To start it out I have a Dell 50 inch HD Plasma (W5001C) TV. Here is the start of the problem. If I use a standard universal remote to learn in some IR codes, for example the channel 1 and 2 button, the first time I hit the 1 button it works ever time. Then if I hit 1 or 2 button following the first press it does nothing. I have to hit the keys many times or wait about 20 seconds to hit the next key for it to work. It is very sporadic on how many times I have to press the next key. Sometimes it is 2 sometimes 10. I always can fix it immediately by using the original remote and send a key press. I am guessing the TV uses some type of toggle bit (RC6 or MCE). I am not an expert by any means on any of this. I have seen many people post about similar problems. I have the Dell IR discrete PDF that has been floating around. The PDF does not seem like it shows the toggle bit, but I could be very wrong.

Now I understand I can get a different remote to learn in the toggle bit (if this is really the problem), but the real reason I am posting is because I use a JDS system that does not learn the toggle bit. The JDS will send learned IR commands through RS-232. So the question I am asking is there anyway with the Dell TV to turn off the toggle bit feature through some maintenance menu, or any tricks anyone could think of to make this work? Basically I have a device that will only send one IR command.

I could be totally wrong about the real issue here, but I am just assuming it has something to do with the toggle bit. Any information will be appreciated.

Thanks

Dell uses MCE (in at least some devices) and the symptoms are what you would expect from the toggle bit in MCE.

I doubt that you could turn off the toggle bit requirement in the TV (the way you can in some other devices using MCE).

I don't know anything about the JDS, but I expect you can make it do the right thing.

For a simple learning remote, a toggle bit isn't some special magic that the remote needs support for. A toggle bit means there are two versions of each command. There isn't just a single '1' command. Think of it as a '1A' command and a '1B' command.

Your learning system can learn '1A' and it can seperately learn '1B'.

The trick is getting it to send the right one when you want it to send a '1'. If I knew more about how one uses a JDS system I might have some good suggestions.

If it is a device for which you write macros, then you can manage that part yourself.

I'm a little surpised at one detail of your description. After hitting '1', I understand another '1' wouldn't work, but you say a '2' wouldn't work either. That's unusual for toggle bit behavior. It's possible the TV is that picky, I just wouldn't expect it to be.

If that really is a toggle bit symptom, you'd need a lot of care when relearning the digits into the JDS to get the two versions of each digit properly sorted.

Just think of the JDS as a simple universal remote with learning and playing features. I have a little more information. I mentioned that if I send the 1 command I have to wait 20 seconds to send it again. I was wrong. I have to wait about 5 minutes before I can send it again. Even for the 2 command button, I have to wait 5 minutes. The only way to use the command again before 5 minutes is to use the original remote and then I can send another command through the JDS. I have also tried this same thing with a regular universal remote showing the same results. I am guessing the TV is very picky on the toggle bit. The weird thing is I tried to program both codes into the remote (button 1 with different toggle bits) and switch back and forth to act like it was the toggle. I could not get this to work. One thing that puzzles me is the original remote always works 100% of the time. How could this be so? If the remote sends 1A, then 1B and alternate and my remote is only sending 1A why would the original remote work 100% when it sent a 1A after the universal remote? It should send the command sometimes and not work, but it always does. This is very confusing and can not explain everything that is going on.

But if it were really a simple universal remote, it would be very inconvenient to allocate two different buttons for each command (the 1A and 1B toggle states of the 1 key).

If it is some program running on a PC, then it might not be very inconvenient to do so.

Also, if it were really a simple universal remote, there would be no practical way to get a "look" at the learned signals. If it is some more advanced device (or a program running on a PC) there is probably some way (if you knew how) to get access to the learned signals as a PC file. There is a fair chance I could look at such a file and deduce how the IR is encoded and figure out the problem.



How did you try that?

The normal way is to be very careful not to press any other button on the original remote between pressing a button to learn one toggle state and pressing the same button again to learn the other.


Your logic is absolutely correct. Those test results seem to rule out the theory that a toggle bit is causing the problem.

But I know Dell typically uses MCE, which I know has a toggle bit.

Those test results seem to rule out almost all theories.

A learning remote would reliably learn the beginning of an overly complex signal. If that weren't enough then it would never work. If the signal were something like an IR keyboard with a press signal and a release signal, you probably fail to learn the release part. Then (like a toggle bit but for a different reason) the signal works but doesn't work again. The real remote sends "release", which fixes things, but only AFTER it sends the beginning of the signal. So you test kills that theory as well.

Given a test that kills all reasonably theories, I have to wonder about user error in conducting the test.

Is there any test that I could perform that would help better understand this problem and possibly help alleviate user error?

The JDS will show codes, but I think it is its own proprietary codes. I can send some to you in about 2 hours.

I feel when I am learning an IR code, if I hold the original remote button down longer it works a little better. I wonder if it has something to do with a “Release” function as you mentioned. Maybe the TV buffers IR commands and the buffer is getting filled. This might explain the activity, but I know nothing about how TV IR works and if this would even be a common practice.

Here is JDS debug info

00,03,7F,07,6D,22,13,10,13,11,12,23,13,22,37,22,13,11,13,10,13,11,12,11,13,10,13
,11,13,10,13,11,12,11,25,10,13,10,13,23,12,11,25,22,13,10,13,11,12,11,13,10,13,1
1,13,10,13,11,12,11,25,22,13,10,EE,00,5F,00,DB,02,D9,20,B3,00,9F,A8,D3,00,FF,84,
FF,04,F5,81,F3,00,AD,80,9E,00,6F,00,DF,02,FF,00,


00,03,7F,07,6D,21,14,0F,14,10,13,23,13,22,37,22,13,11,13,10,13,11,12,11,13,10,13
,11,13,10,13,11,12,11,25,10,13,11,12,23,12,11,13,10,13,11,13,10,13,11,12,11,13,1
0,13,11,13,10,13,11,12,11,25,22,EE,00,5F,00,DB,02,D9,20,B3,00,9F,A8,D3,00,FF,84,
FF,04,F5,81,F3,00,AD,80,9E,00,6F,00,DF,02,FF,00,