08-16-2014, 02:37 PM
(08-09-2014 11:22 AM)Marcus von Cube Wrote: [ -> ](08-06-2014 08:15 AM)Martin Hepperle Wrote: [ -> ]Some Numbers
The frequency of the IR signal is f = 32768 Hz so that one pulse of 50% duty cycle has a duration of t_pulse = 1/32768 / 2 = 15.3 us (microseconds)).
The HP docs specify a burst length of 6 to 8 pulses = 183 to 244 us as valid range. All my test calculators output 8 pulses per burst.
If I understand my code correctly, WP 34S outputs bursts of seven pulses. The generator is just the AND of two square wave clocks: 32768Hz and (32768/14)Hz. If you want to check the code, it's in trunk/main.c, routine put_ir().
Marcus,
yes, you are right. You chose the golden middle of (6+8)/2.
I saw that the HP 48 and 50 use 8 pulses, while my 28S varies between 6 and 7 pulses to get the average timing right (as per explanation from Christoph above)
According to Christophs suggestions, I have now changed my code to measure the time interval between bursts relative to the average half-bit length (from the start bits) and now I can decode all IR signals that I have access to, from the HP 28S to the HP 50g.
My conclusion so far is that the modern TSOP integrating decoders do not work for me and I have reverted to the C4103A Receiver unit which does no integration and just delivers the raw pulses. One could probabaly use any IR receiver composed of a IR Transistor and a simple amplifier.
Now I can backup HP 28 programs on my PC and only need to build a key pusher device to reload the programs back...
Martin