers and even importers
may be reluctant to stock all possible
versions, but we have been assured that
they are available: less common types
may have a longer delivery time, that's
all.

The electrical connection to the speed
transducer consists of two wires. Which
of these is '+' will depend on where it
is mounted (on the speedo or on the
gearbox). If the connections are
transposed, the unit will not work — but
no damage can be done. Swapping the
wires will cure the problem.

There are four wires from the flow
sensor. Two of these (green and
yellow/black) are for the lamp; one is
connected to 'Y' (+12 V) and the other
to 0 V. A normal lamp is used, so
transposing these connections makes no
difference. The other two wires (blue
and red) are connected to the photo-
transistor. Blue is the emitter, so it is
connected to OV; red, the collector,
goes to 'X'.

Calibration

The transducers are factory calibrated.
(Don't touch the adjustment screw on
the flow sensor!) This means that the
circuit can be calibrated by applying
replacement signals to the transducer
inputs, before mounting the whole
lot in the car. A replacement signal for
the speed transducer is the easiest: all
that is required is a low DC voltage.
For the flow sensor and rev counter,
however, low-frequency squarewaves are
required. To avoid the need of buying
a function generator for this one job, a
simple auxiliary circuit can be built
to produce these signals. As can be seen
from figure 10, this consists of a simple
mains supply and a divide-by-ten
counter; it produces 50 Hz and 5 Hz
reference signals. Using these, the full
calibration procedure (obviously, if a
simplified meter is built — with less
ranges — the corresponding steps in the
procedure can be omitted) is as follows:

1 . Disconnect the flow sensor and apply
the 5 Hz signal to point 'X'. With

SI in position 2 (gallons per hour), P2
and P3 are adjusted until the display
reads 4.7 gallons per hour (or 21.3
litres per hour).

2. Leave the 5 Hz signal connected to
'X'; disconnect the speed transducer

and apply +5 V to this input (from the
power supply in the circuit). Set SI to
position 1 and adjust PI until the
display reads 21.3 litres per 100 km.

3. Applying the same signals as above,
switch to position 3. Adjust P6

for a reading of 13.3 miles per gallon (or
4.7 kilometers per litre).

4. Switch to position 4, and connect
the 50 Hz test signal to the rev

counter input (R30). Adjust P4 and P5
for the correct display: 15.0 (xlOO) rpm
for four-cylinder four-stroke engines
and 10.0 (x 1 00) rpm for six-cylinder
types. We assume that those with 8
or 12 cylinders are not interested in
fuel consumption ... H

old hat?

elektor april 1980 — 4-43

Obviously, this is not the first time we have published circuits for use in cars. While browsing through some
earlier issues, we found several interesting little circuits.

A few of these are repeated here, for the benefit of new readers. In general, limited stocks of the printed
circuit boards are still available.

simple wiper delay

For those readers who require a very
small and simple wiper delay control,
the circuit described here is ideal. This
circuit was in fact first published in
Elektor E17 but has been so successful
that its inclusion here is not out of
place.

The circuit is based on the well known
555 timer and can be seen in figure 1.
The maximum output current of this

1C is 200 mA, so it can drive a relay
direct. This circuit will operate on 6 or
12 volt cars. The only circuit change
is the relay (12 V relay for a 12 V car,
6 V car 6 V relay).

Most circuits using the 555 don't give
a wipe immediately after being switched
on. To cure this, R1 is included. This
resistor keeps the timing capacitor (C2)
charged, so that the relay is activated as

soon as the delay circuit is turned on.

A further improvement of this circuit
is the possibility of having the wipers
sweep twice between delays. This
multi-wipe function is adjusted with PI .
The installation of this circuit is exactly
the same as that for the 'intelligent'
wiper delay described elsewhere in this
issue. M

1

Figure 1. The Simple Wiper Delay uses the well known 555 timer.

parts list

capacitors:

Cl = 10n

C2 = 5/16 V

resistors:

semiconductors:

R1 = 10 k

D1 = 1N4001

R2 = 1 M

IC1 = 555

R3 = 2M2

R4 = 4M7

miscellaneous:

R5 = 6M8

Sl/Sb = double pole 6-way switch

PI = 1 M preset

Re = 1 2 volt relay

Figure 2. Component layout and printed
circuit for figure 1. (EPS 9474-2)

4-44 — elektor april 1980

old hat?

automatic battery charger

Recharging lead-acid batteries is
often assumed to be an extremely
straightforward matter. And that is
indeed the case, assuming that no
special demands are being made on
the life of the battery. On the other
hand, if one wishes to ensure that the
battery lasts as long as possible, then
certain constraints are placed upon
the charge cycle.

Figure 1 illustrates the ideal charge
current characteristic for a normal
12 V lead-acid battery which is com-
pletely discharged. During the first
phase (A— B), a limited charging
current is used, until the battery volt-
age reaches approximately 10 V. This
restriction on the charging current is
necessary to ensure that the charger
is not overloaded (excessive dissi-
pation). For the next phase (C— D),
the battery is charged with the
'5-hour charging current'. The size of
this current is determined by dividing
the nominal capacity of the battery
in ampere-hours (Ah) by 5. At the
end of this period the battery should
be charged to 14.4 V, whereupon the
final phase (E-F) starts. The battery
is charged with a much smaller 'top-
up' current, which gradually would
decrease to zero if the battery volt-
age were to reach 16.5 V.

The circuit described here (see
figure 2) is intended to provide a
charge cycle which follows that
described above. If the battery
is completely discharged (volt-
age <10V), so little current flows
through D3 that T1 is turned off.

The output of IC1 will be low, so
that the base currents of T2 and T3,
and hence the charging current, are
determined solely by the position of
PI.

If the battery voltage is between 10
and 14 V, D3 is forward biased and
T1 is turned on. The output of IC1
still remains low, so that the charging
current is now determined by both
PI and P2. If the wiper voltage of P3
exceeds the zener voltage of D1, then
due to the positive feedback via R4,
the output voltage of IC1 will swing
up to a value determined by the
zener voltage of D1 and the forward
voltage drop of D2. As a result T1 is
turned off and the charge current is

1 I

6

once again determined by the pos-
ition of PI. In contrast to phase A— B,
however, the higher output voltage
of I Cl means that current through
PI, and hence the charging current, is
reduced accordingly.

Since D2 is forward biased, the effect
of resistors R2 and R3 will be to
gradually reduce the charging current
still further, as the battery voltage
continues to rise.

To calibrate the circuit, P3 is adjusted
so that the output of IC1 swings high
when the output (i.e. battery) volt-
age is 14.4 V.

By means of PI the 'top-up' charge
current is set to the 20-hour value
(capacity of the battery in Ah divided

c D

M

0 2 4 6 8 10 12 14 | 16 18

14.4V

79517 1

old hat?

elektor april 1980 — 4-45

Parti lilt.

Resistors:

R1 = 12 k

R2 = 10 k

R3 = 82 k

R4 = 1 M

R5,R6 = 8k2

R7 = 100 n

R8 = 3k9

R9 = 4k7

PI = 100 k preset

P2 = 220 k . . . 250 k preset

P3 = 10 k preset

Capacitors:

Cla = Clb = 4700 m/40 V

Semiconductors:

T1 = TUN

T2 = BD138, BD140
T3 = TIP2955

D1 = 6V8, 400 mW zener diode
D2 = DUS

D3 = 5V6, 400 mW zener diode
IC1 = 741

by 20) for voltages between 14.5 and
15 V. Finally, with a battery voltage
of between 11 and 14 V, P2 is ad-
justed for the nominal (5-hour)
charging current.

The initial charging current (phase
A— B) is set by the value of the
'top-up' current, and depending upon
the characteristics of the transistors,
will be approximately 30 to 100%
greater.

Miscellaneous:

Tr = 16 V, 8 A mains transformer
B = B80C10000 bridge rectifier
fuse = 0.5 A slo-blo

Siemens Components Report
Volume XIII, No. 1 March 1978.

W. Ferdinand

When starting a car journey after dark it
is useful to have a device which will
keep the interior lighting on for a while
after the doors have been closed, and so
make it easier for the occupants to
fasten safety belts and insert the
ignition key. This can be done with the
simple time-switch circuit shown.

afterburner

4-46 — elektor april 1980

old hat?

power flasher

Despite the vast array of solid-state
devices now available, the flasher
units for car direction indicators are
still almost exclusively electro-
mechanical. Apart from the obvious
objection of unreliability, these units
suffer from the problem that the
flashing rate is dependent on ambient
temperature, battery voltage and
load. This latter factor means that if
one wishes to wire all four indicators
to flash simultaneously as a hazard
warning, it is necessary to use a
separate flasher unit.

The electronic flasher discussed here
suffers from none of these disadvan-
tages. The repetition rate is practi-
cally independent of battery voltage,
temperature and load, has a built-in
hazard warning switch and is
extremely reliable. Furthermore it

complies with all the legal require-
ments for turn indicators, the
repetition rate of 40 to 90 flashes per
minute being within the specified
range and the circuit being arranged
so that the indicators light immedi-
ately when the turn indicator switch
is operated.

The circuit is basically an astable
multivibrator constructed around
two CMOS NOR gates N 1 and N2.
N3, N4, Tl, T2 and T3 buffer the
output of this astable to drive the
indicator lamps.

When the indicator switch is oper-
ated C2 discharges rapidly through
D1 and the indicator lamps. Pin 13
of N 1 goes high and its output goes
low. The outputs of N3 and N4 thus
go high, turning on Tl , T2 and T3
and lighting the indicators. The

astable then begins to oscillate at
approximately 1 Hz, turning the
indicator lamps on and off.

If the hazard warning switch, S 1 , is
closed then the circuit operates in
exactly the same fashion except that
all four lamps are connected in
parallel and flash in synchronism.

T3, which switches most of the load
current, must be mounted on a heat-
sink. If a metal box is used to house
the unit then T3 can be bolted to the
wall of this using an insulating
washer and bush. The current in the
leads connected to points A and B is
quite large (up to 8 A) so heavy-
gauge wire must be used for these
connections. The positive supply lead
must be fitted with a 10 A fuse if
not already fused.

Parts list.

Resistors:

R1,R3,R4 = 2M2
R2 = 100 k
R5 = 4k7

R6 = 120 n (1 Watt)

Capacitors:

Cl = 1 0 m/1 6 V

C2 = 1 ju/16 V (tantalum)

C3 = 1 n
C4 = 220 n

Semiconductors:

IC1 =4001 (B)

Tl = BC 557, BC 177
T2 = BC 328, BC 327
T3 = FT 2955 (Fairchild) TIP 2955
D1 = 1N4148

market

elektor april 1980 —4-47

Low-cost keyboard has snap feel

A low profile, low cost keyboard with
discrete keytops and snap feel has been
introduced by Chomerics, Inc. of Woburn,
Massachusetts.

The Chomerics Quickey EH keyboard
features contacts and circuitry screened onto
a single folded piece of hinge. The keys pro-
vide a tactile feel and are recessed to prevent
accidental actuation. Keyboards are available
in 12 or 16 key configurations with X/Y
matrix or single pole outputs. Customized key
legends and special output codes are available.

The EH keyboard measures 2 3 / s '' x 1)4 x 14",
is equipped with heat staking corner posts and
may be flush or panel mounted. Flex tail or
pin terminations are available and the contact
rating ist 20 mA, 30 vdc. The keys are on
J4" centers and key life is 10 million actu-
ations. Operating temperature is — 40°F to
150° F and bounce is less than 3 ms.

Chomerics Europe
Thomas VZ. Walsh
Holbeinstr. 62
D-6000 Frankfurt 70
Telephone: (0611) 63 45 35

(1458 M)

DC tachometer generator

A new 'one-piece' design of DC tachometer
generator is now available from Neco Elec-
tronics (Europe) Ltd. The new 7 volts per
1000 rpm (12,000 rpm maximum) unit (ref
ST-7447A-2) is more compact, better sealed
and lower in price than equivalent tachometer
generators of traditional construction. The
generator is in the form of a 'one-piece'
housing, with access only from a moulded
cover at the non-drive end which also retains

the brushes. Construction is therefore sim-
plified and the enclosure is better protected
against adverse working conditions.

The unit is temperature compensated and is
suitable foroperation in ambient temperatures
between — 55° C to +100°C. The voltage out-
put at 25°C will not deviate by more than
0,001 per cent per degree of change within
— 20° C to +75°C. Linearity at any speed from

0 to 12,000 rpm is said to be better than 0.1%
of the output at 3,600 rpm. An rms ripple
value of less than 3% of the DC value at any
speed in excess of 40 rpm, and bi-directional
output within 0.25% of nominal output are
claimed. Brushes and commutators are
guaranteed for 100,000 hours at 3,600 rpm,

1 mA maximum current.

Intended for use wherever there is a need to
convert speed into an electrical analogue
signal for remote indication or control, Neco
say that the accuracy and low driving torque
of these generators make them ideal for use in
precision integrators and as rate generators for
damping in positioning servos. They are also
suited for use as transducers for recording
equipment and for under-or-over speed
indication by direction and magnitude.
Normand Electrical Company Limited,

Walton Road,

Eastern Road, Cosham, Hampshire, P06 1SZ.
Telephone: (0705) 370988.

(1456 M)

Sub-miniature reed relay

Measuring less than an inch long and weighing
only 3 gm, the Erg PM21 high reliability reed
relay can switch 10 W loads. Switching rate is
500 Hz and coil drive power only 70 mW.
Standard versions of this completely encapsu-

lated 1 Form (n.o.) s.p.s.t. relay have contact
ratings of 200 V at 0.5 A (1 A carry). Contact
resistance is 200 mfi and very stable. Despite
their small size (25 x 8 mm) these relays have
a reed/coil insulation resistance of 10 10 n at
500 V r.m.s. minimum, and a dielectric
strength (reed/coil) of 500 V r.m.s. minimum.
Leadout pins are set on a standard 0.1 in
pitch for p.c.b. mounting, and all relays have
internal magnetic screens to minimise inter-
action with adjacent relays allowing compact
and economic p.c.b. layout. The PM21 com-
ponents are available with coil voltages of 5,
1 2 and 24 V d.c. Consistent reliable operation

is assured even with voltage variations if
—10% +50%. Designed to conform with
BS912, each relay is vacuum encapsulated in
epoxy resin with glass-loaded shell and header
to give reliable protection and performance
even under adverse conditions.

Erg Industrial Corporation Limited
Luton Road, Dunstable, Bedfordshire,

L U5 4LJ England.

Telephone: 0582-62241

(1457 M)

VSWR/Power meter

A combined VSWR and power meter offering
direct reading of both functions without
interpolation is available from Zycomm
Electronics Limited.

In operation, the unit is autoranging for
power output, covering 20 W to 2 KW in
three ranges for 1 .8-30 MHz and 50-1 50 MHz,
and 2 W to 200 W for the 430-470 MHz range.
VSWR from 1 :1 to infinity can be measured.

Separate sensing heads are supplied to cover
each frequency range, and these can be
connected at any position in the feed line
— including the mast head for precise radiated
power indication. Press switches on the front
panel allow the selection of the appropriate
head, and the display of forward and reverse
power as either peak or RMS readings.

The electronic comparator included in the
unit allows constant readout of VSWR
irrespective of power variation, i.e. gives true
indication during speech on SSB.

A 240 V, 50 Hz supply is required for oper-
ation.

Zycomm Electronics Ltd.

47, 49 & 51 Pentrich Rd,

Ripley, Derbys. DE5 3DS
England.

Telephone: Ripley (0773) 44281

(1451 M)

4-48 — elektor april 1980

market

time) and (c) conventional ignition, which
in this case remains totally unchanged.

To realise the full benefits of an electronic
ignition it is important that the rest of the
ignition system is in perfect working order, in
other words, fit new plugs and points and
have the ignition timed electronically. The
review 'kit' has now completed about 15,000
miles, most of them on the Continent, in the
rain and at night with only infrequent stops
for black coffee. Both starting and running
performance have been improved by the
fitting of this electronic ignition system and

no adjustments whatever have been needed
since.

All in all, an excellent kit and very good value
for money especially at the Elektor special
offer price of £15-95 for the C300 and
411-95 for the ES200. See the advertisement
elsewhere in this issue.

Suretron Systems (UK) Ltd.,

Piccadilly Place, London Road,

Bath BA 1 6PW.

Telephone: (0225) 23194/313846

(1478 M)

\.

iiutrkH in depth

The Elektor
special offer kit
reviewed

The Surefire Capacitive ignition
kit

It is not usual for Elektor to review products
in depth but this is a special motoring edition,
it is a special offer to readers, and it is a kit,
so all things being equal . . .

It can be upsetting to discover that, on some
electronic equipment, the inside of the box
does not match the outside.

This is certainly not the case with the Surefire
C300 and ES200 electronic ignition systems.
Since these are kits, the 'inside of the box' is
the first thing you see and it will be readily
apparent that component quality is high on
the list of priorities of Suretron systems. It
doesn't stop there either, the case is probably
second to none in both finish and appearance.
It is an anodised aluminium extrusion and so
exceptionally smart that constructors would
not doubt find many uses for the case alone.
Assembly of the kit is simple with the very
explicit instructions supplied. Only one minor
problem requires careful attention. The cable
grommet on the end plate is a somewhat
tight fit. A word of warning, don't be
tempted to open out the hole for an easier
fitting. Persevere, it will go in . . . The
working environment for an ignition system
is far from ideal and the case needs to be
totally proofed against icy and boiling water,
snow, cold and hot oil, petrol, sand and 'most
everything else'. (It’s a wonder anything
works under there really, let alone elec-
tronics).

Vibration proof shouldn't be forgotten either
and for this reason the tie wrap supplied for
the capacitor must be securely fitted and
extra care taken with all solder joints.

When assembly of the board is completed,
two or three thorough coats of polyeurathane
lacquer over everything will pay dividends
(take note of the specific instructions regard-
ing this).

The completed unit should be fitted under
the bonnet as close as practical to the ignition
coil. Again the instructions for fitting are
very complete even for the inexperienced,
and no problems should be encountered.

This electronic system is fitted with a three
position switch allowing (a) electronic ig-
nition, (b) no ignition, (a theft protection, if
the car is to be left for any lengthy period of

market

elektor april 1980 — 4-15

m ■ m

(1480 M)

Leisure Accessories Limited,

6 Guardian Road, Norwich, NR5 8 PF.
Telephone: (0603) 60925

breakers designed to trip within 0.2 seconds
if the connected equipment assumes a danger-
ous electrical condition. Fire protection
against possible earth faults is also afforded.
Continental Electrical Components (UK) Ltd.,
Tranby Croft, Tranby Lane, Anlaby,

Hull, HU 10 7 EE.

Telephone: (0482) 659309/659300

(1479 M)

Weatherproof mains inlet sockets
and couplers

Connecting mains power lines to a caravan or
outbuilding is not an easy job at the best of
times. However, many difficulties may be
overcome by using suitable plug and socket
arrangements. A large variety of these types
of weatherproof exterior plugs and sockets
are available from Continental Electrical
Components (UK) Limited. Primarily in-
tended for touring caravans (for which
purpose they are extremely useful) they could
equally well be used for the outside work-
shop, electric lawn mower supply and even
garden lighting systems etc.

The range includes weatherproof male
sockets of both polarised and unpolarised
(Continental) types and of course line plugs
to fit them. This company can also supply
a small mobile trip unit capable of handling
up to 25 amps. These units are complete
with current operated earth leakage circuit

Caravan electric water pump

Elsewhere in this issue we refer to the growing
trend towards the fitting of electric water
pumps into caravans. Leisure Accessories have
realised this fact and have just released a new
addition to their ever increasing range. This
new water pump has many advantages over
other types and is of superior quality both
in design and construction.

The 1021 is the 12 volt version of the 1000
series. It is very small and compact enabling it
to be mounted virtually anywhere, and yet it
delivers 1.15 imperial gallons per minute
(5.25 litres/minute). A built in pressure
switch makes it fully automatic — just open a
tap, and the pump operates, close it and it
shuts off. It self primes quickly and ef-
ficiently to 4ft or 5ft (1.2 m to 1.5 m) and
is powerful enough to compress and expel
entrapped air in the supply lines — no priming
tap is required. It requires a current of only

3 amps whilst delivering water at free flow
with an initial surge current of approximately

4 A.

Perhaps the most important factor of all is
that it can run dry without damage. So you
know it won't burn out if someone forgets to
fill the water tank. It doesn't relay on com-
plicated switches to achieve this, the capability
is designed in from the start. Installation is
straightforward and uncomplicated. The
company also manufactures an extensive
range of fittings for mobile water systems.

New Instant Fuel Consumption
Monitor

EnviroSystems Ltd have recently developed
an instantaneous digital fuel consumption
monitor for motor vehicles. This display unit,
using half-inch high LED numerals, can be
mounted in or on the dash panel.

Quick and accurate reading of the display is
of paramount importance because the system
provides an instant and continuous reading of
actual fuel consumption in m.p.g. Therefore
the driver can adjust his driving habits and
monitor systems efficiency to minimise fuel
consumption. The display panel is coated
with GLARECHEQ to eliminate reflected
glare and extraneous light. GLARECHEQ is a
unique process whereby acrylics, with a
normally glossy surface, are sprayed with a
special formulation. This imparts a clear hard
matt finish, thus reducing reflected glare to a
minimum.

Although available as a consumer product the
FuelStretcher is basically a scientific
measuring system using the latest silicon semi-
conductor chip technology. Transducers in
both the speedometer cable and fuel line
generate signals which are fed to the semi-
conductor processor; there they are integrated
to produce an accurate and continuous
measurement of the fuel consumption. The
figures are displayed on a dashboard unit in
large red numerals. The system can be fitted
by the average DIY enthusiast.

For GLARECHEQ:

Mr. Bob Burnett,

Chequers Engraving Ltd.

Phone: 01-739 6964

For the FuelStretcher:

Dr. D. van den Broek,

EnviroSystems L td.

Phone: Grange-over-Sands 04484 4233

4-16 - elektor april 1980

market

missing link

RF power meters

Precise RF power measurement from
0.01 watts to 100 watts over the frequency
range 1 MHz to 1 GHz is provided by a new
range of three absorption wattmeters an-
nounced recently by Racal-Dana Instruments
Limited.

The three models, each with dual range
switching and large clear panel meters, cover
six overlapping ranges for additional accuracy.
Good VSWR performance is a major feature
and high reliability is assured by the use of
thick-film resistive elements and specially
designed, lightweight heat sinks.

In addition to the wattmeter function, each
instrument may be used as a highly accurate
50 Ohm load down to DC. An attenuated out-
put is available for connection to other test
instrumentation, such as frequency counters
or modulation meters, enabling measurements
to be made directly from the output of trans-
mitters.

Accuracy of the instruments is within six
percent of full scale while a VSWR better
than 1 .2 to 1 is maintained over the entire
frequency range. Frequency response is
within ±0.3dB referred to 30 MHz and
operating temperature range is — 10°C to
+45° C.

Racal-Dana Instruments Ltd,

Duke Street,

Windsor,

Berkshire, SL4 1 SB.

(1455 M)

The ICU — a mini microprocessor

Elektor E57, March 1979 page 3-28. In the
circuit shown, switch S3 should be normally
closed.

Burglar's battery saver

Elektor 51/52, Summer circuits 1979 page
7-05. The value of Cl is incorrect, it should
be 33 nF.

pcb for variable fuzz box

Elektor 54, October 1979 page 10-20. Diode
D2 is shown the wrong way round in figure 1 .
Figures 3 and 4 of the original article
(December 1978) are correct. The printed
circuit board is also correct.

Charging nicads — fast

Elektor 56, December 1979 page 12-10. In
figure 3 the zener diodes D1, D2 and D3 are
shown as 8 volt versions, they should be 18 V.

Voice operated control switch

Elektor 56, December 1979 page 12-40.
There was no mention of the value of LI.
This can be made by winding five turns of
enamelled copper wire (0.1 . , . 0.25 mm
dia.) around a ferrite bead.

Colour generator

Elektor 57, January 1980 page 1-06. In some
instances we have had reports of flickering
lamps. This can be remedied by increasing the
value of Cl to 470 nF/400 V. Zener diodes
D3 and D4 were shown as 5V6 both in the
circuit diagram (figure 3) and the parts list.
Operation can be improved by reducing this
to a value of 4V7. Finally, we should have
made it clear that presets PI and P2 control
the initial switch on time of lamps La2 and
La3 — after this they have no effect. To
adjust the timing of the lamps the unit must
be switched off, the presets adjusted, and the
circuit switched on again.

Vocoder

Elektors 57 & 58, January & February 1980.
Page 1-28: several times in the text, A5 & A7
are transposed for A3 & A4 and vice-versa
with respect to the circuit diagram, figure 3.
The easiest remedy is to alter the circuit
diagram. Page 1-29: again, several errors
cropped up in the text. The 'line input' for
the circuit of figure 6a is non-existent and the
sensitivity control (P13) has moved! The
circuit diagram itself (figure 6a) is correct
apart from the addition of a 22 mF tantalum
capacitor between R 1 1 5 and 0 V, as shown in

Modifications to
Additions to
Improvements on
Corrections in

Circuits published in Elektor

figure 6, page 2-19 (February issue). Page
2-16: capacitors C79 & C80, referred to on
this page as being mounted on the bus board
are in fact mounted on the input/output
board.

While we're on the subject of the bus board,
many readers may have noticed that points
9'. V, V and ‘j’ are incorrectly connected
(bottom right of figure 9, page 2-23): point 'g'
should be connected to point 'h', while 'i '
should be connected to 'j'. There is no need
to worry, however, as the printed circuit
board (EPS 80068-2) supplied by the EPS
service is correct.

ESS 005 Luna

It has been pointed out to us that there is an
error in the Lunar-lander program. This error
causes the program to function correctly
during one landing only . After this, all landings
(even the successful ones) will be indicated as
'crashed'! The reason for this is that two
variables (stored in locations 0D5B and
0D5C) are not brought back to their initial

value of 0F.

A minor addition to the end of

the program

(see below) will solve the prob-

lem.

$36

0DD8

C109 LD 09 (1)

0DDA

9805 JZ $ 37

0DDC

C407 LDI 07

0DDE

50 ANE

0DDF

C909 ST 09 (1)

$ 37

0DE1

9027 JMP $ 39

$38

0DE3

37 XPAH 3

0DE4

C4A0 LDI A0

0DE6

33 XPAL 3

ODE 7

3F XPPC 3

TAB:

0DE8

07 6D 00 4F 5B 00 7F 5B

0DF0

7F 3F 00

TAB:

0DF3

A0 5E 79 74 6D 77 50 39

TAB:

0DFB

00 5E 79 5E 37 77 38

TAB;

0E02

00 00 77 37 3E 38 00 00

$39

0E0A

C40D LDI 0D

0E0C

37 XPAH 3

0E0D

C45B LDI 5B

0E0F

33 XPAL 3

0E10

C40F LDI 0F

0E12

CB00 ST 00 (3)

0E14

CB01 ST 01 (3)

0E16

C40C LDI 0C

0E18

90C9 JMP $ 38

advertisement

eiektor april 1980 — UK 23

BEWARE ! RADIO AITIUITV Q

ONCE UPON A TIME we used to try and pack the entire contents of our

price list into each advertisement. But our new price list is some 30 pages long (before
editing). And our catalogues now total over 200 pages - so these ads are basically left
to cover new products, and try to persuade you to invest in a set of our catalogues.

Our new catalogue contains an incomparable amount of data on tuning synthesiser systems for all
types of radio applications. We offer a complete set of ICs for a LW/MW/SW/FM frequency synth.
with auto tuneing, scanning, 8 preset station memories - driven with CMOS MPU, for only £16.63,
(includes an LED/ VF display driver, programmable divider/PLL, prescalar). Plus more digital freq.
display systems and ICs than any source in the world. Plus more info on our 100 watt power fet
systems, plus more radio components, ceramic and crystal filters etc.

If you have the catalogues, then you will need our new (much expanded) price list, where we list
quantity prices on most items - with huge reductions in our basic ranges of semiconductors. An
SAE will bring you a copy, and please remember to include sufficient postage (16p).

The new 44378/2 stereo decoder is one of new range of 'Ambistandard' tuner modules, and is
simply the best stereo decoder module for under £100 - and it only costs £22.94 (inc VAT).

New 944378 2, the last
word in stereo decoders
with the KB4437/4438

AMBIT SHOP NOW OPEN

We are gradually getting our caller sales area sorted out.
with displays of the products on offer and a browsers
corner to sit and study data/catalogues. Call in next time
you are in the area parking outside the door.

COMPUTER CAPABILITIES

Ambit has been keeping a low profile on the subiect of
the MPU and its applications. Interestingly enough the
first protect we offer with MPU content does rather
more in the way of processing than simply playing a
daft game, or looking like an enormous calculator. Our
MPU facility and expertise is now for hire on a fully
commercial basis. Z80, 6800, 6809, 2650 etc.

Keyboard switch
SCK41505
typ 6m ops
23p each (1 24)

NEW LINE DC/DC + AC converters for fluorescent displays TOKO CPS senes
12v IN. 20 and 3v AC out at 65mA. Thick film design £2.34 ea Qty. prices OA

CATALOGUES 60p ea . all three for £1.60
PRICES SHOWN HERE INCLUDE VAT
POST/PACKAGE CHARGE NOW 35p

ambit

INTERNATIONAL

CWO PLEASE Commercial MA terms on application
Goods are offered subject to availability, prices subject
to change so please phone and check if in doubt

200 north Seruiie Rood, Brentwood, Essex

TELEPHONE (STD 0277) 230909 TELEX 995194 AMBIT G POSTCODE CM144SG

ASSISTANT FILM RECORDISTS
AND TRAINEES

Would you like to specialise in sound with the BBC TV’s Film Department?
There are vacancies in West London.

ASSISTANT FILM RECORDISTS work
initially in sound transfer and dubbing areas
operating sound recording and reproduction
equipment for a wide range of programmes.
There are prospects of progressing to mobile
Film Recording work in due course. If you have
professional experience in this field, the
starting salary would be £4,185 p.a., perhaps
higher if exceptionally qualified, rising to
£5,605 p.a. An additional allowance is paid for
shift work (not nights). Normal hearing
essential.

EXCELLENT TRAINING is given if you have
ambitions to do this type of work but lack

experience. You will need good O' level
standard of education or equivalent, including
Physics and Maths and a basic knowledge of
electronics. You should be able to demonstrate
a practical interest in sound and recording.
Trainees will start at a salary of £3.800 p.a. at
the end of August 1980, and should qualify for
promotion to Assistant Film Recordists about a
year later.

Salary review date April. Conditions of Service
excellent. Contact us immediately for
application form (quote ref. 2053/EL and
enclose s.a.e.) BBC Appointments,

London W1A 1AA. Tel: 01-580 4468 Ext: 4619.

OJCJB -tv