You're right. After digging through the last couple of pages of an old panasonic (not that old) IR spec sheets, I see that several codes for their amps and some CD players have different values for the 1st nybble of the 1st and 2nd byte.

The potion you were calling device was always 4004 (in these references) and sub device varied (0100 - tv, 0900 - video, 0980 - vcr, 0543 - amp, tuner, CD, and tape). Carrier freq was given as 32khz, but I see now that 36khz is being used also. (I tried 32khz and it seems to work fine if the other codes are adjusted to correct the timing.) The timing for lead-out, pulse and space are all maintained using the higher freq. However the timing seqence for the lead-in was changed slightly from <10T> <4T> to <8T> <4T>; (T being about 400us).

I am fairly new to all this, and have been trying to get the codes correct for a panasonic rear projection lcd. The TOAD codes will turn the set off, but will not turn the device on unless it is within it's cool down period. I found that by repeatedly sending an 'on' code and slowly accelerating how fast the codes are sent, the set will switch on once the codes are sent close enough together. So I set the pronto to send two 'on' codes seperated by about 100ms. This worked fine for 'on'. A bit of a problem for 'off' as it would cycle the set to off, then on again (very quickly). Not a big problem, but irritating.

Is there a problem with the lead-in of some off these code that is keeping the device from having the time to properly set the receiver? Is there a null code that can be sent?