JP1 Remotes

This projector, when setup in some strange way beyond my control, cannot be shut off with a standard Panasonic2(LCD) power command from 8910 nor from the OEM remote. But it can be shut off using a 3-button sequence on the OEM remote: press and hold a trigger, press power, release the trigger.

So I recorded it as a macro in IRSA. The press and release patterns are different. The number of bits differs. There are 4 types of timing pairs (1-1, 2-2, 2-1, 1-8 ) which I have no idea how to even begin to interpret. Leadout (not in IRSA) is roughly 130ms for hold section, based on a IR trial. Data in diagnosis, no rush, this is curiosity item PanasonicPT-L6500U-trigger.zip, somewhat related in my mind to the new thread about UI for learned signals in RMIR.

Would something like this require one executor+protocol? two? Can it be made? I tried enlightning myself via DecodeIR.txt but failed. Jon Armstrong was once helping with this projector but this button wasn't discussed then. I'm actually more interested how all this would look in PB and RMIR than getting it to work.

Have you tried learning the sequence using a JP1 remote, and if so, did the learned signal also work? If so, lets see what you got learned.

Yes I did. It's in the zip file. But the only thing I managed to get is the press section and part of the power signal. So I didn't try actually using the learned signal. I know the Power itself works to turn on, and in some configuration it can turn off, in others it requires that trigger button.

In case it's not clear from the files - I have no problem with any of the Panasonic2 stuff, just the non-Panasonic section. In case you need to see the whole picture, see this upgrade https://www.hifi-remote.com/forums/dload ... le_id=1480

Here's what I got from the text files you posted. It appears that the three events each send a signal, so you ought to be able to learn each of the three evens using your JP1 remote. The test files don't contain the carrier frequency, which the JP1 learns would give you, and we would need that in order to create new executors.

step 1) press bottom button
lead-in: +1200 -600
leadout: -5000
0 = -600 +600
1 = +600 -600

000000001000000011
000000001000000011

gap: 140k

step 2) press power
lead-in: +3496 -1766
leadout: +406 -63193
0 = +405 -461
1 = +407 -1321

01000000 00000100 00000001 00010010 00000000 10111100 10101111
01000000 00000100 00000001 00010010 00000000 10111100 10101111
01000000 00000100 00000001 00010010 00000000 10111100 10101111

gap: 1,768k

step 3) release button button
lead-in: +1200 -600
leadout: -5000
0 = -600 +600
1 = +600 -600

000000000000000011
000000000000000011

The frequency is 36866-37037 for all the non-Panasonic2 samples I've seen. 36866 is most common.

Why on "press" do I see a missing bit that "release" has and your interpretation doesn't?

You're interpreting 1-8 (589-5015 approx) time ratio as leadout before duplicating the same thing? right? But where do you see the sign flip +600-600 and -600+600? (I have no problem with rounding) and how did you get at '11' at the end?

What's a gap? Is the gap, for which you report the time, taken from IRSA recording? That's the time it took ME to push the other button, not a constant, and not quick. IR file tels me it's 130ms, am I reading it wrong?

I'll try to gather more info. but it's rough to get hold of that remote. Transition from power to release would be useful - I tried it before and failed, can try again.

Please don't worry about decoding step2. Bits match what they should be. The protocol exists in jp1 and works fine. I didn't want to confuse this thread by supplying the Panasonic2 side, and the keyboard-like signals from some buttons, should I?

ElizabethD wrote: where do you see the sign flip +600-600 and -600+600?

Experience.

Two durations in a row of the same type are added together by any IR capture device. So if -600 -600 is transmitted then -1200 (plus or minus a small error) will be captured.

So if zero is -600 +600 and One is +600 -600 then anyplace a zero is followed by a one the two +600's get merged into a +1200 and anyplace a one is followed by a zero the two -600's get merged into -1200.

So an expert looks a a collection of +-600 and +-1200 and knows to check whether they would make more sense after unmerging the 1200's into pairs of 600's. If the result doesn't make more sense after unmerging, then you try some other way of interpreting things. I didn't to the grunt work myself, but I assume Rob got this right.

ElizabethD wrote: That's the time it took ME to push the other button, not a constant, and not quick. IR file tels me it's 130ms, am I reading it wrong?

The format can't store values larger than that. So any duration from an actual 130ms up to the max it is able to capture would be stored as 130ms.

ElizabethD wrote:it's rough to get hold of that remote.

Do you not have this remote, are you borrowing it from someone?

No remote.
No projector. Not mine. Rare opportunity to grab hold of the remote. Projectors are setup and controlled from boxes bigger than a fridge, with the exception of 1 or 2 where the remote exists. Using 8910 I've been free to do what I want except shutting off.

Signals
Thanks for the explanations so far. Wow This sure isn't simple.
> The uneven number of bits: I vaguely see now how Rob made them the same. Having seen the splitting explanation, I am connecting it now to what Tommy wrote about this sort of thing in his IRSA instructions. Thanks.
> 130ms - OK, I understand.

Possible scope creep
While I just need only the trigger job, this projector has several other variations on this theme where decoding was rough, but the workaround for me are Panasonic2 codes from a ccf file. Do you want to deal with those or just forget it? I have several IR files where Jon Armstrong initially saw
{600,36.9k,msb}<-1,1|1,-1>(2,-2,F1:8,F2:8,1:1,1,-^53)+
{600,36.9k,msb}<-1,1|1,-1>(2,-1,F1:8,F2:8,1,-^53)+
{600,36.9k,msb}<-1,1|1,-1>(3,-2,F1:4,F2:7,1,-^53)+
I also have a bunch of IRSA files with very long runs. Your call.