Hello,
I bought a receiver (Worldsat DT5100) for terrestrial numerical television (in France) and when I learnt code for the remote with Pronto TSU3000, I had this type of code (key "Up"):
900A 0068 0000 0001 10EF 12ED
but this worked randomly, sometimes I needed to press the same key more than ten times.

So I tried to learn differenty, and the only way was to learn with several key press (for the same key), and then all worked correctly.
But I have different code for a same key, and when I used IRTool, it said "This is an incomplete code. There ar too few data pairs".
For example, for the key "Up", here is three example of the learned code which works perfectly :
0000 0069 0024 0000 0164 00B2 0016 0016 0016 0016 0016 0016 0016 0016 0016 0043 0016 0016 0016 0016 0016 0016 0016 0043 0016 0043 0016 0043 0016 0043 0016 0016 0016 0043 0016 0043 0016 0043 0016 0016 0016 0043 0016 0016 0016 0016 0016 0043 0016 0016 0016 0016 0016 0016 0016 0043 0016 0016 0016 0043 0016 0043 0016 0016 0016 0043 0016 0043 0016 0043 0016 062D 0164 0059 0016 00B2

0000 0069 0026 0000 0164 00B2 0016 0016 0016 0016 0016 0016 0016 0016 0016 0043 0016 0016 0016 0016 0016 0016 0016 0043 0016 0043 0016 0043 0016 0043 0016 0016 0016 0043 0016 0043 0016 0043 0016 0016 0016 0043 0016 0016 0016 0016 0016 0043 0016 0016 0016 0016 0016 0016 0016 0043 0016 0016 0016 0043 0016 0043 0016 0016 0016 0043 0016 0043 0016 0043 0016 062D 0164 0059 0016 0EF5 0164 0059 0016 00B2

0000 0067 0026 0000 0168 00B6 0017 0017 0017 0017 0017 0017 0017 0017 0017 0044 0017 0017 0017 0017 0017 0017 0017 0044 0017 0044 0017 0044 0017 0044 0017 0017 0017 0044 0017 0044 0017 0044 0017 0017 0017 0044 0017 0017 0017 0017 0017 0044 0017 0017 0017 0017 0017 0017 0017 0044 0017 0017 0017 0044 0017 0044 0017 0017 0017 0044 0017 0044 0017 0044 0017 063E 0168 005A 0017 0F1F 0168 005A 0017 00B6

So, is there a way to build or clean correct code ? Could you help me please ?
Thank you in advance.

I can't think why that would fail. It correctly represents the same signal as the three copies you said work well.

Try changing the 0068 to 006D so the whole thing is
900A 006D 0000 0001 10EF 12ED
Does that work better, worse, same, or not at all.

Better is unlikely, but possible and if it does work better that gives you a way to fix things. Otherwise the information about whether it is worse or not may help us understand what was wrong.


That means you didn't copy the whole Pronto Hex string. The dialog for Pronto Hex in ProntoEdit seems to confuse a lot of people that way. If the scroll bar is present and working then the Pronto Hex is longer than what you can see. It takes some extra care to highlight the entire string before copying.


That is typical NEC1 protocol, but at the less common modulation frequency (40Khz).

If there is some reason that the long form of Pronto Hex works and the condensed form doesn't, you can create clean long form signals with MakeHex, using the NEC1_40.IRP file.

Control + A works great for selecting all. Trying to highlight everything by dragging your mouse over what you want is a pain. Ctrl+A is a great shortcut, then Ctrl+C to copy and Ctrl+V to paste :-)

Hello,
sorry for the copy/paste error (what a shame !)

Johnsfine, you're definitly the best : replaced "68" by "6D" works perfectly.
I didn't understand, but here is the reason why you're an expert and not me ;-)

To explore all the way, I would want to try with MakeHex, but the file NEC1_40.IRP doesn't exist, even in your version : [Link: john.fine.home.comcast.net]
Could you give me the file or a link please ?

Anyway, thank you very much again (it's the second time you help me) for your priceless help on this forum.

That value controls the modulation frequency. 6D is the common value, 68 is the rare value.

NG Prontos are fairly bad at learning the right modulation frequency. 6D is correct in far more devices than 68. So the probability of mislearning a 6D as 68 is about the same as the probability of having a device where 68 is correct.

But your three samples of learned codes that worked perfectly had 67 or 69 for that value (close enough to 68 to be considered equal to it). That was why I thought it unlikely that changing 68 to 6D would help. But with an initial set of facts as unlikely as what you gave in the first post here, one must try some long-shot tests to get some idea what is really going on.

I still have no clue why those long form learns you posted initially could work better than the condensed one (with the 0068).


I guess I need to clean up my system for keeping the distributed copy of MakeHex in sync with my local copy.

But you don't need the 40Khz version of the NEC1 irp file. The fact that 6D worked better than 68 must mean that the usual modulation frequency for NEC1 is correct.

OK, thank you for all again.