🚀 go-pugleaf

In <1992Sep2.003540.24065@aio.jsc.nasa.gov> mancus@roger.jsc.nasa.gov (Keith Mancus) writes:

>  We're in the process of building a small robot, and we'd like it
>to respond to a TV remote control.  Can anyone tell us:

>  (1) Is there a standard carrier frequency?  The IR sensor we're using
>detects 40 MHz.

>  (2) Is the encoding scheme standardized?  If so, what is it?
> 
>  (3) Are there any other pitfalls we should know about?

I spent a lot of time investigating IR remotes because I wanted to build
an IR repeater for my house.

I got an IR phototransistor, as many varied remotes as I could find
(Sony, RCA, Mitsubishi, NEC, and some off-brands for stereos, TVs, VCRs,
CDs, etc).  I found that all of them used a digital pulse system
based on a 40Khz IR carrier.  An individual command is constructed
of 6 to 14 "clock" pulses spaced every 8ms.  The number of clocks
depended on manufacturer and type of device being controlled.  The
information portion is encoded as a "data" pulse which appears
(or not) 1ms behind each clock pulse.  Each command was repeated
as long as the key was held down.  There was a timing gap between
command repeats.  I think it was 64ms, but I don't recall for sure.

I'm not saying this analysis will hold true for all remotes, but 
it did for the 10 or 12 I tried.

The circuitry for detecting the 40Khz IR carrier, demodulating it,
amplifying it, and cleaning it up would be a lot.  However, don't
bother.  Radio Shack sells a little silver cube a little bigger
than a sugar cube that does all the work!

This little cube is great, but there are a few caveats.  The spec
sheet that comes with it has several mistakes and ommisions.  (Big
surprise!)  The output signal is inverted from the input signal, not
positive logic like the diagram has.  Also there is no indication of
output voltage swings or the amount of current the output can drive.

I used a simple 2N2222 to reinvert the signal, give it a good voltage
swing, and drive the line.  (Ya, I know, if I do it again, I'll use a
FET, but I had a pile of the 2N2222's lying about.)

Important Tip: connect the shielding of the cube to ground!  Without
the connection, the cube is very sensitive to any and all electical
noise.  I found this out the hard way during prototyping my device. 

For my application, this drives a 555 which remodulates the signal
at the other end.

Feel free to send me mail if you have any further questions.

>| Keith Mancus    <mancus@sweetpea.jsc.nasa.gov>                         |
>|                 N5WVR                                                  |
>|  "If I owned Texas and I owned Hell, I'd rent out Texas and live       |
>|  in Hell." -General Sheridan                                           |
-- 
Quentin Barnes
qbarnes@urbana.mcd.mot.com | ..!uiucuxc!udc!qbarnes