I have done a project to measure RC receiver output pulses. Roughly,
each channel outputs a pulse whose duration varies from 1 ms to 2 ms.
The 1 ms variation in pulse duration commands the servo position. One
channel follows the other in time. The pulses are repeated every 20
ms. I used a PIC microcontroller to measure the pulse durations. The
PIC can then be interfaced to a PC via RS-232 to transmit the data.
Now there are some folks who think you can measure these pulse widths
directly on the PC's parallel port channels. They are correct, but the
measurements will vary from time to time due to the non-real time OS
characteristics of Windows or DOS. A friend of mine did the opposite.
He drove hobby servos directly out of the parallel port, but they
jittered due to the variation in timing.
On Wed, 28 Apr 2004 20:35:51 +1000, <Phil> wrote:
another possibility would be my microcontroller of choice, the IsoPod from
NewMicros Inc. www.newmicros.com
Compiler on board. 12 timer channels accessable in high level code.
read the r/c via timers, output to computer via rs232.
<Phil> wrote in message> This is wonderfull link. Thank you.
I used assembly language. I used the RC data to convert to a dual PWM
for a differential DC motor drive system. I displayed the data on an
LCD display, but did not interface to the PC via the serial port.
A friend of mine just did a PC to PIC RS232 interface with the CCS C
compiler. It is VERY easy to do with CCS. The only additional
component you need is an RS232 level converter like the MAX232.
On Thu, 29 Apr 2004 20:56:28 +1000, <Phil> wrote:
What about using a soundcard mic/line input for realtime pulse width
measurements. You'll have attenuate the signal for input, but sampling
a 1-2ms square pulse at 2.5ms rate should be possible at 44Ksps.
You could use the line out with a comparator to generate pulse trains
You should be able to OR RC servo outputs together on many RC RXs,
since they time multiplexed with missing pulses that are used as
index/start. Also, a long time ago I had a RX module that brought out
the non-demux pulse train on the signal pin on the power supply
connection, since it wasn't used. I suspect it was used for testing.
I don't know if current RX modules do that or what brand/model the
RX was that I used.
See ya, -ingo
Polytechforum.com is a website by engineers for engineers. It is not affiliated with any of manufacturers or vendors discussed here.
All logos and trade names are the property of their respective owners.