JP1 Remotes

Long vs. short press doesn't matter in this case. What matters is when you START to press it. That must be AFTER you press the key on the 8910 to which you want it learned.

Just a couple of suggestions, but have you tried learning in subdued lighting and away from any flickering plasma screens, sunlight, fluorescent lighting etc.? And if it doesn't work completely flat on a table (since some IRs are a little bit "underneath" the remote) then is it any better resting the business ends of the remotes on the edge of a thin book so as to create a slight angle between them?

underquark wrote:have you tried learning in subdued lighting and away from any flickering plasma screens, sunlight, fluorescent lighting etc.?

We're not talking about a subtle learning failure or distortion (as you would get from any interference source). The main part of the signal is totally missing.

underquark wrote:And if it doesn't work completely flat on a table (since some IRs are a little bit "underneath" the remote) then is it any better resting the business ends of the remotes on the edge of a thin book so as to create a slight angle between them?

In one of three learned signals there is a perfect learn of the end of the content after the main signal. When that was learned the remotes were perfectly aimed with no outside interference. If we believe that the remotes weren't moving during the learning process and were perfectly aimed at the end of that process, that rules out aiming problems entirely and makes interference quite unlikely.

checksplay wrote:I've been keeping the button on the original remote pressed continuously while learning.

That's the exact cause of the problem. This device uses the NEC1 protocol, which sends the "data" portion of the signal just once, then when you hold the button down, it sends an otherwise meaningless signal which simply tells the device to keep doing what it's already doing (like raise the volume or whatever).

When you start pressing the OEM button ***before*** the JP1 remote is ready to learn, the OEM remote sends the data portion ***before*** the JP1 remote is ready to capture it, therefore all that the JP1 remote is able to capture is the meaningless pattern that gets sent when the OEM button is held.

The correct learning procedure is to FIRST put the JP1 remote into learn mode, THEN press and hold the OEM remote button. Better still, wait a second after putting the JP1 remote into learn mode before pressing the OEM remote button.

So, it's a bit like yodelling then? All you get is the "ay-hee-hooooooo" bit and not the all-important "yodel"?

If you were to yell "What are you doinggggggggggg?" all we would here is the "gggggg?"

SUCCESS! Thank you everyone for your help and suggestions. I finally go so frustrated that I just took apart the NAD remote to see if there was anything obvious. When I put it back together, I left off the red plastic cover over the LED and bent the LED so that it was pointing more directly out of the remote.

When I tried learning - it worked right away! Didn't have to be careful about position or anything. I downloaded the file and put it in the diagnosis area if anyone is curious. IR.EXE no longer shows a stream of numbers, but seems to have recognized an NEC1 code.

My next question, is one of the columns the 'advanced function'? Is it 'EFC' and a value of 190? I originally tried every value between 0 and 256 for code 0320 and none of them corresponded to TUNE-. Should 190 have worked? Thanks.

David

checksplay wrote: I downloaded the file and put it in the diagnosis area

When saying that, you ought to give the URL. It is

https://www.hifi-remote.com/forums/dload ... le_id=2698

checksplay wrote: is one of the columns the 'advanced function'? Is it 'EFC' and a value of 190?

Yes the "advanced function" is the EFC and it is 190.

checksplay wrote: I originally tried every value between 0 and 256 for code 0320 and none of them corresponded to TUNE-. Should 190 have worked?

Yes 190 should have worked.

Wow, fast response. Sorry, I forgot to put the URL, I had it copied and was all set to press cntl-v, but it slipped my mind.

Well, I went back and checked 190 and it worked, so I must have screwed up first time around. Thanks for everyone's patience for what turned out to be non-issue.

Gotta go to work, otherwise, I'd ask about some other stuff.

David

Question:

checksplay wrote:I left off the red plastic cover over the LED and bent the LED so that it was pointing more directly out of the remote.

When I tried learning - it worked right away!

How are these hardware modifications related to the missing start of the signal?

I think it just allowed me to get more signal strength into the OFA from the NAD remote.

LEDs are almost instant-on but they do draw more current for a fraction of a second as they warm up (an effect more recognised in high-brightness LEDs). Is it conceivable that the brightness of the LED for the very brief first part of the signal was sub-threshold with the cover on but adequate with the cover off?

Those tinted covers are almost perfectly transparent to IR, even though they block the higher end of the spectrum of visible light. In a signal modulated on multiple levels as this one is, there wouldn't be any detectable effect from a hardware characteristic (if there is one) of initially lower brightness. The opposite effect is commonly observed. Old batteries deliver current with lower output impedence at the beginning of the signal than they do later, so the later part of the signal is generated at a lower voltage and is dimmer. But that is not happening in this case.

IF the described hardware change really made the difference (as opposed to some coincident operator behavior change), the only explanation is aiming.

The totally trashed learns were because the original remote's led wasn't aimed at the learning remote's IR detector.

The one short perfect learn of content beyond the important part of the signal was because one of the remotes moved a bit while the button was held down, apparently just enough to bring the aim to acceptable.

I really think it is because I bent the LED so I could point it more directly at the OFA. It worked without any effort at all after that. I didn't have to put it on a table or hold it still or anything.