Subject:
|
Re: IR to FM
|
Newsgroups:
|
lugnet.robotics
|
Date:
|
Tue, 29 Aug 2000 03:36:04 GMT
|
Original-From:
|
Steve Baker <sjbaker1@airmail.^NoSpam^net>
|
Reply-To:
|
SJBAKER1@AIRMAILnomorespam.NET
|
Viewed:
|
683 times
|
| |
|
|
> As I am constructing an RC guided Lego Car (No, I don't want to buy a 5600.
> I want to create it all out of individual Lego bricks.), I thought it must
> be able to make an IR to FM link, like those little things you see to
> transmit IR signals of your remote control to your telly via FM. Has anyone
> constructed such a device? I don't want to use a Cybermaster, because I
> don't own one and it weighs too much. I think I'll have to use two RCX's:
> one in the car and one in the transmitter.
Well, I know a little about the format of Radio control signals of the kind
you get in RC planes and cars.
There is a signal pattern (which can be either FM or AM depending on the
waveband of the radio) that represents a train of pulses that are repeated
50 times per second (roughly). There is one pulse for each 'channel' of
radio signal that the transmitter is capable of. Each pulse has a width
proportional to the position of the joystick - I don't exactly recollect
the width of the pulse - but my memory is telling me that it's 1 to 2
milliseconds. 1ms for minimum stick position, 2ms for max. You'll need
to confirm that though...I'm not certain.
Sanity check: 50Hz is 20ms - a fully-stuffed radio can have ~8
channels - meaning that 16ms or more is consumed
by the pulse train - leaving less than a 4ms gap
for the receiver to realise that the train is about
to repeat...that suggests that 2ms is about the MOST
those pulses could be...perhaps they are somewhat
shorter.
Hence, a decoder has to wait out and detect the LONG gap then take each
pulse in turn, measure the duration and send it off to the next motor (or
whatever) in the sequence.
Some of the fancier (and MUCH more costly) RC aircraft systems are moving
to a pure digital solution, but this pulse-width modulated scheme is by
far the most common.
The nasty flaw in this scheme is that it relies on the clocks in the
transmitter and receiver agreeing - or else the stick position that
the transmitter sends won't agree exactly with the output. That's one
reason why RC transmitters have little 'trim' wheels to adjust the
sticks precisely. I've seen maybe 10% differences in timing from the
same transmitter from one day to the next - so calibration will be
important.
I have used a small 8051 (single chip) microprocessor to decode these kinds
of signals in the past. I converted them into RS-232 so that I could plug
this into a PC's serial port - which I used to drive a flight simulator
program that I'd written for a radio controlled model plane!
So it's certainly not impossible that you could do it in software inside
an RCX with the output of a radio receiver (appropriately
demodulated/amplified/attenuated/whatever) stuffed into one of the inputs.
The RCX is quite a bit faster than an 8051 - so this shouldn't be too
hard.
However, I do think you'd need to use a true machine-code/C approach using
LegOS or something to do the decoding. The NQC byte-code wouldn't be fast
enough - and the task scheduler would do bad things to your high precision
timing.
Getting this to work would actually be rather cool - the radio system
could easily provide up to 8 analog/switched inputs using only one
RCX input.
On the other hand, any filtering/protection circuitry that Lego may
have hung off the RCX inputs might screw things up to the point where
you can't read the input signal anymore.
Using one of the RCX outputs to drive a radio transmitter ought to
be possible too. Some RC transmitters have an optional 'buddy box'
input that lets you slave one transmitter to another so that an
'instructor' can take over and fly your model plane if you are
in danger of crashing it while learning to fly.
The buddy box input is in the same format at the RC output
signal - except that it's not FM or AM modulated. Again, I'm
not sure what voltages you'd need - but in principle one
could connect the output of an RCX to the buddy box input via
some kind of signal conditioning and some careful timing
software and have a radio link!
I considered doing this once before (but with a laptop's printer
port - not an RCX) - so I could teach my computer to fly
an RC plane...
...but then sanity prevailed.
:-)
** WARNING WARNING WARNING WARNING WARNING WARNING WARNING **
PLEASE DON'T DO THIS UNLESS YOU ARE FULLY AWARE OF
THE
EXACT VOLTS, AMPS, OHMS, WATTS, HERTZ,
BITS AND BYTES OF WHAT YOU ARE DOING!
If you do any of these things and wreck either the RCX, your
computer or your radio gear, I will not be accepting responsability!
--
Steve Baker HomeEmail: <sjbaker1@airmail.net>
WorkEmail: <sjbaker@link.com>
HomePage : http://web2.airmail.net/sjbaker1
Projects : http://plib.sourceforge.net
http://tuxaqfh.sourceforge.net
http://tuxkart.sourceforge.net
http://prettypoly.sourceforge.net
|
|
Message has 1 Reply:
 | | Re: IR to FM
|
| <snapped an excellent technical description> (...) kinds (...) I wasn't exactly thinking of the RCX doing the smart job of the FM communication. I imagined more of the RCX just outputting IR signals to an apparatus that translates them to FM. These (...) (24 years ago, 31-Aug-00, to lugnet.robotics)
|
Message is in Reply To:
| | Re: IR to FM
|
| (...) 5600. (...) anyone (...) I am currently working on this very project. I have decided to use 2 RCX's as I want to keep the project strictly Mindstorms. I realize that IR won't allow the car to go very far from the controller, but I'll live with (...) (24 years ago, 29-Aug-00, to lugnet.robotics)
|
7 Messages in This Thread:
  
- Entire Thread on One Page:
- Nested:
All | Brief | Compact | Dots
Linear:
All | Brief | Compact
This Message and its Replies on One Page:
- Nested:
All | Brief | Compact | Dots
Linear:
All | Brief | Compact
|
|
|
|
Active threads in Robotics
|
|
|
|
|
|
