Hello Pronto experts,

I use a Pronto 940 and want to learn the codes of two 15 year old devices: Tuner CT441 and Amplifier CV441, manufactured by "Dual" (Germany). The Remote Control is called "Dual RC50".

The Pronto tells me: Learning is "OK" but the devices don't show any reactions when I use the Pronto as Remote Control afterwards.

The Hex Code of the second word (i.e. frequency) ranges from 000a to 0012!!

e.g: one of the Tuner keys produces the following hex codes:
1st try
0100 000a 0000 000c 0008 094d 0008 0948 000a 0948 0008 0948 000a 0df0 000a 0df0 000a 0946 000a 0df0 000a 0df0 000a 0df0 000a 0946 000a 61c8

2nd try
0100 000c 0000 000c 0008 07bd 0008 0b9d 0008 07bc 0008 07bc 0008 0b9d 0008 0b9f 0008 07bc 0008 0b9d 0008 0b9f 0008 0b9d 0008 07bc 0008 4d99

3rd try
0100 000e 0000 000c 0008 06a1 0008 06a0 0008 06a0 0008 06a0 0008 09f5 0008 09f3 0008 06a0 0008 09f5 0008 09f3 0008 09f3 0008 06a0 0008 45d7


Is it possible that a 15 year old remote control use such a high frequency?
Is the Pronto out of range and cannot learn the IR signals?

Any help is appreciated
ofen

That isn't high frequency. The initial 0100 instead of 0000 means it's unmodulated. The Pronto invents a wavelength value (that second number) when it learns an unmodulated signal. It's normal that the wavelength isn't consistent across different learns.

That protocol has a "toggle bit" meaning the device won't accept the identical signal twice in a row. But I can't see more than that wrong with it. I'd expect you can tell the difference between signals not working twice in a row vs. not working at all.

I haven't checked carefully, but I think this is a version of the RECS80 protocol. I think there may be a 7000 format for it that will get even the toggle bit right.

Hello John,
thanks for your reply!

The learned signals doesn't work at all.

What do you mean with "7000 format"? Is it the first word in the hex code? Is there a CCF to download that solves my problems?

Thanks for your patience and help
ofen

John, I checked it with your ccf2efc tool and here is the result of the tuner preset keys 1 to 4:

panel "Digits"
"1""Learned"(Gap-27-6720-11?):2:98 {12.3}
freq=0.0KHz
0100 000E 0000 000C 0008 06A1 0008 09F3 0008 069F 0008 06A0 0008 09F2 0008 06A0 0008 069F 0008 06A0 0008 09F3 0008 09F3 0008 06A0 0008 4929

"2""Learned"(RECS80):1:4 EFC=045 {T=0 (27/5751/8636)}
freq=0.0KHz
0100 000E 0000 000C 0008 09F5 0008 06A0 0008 069F 0008 06A0 0008 09F2 0008 06A0 0008 069F 0008 06A0 0008 09F3 0008 06A0 0008 06A0 0008 4C7E

"3""Learned"(RECS80):1:3 EFC=141 {T=1 (27/5751/8639)}
freq=0.0KHz
0100 000E 0000 000C 0008 09F6 0008 09F3 0008 06A0 0008 069F 0008 09F3 0008 06A0 0008 06A0 0008 06A0 0008 06A0 0008 09F3 0008 09F3 0008 45D7

"4""Learned"(Gap-27-6720-11?):0:66 {10.2}
freq=0.0KHz
0100 000E 0000 000C 0008 06A2 0008 069F 0008 06A0 0008 069F 0008 09F3 0008 069F 0008 06A0 0008 069F 0008 06A0 0008 09F3 0008 06A0 0008 4FD1


Looks not too bad, but Pronto still doesn't work as remote control.

Regards
ofen

I don't have any good guess why the Pronto can't send these. IIRC the 950 is an "NG" model (like the TSU3000) but the 940 is an older model. I've seen several posts indicating the NG models fail on this sort of signal. But I haven't heard of problems with older models.

The documentation I've read on RECS80 says the first bit is always 1 and the second bit toggles. You have samples with the first bit 1 that my decoder sees as RECS80 and samples with the first bit 0 that it doesn't see as RECS80. But I think all that is just an issue with my decoder and not anything wrong with your learned signals.

When an IR signal has very short unmodulated pulses, it sometimes helps to lengthen the pulses. The 0008 values in those learned signals represent those lengths. You might want to try doubling those (it's hex so double 0008 is 0010). In theory, any amount added to the those numbers ought to be subtracted from the following number (so 0008 09F5 would become 0010 09ED) but all those numbers are approximate anyway and the relative difference between 09F5 and 09ED is too small for the device to notice).

I still haven't had time to reread Eigeny's documentation of 7000 format to see if that can be used.

John

I checked the "toggle bit" behavior once again:

Here are the results of a 5 fold learning process of the "CD" input key on the Amplifier.

I tried to rebuild the bit structure: see each second line (please correct me, if I am wrong)


"CD" 1st try:
0100 000c 0000 000c 0008 0ba0 0008 07bb 0009 07bb
1 0 0
0008 07bb 0009 0b9c 0008 0b9c 0009 07bb 0009 07b9
0 1 1 0 0
0009 0b9c 0009 0b9c 0009 0b9c 0008 4d98
1 1 1 1

"CD" 2nd try:
0100 000c 0000 000c 0008 0ba0 0008 0b9c 0009 07bb
1 1 0
0009 07b9 0009 0b9c 0009 0b9c 0008 07bb 0009 07bb
0 1 1 0 0
0008 0b9c 0009 0b9c 0009 0b9c 0009 49b5
1 1 1 1

"CD" 3rd try:
0100 000e 0000 000c 0008 06a1 0008 06a0 0008 069f
0 0 0
0008 06a0 0008 09f3 0008 09f3 0008 06a0 0008 06a0
0 1 1 0 0
0008 09f3 0008 09f3 0008 09f3 0008 45d6
1 1 1 1

"CD" 4th try:
0100 000e 0000 000c 0008 06a1 0008 09f3 0008 06a0
0 1 0
0008 06a0 0008 09f3 0008 09f3 0008 06a0 0008 069f
0 1 1 0 0
0008 09f3 0008 09f3 0008 09f3 0008 4282
1 1 1 1

"CD" 5th try:
0100 000c 0000 000c 0008 0ba0 0008 07bb 0009 07bb
1 0 0
0008 07bb 0009 0b9c 0008 0b9c 0009 07bb 0009 07b9
0 1 1 0 0
0009 0b9c 0009 0b9c 0009 0b9c 0008 4d98
1 1 1 1

As far as I can recognize, the first two bits are "toggling". Is this possible?

Do you have a link to "Eigeny's documentation of 7000 format"?

Thanks again for all your answers
ofen

I think Eigeny's document is here
[Link: home.tiscali.nl]

I think you're decoding things exactly right.

I remember some protocol like this in which two bits toggled. Your first post in this thread made me think only one bit was toggling, but now that you tested it so well I see both toggle.

A few other people with devices that toggle two bits have had good results with a universal remote that toggles just one bit. I don't know why the original remote was designed to toggle two, but I think the IR receiver is happy with just one bit toggling.

I think 7000 format can toggle two bits, but will toggle them in sync rather than as a two bit counter. I'm not sure of that, but if true it wouldn't have any advantage over toggling just one bit.

I suddenly realized there is another possibility for what's going on here.

There is an unmodulated form of RECS80, but you might have the modulated RECS80. The IR receiver in the Pronto is much less sensitive than IR receivers in typical devices. If the batteries in the original remote are in poor shape, the first cycle of every modulated pulse is normal but the rest of the pulse is weak. The Pronto will only see the first cycle, which makes the modulated signal look unmodulated. In most protocols I'd recognise that easily as a battery problem, but RECS80 has a format that looks just like that by design, so I'm just starting to guess batteries because your learned signals don't work.

I have a CCF file with some samples of the 7000 format of RECS80 that has two toggle bits and the same overall timing of your signals, but it's modulated. If your RECS80 is actually modulated those would be a good starting point, but we'd need to edit the hex to change the device number.

I don't have any CCF files with RECS80 device 1. Is it coincidence that suddenly someone is asking about RECS80 device 1 in the OFA forum and you are asking about it here at the same time after I'd never heard mention of it before?
[Link: remotecentral.com]

I'll post more about those 7000 format codes later.

Thank you, John, for your thoughts and efforts!

Strange things happened in the meantime...

First of all, I checked the battery of my original remote: it is in good shape (nevertheless: thank you for the tip!)

Then I took another learnable remote control unit that came with my AV-Receiver: it is an ONKYO RC-390M.

I tried to learn the DUAL codes into the ONKYO RC and the ONKYO RC tells me: learning is OK but the signals didn't function (same behavior as with my Pronto).
I tried the same procedure over 100 times (different distances and angles) and suddenly, one of the ONKYO buttons has learned one of the original functions. The learned key is not repeatable like the original key but it works!

Then I took the Pronto and tried to learn the just learned ONKYO RC signal.
First of all the Pronto told me learning is OK, but a test with the amplifier showed: the signal didn't function.

The Hex Codes of the learned ONKYO signal were all of the following kind:
0100 000a 0018 0000 0008 094b 0008 094b 0008 094a 0008 0948 0008 0df3 0008 0948 0008 0df2 0008 0948 0008 0948 0008 094b 0008 0df2 0008 6672 0008 094a 000a 0948 0008 0948 0008 094a 0008 0df3 0008 0948 0008 0df2 0008 0948 0008 094b 0008 0948 0008 0df2 0008 094b

After about 30 tries I suddenly got another Hex code:

0000 0053 000f 0000 0002 011e 0003 01ac 0002 0c57 0002 011d 0003 011c 0003 011c 0003 011c 0003 01ac 0003 011c 0003 01ac 0003 011c 0003 011d 0002 011d 0003 01ac 0002 011e

And this strange code works!!
It reacts a little bit slowly and is not repeatable but it works.

Now I have the first code that is working - and it is completely different from that ones I have seen all the time before.
(can you imagine what is the coherence here?)

Maybe you are right with your assumption that my Pronto shows an unmodulated signal and in reality it is a modulated one!


Can you recognize the protocol of that signal?

Here is a sample of RECS80 with two toggle bits in 7000 format
7000 0018 0000 0008 000E 000D 0010 0016 001A 0002 0002 0002 0003 0003 0003 0003 0003 0002 0000 03D9
It is device 7, command 1.
I got it from the CCF at
[Link: remotecentral.com]

Each of the 0002's is a '1' bit and each of the 0003's is a '0' bit. The 0002 0002 0002 is the device 111 binary is 7 decimal. The 0003 0003 0003 0003 0003 0002 is the command 000001 binary is 1 decimal.

In that form, your digit "2", which is device 1, command 4 should be

7000 0018 0000 0008 000E 000D 0010 0016 001A 0003 0003 0002 0003 0003 0003 0002 0003 0003 0000 03D9

Here is a 7000 format for RECS80 with only one toggle bit, but it is for a device in which most of the RECS80 codes are in 0000 format and have two toggle bits
7000 006b 0000 0008 0011 000d 0010 0002 0017 0002 0003 0003 0002 0002 0003 0002 0003 0002 0000 00be
It is device 4 (the 0002 0003 0003 right after the 17), command 53 (0002 0002 0003 0002 0003 0002).
It is from this CCF (it is the info button in Menu1 in Projektor in Macro Groups):
[Link: remotecentral.com]

In that format I think your device 1 command 4 would be
7000 006b 0000 0008 0011 000d 0010 0002 0017 0003 0003 0002 0003 0003 0003 0002 0003 0003 0000 00be

I still haven't had time to review Eigeny's document to see how well the timing of those matches your RECS80 nor to see if there is a way to do unmodulated RECS80 by 7000 format (both of those are modulated).

Really, John, the second 7000 format codes worked on my system and executes a DIGIT "2"!!!

Excellent informations!
I will now try the second CCF file...

Can you please explain me, where I can see that DIGIT "2" is "command 4 of device 1"?

Best Regards
ofen

This message was edited by ofen on 11/10/03 20:03.

I didn't see it earlier (for the reason you guessed).
That semi working code you posted looks like a bad learn of a signal that should have been

0000 0053 0000 000c 0003 011c 0003 011c 0003 011c 0003 011c 0003 01ac 0003 011c 0003 01ac 0003 011c 0003 011d 0002 011d 0003 01ac 0003 0c57

I don't have the right files or programs on this computer to check further. I'll be on my other two computers at various times tomorrow and will see if I can deduce more.

unbelievable, John!!
I just checked your last suggestion, i.e.

0000 0053 0000 000c 0003 011c 0003 011c 0003 011c 0003 011c 0003 01ac 0003 011c 0003 01ac 0003 011c 0003 011d 0002 011d 0003 01ac 0003 0c57

and it works perfectly: reacts fast and is even repeatable!!

Many, many thanks, you are fantastic!
It is a little bit curious that your "human" IR code recognition is sooo much better than that of my Pronto ;-))
ofen