This is only a preview of the February 1990 issue of Silicon Chip. You can view 51 of the 104 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:
|
REMOTE CONTROL
By BOB YOUNG
Transmitter encoders & black
•
wire syndrome
All remote control transmitters use some sort
of encoding which then must be decoded at the
receiver. This month, we start on the subject of
encoders and then pass on to practical matters
such as the effect of battery fumes on PC
boards.
If we think of an RIG set in human
terms, then the transmitter encoder
is the "brain" that digests the incoming information from the control sticks and knobs. This raw information, in the form of parallel
varying voltages, resistances,
capacitances or inductances, must
then be processed and converted
into serial code format or "digital
speak", if you will pardon the
Orwellian jargon.
In the PPM (pulse position
modulation) system, the Rx and servo electronics are all passive
elements and derive their timing
from the master clock in the encoder, thereby making it possible to
run as many receivers as you wish
from the one transmitter.
If we carry the human analogy
further, the transmitter RF section
then becomes the "voice", transmitting the processed serial control
data over great distances.
There is an interesting aside to
this analogy, for in the "Book of
Zohar", an ancient Jewish book
said by some to contain the secrets
of Moses, there are passages in
which the "The Voice" is quite carefully distinguished from "The
Speech". It appears that the ancient priests converted speech into
"The Voice" . "The Voice" was then
described as having an action
lncoder Output
similar to that of water in a pond
when a stone is dropped into the
middle of it.
The stone disturbs the water and
sends out waves which reach the
distant shore. This is quite an up-todate description of the effect of
Hertzian waves. So ends this
month's sermon.
First generation digital sets used
a 2-transistor monostable multivibrator as the pulse generator.
These very quickly gave way under
the unrelenting pressure of cost to
the "half shot" circuit - see Fig 1.
This was the standard pulse
generator circuit found in almost
every set produced between 1969
and 1985. It is still very common
and is very reliable and stable.
This is an 8-channel halfshot encoder
board which is actually half of the
encoding circuitry used in the
16-channel transmitter shown in our
December issue (page 15). This
transmitter was developed for John
Grant of Custom Model Cars.
r------ ---, r--------,
Audl1ary fl
Auxiliary 12
I
I
._
________ J
14
SILICON CHIP
L--------'
Fig.1: this early R/C transmitter circuit uses a number of "half shots" (eg, Q3,
Q4, Q5) to perform encoding of the control settings. Qt and Q2 provide the
clock circuit while Q7 and Q8 provide the modulator stages. The half shot
pulses are fed to the modulator via the associated diodes.
Eventually, this circuit gave way
to the unrelenting pressure for
gadgetry and sales gimmicks.
The half shot is not very flexible
when it comes to the addition of ancilliary control effects such as servo reversing, half rate, exponential
etc. Thus we saw the advent of the
discrete symmetrical encoder,
usually balanced around a midpoint reference voltage. This allowed the servo control voltages to be
reversed without retrimming the
neutrals.
This development was eventually
incorporated into standard encoder
chips, the Signetics NE5044 being a
good example.
This type of chip is very flexible
and allows all kinds of gadgets to be
incorporated into a model Rf<:;
transmitter. There are drawbacks,
of course, and two of note are RF inter£erence from the Tx RF section
and shifts in neutral due to contamination on the printed circuit
board.
This latter problem shows up in
some older "half-shot" sets quite
regularly, especially those using
high impedance IC encoders. The
most common source of contamination is the gas vented from overcharged batteries. This produces
quite pronounced shifts in neutral
positions with temperature or
humidity changes.
If this occurs, scrub the Tx PC
board with methylated spirits and
blow dry with a warm hairdryer
(not hot). When the neutral shift is
eliminated, spray the board with
Electrolube Clear Protective Lacquer (CPL200) or something similar.
It really is poor design procedure
to include the Nicad battery inside
the main electronics chamber or
housing. In all of my new industrial
designs, the battery is enclosed in a
separate housing and is removable
for charging. The gas from venting
batteries does terrible things to
electronic components and leaking
electrolyte will eat its way through
copper circuit boards and aluminium cases in very short order.
As is always the case, do not
leave batteries (dry or nicad) in
elecronic equipment if it is not being used. If the transmitter has a
removable battery, then take it out
for charging. This procedure may
be somewhat tedious but is well
worth the effort in the long term.
The "black wire" syndrome
An associated problem is the
dreaded "black wire" syndrome,
the bane of all modellers and probably all users of nicad batteries.
Nobody seems to know just what
causes this problem but on the surface it appears to be a process by
which gremlins, gradually and using great stealth, exchange the copper in the negative lead from the
battery for some kind of black garbage. The transformed conductor is
shiny blue-black in colour, devoid of
any tensile strength and with a con-
ductivity approaching that of air.
The transmitter thus suffers, in
human terms, coronary occlusion,
with its supply of much needed electrons gradually strangled off. The
results can be horrific but milder
effects range from from low
transmitter power to lack of
decoupling. The latter can give rise
to RF inter£erence in the encoder
and this can cause shifting or jittery neutrals.
Eventually, a complete breakdown will occur. If this happens
during a flight, then it's a serious
matter indeed.
What to do
Fortunately there are very simple
solutions to this problem. Standard
insulated hookup wire in which the
conductors are tinned (ie, normal
hookup wire in which the strands
are bright and shiny rather than
plain copper) will slow the process
remarkably. The effect shows up
mainly on plain untinned copper
conductors. Keeping the battery
ends moist with CRC 2-26 will also
help.
You should examine all battery
wiring every 6 months or so. Pull
back the insulation and examine
the copper conductor. If there is
"black wire", the lead will probably fall off in your hand. Re_Q_lace
all leads showing the faintest
traces of this black contamination.
It usually stops at the switch, but
not always.
A word of advice here: use an old
soldering iron tip to disconnect the
lead. The black stuff will contaminate the tip and render it
FEBRUARY1990
15
VIRTUALLY NO UP - ALL OOWN
-UP
45'
--....
OFFSET
ZERO REFERENCE
SERVO 0/P ARM
PUSH ROD
TD CONTflOL
SURFACE
Fig.2: the servos should be set up so that they give equal travel about
the neutral point. If this is not done, the control will move faster in one
direction than the other.
useless for soldering thereafter.
If the conductor is clean and
bright but plain copper, then
replace the lead anyway. If it is
silver (plated) then you have few
worries, but you should still check it
periodically. A leaking battery can
chew up good leads very rapidly.
Battery damage in transmitters is
a problem that frequenntly confronts the R/C serviceman. Care in
this area will pay dividends.
· ·One final word on the "black
wire" business. It has been suggested that the black dye in the PVC
insulation affects the copper. I have
also seen this effect on the positive
(Red) lead on rare occasions and on
wires of all colours when used
around earth terminals.
It has also been suggested that
overcharging plays some part. I
have seen brand new sets which
have been charged once or twice
and left to stand for several years
exhibit the problem. As stated
previously, I have yet to see a
satisfactory explanation for the
problem. That does not mean that
an explanation has not been
published but if it has I have yet to
read it.
16
SILICON CHIP
Perhaps all I can say for sure on
this subject is that leaving sets standing unused for any length of time
is very poor practice. The batteries
should be cycled regularly and left
in the discharged condition.
Control memory
With the advent of LSI, the
microprocessor eventually found its
way into R/C equipment for models
and thus we saw the development
of the PCM (pulse code modulation)
system.
I am not a great fan of PCM for
models and prefer the . old PPM
system. However, the processor has
revolutionised the PPM system and
we now have transmitters offering
some very nice features as a result.
One of these features is the ability to store the trim and throttle settings for several different models in
the transmitter memory. A PPM
transmitter capable of storing trim
locations for three models is a very
useful item.
But here again I must stress that
before you rush out to buy such a
transmitter, remember that this all
costs money and is not really essential. A well trimmed model should
fly with all controls set at neutral
and the control throws set in such a
manner that full throw gives just
enough control response to accomplish the tasks required of that
model. Smooth, precise flying
begins with the correct setting up of
the controls.
There is a very practical reason
for this and it has to do with
mechanical advantage and servo
flutter. Using less than full servo
throw is poor engineering practice
indeed.
For many years I flew aerobatics,
pylon and helicopters on the same
day using the same transmitter
without memory and without undue
difficulty. However, I was very
careful about the trimming and setting up of the controls in each
model. Every model flew with the
controls set at neutral. If they did
not, then I landed and retrimmed
the aircraft.
At first glance this may seem odd
and dated, but the concept is as
valid today with the processor
memory encoder as it was then with
the old half shot encoder. The
reason is very simple.
Servo output arms are a rotary
device and thus will only give equal
throw about neutral if the neutral
reference point is goo to the control
push rod. If this angle is not goo,
then some amount of differential
control throw will be introduced into the system and the model will
respond faster in one direction than
the other. Fig.2 illustrates this
point.
Now this can be a useful feature
and is used quite often in setting up
control throws, particularly on the
ailerons where large amounts of
differential control are desirable at
times. It is not in the least bit
desirable if it gives an unwanted
faster roll to the left than the right
or more down than up.
Control geometry is a very complex subject and will form a column
of its own in due course. For the moment I will just confine myself to
pointing out that everything is not
as simple as it looks and that hi-tech
gimmickry is useful but no substitute for the careful and studied
application of basic principles.
All of the foregoing aside however, the new encoders do have
Problems?
.. .and you
don't have our
NEW
1990/91
This is a 7-channel symmetrical encoder developed by Proportional Systems
Australia (PSA).
some very nice features and the
most important are listed below:
Servo reversing: a slide switch is
provided on the Tx to invert the
pulse width on each channel,
thereby reversing the direction of
travel of the servo. The alternative
is to rewire the servo by reversing
the two outside pot wires and the
two motor leads.
This feature calls for a deal of
caution on the part of the user in
case take-off is made with the servos reversed. Let me stress right
now that this is all too easy. All control throws must be checked before
the first flight of the day for correct
direction of travel. This applies on
any Tx, even those without servo
reversing.
It only takes the pushrod to be accidentally replaced after adjustment on the wrong side of the servo
to wreak havoc. With servo reversing, it is even easier to come undone, especially if two models are
being used on the one transmitter
and reversal is required for only one
of the models. I have seen the odd
pilot who is clever enough to fly
with reversed controls but they are
rare indeed.
Dual rate: this is somewhat dated
in my mind by the exponential
system. A switch is provided for
one or more channels on the front of
the Tx with an associated potentiometer. The pot is adjusted to set
the overall percentage of servo
travel available (O - 100%) with
full stick throw. On half rate, full
stick throw will only deliver 50% of
the available servo travel. Returning the dual rate switch to the off
position restores 100% of the servo
travel.
This is useful for high speed
flight where the controls become
very sensitive around neutral. It
must be remembered where this
switch is before commencing any
manoeuvre, particularly outside
loops. I have seen models crashed
by pilots starting too low to the
ground in the belief that they were
in "high rate" when in fact they
were in "low rate" . It is very
awkward to get to the rate switch in
time if this error is made.
Another drawback is that two
distinct sets of control reflexes
must be developed, one for high
rate and one for low rate. Learning
to fly is hard enough without added
handicaps such as this.
Exponential control: often switched in by an external or internal
switch and gives electronic damping of the servo throw 1;1round
neutral. As the name implies, the
control throw follows an exponential curve, with less throw at
neutral and increased pace as the
stick moves to extremes. The advantage of this scheme is that the
control response of the aircraft is
alwavs constant whereas with dual
rate, two sets of reflexive responses must be developed.
Servo end point adjustment
(EPA): a very useful feature and
quite safe to use. It is especially
useful for throttle adjustment
where it is not desirable for the servo to run up against the end stops.
If this happens, it increases servo
drain and can burn out servo
motors and amplifiers. It can also
lead to Rx failure due to flat batteries in the model.
A potentiometer is included in
the channel for fine adjustment of
one end of the servo travel. This is a
very important point and sets not
fitted with EPA must be set up very
carefully to avoid these problems. ~
148 page
electronic parts
and accessories
catalogue ...
Its our latest
TRADE
catalogue for
the consumer
ARISTA ... Your one-stop
problem solver.
...Stylus ...
... Plugs, Jacks and Sockets ...
...Batteries .. .
... Cable .. .
...Tools and Technical Aids ...
...Plug and Power Packs ...
... Car/Auto Accessories .. .
... Boxed Hi Fi Speakers ...
... Raw Replacement Speakers ...
...Speaker Accessories ...
... Telephones and Intercom .. .
...Public Address Accessories ...
.. .Security and Alarm Accessories ...
...TV/Video/Antenna Accesories .. .
... Videocam Accessories ...
... Audio Accessories ...
...Headphones ...
... Computer Accessories .. .
...Microphone Accessories .. .
...Mixers, Amplifiers. Equalizers ...
Just about anything
you want. ..
Get your catalogue
complete with
"Recommended Retail Prices"
free from your local ARISTA
dealer or send $2.50 P & H
and your return address to:
AR~
ELECTRONICS PTY LTD
PO BOX 191,
LIDCOMBE, NSW, 2141
FEBRUARY1990
17
|