This is only a preview of the October 1989 issue of Silicon Chip. You can view 46 of the 112 pages in the full issue, including the advertisments. For full access, purchase the issue for $10.00 or subscribe for access to the latest issues. Articles in this series:
Articles in this series:
Articles in this series:
Articles in this series:
Articles in this series:
Articles in this series:
Articles in this series:
Articles in this series:
|
REMOTE CONTROL
By BOB YOUNG
The beginnings of remote control
Bob Young has had over thirty years experience as
a designer and manufacturer in the field of remote
control, particularly radio control. He is the
proprietor or Silvertone Electronics, which has
been notable for radio control products for many
years. In this, his first article in SILICON CHIP, he
introduces the subject of radio control.
When Leo Simpson approached
me to do a monthly column on
remote control, I was a little hesitant at first. While I have written
quite a lot for radio control model
aircraft magazines during my 26
years as a radio control manufacturer, I have never had the chance
to discuss all aspects of remote con-.
trol, either model, industrial or ·
commercial.
, Not that this presents any real
problem, for during those years I
have undertaken projects covering
:an of those aspects and many more,
including some low key military
applications.
I Perhaps by way of introduction
then, a brief description of some of
the more interesting projects in
remote control I have been involved
in would be in order. But first to the
name.
Originally, or as the Bible puts it,
" In the beginning", remote control
was generally accomplished by the
"wireless" transmission of one or
more frequenies in the electromagnetic spectrum. This was the
system used by Tesla in his model
submarine, the very first R/C [radio
control) model, demonstrated in
1896.
Tesla used two separate carrier
frequencies and an AND gate
[another of his inventions). Thus, as
a general rule, remote control
became "radio control". Technology has rendered this term obsolete with the development of
ultrasonic, infrared, laser and long
range wire-guided systems, and a
host of other mediums of transmission.
To further complicate things we
now have radio controlled people,
as many service vans tell us on
their advertising panels: "Radio
Controlled, 24 Hour Service".
Thus our field of interest is better
described as the ''Remote Control
of Mechanical Devices" or if a vehicle, the more common "Remotely
Piloted Vehicle" or RPV. This now
leaves the way open to employ any
transmission medium we choose.
The early days
It appears that Leo remembers
me from the early days at Silvertone Electronics. Around 1965 or
thereabouts I developed a single
channel TX (transmitter) and Rx
[receiver) for use in models.
This early super-regenerative receiver from Silvertone
Electronics was a hybrid unit employing an XFY34 valve
and three germanium transistors.
4
SILICON CHIP
Before servos became available, model aircraft were
controlled by rubber-driven escapements which gave left
& right rudder control.
This wa:s very popular with radio
control modellers of the time and
was regarded as state of the art. It
had a one valve super-regenerative
front end using an XFY 34 to give a
bandwidth of 300 to 400kHz.
Following the front end were three
transistors, the final stage driving a
miniature relay, which in turn
drove an escapement.
The escapement was a rubber
band driven actuator which was used to turn the rudder; the only control available.
It was operated in sequence: first
left, then right. There were no steps
in between. If you forgot which you
used last, left or right, you just
found out the hard way. All this
was driven by a 1.5V penlite cell
and a 22.5V hearing aid battery
supplying the HT (high tension).
Oddly enough, these sets were
extremely reliable and gave good
results in practice. We did some
very satisfying flying with them and
the only reason we stopped servicing them was that hearing aid
technology zoomed ahead and
valves and 22.5V batteries became
difficult to obtain.
Over the years we gradually
developed these sets into a fully
transistorised unit which was much
nicer to handle, with its single 6V
battery and no delicate valve filament to worry about. And of course
super-regen gave way to superheterodyne.
Electrically-driven servo units eventually took over from
escapements. This early unit is Japanese made and is big
& bulky by modern day standards.
Reed units
Single channel receivers gradually gave way to tuned reed units.
These were a real eye-opener and I
tend to feel sorry for any modeller
who missed this era. Here we used
a bank of 10 or 12 tuned reeds to
filter out the audio tone modulation,
thus giving true multi-channel
operation. Heaven had arrived on
Earth at last.
These 12 tones had to be tuned
inside a full octave to avoid har-
Tuned reed units were used to fiter
the audio tone modulation on the
receiver output to give multi-channel
control. This unit employed 10
separate reeds.
monies and placed extraordinary
demands on the tone generator, for
they had a bandwidth measured in
just a few Hertz (typically 4-BHz), or
as we used to say in those days, a
few "cycles per second".
If there was any drift in the
receiver, it would be all over for the
model. Modern solid state oscillators have no problem achieving
this degree of stability but in the
late 50s it was almost impossible to
achieve in a portable unit.
The reedbank itself was very
cleverly constructed right from the
very beginnings and changed little
in the 10 or so years they were in
commercial production. Practical
results were poor though, until Bob
Dunham of Orbit in America, produced a toroid stabilised tone
generator which revolutionised the
sport of radio controlled model
aircraft.
Reliability became accepted as
the norm and the really good flyers
produced results which were
almost indistinguishable from those
seen from modern proportional control units in use today. This was
quite a feat, keeping in mind that
we only had on or off servos, no proportional. You learned to pulse the
controls for half throw.
The other major problem and one
that still shows up today in modern
remote control systems, was the
lack of simultaneous control.
Modern garage door units, for exOCTOBER 1989
5
This view shows the works of a modern radio-controlled model aircraft. The multi-channel receiver at left drives a
number of servo units for full control over throttle, ailerons, flaps, elevators, rudder & nose gear steering.
ample, now give up to 10 separate
switching channels (in remote control a decoded data stream is called
a channel) but not simultaneously.
This is no problem when switching
on lights but in a model aircraft
travelling at 50 metres per second
there is no time for queued
commands.
Also some manoeuvres call for
the application of three or more
controls simultaneously. Thus a
successful remote control system
for vehicular work must be capable
of simultaneous multi-channel
transmission with a response time
of less than 100 milliseconds.
Modern digital systems can
deliver 32 channels very easily in
less time than this. Interestingly
enough, despite this being faster
than the human response time (200
milliseconds), you can still notice
the slight lag in control response.
For competition flying, 50ms is the
absolute maximum cycle time with
the typical figure being around 14
to 20ms.
To arrive at this modern system
was a constant technological battle
every inch of the way. I started by
producing a super-regen valve/
transistor hybrid unit, using 10
miniature relays and a 22.5V HT
6
SILICON CHIP
battery, which had to be large
enough to drive the relays. The
whole airborne system including
batteries came out at about 22
ounces (this was before the days of
metrics, remember) and it was big
and bulky.
To give simultaneous control of
two channels, we mixed two tones
which halved the power to each
reed and made tuning even more
difficult. Reed gaps had to be adjusted and in the early days before
transistorised amplifiers to drive
the servo motors, you had the inductive load of the relay on the reed
points which resulted in pitting.
The relay contacts had to be adjusted and kept clean as well.
No instructors
On top of all of this you had to
teach yourself to fly. No RCAS
(Radio Controlled Aircraft Society
of NSW) qualified instructors on
flying fields in those days. Ah ...
they were the good old days.
But we really did enjoy it. The incredible difficulties that each of us
faced in bringing home the model
intact after a day's flying generated
a sense of camaraderie no longer
seen these days.
And the sense of achievement -
it was indescribable. However
above all of this the thing that sustained our interest was the constant dream of the big one:
simultaneous proportional control.
When we got that, we could build
scale Spitfires with retracts.
To cut a long story short, we did
eventually get just that and much,
much more but very few of us ever
built that scale Spitfire with
retracts. Such is the stuff that
dreams are made of.
My first production proportional
set was a fully transistorised,
5-channel PPM (Pulse Position
Modulation), superhet unit with an
airborne weight of 450 grams (I
measured it as 16oz but I've converted it for the benefit of those
who have forgotten ounces).
I could not believe that I had
finally achieved true three dimensional flight with five simultaneous
controls and the freedom and accuracy of manoeuvres that accompanied this amazing technological
achievement. And it was amazing;
make no mistake about it. I still
look in wonder at the incredible
cleverness of the men who made all
of this possible.
I dislike intensely the modern jaded consumer who has everything
This modern radio-control transmitter uses pulse code modulation (PCM)
techniques to provide simultaneous 7-channel control. The aircraft is flown by
manipulating two joystick controls on the front panel while optional channel
mixing is provided by the front-panel switches.
and never stops complaining, or
worse still, never stops to consider
the magic of it all. They were heady
days, for we were blazing trails
where no.man had gone before. But
the most amazing thing of all, from
my point of view as a technician,
was the lack of tuning resulting
from the use of .Pulse Position
Modulation (PPM). It was unbelievable.
These days it seems incredible
when I look back at just how difficult it all was. When I look at the
modern control unit with microprocessor control, inbuilt memory,
voltage monitor, liquid crystal
display and everything else that
opens and shuts, it takes my breath
away.
It all seems so long ago and I
guess it was, for I flew my first R/C
model in 1955. Yet even today I
have not lost my fascination for the
concept of wireless control of model
aircraft. To me, it is the ultimate in
technological achievements and yet
Tesla was sailing his multi-channel
radio controlled model submarine
long before Marconi ever achieved
his reputation. That man was absolutely incredible. But that's
another story.
Other applications
As a result of my experience in
radio control, I gradually became
interested in the wider applications
and thus began to involve myself in
industrial and commercial projects.
Some of the early ones included
radio control of flag poles, the window washing robot for the Sydney
Opera House, a full size Volkswagen 1600 TLE, toasters, chairs
and so on. You name it, I have done
it or at least, thought about doing it.
It was all good fun but very
dangerous at times. Hanging off a
flag pole in a high wind, 22 storeys
above the ground is not my idea of
the best way to spend an afternoon.
Flying military target models, in
which people are constantly putting
bullets through receivers and batteries and servos, and which
results in aircraft digging furrows
near where you are standing, is
another.
Nearly running over a curious
policeman with a driverless Volkswagen rates low on the scale also.
And of course there was the day we
had to test the droparm switch on
the Opera House window washing
robot, only to see it skid to a stop,
totter at the end of the glass roof,
and vanish over the edge of a 20
metre drop, nearly taking an apprentice with it.
When we peered cautiously over
the edge, there was 90 kilograms of
robot dangling in space at the end
of a rubber hose, with a quick
disconnect fitting the only thing
between it and oblivion. All of this
on a wet, glass roof, pitched at a
very unpleasant angle.
But all in all, I would have to be
listed among those who really did
enjoy my work, that is until the
economic system fell over in
Australia and manufacturing became a dirty word, particularly in
consumer electronics. Thus, in
keeping with all of the survivors,
we fled to the niche market.
Here we found much of interest
including radio control of military
targets, robots for film companies,
radio controlled concrete placement booms and so on.
So from the occasionally dangerous we now moved into the really
scary: machines six storeys high,
fullsize pilotless aircraft that fly
over the horizon, robots that require 24 channels of simultaneous
control to do the job (and operators
with only two hands), trolleys
weighing 80 tonnes, and much
more.
New technology
Now we began to call upon the
very best technology had to supply.
The new techniques include pulse
code modulation (PCM), microprocessor coding and decoding, input noise algorithms, multiple
OCTOBER 1989
7
'
I
I
This remote-controlled model ship
was designed by Nikola Tesla in the
mid 1890s and relied on a two-carrier
system. Another of his designs was
submersible.
verification of valid data, stringent
fail safe requirements and many,
many more safety features.
Which leads us nicely into the
item at the very heart of R/C applications. That is, just how applicable is R/C and the technology
involved, to your application? For
example, PCM (Pulse Code Modulation) is ideal for industrial control
of machinery but can be too rigid
for model aircraft work.
This sounds a little confusing so
let me explain. When we began flying proportional control in 1964, all
of the first generation PPM sets had
a built in "fail safe". Thus the incoming pulse train was examined
and any spurious pulses a hove a
limit previously defined as tolerable
resulted in the set being shut down,
the servos neutralised and the
throttle run back to low.
All this sounded highly desirable
until the practical results came in.
Very quickly fail safe was defined
as " That circuit which neutralised
the controls on the way to the
crash". What the designers had
forgotten was that the modern
aerobatic model aircraft was doom8
SILICON CHIP
ed once the controls were neutralised. Cutting the throttle wa s a very
big help however and reduced
damage significantly.
Second generation systems did
away with the failsafe system and
relied upon random noise to
average out the controls, or even
the odd snatch of restored control,
to keep the aircraft flying until the
interference passed. Sometimes the
interference did not pass and you
just fought the model all the way to
the ground but at least you could
fight.
Once that failsafe locks out, it' s
all over, that is unless the model
has good inherent stability. Then
you just lost it, for it flew away. I
once lost a model in Sydney and had
it returned from Leeton (in Victoria)
months later.
Flying through noise and interference was the system used until recently and served us well for
over 20 years, giving excellent
reliability in use. Someone then r ediscovered the failsaf e when PCM
sets arrived on the market and even
now after several years, the correct
way to use PCM sets is still causing
confusion in the model aircraft
field.
The big difference between the
two systems is that PCM uses
microprocessors for encoding/
decoding and the old PPM (Pulse
Position Modulation) system uses
logic. Thus, you must choose your
technology very carefully to suit
your application. At least the
modern PCM set now gives the option to configure your own fail-safe
parameters, even to switching it out
completely if desired.
The situation in industrial control
is quite different. As zero deviation.
from the norm is required, any interfe rence can be made to shut
down the system immediately until
the signal is once again valid. PCM
is ideal in this application.
Wired controls
But is R/C the correct technology
for your application? There is
nothing cheaper and more reliable
than a piece of wire (except
perhaps two pieces of wire in
parallel) and designers of wire
guided missiles realise this only too
well.
The Argentinian Cruiser " General Belgrano" was reportedly sunk
from a range of 45 kilometres by
two Mark 24 "Tigerfish" wire guided torpedoes. Now that is a lot of
wire but it is one way of ensuring
the torpedo is not interfered with
on its way to the target.
As it turned out the report was
incorrect and the torpedo was an
old Mark 8 non guided. One interesting sidelight here is that the
Tigerfish was thought to be unsuitable for sinking a surface
vessel.
By introducing a radio link, you
increase the complexity and cost
and reduce the reliability. There
are many ways to make the link
continued on page 111
This early single-channel superhet receiver was designed for use with motordriven escapements or could be used to drive a single servo.
Notes & Errata
Touch Lamp Dimmer, June 1989:
As noted in the article, the revised version of the SLB0586 does
not require diodes D1 and D2 but
they will not affect circuit operation if they are left in. However,
now that the revised chip has
been released, as SLB0586A, it
has been found not to work in the
circuit as published. To make it
work, two components must be
changed. The 680k0 resistor
must be swapped for a 0.33µF
capacitor while the .0022µF
capacitor should be swapped for
a 1ookn resistor. These components will fit without any
modifications to the PCB pattern
being required. If your kit has
the original SLB0586 IC, no circuit changes are necessary.
Studio Series 32-Band 1/3 Oct-
ave Equaliser, March, April
1989: one of the four 220µF electrolytic capacitors shown on the
main equaliser in the wiring
diagram (Fig.1(b} on page 48 of
the April issue} is reversed in
polarity. The capacitor in question is connected to pin 4 of IC2.
Ultrasonic Car Burglar Alarm,
July 1989: the wiring diagram on
page 63 shows one side of the
siren connected to chassis. It
should connect to + 12V, as
shown on the circuit on page 61.
Garbage Reminder, August
1989: the .047 µF supply bypass
capacitor shown between IC6
and IC7 on the wiring diagram of
page 50 is shown as 0.1µF on the
circuit of page 49. The value is
not critical though and either
0.lµF or .047µF is OK.
Advertising Index
Allied Capacitors .... .... .. .... .... 65
Altronics ...... .. ...... .. ....... 46-49
Arista Electronics ............. .. .. 1 7
Banksia Information Tech ...... 89
Board Solutions .. .... .. ... .......... 9
Dauner Electronics .. .. .. .. . .. ... 1 3
David Reid Electronics ..... 56,57
Dick Smith Electronics ..... 68-73
Electronic Solutions .. ..... . 14, 15
Elmeasco .... ..... .. .. ..... .... .. OBC
Geoff Wood Electronics ..... .. IFC
Harbuch Electronics .. ... .. ... .. . 94
Hycal Instruments .. .... .. .. ...... 13
Jaycar Electronics .. .. .. .... 36-39
80-83
J.V. Tuners .. ... ............. .. ..... 94
Kepic .... ... .. .... ... .. ........ .. .... . 95
Novocastrian Electronics .... .. 1 9
PC Marketplace ....... .. ..... ..... 31
Pelham .. .. .. .................... .. . 112
Power-Sonic Corp .... .... ..... . IBC
Philips Test & Measurement .. 89
RCS Radio ... .... .. ...... .. .. .. ... 102
Rod Irving Electronics .. ... 96,97
WIA .. .. .. .. .... .. .... ............... 103
tell me where I can get one?
SILICON CHIP is a great magazine.
I like the Vintage Radio section and
the Serviceman's Log. What happened to the Technology Letters?
(A.R., St George, Qld}.
• Tunnel diodes have become
very rare. In the 1960s they were
regarded as the wonder diode but
very largely they look to have been
a solution looking for a problem.
Their special claim to fame is that
they have a "negative resistance"
characteristic which enables them
to be used as amplifiers or
oscillators right up to microwave
frequencies.
Now, most of the applications for
tunnel diodes appear to be fulfilled
by more conventional devices such
as microwave transistors and Gunn
diodes.
General Electric was at one time
the major manufacturer of tunnel
diodes but this is no longer the case.
However, we have been in contact
with the agents for General Electric, GEC Components, and they
have indicated that they may be
able to source tunnel diodes, depending on the type number and quantity required. If you want to make
further enquiries, you can contact
GEC Components by phone on (02}
887 6222 or by fax on (02} 805 0272.
Remote Control ctd from page 8
secure and military designers are
constantly searching for better
ones. Some of the more complex included frequency hopping, chopped
chirp, exotic encoding and above
all else, making the vehicle as intelligent as possible, so that it can
perform its task with as little outside assistance as possible.
All of this and more will be
discussed in columns to come.
There'll be simple explanations
covering the installation of the
equipment into models, care and
feeding of nicads, and the correct
use of servo arms, plus the exotic. It
should be a lot of fun. See you next
month.
~
The Way I See It ctd from page 90
ly turned up in disposals stores for
50c each or three for a dollar!
"As far as I could see, the only
thing the power transformer did
was to operate the pilot light. A
single wire ran to the nest of pots
and another to the circuit board,
PC Boards
Printed circuit boards for SILICON
CHIP projects are made by:
• RCS Radio Pty Ltd, 651
Forest Rd , Bexley, NSW 2207.
Phone (02) 587 3491 .
• Jemal Products, 5 Forge St,
Welshpool, WA 6106. Phone
(09) 350 5555 .
• Marday Services, PO Box
Avondale, Auckland,
NZ. Phone 88 5730 .
19-189,
being soldered, of all things, to the
piece of PCB foil spelling out the
manufacturer's part number!
"What does a friend do in this
situation? Does one protest that the
$300 miracle healing machine is
just a con and refuse to proceed
any further with the farce?
"Knowing the lady as I did, I
replaced the crunched diodes with
three new ones and assured her
that they were a close equivalent to
the originals. And of course, as
soon as she turned it on, she noticed
the difference!"
K.W. observes that "thousands
of electronic enthusiasts have pulled apart discarded computer
boards. How many can claim that
they've actually repaired one?" ~
OCT0BER1989
111
|