| Thanks for the URL leads on what is available. You guessed it in your
| last paragraph. I am looking for a way to possibly "piggyback" (or
[quoted text clipped - 4 lines]
| Still thinking further out of the box, making the channel two way would
| allow commands to be sent and results to be transmitted back.
Well, the list of things you described can all be done with the sort
of thing I described and no changes made to your TX at all. The only
limitation is that you'd have to use one channel for each item or each
two items.
For example, if you have a three position switch, that will send a
signal of either 1.0 ms, 1.5 ms or 2.0 ms. So you can have 1.5 ms do
nothing, 1.0 ms toggle the lights and 2.0 ms take a picture. Or if
you have a two position switch, 1.0 ms = lights on, 2.0 ms = lights
off. Landing gears are often done with retract servos that are all or
nothing -- they're not porportional at all.
You could also do this on an unused stick -- middle do nothing, up do
one thing, down do another way.
With a few channels dedicated to this, you could do all the things
you've described. But yes, it sounds like what you want might be
easiest done with a completely seperate RF link. Do take lots of care
to make sure any TX in your plane doesn't desense the 72 MHz RX!
| > | I understand that there may be receiver/transmitter pairs available
| > | which would allow a user to send custom command information to the
| > | receiver. Conceivably a small microprocessor onboard the aircraft would
| > | be able to decipher these commands and carry out special functions
| > | (turn on/off lights, landing gears up/down, trigger small digital
| > | cameras, deploy parachutes, etc.)

Signature
Doug McLaren, dougmc@frenzied.us
"I'd rather be lucky than good, any day" --well known auto-racing saying
JoelKatz - 21 Sep 2006 04:36 GMT
> For example, if you have a three position switch, that will send a
> signal of either 1.0 ms, 1.5 ms or 2.0 ms. So you can have 1.5 ms do
[quoted text clipped - 10 lines]
> easiest done with a completely seperate RF link. Do take lots of care
> to make sure any TX in your plane doesn't desense the 72 MHz RX!
You should be able to encode a fairly large number of functions onto a
single channel with a PIC on each end. You should easily be able to
transmit and receive 16 different pulse widths, allowing for 4-bits to
be encoded in each pulse. You could reserve one bit to make the
beginning of a block, and one bit to alternate between high and low (so
the receiver will know when the data has changed) and pass two bits of
data in each framed chunk.
For example, if you want 6 on-off switches, you can simply send this
block:
1 0 S1 S2
0 1 S3 S4
0 0 S5 S6
Notice how the first bit is set on the first block. That allows you to
identify the first block. Notice how the second bit alternates. That
allows you to know when you have a new block (otherwise you might not
know of there are two blocks with the same value or one block with that
value). The last two bits hold the data.
Each combination of 4 bits would be encoded as one of 16 pulse widths.
If you just need to send changes, you can reduce the latency with a
protocol that consists of so many bits of data to indicate which
control has changed and so many bits to indicate the new position. You
can encode the beginning of the block and when there's new data the
same way.
How much data do you need to send? How many signals and how many values
for each signal? I can whip up an encoding scheme for you very easily.
You should try to confirm two things. First, what is the update rate
with your receiver and transmitter? That is, how many pulses per second
does it send (minimum and maximum, it may vary based on the input).
Next, what is the resolution of your receiver. That is, how many
distinct pulse widths can it reliably send? I believe most modern
transmitters and receivers can handle at least 32 positions, so you
should be able to get 16 reliably (with safety windows around each
valid position to detect an invalid signal).
DS