december 1979
volume 5
number 12

page 1 2-02 selektor

Vocoders are becoming >ocoders today „ Vi „. rl
increasingly well-known.

The 'talking music effect, electronic nuisance (w. v
in particular, has caught

on - witness the charging nicads - fast .

astonishing increase in Nicad cells have the advantage
the number of manufac-

turers Of popular music and this can be a nuisance wf
vocoders. The next, almost immediately. Rapid ch

logical, step is a 'build-it- musl 66 done properly

yourself vocoder. First t o pp ,eamp

however, a brief recap of No run-of-the-mill-preamp, thi
the basic principles. are included, making for a sm

board'. The size of a Mini and

And for a reasonable outlay, at . —

charging

performance of t
at A perfect from

steam train

cumulative index 1979

link 78/79

talk funny? .........

page 12-08

The electronic nuisance
is like a cricket at night:
infuriating. A few
minutes after you switch
off the light, it starts to
make a noise; when you
turn on the light to look
for it, it stops.

on of speech
Professional i

designed as a 'front end' to the Analogc
e purpose of allowing greater flexibilit
It produces a variable rate clock sign;
fferent modulation waveforms that ca
, vibrato and other effects. A randor
o included for chorus effects. The con
s intended to be connected to the exte

tailoring potentiometers ig. Reinhoidi

Most potentiometers are supposed to have a f
forward linear or logarithmic characteristic. Thi
most applications, but sometimes the particular
required is not readily available. Fortunately,

page 1 2-20

The steam train sound effects generator fits inside
the model. It provides the steam sound, varying with
the speed of the engine; a steam whistle is also
included.

12-40

voice operated control switch .

Amateur radio operators normally use i
button to switch from 'receive' to ‘trans

the speech signal from the microphont
mated PTT button is usually referred to

market

missing link . . .
advertisers index

elektor decer

1979 - 12-01

Microcomputers as a hobby
For more than half a century there have
been radio hobbyists and amateur radio
enthusiasts. Originally they used crystals,
valves and discrete component semi-
conductor technology. The advent of
microelectronics greatly expanded the

opportunities open to hobbyists and
consequently, for example, led to a
boom in radiocontrolled models. Today
the number of electronics hobbyists is
estimated to be millions in Europe alone.
A significant percentage of this number -
— estimated to be as high as 1 5 per
cent - dedicate most of their spare-time
to the new technology of micropro-
cessors. Some hobbyists are more
concerned with basic hardware, while
others are searching for innovative
applications. Both activities are often
pursued collectively in clubs.

One of the largest groups in Germany
has been formed at Siemens in Munich.
The 250 members of this group have
been meeting on a regular basis for two

years; mainly to attend lectures and
participate in training courses. Pro-
grammes and systems are being pre-
pared as joint efforts in order to 'har-
monize' the activities of the group, for
example collective orders for com-
ponents and related equipment reduce
costs: Uniform p.c. boards provide a
simple basis for all kinds of circuits.
Altogether 700 Siemens employees co-
operate with each other. There are also
some 'freelancers' in the group. All
members are kept in touch by a news-
letter, which reports about the most
recent microcomputer programmes,
device developments and technical
literature.

The main interest is, of course, centred
on the application of microcomputers,
ranging from intrusion detection; word
processing for personal invitation cards;
electronically composed, recorded or
reproduced music to novel circuits for
cameras. Again and again rolling stock,
points and signals of model railways are
ingeniously controlled to simulate the
real thing. While another member is
endeavouring to make telegraph charac-
ters appear noiselessly on a screen; and
several other members are tracking
Earth satellites. One enthusiast in
particular is developing a VOR radio
navigation aid in which the frequencies
of all European ground stations are
stored. The pilot can now do without
the standard reference lists. It is es-
pecially this last application which is on
the brink of going beyond a spare-time
activity. The innovation has already
aroused the interest of some manufac-

Siemens Limited,

Siemens House,

Windmill Road,

SUNBUR Y-on- THAMES,

Middlesex TW16 7HS.

Tel.: Sunbury-on-Thames
(09327) 85691.

(505 SI

vocoders
today

talking music
is here to stay

It's not surprising that vocoders have
become so popular in such a short
time. Certainly in the popular music
field, where interest in all kinds of
artificial effects has increased rapidly
over the last few years. Add to this the
undeniable fascination of anything
associated with artificial speech pro-
duction (nothing new: this has been
going on for centuries!) and you have
two solid foundations for this vocoder.

History

Although artificial speech production is
not really a job for a vocoder, the first
experiments in that direction can still
be seen as the earliest stage of vocoder
history.

A Mr. von Kempelen was the first to
experiment successfully in this field.
Around 1790, he produced a compli-
cated machine consisting of an amazing
array of bellows, membranes, resonators
and pipes. Believe it or not, it produced
'human speech' sounds!

At the beginning of this century,
Stewart succeeded in constructing the
first electrical synthesiser of simple
simple speech sounds. This speech
synthesiser inspired Homer Dudley, at
the Bell labs in the United States; his
invention was patented in 1936. He
called his speech analyser/synthesiser
a 'Vocoder' — from VOice enCODER-
decoder. This vocoder was intended for
transmitting speech over a transmission
link with the smallest possible band-
width. Purely for telecommunications,
in other words.

Inevitably, the military showed great
interest in the vocoder. Not only did
it have the advantage of requiring only

When we first discussed vocoders in Elektor, a few years ago, they were
still relatively unknown. Since then, interest in this type of sound-effect
system has grown at an astonishing rate. Especially where the popular
music vocoder is concerned, the number of different manufacturers and
types seems to be increasing exponentially and the end is nowhere near
in sight.

There is every reason, therefore, to take another look at the vocoder
phenomenon — especially since we have now reached the point where
we can describe a vocoder circuit specifically designed for the home
constructor! More on that next month; first, we will recap the back-
ground and basic principles of vocoders briefly, so that everyone knows
what we're talking about.

a narrow transmission bandwidth; it also
offered the possibility of speech coding
- 'scrambling'.

Around 1950 one of the first musical
applications of the vocoder, the 'talking
piano', appeared on a gramophone
record ('Sparky'). The effect was excep-
tionally effective, certainly when one
considers the state of the art at that
time, but is was accepted without a stir.
It was merely another byproduct of the
'mysterious art of electronics'. The same
casual, if mystified, acceptance was
widespread when Radio Luxemburg
first introduced their well-known jingle,
and again when the Beatles used an EMI
vocoder to produce some extremely
sophisticated effects.

It wasn't until 1975 that the mystery
surrounding the vocoder started to
dissolve. Until then, it had been used
only in a few large laboratories (Bell,
Siemens, EMI, Philips, Sennheiser). With
good reason: those vocoders were so
big that some of them filled a whole

1979-12-03

It is interesting to compare the devel-
| opment of the vocoder with that of the
computer. The latter was initially seen
as a rather frightening and very power-
ful machine. Only 25 years ago, it was
thought that two computers would
suffice for the whole of the United
States: one on the East coast and one
on the West coast. In fact, we are now
rapidly approaching the point where
there will be a computer in every home!
It is unlikely that the popularity of
vocoders will go quite that far. However,
like earlier 'revolutionary' inventions
j (railways, cars, computers, electronic
music synthesisers), it is likely that it
will become far more commonplace
than was originally expected. Speech
analysis, speech synthesis, speech recog-
nition, speech input and output for
computer systems, and - last but not
least - applications in (electronic)
music: vocoders are used in all these
fields, and the end is nowhere near in
sight.

What's on the market?

1975 can be considered a turning-point
in the history of the vocoder. In that
year, a British manufacturer of music
synthesisers and similar specialised
equipment introduced a vocoder de-
signed by Tim Orr. EMS was already
known as a company with 'vision'; it
was one of the leaders in the field of
electronic music. In this case, they were
again the first to launch a completely
new instrument: the vocoder.

It is outside the scope of this article
to analyse the marketing philosophy of
all present-day manufacturers of

Table 1 Approximate price

(exd. VAT)

Bode Vocoder £2300

Electroharmonix £ 400

EMS Vocoder £ 8500

EMS 2000 Vocoder £2000

EMS 1000 Vocoder £ 945

Korg Vocoder £ 726

Moog Vocoder £3081

Musicoder £ 1630

Roland VP 330 £1143

Roland SVC 350 £ 507

Sennheiser VSM 201 £5000

Syntovox 221 £ 2950

Syntovox 222 £ 495

Syntovox 232 £1050

Syntovox 202 £ 275

Table 1. A list of vocoders that are presently

vocoders, but a single example may
serve to illustrate the confusion and
hesitation - both on the part of the
manufacturers and on the part of
musicians - which has become apparent
since the EMS Vocoder first appeared.
Dr. Robert A. Moog, the 'father' of the
music synthesiser, first built a channel
vocoder in 1970. It cinsisted of a
multitude of filters, envelope followers
and voltage controlled amplifiers, and it
was used for an adaptation of a
Beethoven chorale by Walter Carlos for
the film 'Clockwork Orange'. At the
time, Moog apparently failed to see any
commercial future for a more practical
version of this device. It wasn't until
the fearfully expensive EMS vocoder
appeared that a few other manufacturers
suddenly showed interest (Sennheiser,
Synton, Bode). This forced Moog to
face facts: his extensive range of pro-
ducts was incomplete without a vocoder.

However, the presently available Moog
vocoder is not his own design: it is
manufactured under licence. The rights
belong to Harald Bode, who has had his
own (patented) vocoder on the market
for some time. This patent will be
discussed later.

The growing competition and falling
prices since 1975 are clearly illustrated
in figure 1. The last two years, in par-
ticular: a new manufacturer — or a new
type, at least — every few months!
For those who are more interested in
price than in date of introduction, the
available types with approximate prices
are listed in table 1.

Applications

The first large vocoder systems on the
market (EMS Vocoder, Sennheiser VSM
201, Syntovox 221) were aimed at the
'high end' of the market. They were
expensive — well above the means of
musicians or even small sound studios
— and so complicated to operate that it
was difficult to attain high levels of
artistic achievement . . . Their use was
limited to large studios, radio stations,
film studios and a very few well-known
pop groups or composers with their
own studio. Furthermore, a system
that offered good intelligibility and
speech precision was useful for speech
research.

A large potential market remained
unexploited: the musicians and groups
who are always on the look-out for new
effects, a new 'sound'. It was to be ex-
pected that Japan would be the first to
introduce a vocoder at a price that the
average musician could afford. It was to

be expected . . . but it didn't happen!

In November 1978, at an Audio Engin-
eering Society exhibition in New York,
the American manufacturer Electro-
harmonix introduced a vocoder system
priced at about 800 dollars. Admittedly,
a Japanese manufacturer (Korg) also
had a vocoder on show — but it was
much more expensive. Both of these
vocoders were quite obviously rush
jobs, and the commercial departments
were unexpectedly faced with the task
of explaining this highly complex unit
to a very broad group of potential
customers. To make matters worse, the
few people who did know anything
about it by and large failed to realise
its full potential: they were interested
mainly in the 'talking music' effect.

There is, however, a completely differ-
ent field of applications for the vocoder: onto the sound of a musical instrument singing louder or softer; instruments
speech training for the handicapped. (Electric Light Orchestra, Herbie that would normally have a relatively
Speech sounds, or even complete words, Hancock) or any other basic sound. But slow 'attack' can be made more percuss-
can be produced by a vocoder. These there is more. It is also an ideal aid for ive by vocalising the desired 'explosive'
can serve as an example for the learner, modifying the timbre of a sound, for effect; chords played on an organ,
and his own attempts can be compared instance by superimposing vocal 'colour- polyfonic synthesiser or by a string
with the original. ation'. ensemble can be coloured and rhythmi-

A further, possibly highly important. There are a few restrictions that must be cally articulated by singing short tones
application of vocoders is in 'expression considered. Two points in particular at the desired pitch,
training'. Modifying sounds by making limit the choice of sound sources. In Obviously, all this calls for some
other (vocal) sounds often proves to the first place it is essential that the two practice. The musical effects that can
have a most beneficial effect for those sounds occur simultaneously — vocoding be obtained by means of a vocoder
who join in this kind of (group) therapy, is a 'live' process - and furthermore the depend entirely on the vocal capabilities
The most interesting — and funny! — spectra of the two sound sources must (and the long wind!) of the vocoder
effects are obtained when one succeeds overlap as much as possible. Some player.

in overcomming initial inhibitions, when examples are given in figure 2 and 3. One of the most important character-
faced with a group. Colouration of the sound from a musical istics of the vocoder in musical appli-

instrument is not the only possibility, cations is that it is a kind of interface
The loudness of the final output is also between the musician and the musical
determined by the loudness of the instrument. A vocoder is an ideal aid
Musical applications speech signal. This can be extremely to musicians who wish to achieve a

A vocoder offers the possibility of useful in itself. The attack and decay personal 'sound', a unique 'signature',
superimposing speech characteristics of the musical sound can be varied by in their performance. The musician has

elektor decamber 1979 - 12-05

a 'real time' tool that he can use to
modify the complete tonal structure
immediately, while he is playing. He can
make the sound harsher, fuller, softer,
more percussive. The results are immedi-
ately obvious, so that a kind of feedback
mechanism occurs: the musician can
hear exactly what he is doing and
modify his vocal control accordingly.
The result, as far as 'playing' the instru-
ment is concerned, is similar to playing
a conventional instrument; for example,
the light touch on a keyboard instru-
ment or the precise lip control and
embouchure for wind instruments. In
these cases, the final result is also
determined by a similar 'feedback'
mechanism. It is worth nothing that this
effect is almost absent when playing
other electronic instruments, since the
programming, presets and so on can
only be modified by means of a separate
hand or foot control. This control does
not lend itself to such immediate and
precise control of the total sound, with
the result that it is extremely difficult
for the musician to produce exactly
the desired effect.

Designing a vocoder

It is no easy matter to design a vocoder
that is suitable for (mass) production.
Before going into the problems, how-
ever, it is essential to take a closer look
at the basic principles involved. For a
more extensive discussion, readers are
referred to the two articles on vocoders
in the April and May 1978 issues of
Elektor. In this article, we will keep the
explanations as brief as possible.
Basically, then, a vocoder consists of
two groups of identical filters; one of
these is used to divide the speech spec-
trum into narrow bands, from each of
which a voltage is derived that can be
used to control the other group of fil-
ters, which reconstruct the speech spec-
trum. This would seem rather pointless
- using speech to make speech - but
the difference is that the second group
of filters receive a completely different
input signal as a basis for the recon-
structed speech. The first group of
filters is the 'analyser' section, the
second is the 'synthesiser'. The input
signal to the synthesiser section is
called the 'carrier', 'excitation' or
'replacement' signal.

As the block diagram in figure 4 shows,
the analyser section is basically similar
to a graphic equaliser, with one major
difference: the outputs of the various
filters are not summed. Each is followed
by its own rectifier and low-pass filter;
together, these form an envelope
follower. In this way, an audio signal
can be converted into a set of control
voltages (V c ) for driving the synthesiser
section.

The second group of filters, the syn-
thesiser section, could also consist of a
graphic equaliser (figure 5). In this case,
each of the filters is followed by a
voltage controlled amplifier; the outputs

ENVELOPE

FILTERS FOLLOWERS

low-pass networks (envelope followers). This section derives a set of control voltages from the
input (speech) signal: one Vc for each frequency band.

of these VCAs are summed to produce
the final output. This system, in its
simplest form, would seem to fulfil the
requirements for a vocoder. In all
probability, the results obtained would
indeed be faintly reminiscent of the real
thing . . . However, intelligibility and
dynamics would leave a lot to be

Numerous tests and intensive investi-
gation have led to a list of requirements,
relating to the various sections of the
block diagrams discussed above. The
exact requirements depend to some
extent on the application for which
the vocoder is intended.

In general, if vocal sounds are to be

superimposed on some other sound,
filters covering the range from 300 Hz
to 3 kHz will usually suffice. Obviously,
using more filters and covering a larger
total bandwidth will lead to better
'definition'. The large EMS, Sennheiser
and Synton vocoders use about twenty
filters, covering a range from approxi-
mately 200 Hz to 8 kHz. Within this
range, bandpass filters are used for both
analysis and synthesis. Frequencies
below 200 Hz and above 8 kHz are
covered by a low-pass and a high-pass
filter, respectively, so that the complete
audio band from 30 Hz to 16 kHz
is processed by the vocoder.

When a large number of filters are used,

12-06 -elektor

aders

FILTERS

Figure 5 . The other section in a basic vocoder is the synthesiser. A group of filters is used to
split up the 'carrier' signal (music, for instance) into small frequency bands. The output level
in each band is determined by the control signals applied to the voltage controlled amplifiers
(VCA); these control signals (Vc) are normally derived from the analyser section.

deciding how to subdivide the audio
band is no real problem. However, in
this case design of the filters is critical:
a fairly narrow and well-defined pass-
band is required, and the centre fre-
quencies must be accurate. In large
vocoders, like those mentioned above,
it is customary to use third-octave
filters (or an approximate equivalent).
Vocoders that use less filters must
obviously use a wider spacing of the
centre frequencies - the same total
range must be subdivided into fewer
pass-bands. Furthermore, the filters
may cover different bandwidths, giving
more precise analysis and synthesis
in the frequency range that is important
for speech intelligibility.

The number of filters used (and the
spacing) determines the required band-
width and the filter steepness outside
the band. If filters are set close together
but with an insufficiently steep cut-off,
there will be a large frequency overlap.
The result is that the speech becomes
indistinct and 'woolly'. This will almost
invariably happen if two graphic equal-
isers are used, as suggested in the basic
example given earlier. Equaliser filters
are just not good enough for this
application.

The easiest and cheapest way to obtain
a filter with a sharp cut-off is to use a
gyrator, but this has other drawbacks.
This type of circuit tends to 'ring'
noticeably and unwanted frequencies
do leak through; both of these effects
severely affect the intelligibility. We
could go on like this, crossing off the
various types of filter, but there is
little to be gained by beating around
the bush: in practice, there is really
only one filter type that is suitable.

As you would expect, it is by no means
the cheapest.

For optimum intelligibility, the initial
slope of the filter should be in the order
of 50 ... 54 dB/oct. This type of
filter is used in the Synton Syntovox
221. Regrettably, the large number of
close- tolerance components required
precludes its use in low-cost vocoders.
The Sennheiser VSM 201, for instance,
uses 36dB/octave filters; in the large
EMS vocoder, about 30 dB/oct. is used.
The high price of professional vocoder
systems is a direct result of the high
component and assembly costs involved
in the large number of high-precision
filters.

But good filters aren't the only problem.
In the analyser section each filter must
be followed by an envelope follower,
consisting of a precision rectifier and a
low-pass filter. Output offset voltages
are the headache here: they can ruin
the dynamics of the whole system.
There are only two alternatives: either
use very carefully selected components
or else include a calibration facility.
Another point to watch is the cut-off
frequency of the low-pass filter. It's
not a good idea to use identical filters:
the cut-off frequency should be related
to the centre frequency of the corre-
sponding analyser filter.

Hold on: we're not out of the woods
yet. Things get worse before they get
better; the synthesiser section poses
even more problems.

Each filter in the synthesiser section
must be followed by a voltage (or
current) controlled amplifier. If you
draw up a list of all the ways to make
a voltage controlled amplifier (VCA),
the OTA (operational transconductance
amplifier) turns out to be the best bet.
This is not to say that it is ideal — it
most definitely is not. The transconduc-
tance (gm) tolerance is bad enough,
but there are two more problems. In
the first place, OTAs are noisy. They
hiss. This is not quite fair, perhaps -
there are other noisy opamps — but
the problem is that only very low
signal levels can be used if the distortion
is to be kept within reasonable limits,
so the signal-to-noise ratio suffers.
Furthermore, the signal leakage from
control input to signal output is often
considerable. Not that you can blame
the manufacturer of the OTA (CA 3080):
this leakage is not included in the
specifications, and in most applications
it is relatively unimportant. For a
vocoder, however, it is essential that this
leakage is minimal; otherwise the
control signals from the analyser can
break through to the output, even in the

elektor december 1979 - 12-07

absence of a 'carrier' signal. This is a
nuisance, to put it mildly . . .

As before, the solution is to either
select the components carefully or else
provide a calibration point. For really
good results, you really have to do both.
In the constructional project that will
be described next month, a large
number of adjustments are included for
this reason; even so, a test procedure to
reject really 'bad' OTAs will improve
the final performance.

So far, we have only considered the
most essential parts of a vocoder sys-
tem: the analyser and the synthesiser.
Using these two, speech sounds can be
superimposed on other signals. Some
speech sounds, that is: the so-called
'voiced' sounds (vowels, for example).
Complete speech synthesis, including
'unvoiced' sounds (s, f, p, and so on) is
not possible with this basic system. For
this, a noise generator and a voiced/un-
voiced detector are required; the latter,
in particular, is quite a complex circuit.
It is the intention to describe it in
greater detail at a later date. However,
if the vocoder is to be used for musical
applications, the basic system discussed
so far is perfectly adequate. For that
matter, most low-cost vocoders pres-
ently available also lack a voiced/un-
voiced detector, mainly for reasons

If the vocoder is used in conjunction
with musical instruments that produce
a broad spectrum, with plenty of
higher harmonics, a reasonable approxi-
mation of the unvoiced sounds will be
obtained without a voiced/unvoiced
detector and associated noise generator.

Patents

A search through the files in the patent
office shows that there are hundreds
of patents directly related to the vocoder,
and even more that have some bearing
on it: patents in areas like speech
recognition, detecting the fundamental
speech frequency, etc.

The most recent patent relating to
vocoders is in the name of Harold Bode,
the manufacturer of the Bode vocoder
(that is also manufactured under licence
by Moog). The main point in this
patent is a clever little trick that Bode
uses in his vocoders to increase the
intelligibility of speech - the filters
used in the vocoder have a slope of
only 24 dB/octave.

As explained earlier, the intelligibility
of synthesised speech depends on the
type of filter used: its general perform-
ance, and the slope outside the passband.
If a vocoder is not intended for speech
synthesis in the full sense — where
external control voltages can be used to
create intelligible speech - then the
intelligibility for musical applications
can be improved by adding the high
frequency portion of the speech signal
(above 3 kHz) to the output signal
from the vocoder. This high frequency
signal only contains the noise signal and

transients, for consonants like k, p and

The main disadvantage of this system
is that a real voice must be used to drive
the vocoder: if artificial control signals
are used, the high frequency content
will be missed in the output. Further-
more, this 'high frequency bypass'
system produces a similar effect to
'signal breakthrough' in the vocoder.
Despite these disadvantages, the effect
is interesting enough; it is worth exper-
imenting with when you are building
your own vocoder.

The future

It is difficult to estimate future devel-
opments in vocoders. At present, it
seems unlikely that a digital version
will be produced. The conventional
analog vocoder has the unique feature
that it works 'real time'. The incoming
signal is analysed immediately, and the
output from the analyser can be used
for simultaneous synthesis. In spite of
the problems involved in using sharp
analog filters (phase shift), it seems
unlikely that a digital alternative with
a reasonable price will be found in the
near future. Synthesising speech arti-
ficially is another matter, of course.
There are several digital approaches to
this. The problem facing the would-be
digital vocoder constructor is to analyse
complex signals, like speech, sufficiently
rapidly and accurately to make a
workable vocoder.

The popular music vocoder has a bright
future. The number of manufacturers
and types will increase rapidly, and this
is bound to lead to falling prices.
However, it is unlikely that the near
future will see vocoders in the same
price range as 'effect boxes'. A vocoder
is too complex for that, using large
numbers of close-tolerance components
if optimum performance is required.
That, and the number of man-hours

required to build one unit, precludes
the appearance of a mass-produced
low-cost vocoder for some time to

It is to be expected that vocoders will
be incorporated in electronic organs in
the not-too-distant future. In a few
years time, most organs should have a
'vocoder' button — offering one of the
most intriguing and creatively-inspiring
effects of our time at the touch of a
finger!

What of the near future? Next month?
That, at least, can be foreseen with
great certainty: for the first time, as
far as we know, a vocoder designed
specifically with the constructor in
mind. Build your own vocoder! H

Lit.:

Elektor, April and May 1978:
Vocoders.

Elektor, January 1978:
Elektor Equaliser.

12-08-1

1979

electronic
nuisance

an infuriating little circuit

Practical jokers will want to hide the
circuit in such a way that it will take
some time to find it. For this reason, it
must be small; furthermore, it will have
to be battery-powered — a mains cable
would be a dead give-away. The circuit
described here fulfils both requirements:
it fits on a small p.c. board and is
powered by a small 9 V battery.

The light sensor is an LDR. In the dark,
its resistance is quite high; preset
potentiometer PI is adjusted so that the
inputs of the CMOS gate N1 are just at
logic zero under these conditions. The
calibration procedure will be described

The two CMOS gates, N1 and N2, are
connected as a 'trigger' circuit. When
the voltage at the inputs of N1 falls
below the trigger threshold, the output
of N2 switches to logic zero. Transistor
T1 is turned off. and Cl can now charge
up through R5.

The voltage across Cl rises so slowly
that it takes a few minutes for it to
reach the upper trigger threshold of the
second trigger circuit, N3 and N4. At
that point, the output of N4 swings up
to logic one — i.e. practically the full
supply voltage. This takes the reset
input of the 555 timer (IC2) high,
enabling this 1C. The 555 is used in an
oscillator circuit, driving a loudspeaker,
so that an irritating tone is produced.

Have you ever been kept awake by a cricket? You switch off the light
and snuggle down, and just as you're drifting off to sleep the insect
starts to make an irritating noise. As soon as you switch on the light to
look for it, it stops again. Tracking down this type of noisy nocturnal
nuisance can be infuriatingly time-consuming. The same result can be
obtained electronically. What's the point? Well, just for the fun of it.

When the victim turns on the light to
hunt for the source of the noise, the
resistance of the LDR decreases sharply.
The trigger circuit (N1/N2) changes
state, turning on T1. Cl discharges
rapidly through R4, the output of the
second trigger circuit goes 'low' and the
oscillator is turned off.

When the light is switched off again, the
circuit again waits a few minutes before
making a noise. Very infuriating . . .

Calibration

Preset potentiometer PI must be
adjusted so that the inputs of N1 are at
logic zero when the circuit is in the
dark. The easiest way to do this is to
connect a voltmeter to the output of
N2. First, PI is adjusted so that this
output swings up to nearly full supply

voltage; then PI is turned back until the
output switches to the 'low' level
(practically 0 V) - with the LDR in the
dark, of course. This completes the
calibration.

The time delay, from the moment the
light is turned off to the first squeak
from the oscillator, can be modified
according to personal taste by altering
the value of Cl. In the same way, a
different frequency can be obtained by
selecting a different value for C2. The
ratio of resistor R9 to RIO determines
the type of sound obtained.

Finally, the sound level depends on R8.
Note, however, that this resistor should
not be less than 100S2. Any loudspeaker
impedance from 4 f2 up can be used;
the higher the impedance, the louder
the output. M

Figure 1. Not much is needed for a

charging uR*»(ls-fhsl

more haste? more speed!

Rapid charging (within one hour) of discharge them completely, and then
nicads is a popular theme. You regularly charge them with a known current for
see circuits for charging these cells with the correct length of time. In this way,
a constant voltage. This is a very poor there is no danger of overcharging a
solution, since the total charge is semi-charged cell, with all the associated
completely unknown in this case risks.

(although this system can be used to Figure 1 gives the basic relationship
charge open cells). between cell voltage, temperature and

All problems associated with charging pressure, as the cell is charged from zero
nicads are aggravated when you start to 100% — and above. Initially, voltage
rapid charging. On the one hand, you temperature and pressure all increase
want to be sure that the cells are fully slowly. As the cell nears the full-charge
charged when the charging cycle is limit, the voltage starts to rise more
terminated; on the other hand you rapidly. At the same time, more and
know that the cell will only tolerate a more of the energy being pumped into
limited amount of over-charging. If they the cell goes into the production of gas
are charged beyond the safe limit, gas (oxygen) instead of being stored as
pressure builds up very rapidly inside chemical energy in the electrodes. This
the cell. A safety valve may open, if causes the pressure to increase; as a
there is one; otherwise the cell is likely result, some of the oxygen is reconverted
to explode. Even when a safety valve is at the negative electrode — producing
provided, this cannot do more than heat. As the temperature increases, the
limit the damage: the capacity of the cell voltage drops: nicads have a negative
nicad cell (in mAh) is reduced — perma- temperature coefficient, approximately
nently. — 4mV/°Cr It is this effect that causes

Until recently, the only safe way to the hump in the voltage plot: initially
charge nicad cells rapidly was to first the voltage rises, but when the cell is

Nicad cells have the advantage
that they can be recharged, so that
they don't have to be replaced as
often as normal dry batteries. The
only disadvantage is that charging
takes time, and this can be a
nuisance when you want to re-use
them almost immediately. Rapid
charging is the solution, but it
must be done properly.

charging nicads-f

elektor december 1979 - 12-11

fully charged it begins to fall again.

This principle is valid for all nicad cells.
The actual values given in figure 1 are
only intended as a general indication, of
course; they depend on the construction
of the cell, and so different values will
be obtained for different types. Manu-
facturers always specify whether their
cells are suitable for rapid charging,
what maximum current may be used
and how much over-charging is permiss-
ible.

To avoid explosions, or opening of the
safety valve, the safe limits specified by

S the manufacturer must not be exceeded. .

’ The charging cycle must therefore be
stopped in time. One or more of the
three parameters given in figure 1 may
be used to determine the end of this
cycle. Measuring the pressure build-up
inside the cell is not very practical, so
we may as well forget it. Measuring the
temperature is possible, but rather
clumsy. Which leaves us with the cell

Back to square one? No, not quite.
Because of the effect of the temperature,
it is not possible to use a certain, fixed
voltage level to determine the cut-off
point. However, the shape of the plot is
generally valid — and it has the makings
of a reliable indication.

The circuit given in figure 2 reacts to
the rate at which the cell voltage rises.
From figure 1, it is apparent that the
voltage starts to rise rapidly as the fully-
charged limit is reached. When the slope
becomes sufficiently steep an LED
lights. Alternatively, a relay can be used
to disconnect the cell at this point.

The circuit itself is quite cunning. An
oscillator (A4) gives one short pulse
every 10 seconds or so, closing the
] (electronic) switches SI and S2. When
these switches are closed, A1 operates
as a voltage follower (and C2 is
discharged), so that Cl is charged to the
input voltage at pin 3 of A1. The input
offset voltages of A1 and A2 are auto-
matically compensated for by the
circuit, so that the output voltages of
A1 and A2 will be identical at this stage.
At the end of the pulse from A4, the
two switches open. A1 now becomes an
integrator, and Cl is disconnected from
' its output. At this point, the output
voltages of A1 and A2 are still identical.
If the input voltage (derived from the
voltage across the nicad cells!) rises,
however, this voltage increase will be
integrated by A1. The faster the voltage
rises, the higher the output voltage of
A1 will be. If the voltage difference
between the outputs of A1 and A2 .
. becomes greater than the trigger
threshold of A3, its output will swing
high and LED D3 will light.

The trigger threshold of A3 depends on
the value of R14 and on the initial
output voltage of A1 and A2. A higher
initial voltage (corresponding to a larger
number of nicad cells in series) will lead
to a higher threshold. This means that it
is the relative rate at which the voltage
increases that determines the cut-off

Figure 1. Voltage, pressure and temperature varations in a nicad cell, during a rapid-charging

2

Figure 2. This rapid-charge cut-out circuit reacts to the more rapid increase in cell voltage as the
100% charged point is reached.

point — the shape of the plot in figure 1 ,
in other words. The circuit can there-
fore be used, without any readjustment
or switching, for anything between 4
and 12 cells — provided a suitable
supply voltage is chosen (between 12 V
and 18 V; the voltage divider R2/R3 is
included so that the supply voltage can
be equal to the voltage across the nicads,
provided it remains within the range
mentioned).

This circuit has been tested extensively,
and it works perfectly as long as all the
nicad cells being charged at the same
time are initially discharged by about
the same amount. This will normally be
the case if they are all used together to

12-12 -elektor dece

1979

Figure 3. An industrial rapid-charge circuit that reacts to the falling cell voltage when the cell is
fully charged (see figure 1 1.

\

power one model, for instance. How-
ever, we have not done comparative life
tests to determine the effect of this type
of rapid charging on the nicad cells.

It seems fairly safe, however, because
the industry is quite prepared to go a
step further! A well-known German
manufacturer supplies the unit given in
figure 3, for about £20. In this case a
reliable, low-drift opamp is used to
sense the rate of change of the input
voltage. As long as this voltage is rising,
the voltage at the inverting input will
lag behind — remaining slightly lower —
because C2 has to charge through a high
resistance (R3). The output voltage of
IC1 will be high, and the relay is pulled

Once past the highest point on the cell
voltage curve (see figure 1), the input
voltage starts to drop again. The voltage
at the inverting input will still lag
behind, but now the result is that it
will be higher than that at the non-
inverting input. The output of the
opamp swings negative and the relay

It is apparent from figure 1 that this
circuit will cut out a good deal later in
the charging cycle than the circuit given
in figure 2. The advantage is that the
cut-off point is more reliable; further-
more, the cells will be more fully
charged. On average, a cell must be
charged to 120% if it is to reach 100%
capacity; charging to 100% gives only
80% capacity. Strange, but true.

Going back to figure 3: preset PI is
adjusted so that the opamp output
swings low when the voltage at the
inverting input is 4 or 5 mV higher than
that at pin 3. When the relay drops out,
one of its contacts opens the connection
from the emitter of the transistor (so
that the relay cannot pull in again) and
discharges C2, ready for charging a new
set of cells. The other contact dis- i
connects the cells from the supply.

For both circuits, the same restrictions
apply:

• All cells should have approximately
the same capacity (this will always be '

the case if they are supplied as one
complete unit).

• The cells must be suitable for rapid j
charging — see the manufacturers

recommendations.

• The temperature of the cells must be
approximately equal to ambient tem-
perature before starting to charge them, j
'Hof cells would cool down initially,
the cell voltage would change and the
cut-off point might be incorrect.

• The cells should all be discharged by
approximately the same amount. If

they have been lying unused for some
time, they will all have 'self-discharged'
to some extent. The discharge level may
vary quite considerably from one cell to
another under these conditions. When [
charging, they will not all reach their
full-charge level at the same time. The
cells that were originally 'fuller' may be '
damaged by rapid charging in this case.

A similar situation may occur after
repeated rapid-charging. Since the j
capacity of the cells can never be
identical, some of them will gradually I
become less fully-charged than others
after several charge-discharge cycles. For I
this reason, it is advisable to charge in I
the normal way first (7 hours at a
current equal to 20 . . . 30% of the
capacity of the cells) . The next time the
cells must be charged, rapid charging
will be permissible; after about five
'rapid charges', it is time for another
'normal charge' cycle.

• For rapid charging, the current
should be equal to twice the cell

capacity. At lower currents, the shape I
of the voltage curve will not be suffi- I
ciently pronounced. H

iber 1979- 12-13

'Small is beautiful' is the modern slogan,
especially where electronic equipment
is concerned. One advert shows a baby
sitting on a complete 'hi-fi rack', in a
familiar position. Although the symbol-
ism is probably unintentional, it
illustrates the size of the equipment.
Then there are midget TV sets, with
a screen about the same size as the same
baby's hand. Apparently, somebody has
decided that all that empty space inside
a cabinet serves no useful purpose.
Not only 'small' is beautiful: simplicity
is another key word. In audio equip-
ment, for instance, the number of
controls (and in- and outputs, for that
matter) is being reduced to the essential
minimum. There are even amplifiers
without tone controls on the market.
For the same price, believe it or not.

integrated preamp for the topamp

No run-of-the-mill-preamp, this. Only truly useful controls are included,
making for a small and easy-to-operate 'dashboard'. The size of a
Mini and the performance of a Jaguar. And for a reasonable outlay, at
that. A perfect front-end for the topamp power amplifier published
last month.

Pruning

Reducing the number of knobs,
switches, inputs and outputs makes a
preamp less expensive; at the same
time, a smaller p.c. board and cabinet
will suffice - making for a further price
reduction. Some of this profit can be
re-invested, as in the design described
here, by using better components to
obtain better performance — special
low-noise opamps, for instance.

The first question, obviously, is: what
can we do without, what is essential,
and what is maybe-yes-maybe-no? What
do you really need in a preamp?

• output to power amplifier? Yes,
obviously.

• output to tape recorder? Yes, if
you've got one; no, if not. Con-
clusion: the option must be available.

• input from tape recorder? Again,
yes or no. Optional; with a 'monitor'

switch, if it is included.

• inputs from other signal sources?
Yes, obviously. But which ones?

Dynamic pickup? Yes. Tuner? Yes, that
too. Microphone? 'Auxiliary'? Nine
times out of ten they remain unused,
so let's be democratic and leave them

Table

Specifications

• input sensitivity (for 500 mV output ii
dynamic pickup:

e output impedance:

e balance control:

e signal-to-noise ratio (referred to 500
dynamic pickup:

e maximum input voltage, dynamic pic

2.6 mV (50 kfi, 1 kHz)

130 mV O 50 kfZ)

130 mV OSOkn)

< 1k2

2 10dB at 50 Hz (bass)

2 10 dB at 10 kHz (treble)

+3.3 dB. — “ dB (10 k load)

+2.3 dB, — dB (no load)

RMS):

65 dB (1 k!i in series with 100 mH
across the input)

75 dB

> input, at 1 kHz: approximately 200 mV RMS

e frequency response (tone controls 'flat'): 15 Hz ... 100 kHz dB
a crosstalk (20 Hz ... 20 kHz): <-60dB

e gain (balance control in mid-position, output load 10 k):
from dynamic input to tape output: 34 dB (x50)

from dynamic input to preamp output: 45.5 dB (x188)

IC1.IC1' - TDA1034BN,
NE5534N (Philips/Signetics)
IC2.IC2'.IC3,IC3' = TDA1034B.

NE5534 (Philips/Signetics)

IC4 = MC78L15CP (10%) or
MC78L15ACP (5%) (Motorola)

out. Moving coil input? Better not. It
raises the price for the majority who
don't want it, and the minority who do
can add a separate preamp. Democracy
again.

• volume control? Yes, carried by
majority vote.

• physiological volume control? Oh no,
please! An awkward potentiometer

with tap, a handful of R's and C's, a
sidelong glance at the Fletcher-Munson
curves (if you know where to find
them) and the result is ... a mess.
Those curves relate to actual sound
pressure, and that in turn depends on
the '0 dB level', the loudspeakers and

the living room. No, the only way to
do the job properly is to provide suitable
bass and treble controls. Which answers
the next question:

• tone controls? Yes, we'd better have
them. Bass and treble both. Not the

vicious kind of course, but say ± 10 dB
with well-chosen turn-over frequencies.
And with a nice and smooth control
characteristic, not the kind that does
nothing for a while and then suddenly
gives maximum cut or boost like a
switch. A 'cancel' switch might be
useful, but a 'flat' centre position is just

• rumble and scratch filters? If so.

with a switch? Yes, not really and no,
in that order. A fixed rumble filter is
essential, but at a fixed, low frequency
and as sharp as possible. The idea is to
protect the loudspeakers (and the
amplifier, for that matter) from high-
level subsonic signals. Scratch filters are
another matter. They're useless unless
they operate within the audio band, so
leaving them permanently in circuit is
out of the question. On the other hand,
signal sources are improving so rapidly
that a scratch filter is likely to be left
permanently out of circuit if a switch is
provided. This being the case, it is
simpler and cheaper to leave them out

12-18 - elektor december 1979

Dppreamp

5e

Figure 5. Tone control characteristics.

6

Figure 6. The frequency response of the dynamic pickup preamplifier (RIAA/IEC

of the circuit entirely. ' ~

• balance control? Yes, unfortunately. I
More often than not, the mid-
position of the balance control is not

the best setting — even though it should
be, in the ideal case. An effective
balance control is desirable; it is useful
if it can suppress one channel com-
pletely. If nothing else, this can be
useful for test purposes.

• mono/stereo switch? The only real
use for this is to reduce the hiss

when listening to a weak VHF-FM
stereo transmission. For this reason
it belongs in the tuner (and often is
built in). No need for two of them, so:
omit.

• other gimmicks? No, we're only
looking for essential controls.

Clean lines and attractive
performance

After this pruning operation, we are
left with only those features that are -
necessary and sufficient. A preamp
designed according to this principle will
do exactly what it is meant to do: help j
to give listening enjoyment without
leading to knob-blindness.

The block diagram of the toppreamp I
is given in figure 1. The input selector ;
switch. SI, has only two positions:
tuner or MD preamp. The selected
signal is passed to the tape output
and to the monitor switch, S2. This is
followed by the volume control, and an
amplifier stage that boosts the signal j
level to that required to drive most
power amplifiers (500 ... 1000 mV). f|
The following tone control stage has a _
'flat' gain of 0 dB - times one, in other J
words. 1

The last link in the chain is the balance j
control. o

From block diagram to design | P

The circuit diagram is given in figure 2. j L
One channel is shown, with the com- c
plete supply circuit. Since the circuit is r,
simplicity itself, only a fairly brief r,
discussion should suffice. 7

The preamp for dynamic pickups a
consists of one opamp (IC1) and a hand- p
ful of passive components. The only t,
peculiarity in this circuit is R4: this o
flattens out the frequency response s
above approximately 35 kHz (instead f
of carrying on down ad infinitum, as 7
specified by the RIAA equalisation it
curve). Frequency compensation for «
IC1 is now unnecessary, so that better n
dynamic performance of the opamp 7
(slew rate) can be achieved. I

The main amplifier stage (IC2) is a It
standard circuit. With the values given j
for R9 and R10, the gain is set at x5. «

The tone control stage (IC3 with its a
surroundings) is rather less conventional. ?
Two capacitors, CIO and Cl 1 , deter- n
mine the turn-over frequency for both p
■ bass and treble controls. A more com- /;
mon circuit would use four capacitors. A
The electrolytics C9 and Cl 2 keep DC t
voltages well away from the potentio-- a
meters P2 and P3. By now this pre- .

The opamps: worth a closer look

The NE1034 ( TDA1034 ) is a bipolar
opamp — in other words it contains NPN
and PNP transistors, just like its
predecessors (741, TBA221 , LM301,
LM307 and so on). Another feature in
common with many of its brethren is
the pinning: identical to the 741. But
that is where the similarity ends.

The inner life of the 1C is shown in the
accompanying diagram. There is no
I point in going into all the details, but
three points in this studio-audio-opamp
deserve some attention. The output
stage is capable of handling up to 10 V
RMS, with a power bandwidth of
70 kHz and without crossover nastiness,
into a 600 SI load. Furthermore, the
input stage is designed for very low
noise: the equivalent input noise is
7 nV /Hz at 30 Hz and 4nV/Hz at
I kHz. There is even an extremely
I low-noise N version, specified at 5.5 and
3.5 nV/Hz, respectively, its noise figure
is only 0.9 dB (at 20 kHz bandwidth
and a 5 k source resistance).

The unity gain bandwidth is approxi-
mately 20 MHz; with frequency com-
pensation (22 pF between pins 5 and 8)
it is still a quite respectable 10 MHz.
A cunning arrangement of four capaci-
tors (Cl .. . C4) provides high bandwidth
and high slew rate (13 V/ps, uncom-

caution, fortunately, is fairly common:
without it, the controls invariably
become very noisy.

Finally, the balance control. A linear
potentiometer is used. The mid-position
must give equal gain for both channels,
but it's a pity to throw away 6 dB of
gain in both channels. The solution is
to add R17: with an open output,
only 2.3 dB is lost in the mid-position;
loaded by a 10 k input impedance (that
of the topamp, say) the loss is still
only 3.3 dB. As an additional bonus,
adding R17 provides a more 'comfort-
able' control characteristic — see
'Tailoring potentiometers', elsewhere in
this issue.

The supply must be stabilised and
adequately smoothed. IC4, IC5, and lots
of capacitors take care of these require-
ments.

Construction

Two hundred and ninety-three holes
in 137 ’/j square centimeters of copper-
laminate board provide space for all
the components requirement for a
stereo version. The result is given in
figure 3; components marked with an
accent are for the right-hand channel.
The potentiometers and switches are
not mounted on the board. This keeps
the size (and price) down and gives
more flexibility in the construction.
A complete wiring diagram is given in
figure 4. Although 'cinch' plugs are
shown for in- and outputs, other types
can obviously be used as required.

pensated, 6 V/ps with compensation).
Frequency compensation is needed for
closed-loop gains of less than three.
Fjnally, some other important specs:
open-loop gain: x 100,000
open-loop bandwidth: approximately
1200 Hz (uncompensated)

approximately 600 Hz (compensated)
By way of comparison: fora 741, this
is less than 10 Hz!)

supply voltage range: ±3 V ... ±20 V
common-mode rejection: lOOdB
current consumption: typical: 4.2 mA
maximum: 7 mA

12-20 - elektor december 1979

steam

letting off steam - electronically, of course!

Electronics can be used to simulate the most amazing range of different
things. Cybernetic models, sound effect generators, electronic noses
- you name it, it's been tried! Somethings, obviously, are more
difficult than others; the sound of a steam engine is certainly easier to
imitate than the taste of certain types of coffee. However, it can be a
problem to fit a realistic sound effects generator inside a model engine.
It's possible, though, using miniature components and a little p.c. board.

This design is intended for use in HO I
models. These are big enough to provide
adequate room for the electronics -
either in the boiler or in the tender. In
smaller models, the same design may
fit . . . but not on the p.c. board given
here! The circuit can be used on both
AC and DC systems.

What, exactly, does this steam train
simulator do? First off, it imitates the
bursts of escaping steam from the
cylinders. To be even half-way realistic,
this must obviously be related to the
speed: the faster the engine goes, the
faster the steam bursts must come. The
different sound going up or down a
gradient would be a neat extra, but the
electronics required took up too much
room . . . Then, of course, there's the
steam whistle. That is included.

The circuit is powered by a battery or
nicad cell, so that the engine will still
make suitable noises at low speeds or j
even when stationary.

The block diagram

As you would expect, the steam sound
is derived from a noise generator (see
figure 1). No problem for an electronic
system. (It's usually more of a problem
to get rid of it!) The desired rhythm is
obtained by means of a modulator
driven by a VCO (Voltage Controlled
Oscillator). This VCO produces a low
frequency signal that varies with the
engine speed; its control voltage is
derived from the supply to the motor.
The steam whistle sound is also derived
from the noise signal. In this case the
noise is fed to a low-frequency oscillator
(LFO), producing the characteristicly
'hoarse' steam whistle sound. A power
amplifier (A) boosts the outputs from
the modulator and the LFO, to drive
the loudspeaker.

The steam whistle is turned on by a
switch. This can be done by hand, of
course, but that's not so realistic. A
better system is to mount a micro-
switch under the engine, and add
'humps' between the rails to operate it
at suitable points.

The circuit

At first sight, the circuit given in figure
2 may be a bit frightening. It may seem

incredible, but it all fits on the p.c. board

shown in figure 3! However, let's forget 1

take a closer look at the circuit.

The original noise source is a zener

f

modulate the noise signal, producing the *—

'bursts of steam'. This is done by A2; \/V/ vco

the control signal for this modulator is F

L.

derived from a low-frequency VCO (A3).

Potentiometer PI sets the modulation
depth. P2 determines the DC bias for

A2; this varies the noise level and

'sound'. With the train Stationary, P2 is Figure 1. Block diagram of the steam tri
adjusted for the desired 'parking hiss'. steam sound and adds the characteristic!

i

lin sound generato
illy 'hoarse' sound

eh?*"

r-v

12-22 - elektor decemt

When the engine starts to move, there
must obviously be a drive voltage
(AC or DC) across the motor, M. This
voltage is rectified by D4 . . . D7, turning
on T2. The VCO (A3) starts to oscillate,
modulating the noise signal. Including
diode D3 has several interesting effects:
the voltage across Cl 5 is pulled down
more rapidly than it can rise, so that a
sound more like sudden bursts of steam
is obtained; as the speed increases, the
average DC voltages across Cl 5 will tend
to increase, so the noise level goes up;
finally, when the engine stops the voltage
across Cl 5 rises slowly to the final
'parking' level.

As the engine speeds up, the voltage
across the motor rises. This increases the
frequency of the VCO, so that bursts of
steam occur more rapidly. There is, of
course, a slight -delay: if the voltage
across the motor increases suddenly, it
takes a while for the engine to pick up
speed. A similar delay is therefore
incorporated in the control circuit: C14.
If necessary, the value of this capacitor
can be modified until the rate of the
bursts of steam corresponds sufficiently
accurately to the actual speed of the
engine even when it speeds up or slows
down. A fixed 'calibration' of this type
is only an approximation, obviously:
coupling more or less coaches to the
engine will upset the synchronisation
slightly. In practice, however, this effect
was hardly noticeable.

The steam whistle sound is produced by
A4. Basically, this is a low-frequency
oscillator. Some noise signal is added,
via Cl 7, to produce the characteristic
sound. The whistle is turned on and off
by switch SI. As mentioned earlier, it's
a good idea to use a micro-switch under-
neath the engine, operated by raised
humps between the tracks.

The 'steam' and 'whistle' signals are
both fed to IC2: the output amplifier
(you can hardly call it a 'power' ampli-
fier! . . .). The levels of the two signals
can be modified by altering the values
of R1 2 and/or R14.

Construction

The p.c. board and component layout

down, the (1/8 watt) resistors and the
diodes are mounted vertically. For the
same reason, tantalum electrolytics are
used - they're much smaller than
the normal type.

On the component layout, there was
only room for the resistor and capacitor
numbers (without the R or C). Be
warned: don't mix them up!

It may be a problem to obtain a suitable
loudspeaker, small enough to fit inside
the engine or tender. If it's any help,
any impedance between 4 £2 and 16 £2
is permissible.

Finally, the supply. Three 1 .5 V batteries
in series will do the job, but nicad
accumulators are a more practical
proposition. They can be charged from
the main motor supply , when the engine
is running. A suitable connection is
provided ('N' on the p.c. board, and in
figure 2, for that matter); this is con-
nected to the '+' of the nicad cells.
Don't forget the positive supply con-
nection ('+') to the rest of the circuit, in
this case! 'N' is not connected to '+' on
the board. The value of resistor R27
depends on the maximum motor voltage
and the capacity of the nicad cells: the
maximum charging current, in mA,
must be limited to one-tenth of the

capacity of one cell in mAh. In other
words, the maximum charging current
for a 500 mAh cell is 50 mA; this limit
is set by the value of R27 and the volt-
age difference between the maximum
motor voltage and the total cell voltage
(4.5 V).

If normal dry batteries are used, R27
and D8 can be omitted.

Note that connection N and the con-
nection to switch SI are both on the
copper side of the p.c. board. H i

R1.R11.R16 = 10k
R2= 1 M

R3,R4,R6,R7,R8,R9,R20,
R23 = 100 k
R8= 120 k
R5 = 1 k
R10 = 150k
R12.R14 = 33 k
R13.R18 = 2k2
R15- 39 St
R1 7 = 2M2
R19 = 47 k
R21 = 22 k
R22 = 470 n
R24.R25.R26 - 220 k
R27 - see text
PI = 47 k preset
P2 - 100 k preset

Capacitors:

C1.C3-0.1 m/ 3 V tantalum
C2.C6.C7.C8 = 1 m/ 3 V tantalum
C4.C17- 10 n

C5,C12,C13,C15 = 10 m/6.3 V

C9 = 2m 2/3 V tantalum
CIO - 47 m/6.3 V tantalum
Cl 1.C16 = 1 m/6.3 V tantalum
C14 = 10 m/12 V tantalum

Semiconductors:

T1 = BC 549C, BC 109C or equ.
T2= BC547B, BC107Bor equ.
IC1 = TL084
IC2 = LM386N

D1 = 2V7/100 mW zener diode

D2 = 3V9/100 mW zener diode

D3 . . . D8 = DUS

Audio

assistentor 4-20

audio analyser 9-38

audio sectioner 8-03

chorosynth 8-07

class tells 1-22

crosstalk canceller 3-04

current dumping amplifier 7-37

digitally-controlled phaser 7-67

d.j. killer 7-51

improved DNL 7-08

monitor output 9-17

noise level meter 7-71

octave shifter for electronic guitars 7-10

opto-receiver for speech 7-40

opto-transmitter for speech 7-40

parametric equaliser 9-26

programmable sequencer 10-38

sequencer 7-80

sound effects with analog reverb 12-30

sound processor 7-42

stentor 4-14

talk funny 12-27

topamp 11-10

toppreamp 12-13

ultrasonic receiver for headphones 7-57

ultrasonic transmitter for headphones 7-54

using an equaliser 9-04

vocoders today 1 2-02

256-note sequencer 7-96

Car and bicycle

automatic battery charger 7-50

automatic battery charger 7-72

automatic windscreen clearer 7-70

battery monitor 7-48

bicycle speedometer 7-69

car anti-theft protection 7-23

car collision alarm 7-89

car light reminder 4-13

digital milometer 7-91

digital rev counter 10-15

dwell meter 7-30

fuel economiser 11-22

heated rear windscreen 7-22

motorcycle emergency lighting 7-31

varispeed windscreen wiper delay 7-78

Design ideas

capacitance and inductance meter 7-68

emergency break 7-74

FM stereo noise reduction 7-14

metal detector 7-77

news detector 7-15

non-stop Newton's cradle 7-77

sawtooths up or down an octave 7-62

TV programme multiplexer 7-34

Domestic and hobby

aquarium thermostat 6-28

barometer 7-34

clap switch 2-27

digital contrast meter 7-43

doorbell drone 7-39

electronic weathercock 7-35

elekdoorbell 6-12

fermentation rate indicator 7-95

flash sequencer 11-14

flexible intercom system 7-94

I R lock 7-79

liquid level sensor 7-10

moisture sensor 7-20

monoselektor 6-32

musical doorbell 547

photo-flash delay 7-38

pools predictor 2-06

programmable timer/controller 5-08

random tune doorbell 5-17

slave flash 7-38

sun lamp timer 7-11

tape-slide synchroniser 7-92

thermometer 7-62

UFO detector 7-36

Fun, games and model building

bio-control 7-33

building the TV games computer 4-26

digital wooing aid 7-17

electronic horse 8-03

electronic nuisance 12-08

electronic poker dice 7-66

emergency flight controller 744

flashing badge 747

I see your point 11-39

12-24 — elektor december 1979

miniature traffic lights 7-04

model railway block controller 7-04

pachisi 7-06

quiz master 4-24

remote control motor switch 11-18

robot with reflexes 7-88

servo controlled motor 1140

simple sound effects 5-32

speed controller for model railways 7-84

steam train 12-20

Generators

AC millivoltmeter and signal squirt 1-16

aircraft sound effects generator 7-90

electronic horse 8-03

FM stereo generator 1-26

frequency synthesiser 7-28

IHF toneburst generator 1-06

programmable function generator 7-75

programmable melody generator 8-04

simple sound effects 5-32

sinewave generator 3-21

sinewave oscillator 7-22

spot sinewave generator 2-20

sweep generator 5-28

variable pulse generator 3-18

video pattern generator 8-00

HF

FM IF strip 6-22

FM PLL using CA 3089 745

FM stereo generator 1-26

shortwave converter 11-34

stereo decoder 6-16

touch tuning 10-06

voice operated control switch 1240

Informative articles

applikator, heavy stuff: audio at 200 Watts .... 9-34

applikator, programmable sound generator .... 1-32

applikator, universal counter, type ICM 7216 . . . 3-24

class tells 1-22

computers and chess 1-34

crosstalk canceller 3-04

delay lines 2-11

delay lines (2) 5-18

ejektor, electronically variable resistance 1-25

ejektor, self oscillating PWM amplifier 9-46

ejektor, squelch for FM radio receivers 2-30

electronics the easiest way 11-16

goodbye E300/E310, hello J300/J310 6-27

how I beat the monster 140

ionosphere 11-34

I played TV games 10-28

I played TV games (2) 11-24

measuring by the book 1-02

missing link in audio systems 2-39

nicads 6 ' 04

one-nil for audio 9-22

optical memory disc 2-03

switching mains powered equipment 5-13

tailoring potentiometers 1 2-36

using an equaliser 9-04

using Elbug 2-32

VHF stereo test transmissions 6-20

Microprocessors

BASIC microcomputer 5-34

building the TV games computer 4-26

capitals from the ASCII keyboard 5-24

D/A for pPs 2-10

FSK modem 7-30

interface for pPs 5-25

I played TV games 10-28

I played TV games (2) 11-24

new programs for the SC/MP 10-12

NIBL-E 5-34

page extension for the Elekterminal 9-18

pseudo PROM 7-76

shift-lock for ASCII keyboard 7-27

speed controller for model railways 7-84

the ICU, a 'mini' microprocessor 3-28

using Elbug 2-32

pP TV games 4-06

I

I

Miscellaneous

automatic emergency lighting unit 1-48

battery saver 10-09

burglar's battery saver 7-05

calculator as a chess clock 7-41

charging nicads — fast 12-10

chess challenger 10 plays like a human 1-39

DC polarity protection 1-19

Formant - an invitation to our readers 2-39

home trainer 11-20

inclusive always/exclusive never gate 7-61

LED lamps 7-15

low voltage dimmer 11-39

metronome 7-06

oscillographics on board 1-13

quick starter for fluorescent lamps 7-46

reliable nicad charger 1-08

right-up and left-down 640

short interval light switch 10-20

solar tracker 7-07

TAP switch 4-19

TAP thieves on the head 9-45

ten channel T AP 7-19

variable logic gate 6-21

vicious chess buzzer 7-25

zero pF screen 4-23

2 switches — 2 lamps - one wire 7-27

5-minute chess clock 7-73

7-segment displays on a 'scope 7-85

Music

audio sectioner 8-01

cartridge life-expectancy counter 7-26

chorosynth 8-07

digitally controlled phaser 7-67

disco lights 7-02

octave shifter for electric guitars 7-10

p.c.b. for variable fuzz box 10-20

programmable melody generator 8-04

ring modulator 3-14

sequencer 7-80

sound effects with analog reverb 1 2-30

sound processor 7-42

talk funny 12-27

vocoders today 1 2-02

256-note sequencer 7-96

1

I

i

1

Please note:

Our offices will be closed

from 22-12-1979 through 1-1-1980.

Power supplies

PSUsonPCBs

robust lab power supply

Test and measuring equipment

AC millivoltmeter and signal squirt

analogue frequency meter

autoranger

autoranging peak meter

'de luxe' transistor tester

digifarad

digisplay

digital heart beat monitor

floating input for DVM

FM stereo generator

four quadrant multiplier

frequency counter for synthesisers .

frequency multiplier

frequency ratio meter

frequency synthesiser

f-to-v converter for multimeter . . .

gate-dipper

harmonic distortion meter

IHF toneburst generator

impedance bridge

improved LED VU/PPM

linear thermometer

logic analyser

oscilloscope light pen

passive oscilloscope probe

pH meter circuit for DVM

resistance bridge

ribbon cable tester

simple synthesising of PPMs

sinewave generator

sixteen logic levels on a 'scope

spot sinewave generator

strain gauge

sweep generator

transistor tester

transistor tester

universal digital meter

variable pulse generator

voltage comparison on a 'scope . .

voltage prescaler

voltage trend meter

3-state CMOS logic indicator ....

the Elektor staff
wish all our readers

an index to missing links

The intent of the Link is to assist the home constructor by listing
corrections and improvements to Elektor circuits in one easy to find
place. A simple check of the Link will show whether any problems were
associated with a project.

Don't forget to check previous Links if the project in question was
published before January 1978.

talk

ninny?

ring modulator,

chopper and frequency modulator

Deliberate electronic distortion of speech and music signals can give
fascinating results. Professional musicians use extremely expensive
equipment to obtain their very own weird and wonderful 'sound'. For
electronics enthusiasts, it is much more fun to get the same sort of
results from very simple circuits. Which is what this article is about:
getting effective effects using a single 1C, the 2206.

One of the best known and most
impressive distorters for audio signals is
the ring modulator. Normally speaking,
a ring modulator circuit has two inputs:
one for the audio signal (speech, for
instance) and one for a 'carrier'. The
weirdest effects are obtained when the
carrier frequency is within or just above
the audio range; using different carrier
shapes (sinewave, squarewave or triangu-
lar waveform) can produce different
effects.

The circuit can be drastically simplified
by using a 2206. This 1C contains a suit-
able generator for the 'carrier', and a
multiplier circuit that is ideally suited
for use as a ring modulator. The internal
block diagram is shown in figure 1 .

The oscillator (VCO) is already con-
nected internally to the multiplier. This
means that, basically, applying an audio
signal to the other multiplier input
(pin 1 ) will produce a 'ring-modulated'
output at pin 2. Simplicity itself I
Obviously, a few other components are
needed in a practical circuit. Not many,
though, as shown in figure 2. A single
capacitor. C4 (C e xt ' n figure 1), deter-
mines the frequency range of the VCO.
With the value given, the 1M potentio-
meter (PI ; Rext figure D can be use ?
to set any frequency between approxi-
mately 10 Hz and 10 kHz. The wave-
shape is selected by means of SI : switch
closed for sinewave, switch opened for
triangle.

The audio input signal is fed to the
modulation input via Cl. A voltage
divider circuit (R1, P2, R2) sets two
DC bias levels: the voltage across C2
provides the basic internal DC reference,
and P2 is used to adjust the operating
point of the multiplier. This adjustment
is important: it determines the 'carrier
level' (the output from the oscillator)

12-28 - elektor decamber 1!

talk funny

present in the final audio output. The
easiest way is to short the audio input
and then adjust P2 for zero audio
output. Only then is the circuit operating
as a true ring modulator. If P2 is incor-
rectly set, the oscillator frequency will
appear at the output, amplitude
modulated by the input (speech) signal.

This can give interesting effects, but it
isn't really the intention!

A stabilised supply must be used, other-
wise the DC settings may drift. This ,
would mean regular re-adjustment of P2
- which is rather a nuisance.

Chopping and frequency
modulation

The circuit can be extended, as shown
in figure 3. Only a few additional com-
ponents are needed to really use the 1C
to the full. Apart from adding the
'chopper' and 'frequency modulator'
features, a useful linear frequency scale
for the oscillator control is obtained as
an additional bonus.

The basic ring modulator circuit is I
virtually identical to the circuit given in
figure 2. The main difference is that the
multiplier bias adjustment is improved:

P3 is used for initial coarse adjustment,
with P2 in the mid position; then P2 is
■ used to tune out the last traces of the
carrier.

The chopper circuit makes use of a
squarewave output available at pin 11.

To be more precise, this is the collector
of an internal switching transistor (see
figure 1). With S5 in position 'chopper',
this point is connected to the signal I
output. When the transistor is turned i
on, the output is shorted; since the
transistor is turned on and off periodi- I
cally by the internal oscillator, the
chopper frequency is determined by the
setting of P5 (the VCO frequency ;
control). Switch S2 can be used to
select the audio signal before or after
the ring modulator; note, however, that
in the latter case the 'carrier' frequency ,
for the ring modulator and the chopper

frequency are identical — they are both 7 DIN ring modulator 'sound' is perhaps the

derived from the same VCO. best known: all kinds of additional fre-

The main reason for modifying the Q — 1 w — | (° 4 { — * — 4| quencies are added to the original signal,

frequency control circuit for the VCO is without any harmonic relationship. If

to obtain a linear voltage control *{ • OOM ’ really sharp dissonances are what Y ou

point. The frequency of the VCO varies Figure 7. A combinedin- and output can be want, the 2206 ring modulator is just
linearly with the voltage at the base of w "ed as shown here. the trick!

T1; this voltage is determined by the ■ The effect can be 'improved' by

setting of P5, but a frequency modu- A simple supply using a 78L12, say, is switching from sinewave to triangle: if

lation signal can be superimposed via adequate. A suitable circuit and you're not careful, you end up with a

C7. PI sets the modulation level; SI is p.c. board were given in Elektor, July/ completely scrambled signal. On the

used to select either the audio input August 1978, p. 7-75. other hand, using a low-frequency

signal or the output signal. A basic printed circuit board layout for sinewave produces a more 'pleasant'

I The frequency control range is set by the circuit itself is given in figure 4, and sound - the ring modulator adds an

P4. The procedure is as follows. Turn the two sides of the front panel with the interesting rhythmic effect to the

P5 right up (lowest frequency) and set controls are shown in figures 5 and 6. original.

P4 to maximum resistance. C5 is Finally, a suggestion for a combined in- The chopper facility can be useful on its

. switched into circuit via S3 and P2 is and output connection is shown in own, producing a kind of 'robot' or

offset so that the oscillator signal figure 7. All of these drawings are 'computer' sound. When used in com-

1 appears at the output. P4 is now slowly included as suggestions only; the final bination with the ring modulator, the

» turned down until the oscillator stops, design may be modified according to most weird results can be obtained. In

| and then turned back until it starts personal taste. the same way, combining frequency

I again reliably. This is the optimum modulation with the ring modulator can

I setting. Once again, it depends on the How funny does it sound? be interesting: low modulation levels

* supply voltage — so the latter must be Sound effects are always difficult to produce a kind of vibrato effect, and

' stabilised. describe - you've got to hear them. The high modulation levels - well. Try it! H

We have recently published a number of articles featuring delay lines
and the most popular was the Analogue Reverb Unit in Elektor 42
(October 1978). It would appear that this article was greeted with
such enthusiasm by our readers that many have been encouraged to
experiment further.

The following project has been designed as a 'front end' to the reverb
unit with the purpose of allowing greater flexibility with reverb effects.
It produces a variable rate clock signal together with five different
modulation waveforms that can be used for phasing, vibrato and other
effects. A random signal generator is also included for chorus effects.
The composite output signal is intended to be connected to the
external clock input of the analogue reverb unit.

elektor december 1979 — 12-31

It will be seen, when referring to
Elektor 42, that the Analogue Reverb
Unit (ARU) uses the well known
SAD 1 024 shift register. As most of our
readers will know, this device operates
on the 'bucket brigade' principle.
Briefly, this is analogous to a chain of
buckets from input to output. The
sampled signal at the input corresponds
to the level of water in the first bucket.
At the 'word of command' (clock pulse)
this bucket is poured into the second
bucket (which was of course empty). At
the next word of command the second
bucket is emptied into the third and so
on for 512 times, the number of stages
in one half of a SAD 1024. We should
explain to newcomers to electronics
that we don't really use water (at least,
not yet) and the water level in our
mythical bucket is in reality a charge
packet on an almost mythical capacitor
(they are physically very small).

Back in the real world, it will be appar-
ent that the delay time is dependent
mainly on two factors: the number of
stages in the shift register (or registers),
and the clock frequency. The first is a
hardware design parameter and not

1

Figure 1 . The oscillograph shows the comb
like structure of the phasing effect produced
by adding a delayed to an undelayed signal.

easily altered, but the clock frequency is
something that can be varied - and that
is where we get to the point of this
project.

Figure 2. Block diagram of the clock pulse generator. Five different waveforms plus a random
modulation signal can be selected.

A variable clock frequency has rather
more going for it than might at first
appear. If the output of the delay line is
mixed with a 'clean copy' of its input
signal the resulting periodic phase can-
cellation and addition will produce the
so-called comb frequency response
shown in figure 1 . Now, if the clock
frequency is raised and lowered the
comb will 'open and close'. This in
audible terms produces the phasing (or
flanging) effect. A chorus effect is
obtained by an entirely random vari-
ation of the clock frequency. The range
of possibilities will now be apparent.
Before getting too deeply involved in
this circuit some readers may prefer to
become better acquainted with 'bucket
brigade' shift registers, and for this
the previously mentioned article in
Elektor 42 should prove useful.

The external clock

The design target was to develop the
maximum in sound effect possibilities.
The final concept is shown in the block
diagram of figure 2.

The low frequency oscillator (LFO) is
variable between 0.1 and 10 Hz and
produces five different waveforms:
sinusoidal, triangular, rising sawtooth,
falling sawtooth and square. As a sixth
possibility a noise source generates a
random signal which is low pass filtered
to limit the passband. The filter roll-off
frequency is adjustable for variation of
the average speed of the random signal.
Switch SI selects the required modu-
lation waveform and the modulation
depth is varied by the intensity control.
After amplification the resultant signal
controls the frequency sweep of the
voltage controlled clock pulse gener-
ator (VCCPG?). Figure 3 shows the
VCO output frequency as a linear
function of the modulation control
signal. The frequency modulated output
of the VCO is connected to the input
of the analogue reverberation unit
thereby producing the various sound
effects discussed in previous paragraphs.

Sewar Circuit

As can be seen from the circuit diagram
of figure 4, the unit is built around
three integrated circuits, a function
generator (XR 2206), a VCO (XR 2207)
and four FET input op-amps housed in
one package (TL084 or TL074). The
circuitry around the function generator
(IC1) may be familiar to regular Elektor
readers. The oscillation frequency is
determined by components C2 + C3,
R3, R4 together with potentiometer P4.
Since availability of bipolar electrolytic
capacitors may be limited, the required
capacitance is made up from two
220 fiF types connected back-to-back.
The resultant 110 /jF suffices to bring
the frequency down to 0.1 Hz, the
upper limit being 10 Hz.

The output waveforms and amplitudes
are defined by the networks connected
to various pins of the waveform gener-
ator. Switch Sla . . . Sid functions as
follows.

Switch position 1 connects the filtered
noise generator output to the VCO. The
waveform generator is switched off and
Sic contacts cl shorts pin 11 to ground
to suppress any stray radiation.

Switch position 2 corresponds to a
sinusoidal waveform which is available
from pin 2 of the waveform generator.
The sinewave is produced by connecting
resistor R2 across pins 13 and 14 via
contact b2 while contact a2 shorts pin 1
to ground. The amplitude of the sine-
wave can be adjusted by means of preset
potentiometer P3.

Switch position 3 corresponds to a
triangular output at pin 2, by dis-
connecting R2 from pin 13. The ampli-
tude of the triangular waveform can be
adjusted by means of PI which is con-
nected to pin 1 via contact a3.

Switch position 4 corresponds to a posi-
tive going sawtooth waveform by
removing the short from pin 1 1 and
connecting this pin to the FSK input
(pin 9) via contact c4. The positive
going ramp of the sawtooth lasts for
half of the triangle period, the negative
going slope is determined by the resist-
ance of R1 and is much steeper. The
sawtooth frequency is therefore, prac-
tically twice that of the sinusoidal and
triangular waveforms. The amplitude is

again adjusted by means of PI .

Switch position 5 corresponds to a nega-
tive going sawtooth waveform by
moving the bias at pin 1 from PI to P2
via contact a5, thereby inverting the
sawtooth polarity. The output ampli-
tude is now controlled by P2.

Switch position 6 corresponds to a
squarewave output. The generator out-
put is now taken from pin 1 1 via R6
and Sid contact d6. It is clipped to
1 ,4 V p-p and made symmetrical with
respect to ground by the network
composed of R5, R6, R7 and the
reverse-parallel connected diodes
D1/D2. This symmetry obviates the
need for a coupling capacitor which
would otherwise distort the square pulse
shape, especially at low frequencies.

Any DC component at pin 2 of the
function generator 1C is blocked by the
coupling capacitor Cl. This DC com-
ponent is apt to surge when SI is
operated, and these surges cannot be
sufficiently bled via the high resistance
of P5 alone. However, the reverse-
parallel connected diodes D3/D4
become conductive only on these
surges and together with R8 speed up
the discharge rate of the capacitor.

The random signal is generated as
follows. Transistor T1 is used as a noise
source. Its base-emitter breakdown
comes into effect at around 8 V and
makes the transistor behave like a very
noisy zener diode. The resultant noise
signal is greatly amplified by A1 and A2

in cascade which function as active low- |
pass filters due to capacitors C6 and C7 |
in their feedback loops. This combi- I
nation gives a roll-off frequency of I
about 10 Hz. The random signal zero
frequency component is offset by the I
bias control P8 at the non-inverting
input of A2. The filtered output of
A1 + A2 is passed through a further
active low-pass filter, A3, with a 12 dB
roll-off at an adjustable frequency
controlled by P6. This sets the average
fluctuation speed of the random signal. 1
The final output is available at selector
switch contact dl.

The sweep control signal from the
modulation mode selector switch, SI,
is attenuated by P5 to the modulation
depth desired. This is applied to the
non-inverting input of the 16 dB ampli-
fier, A4, whose output determines
the oscillator frequency of the VCO,
IC2, as shown in the graph of figure 3.
The VCO control signal is composed of
the periodic or non-periodic waveform
from the mode selector switch, plus a
zero frequency component introduced
at the inverting input of A4. The centre
frequency of the VCO is then adjustable
by P7 to between 20 kHz and 250 kHz.
The stabilised voltage required for this is
supplied by the network R20, D5 and
D6. Capacitor C9 is the reactive com-
ponent of the oscillator circuit and this
capacitor determines the free-running
frequency of the VCO. The power
supply for the VCO is stabilised intern-

ally with the help of capacitor CIO. The which P7 is set to give an output of 5 or ing switch positions and control adjust-

I final squarewave output signal to be fed 6 volts. The actual figure will serve as a ments.

to the reverberation unit is taken from reference around which the modulation The final adjustment to complete the

pin 13 of the VCO. The power supply signals will swing symmetrically. setting up procedure is the random

for the clock generator (± 15 V 50 mA) The first output test is on the square- signal setting — with SI in the first

can be derived from the supply for the wave, for which SI is moved to the position and P5 at maximum. To reduce

reverberation unit. sixth position and P5 set to maximum the noise amplitude to a comfortable

output. With P4 set for the lowest oscil- level, a 1(iF capacitor is used to bridge

Construction and setting up lator frequency (its wiper fully towards the emitter of T1 to ground (capacitor

The printed circuit and component R3) the meter reading will fluctuate positive terminal to emitter). Poten-

layout for the ARU 'front end' is shown between a low and a high reading, in a tiometer P8 is used to adjust the DC

in figure 5. Assembly of the printed 3 to 5 second period, symmetrically output component to match the refer-

circuit board should not present any about the reference level established ence level established in the preliminary

problems if suitable sockets are used previously. The peak-to-peak amplitude operation. If the meter reading appears

for the integrated circuits. Electrolytic of the squarewave should be some 7 or to be somewhat erratic, due to the

capacitors, particularly Cl, C2, C3 and 8.5 volts. The actual voltages obtained extremely high gain in the noise ampli-

C8, should be low leakage types. should be noted, since they will have to fication circuit, the output should be

Special attention should be paid to the serve as a standard for the other wave- adjusted so that its average reading

selection of transistor T1. With the form measurements. approaches that of the reference level.

' circuit parameters given, its standing Should the squarewave oscillation stop The 1 fiF bridging capacitor is now

emitter voltage must lie between 6 V or the frequency rise too high when P4 removed, and the circuit is ready,

and 9.5 V, this voltage is the same as is turned to the fully clockwise position,

that of the DC component at the out- then the value of R3 should be altered. ARU + Sewar

put of the unity gain amplifier A1. If This can be done with the aid of a 47 or So far, the circuit is just a front end

the reading obtained lies outside this 50 kJ2 trimmer and, once the correct that supplies a sequence of clock pulses

range a different device must be tried, value has been found, a fixed resistor at a controlled variable rate. Its effect

A multi-meter can be used for setting up can be substituted. will only be audible when connected to

the circuit parameters although an The next test is on the sinewave, for an electronic reverberation system and

oscilloscope may be preferable. Test DC which SI is set to its second position associated equipment, such as that

levels are indicated at a number of and P3 adjusted to give a sinewave out- described in the Obtober 78 issue of

points on the circuit diagram to simplify put equal in amplitude to that of the Elektor. Consequently, some adap-

setting up. squarewave. tations are necessary to the reverb

Prior to further measurements, the To test the triangular waveform, with circuit board.

working range of P7 should be tested. SI in position three, PI is adjusted for The reverberation unit must use the

This is done with P5 set to zero output, correct output amplitude. A similar SAD 1024 integrated circuit. To prepare

The voltage on the wiper of P7 should procedure is followed for the two the unit for a high clock rate, a wide LF

vary from 0 V to around 8.5 V, after sawtooth amplitudes with correspond- band is required, which is made possible

by adapting the low-pass filters to a
15 kHz roll-off. The method of doing
this has been explained, together with
other modifications, in the October 78
article. The cable connecting the clock
unit to the reverberation unit must, of
course, be screened.

In order to obtain the desired phasing
effect, an additional control is required
for blending the delayed to the
undelayed signal. This modification is
suggested in figure 6, for mono, and
figure 7 for stereo operation, the latter
featuring a mono/stereo switch and a
500 (470) kfl tandem volume control.
The phasing effect is most pronounced
when the delayed and undelayed com-
ponents are of approximately the same
intensity.

Selecting and setting the clock rate and

its frequency sweep is a fairly simple
matter. The first action is to set control
P5 to minimum, cutting out all fre-
quency modulation, and to adjust P7 to
set the clock rate to the delay required.
The required modulation mode is then
selected and the modulation depth can
then be adjusted by increasing P5. If
the sweep gets too wide with respect
to the centre frequency, which shows
up as an audible whistle, the setting of
P7 will have to be altered - normally
around halfway. For some effects the
equipment may be used without any
modulation at all i.e. with P5 set at
minimum.

The effects obtainable are described in
more detail in the May 79 issue of
Elektor, pages 5-18 .. . 5-24. They are
recapped in Table 2. Quite unusual

reverb/phasing and reverb/vibrato
effects can be found by using the
variable feedback possibility of the
reverberation system. Apart from these,
the triangle and sawtooth modulation
waveforms permit a wide variety of
experimental sound effects, which must
be heard to be believed.

References:

Formant (4)

Elektor E30 October 1977, 10-40 etc.
Analogue reverberation unit
Elektor E42 October 1978, 10-44 etc.
Delay lines (2)

Elektor E49 May 1979, 5-18 etc.

Simple function generator

Elektor E33 January 1978, 1-40 etc. H

tailoring

potentiometers

potentiometer + resistor(s) = modified potentiometer

I Most potentiometers are supposed to have a fairly straightforward
i linear or logarithmic characteristic. This is all right in most applications,
I but sometimes the particular characteristic required is not readily
available. Fortunately, it is not too difficult to obtain various modified
characteristics by adding one or two fixed resistors. Which is what this
article is about.

G. Reinhold

ailoring potentiometers

alektor dece

1979- 12-37

The indications 'lin' or 'log' on a poten-
tiometer (or potmeter, as they are often
called) refer to the intended effect of
moving the wiper along the track.
The resistance measured between the
wiper and one end of the potmeter is
supposed to increase in linear or logar-
ithmic fashion as the wiper is moved
along the track. This type of character-
istic is usually drawn in a graph, where
the resistance between the wiper and
the end of the track is expressed as a
percentage of the total resistance, and
plotted as a function of the wiper
position.

There are applications where the
characteristic is unimportant. Not many,
though. In most cases, the type of
adjustment required dictates the 'ideal'
potentiometer characteristic for that
application. The next step is to find
out whether it exists . . .

The three most common characteristics
are shown in figure 1. The wiper pos-
ition (for either a rotary or 'slider'
potentiometer) is plotted along the
horizontal axis as a percentage of the
total track length: x = 0 corresponds
to the 'low' end (fully anti-clockwise
for a rotary potentiometer) and x = 100
stands for the other extreme position.
The vertical axis gives the percentage
resistance between the wiper and the
'low' end of the track.

The 'linear' characteristic is the easiest
one to draw: it goes in a straight line
from zero resistance at the low end to
maximum resistance at the other. (Note
that this is the theoretical characteristic:
we have yet to find the potmeter that
will give zero resistance at one end . . .).
Potentiometers marked 'log' are
supposed to have a so-called positive
logarithmic characteristic; this is the one
marked 'pos-log' in figure 1. In this case,
the attenuation in dBs varies linearly as
a function of the wiper position — just
the job for volume controls, for instance.
Finally, a less well-known characteristic
is the 'antilog' potmeter ('neg-log') in
figure 1, As can be seen, it is a mirror
image of the normal logarithmic plot;
this can be useful in certain tone-control
circuits, for example.

So much for the theoretical character-
istics. What about real-life potentio-
meter? Well . . . Figures 2 and 3 give the
results for a whole series of logarithmic
and linear potentiometers, respectively.
The linear plots are bad enough, but the
log versions are hopeless!

Add a resistor or two . . .

Fixed resistors can be added between
either or both ends of the potentio-
meter and the wiper, as shown in
figure 4. The result is still, basically,
a potentiometer — but its characteristic
can be weird or wonderful, depending
on the ratio between the total potentio-
meter resistance and that of the fixed
resistor(s).

The possibilities are plotted in a fasci-
nating array of graphs. Figure 5, for
example, shows what can be achieved
by adding one fixed resistor to a linear

o so ► x 100

Figure 3. Linear potentiometers are usually better. The main problems occur at the two

potentiometer. The potentiometer re-
sistance is taken as 100 'units'; the fixed
resistor value can then be given as a
percentage. 'R = 25', say, means that
the value of the fixed resistor is 25%
of that of the potentiometer - a 470 k
potmeter and a 120 k fixed resistor is
a close approximation. In figure 5, the
full lines in the upper left-hand half
correspond to the situation where the
fixed resistor is mounted between the
top of the potmeter and the wiper; the
dotted lines show what can happen
if the resistor is mounted in the position F, «“ r

shown for R3. Note that the two plots

for R = 10 (i.e. one-tenth of the total g
potentiometer resistance) are fairly
close approximations of the anti-log
and log characteristics. This means
that a 4k7 lin potmeter can be modified
to 4k7 log by adding a 470 Si resistor
between the wiper and the Tow' end!

For what it's worth, the theoretical
results of 'padding' a log potentiometer
with one resistor are given in figure 6.

The upper plot for R2 = 10 is a reason-
able approximation of a linear character-
istic. Anybody who feels like trying it
is referred to figure 2 . . .

What about adding two resistors? Why
not. The results (see figures 7 and 8 for
lin and log potentiometers, respectively)
are intriguing , to say the least. In these
plots, one resistor is taken as 25% of
the full potmeter value and plots are
given for various values of the other: the
circuits given in the top left and bottom
right-hand corners correspond to the Fist
full and dotted line plots, respectively. ehai

Finally, figures 9 and 10 give some idea

of what can be achieved if the two -j
resistors have the same value, varying
from 10% to 100% of the total pot-
meter value. Obviously, all these plots
must run through the point where the
wiper is set to half of the total resistance
value. Anybody who wants fine control
in the centre of the potentiometer range
and coarse control toward the ends
should take a look at the plot for
R2 = R3 = 10 in figure 9.

'Add a resistor or two', we said. And
look what happened! Two more things
can happen, not so obvious from the
plots. The total resistance of the
modified potentiometer is no longer
constant, or it is reduced. The circuit
driving it may not like this . . . Also,
the plots given for fixed resistors
between the wiper and the Tow' end of
the potmeter may be taken as dire
warning . . . The same sort of thing
will happen if a relatively high-value
potentiometer is followed by a relatively
low-impedance circuit! H

Figure 7. Using two ret
situation where R2isf
are obtained when R3

12-40-1

ar 1979

Most commercially available VOX units
have the disadvantage that they react to
any sound above a certain level. Back-
ground noises can easily cause a VOX to
switch over to 'transmit', with the result
that any message coming in is 'cut to
pieces’ and may become completely
unintelligible.

Even so, a VOX is a useful little gadget.
It's nice to have both hands free when
transmitting - for making notes, ad-
justing knobs, or pouring a cup of tea. If
only the VOX was better behaved! It
would make life so much easier.

The VOX described in this article may
be the answer to a prayer. It's intelligent
enough to do what it's told - ignoring
chairs scraping on the floor and things
like that.

control switch

next section is the band-pass filter, with
adjustable centre frequency and Q. The
output signal (if any) from this filter
goes to an amplifier stage with a gain of
200. Even fairly small signals will drive
this stage into clipping, so that the
output becomes more like a squarewave
than anything else. This signal is used to
trigger a monostable multivibrator,
which provides an output pulse of (you
guessed it!) adjustable length -from
0.5 to 2.5 seconds, to be precise. The
monoflop is retriggerable; in other
words, as long as trigger pulses keep
coming within the selected period time,
the output will remain 'high' continu-
ously.

Finally, a buffer stage is used to drive
the relay.

vok*<‘ operated

control switch

transmit from
the word 'Go'.

independently variable, so that a

I 'intelligent' VOX. The centre frequency and Q of

Amateur radio operators normally
use a Push To Talk (PTT) button
to switch from 'receive' to
'transmit'. This changeover can
also be done automatically, using
a circuit that detects the speech
signal from the microphone. This
kind of automated PTT button is
usually referred to as a VOX.

Block diagram

The VOX is connected behind the
microphone. This means that any sounds
picked up by the mike are passed to the
input of the VOX. To avoid the disad-
vantages outlined above, the VOX must
be able to discriminate between 'His
master's voice' and other loud noises. A
good solution in practice is to pass the
signal from the microphone through a
fairly narrow band-pass filter. This filter
is tuned to a frequency band that proves
typical for that particular speaker; all
sounds outside the band are ignored.
The output from this filter can be used
to control the PTT switch.

Figure 1 is the block diagram of the
intelligent VOX. The signal from the
microphone is fed to an input amplifier
stage; the gain of this stage can be set
anywhere between x 1 and x 100. The

The circuit

As can be seen from figure 2, the input
impedance of the circuit is determined
almost exclusively by R2: 47 k. This
means that the circuit provides a negli-
gible load, and so it can be connected in
parallel with the microphone amplifier
in the transmitter.

The gain of the input stage (ICIa) is
equal to P1/R1 + 1. With PI at mini-
mum, the circuit gives unity gain; the
other extreme setting corresponds to
x 101. It is advisable to keep the gain
down as far as possible, while still
maintaining reliable operation; too high
a setting will not make the circuit react
any faster, but it will increase the danger
of unwanted background noises getting
through! LI and Cl are included to
block high frequency input signals - the
circuit is intended for use with a trans-

mitter! The gain of the input stage can
be varied over such a wide range that
the type of microphone used is not
really important.

The following three opamps, ICIb..

. . ICId, are connected as a 'state
variable’ filter. The tandem poten-
tiometer P2a/b sets the Q of the filter -
the relative width of the pass-band, in
other words. The Q can be varied
between 1 and 50. The other pair of
potentiometers, P3a/b, adjusts the centre
frequency. By manipulating P2 and P3,
the filter can be tailored until it corre-
sponds to the desired voice band.

The output from the filter (pin 8 of
ICIc) is taken to a single-transistor
amplifier stage (T1), and from there to
the trigger input of the monostable
(type 4528). The latter, in turn, drives
the transmit/receive relay via T2. The
period time of the monostable is deter-
mined by P4, R20 and C7. With the
values given, any period between 0.5
and 2.5 seconds can be set. If desired, a
different range can be obtained by
modifying the values of any or all of
these components.

A stabilised 12 V power supply must be
used. The current consumption will
depend on the relay more than anything

else; a 500 mA supply will normally be
more than adequate. The opamps require
a symmetrical supply, and this is
obtained by including an 'artificial
centre-tap', consisting of T3 and T4.
Obviously, if a symmetrical +/— 6 V
supply is available, this part of the circuit
(T3, T4, D2, D3, R21 and R22) can be
omitted. The capacitors C4, C8, C9 and
CIO should be included, no matter what
type of supply is used.

type of transistor used for T2 will
depend on the relay. It may be necessary
to use a BC141 — possibly even with a
cooling fin. Note that the base current
to this transistor is limited to about
0.5 mA, so if a really 'heavy' relay is
used, the transistor must have sufficient
current gain or else a Darlington pair
must be used. M

Construction

No p.c. board was designed for this
circuit - amateur radio operators are
usually sufficiently experienced to do
without! Screened cable should be used
for the connection to the microphone,
and the shorter the cable the better.

It's always a good idea to use sockets
for the ICs. There are no other construc-
tional points that require special
attention. As far as the choice of
components is concerned, only two
points are worthy of note. The two
tandem potentiometers should prefer-
ably be selected for good tracking
between each pair — this makes the
filter easier to adjust. Furthermore, the

579 Kingston Road. Raynes Park.
London, SW208SD.

Tel.: 01-5430077

yourself' course in microprocessing.

The Nanocomputer uses a calculator-style
hand-held hexadecimal data input and display
station. Using the Nanocomputer, a student is

computer systems, and a conversion kit and
additional interface boards allow the system
to be expanded to a full-scale CLZ80 micro-
computer.

For tutorial use. the Nanocomputer can be
interfaced with a low-cost audio cassette

Plastic control knobs

A new range of h

iduced by Argo Electroi
Components Ltd. These control knobs, wh
have an attractive modern design, are p

d 19m

r. The b<

and typesetting

is offering

Cambridge Micro Computers L
one-day 'hands-on' introductory
on the Nanocomputer at a cost of £ 40 (plus 1
V.A.T.); the course is offered free of charge
to every purchaser of a Nanocomputer.
Cambridge Micro Computers Limited.
Cambridge Science Park,

it black phenolic
has fine knurling
:ontrol. They are
having diameters
% inch. Coloured

iber 1979 - 12-43

Frequency counter

A portable 100 MHz frequency counter, the
MAX-100, equally suited to the needs of
hobbyists or professional electronics users is
available from Continental Specialties Corpor-
ation. The instrument's range can be extended
to 500 MHz by the use of the PS-500 pre-
supply options make the MAX-100 extremely
versatile for laboratory, workshop or field

The unit gives continuous readings from
20 Hz to a guaranteed 100 MHz, and has a
0.6 inch high 8-digit LEO readout. The input
is sampled for one second with 1 V* second
updates, and the crystal-controlled timebase
has an accuracy of 3 parts in 10 6 . A high-
sensitivity preamplifier gives readings from
signals as low as 30 mV, and the input is
diode protected to a peak of 200 V. The
extreme left-hand digit flashes automatically
when the input signal exceeds 100 MHz.

The MAX-100 can be used with a clip-lead

cable tap (for use with UHF connectors),

and an optional 'mini-whip' antenna is avail-

able for use where direct coupling is not
feasible. A choice of four power sources is

available: internal battery; AC mains (110 or

220 V); a mobile 12 V DC supply; or an

external DC supply. Battery -c ha rger/elimin-

ators are available for AC or 12 V supplies.

Applications for the MAX-100 frequency
counter and the PS-500 pre-scaler cover all

a detailed applications brochure is provided.
Continental Specialties Corporation,

Shire Hill Industrial Estate,

Saffron Walden,

Essex. CBI I 3AQ.

Tel.: (07991 21682

(1342 M)

Texas bubbles

A family of physically and electrically inter-
changeable magnetic bubble memories with
the largest capacity device having one-million
bits of storage has been announced by Texas
Instruments. Bubble domains for these new
memories are two-micron diameter.

The first two devices — to be available as
board-level systems - are the T1B1000. a
binary megabit device organised as 51 2K x 2;
and the T1B0500, a half-Megabit device with
512K x 1 organisation. In the second quarter

compatible with the two larger devices will be
available. The family approach will allow
designers to vary system storage capacity by
interchanging the bubble devices.

The T1B1000 has a maximum non-volatile

storage capacity of 1 ,229,400 bits. A portion

of this storage is used for redundancy handling

and error-correction. The available data-
storage capacity with error-correction capa-
bility is a full 128K bytes. It uses a block
replicate architecture and is organised as two
identical sections of 51 2K bits each. There are
300 minor loops per section with 2049 bits
per loop. A page of data consists of bubbles
from 256 of the loops. Of the remaining loops

and as many as 26 are allowed to be defective.
At 100-kilohertz bubble field frequency, the
T1B1000 has an access time of 11.2 milli-
seconds. Data rate is 160K bits per second.
All members of the new bubble memory
family are packaged in a 24-pin 3.3 x 3.56 cm
package with pins on 100 -mil centres.

Texas Instruments Limited.

Menton Lane,

Bedford, MK4I 7PA.

Tel.: 0234 67466

(1339 M)

Large area clock/panel meter LCDs

A series of large area liquid crystal displays
has recently been announced by the Opto-
electronics Product Group of Fairchild
Camera and Instrument (UK) Ltd. They can
be supplied in 3% and 4 digit versions and are
suitable for clock and digital panel meter
display purposes. Digit height is 0.5 inches.

im

with logic circuitry.
GRA YHILL. INC.,

561 Hillgrove Avenue,
La Grange. Illinois 6052

compatil

Fruity box?

A new moulded

LCD thermometer

A pocket-sized LCD F

UK 16 - elektor de

1979

□ B3«
• •••

The CBP-1 case, which measures 1.75x5.63x
7.75 inches (44 x 143 x 197 mm), comes
complete with a battery compartment cover,
a red transparent plastic front panel, four
rubber feet, all necessary screws, a power jack
socket and two fitted switchplates.
Continental Specialties Corporation.

Shire Hill Industrial Estate,

Saffron Walden.

Essex CB1 1 3 AO.

Tel.: 1 0799 ) 21682

(1348M)

UHF modulator

Astec have announced the availability of a
new UHF high performance modulator, type
UM1 286. The modulator is intended for use
in computer graphics or VCR applications.

The vision carrier is pre-tuned to channel E36

oscillator may be pre-tuned to 5.5 MHz or

6.00 MHz. Separate balanced modulator cir-
cuitry is used to provide excellent linearity

products. The colour sub-carrier/sound sub-
carrier beat product is -55 dB with respect to
carrier thus resulting in interference free

10 20 30 40 50 60 70

The UM1 286 is designed to operate from a

5.0 V • 10% supply and consumes only

9.0 mA. It is housed in a robust screened box
measuring 71 mm x 37 mm x 20 mm that can
be PCB mounted. R.F. output is via a co-ax

Astec Europe Ltd.,

4a. Sheet Street.

Windsor, Berkshire.

Tel.: 1075351 55245

(1353 M)

also generate tunes from data held in external
'plugged-in'.

The AY-3-1350 may operate in a number of
different modes, making it suitable for a wide
variety of different applications. In a door-
chime for instance, it can be connected to
play anyone of 25 pre-selected tunes from the
front door bell push, with one of 5 tunes
from the back door. In addition a third bell
push can be wired to play a simple chime.
All the tunes would be selected by switches

The device also has applications in low cost
paging systems, where key personnel are each
allocated one tune. A brief tune played over
loudspeakers in a noisy factory would be
much easier to recognise than a spoken name.
To conserve power, the circuitry may be
connected so that when a bell push is acti-
vated it 'powers-up', plays a tune and then
automatically powers down. Releasing the
button and repressing would cause either the
same tune to be played again, or the next
tune to be selected, depending on the precise
operational mode of the device. Alternatively,

again until the button is released. The pitch,
tone and speed of tunes may be indepen-
dently set by simple external components.
These may be either preset or brought out as
potentiometers as a user control. What now

1-4 Warwick Street.

London, W1R 5WB.

Tel. : 01-439 1891

((1343 M)

Melodic chip

A new single-chip tunes synthesizer, which
can be programmed to generate up to 28 dif-
ferent tunes, has been introduced by General
Instrument Microelectronics Limited. Desig-
nated the AY-3- 1350. the 28-lead N-MOS

supply, and is suitable for use in toys, musical
boxes, doorchimes and other ’novelty'
products. The chip is based on a standard
GIM microcomputer circuit and will normally
be mask-programmed during manufacture. Its

tunes selected for their international accept-
ance. The standard circuit is pre-programmed

this may be altered to suit the application.
It is possible for instance, to program just a
single tune of up to 251 notes. Thz chip can

Modifications to
Additions to
Improvements on
Corrections in

Circuits published in Elektor

Digital rev counter

Elektor 54, October '79, page 10-15. The !
binary outputs of IC3a and IC3b have been
reversed on the circuit diagram. The outputs |
from IC3a should be taken from pins 3, 4. 5
and 6 (A, 8. C. D.) while the outputs from I
IC3b should be taken from pins 11. 12, 13 I
and 14 (A, 8. C. D.),

Metronome

elektor december 1979 - UK 17

This kit

for new subscribers

(offer valid until December 31st 1979)

A complete kit (including loudspeaker) for the simple sound effects
generator will be sent free to all new subscribers. This offer is valid for all
applications postmarked up to and including December 3iste 1979.

As the name suggests, the simple sound effects generator will produce a
range of sounds from that of an American police siren to one closely
resembling the 'twittering' of birds.

You can become a subscriber by filling out the order card included in this
issue and including the text 'subscription and free kit'. The kits will be sent
out at the end of January.

elektor

up-lo-dale electronics fa lab and leisure

Elektor House
10 Longport,

Canterbury, Kent.
Tel. (0227) 54430

elektor decen

-UK 25

A range of 3’/2 digit LCD multimeters
offering high precision and extended
battery life. All feature 0.5" LCD readout
with battery low' warning, inputs
protected against overloads and tran-
sients, Auto- polarity, Auto-zero, rugged
ABS cases and a full 1 -year warranty.

The LMM-200 is a compact handheld
multimeter with 0.5% basic accuracy and 1 5
different ranges. It measures voltage from
0. 1 mV to 500V, current from 0. 1 uA to
2 Amps, and resistance from 0. 1 A to
2MA.

The LMM-2001 is an identical
instrument but with 0.1% basic accuracy.

The LMM-100 has an adjustable
handle, a 2,000 hour battery life and is ideally
suited to field or bench use. It measures
voltage from 0. 1 mV to 1 KV, current from
0. 1 u A to 2 Amps, and resistance from 0. 1 A
to 20M A . 0. 1 % basic accuracy.

The NEM/ /Marshall’s 79/80 catalogue
is just full of components

and that's not all . . .

60 pages are details and prices of the complete range of
components and accessories available from Marshall's
These include Audio Amps. Connectors. Boxes. Cases, Bridge

Rectifiers. Cables, Capacitors, Crystals. Diacs, Diodes Dis- 5T51 1

plays. Heatsinks I Cs. Knobs LEDs. Multimeters. Plugs. "m Iw

Sockets. Pots, Publications. Relays. Resistors. Soldering I fr ,■*

Equipment, Thyristors. Transistors. Transformers. Voltage 2 ' ”

Regulators, etc . etc 9 '

Plus details of the NEW Marshall s budget' Credit Card. We 2

are the first UK component retailer to offer our customers our S - A—/

own credit card facility. ’S' ” V ,^^5, w

Plus — Twin postage paid order forms to facilitate speedy By |

Plus — Many new products and data. Bjf *». _ f

Transistors, Resistors and many more 9 B^ v

Catalogue

House, Kingsgate Place. London NW6 4TA. Also avaite ble