Possible signal duration problem

fichmana2003 · Post by **fichmana2003** » Wed Jan 14, 2004 5:41 pm

I am experimenting with Greg's example for the 8811 and Extender 3.1. I've implemented the following for my TV:
Shift-L4 ($A8 ) - TV on/off code
Phantom1 ($3D) - TOAD off (toadtog - $09 $A8 )
Phantom2 ($3E) - TOAD on (toadtog - $88 $A8 )
Phantom3 ($3F) - Toggle on/off ($01 $A8 $A8 )
Phantom4 ($40) - All OFF macro (using X_TV;Phantom1)
Power: LKP ($61 $3F $40)
Well, to put a short story long, it doesn't work

When pressing power, the LED stutters briefly and nothing happens. Pressing Shift-L4, even a short press, immediately powers on/off the TV and the transmit LED has a clear and healthy signal indication. I replaced Phantom3 with the power button for testing, and the toadtog works *IF* I keep pressing the power > 0.5 sec; a short press again does nothing. The LKP doesn't work either (I guess related to above problem).
So... anything I can do about it?

Thanks

fichmana2003 · Post by **fichmana2003** » Thu Jan 15, 2004 12:29 pm

Isn't there anyone having a clue as to what can be done?

DGG · Post by **DGG** » Thu Jan 15, 2004 7:09 pm

What do you mean when you say you replaced Phantom 3 with the Power button? The LPK on the Power button calls Phantom 3 on a short press. BTW, 0.5 seconds seems about right for a LPK time of $6.

fichmana2003 · Post by **fichmana2003** » Fri Jan 16, 2004 2:54 am

I deleted the LPK power button keymove. I then editted TV/Phantom3 and re-assigned the button to Power. Now, when I press power it executes a toadtog (which was previously Phantom3).
simple, isn't it?

DGG · Post by **DGG** » Fri Jan 16, 2004 2:31 pm

The original code looks OK (but you already knew that) - provided the bound device in each case was TV and the OTHER keypad was assigned to the TV when you attempted to execute it (and all the other usual provisos). But, I still don't understand what you mean by "replacing Phantom3 with the Power button" or the later "reassigning the button (which button, Phantom3?) to Power".

If you moved the Toadtog toggle from Phantom3 to the Power button, I'd expect the result you indicate you're now getting. If you put a keymove on Power to call Phantom 3, I'd also expect a simple toggle based on Phantom3's original contents. (But you said you edited Phantom3. What was its keymove after the edit?) In any case, it's curious that, after you deleted the LPK from the Power key, you still have the long keypress symptoms you reported in your original message. Something doesn't "add up".

Maybe we need to start again, i.e, what are the keypresses you're now testing - after all your edits - and what's now happening?

Or, if I'm making this more difficult than necessary, maybe somebody else will jump in.
Don

fichmana2003 · Post by **fichmana2003** » Fri Jan 16, 2004 4:32 pm

At first, we had an LPK that upon a short press, executed Phantom3 - which is a toadtog that executes Shift-L4 then flips toggle-bit.

Power (LPK) -> Phantom3 (toadtog) -> Shift-L4

This didn't work, with the symptoms I described above. Since this is a rather complex operation, I decided to break it up and see which part gives me problems. I reconfigured the power button key-move to be the same as phantom3 (a toadtog), so I can see wether the toadtog is the problem. It seems it is, since a short press on the reconfigured Power key-move, which then should execute shift-L4 (which has the TV on/off code), did not work. The signal duration is too short for some reason.
A long press on the button did work (it seems a long press prolongs the signal duration), but when the toadtog is nested (as is the case with the LPK) it exhibits short-press symptoms.

I hope this is clearer...

DGG · Post by **DGG** » Fri Jan 16, 2004 5:56 pm

When you hold down a key, the remote continues to repeat the last command - in this case a power toggle code - until released. Presumably, if you continue to hold down the Power key, the device would cycle on/off (slowly, since it seems to need a long duration signal.)

I inferred from your first message that you thought you had a coding problem, though the keymoves you set out appeared correct. I doubt I can be of much help on protocol issues. Hopefully, one of the "protocol guys" will jump in. As well, there has been a lot of discussion in this forum regarding signal duration. Perhaps there's some clues in there.

Failing that, I'd be inclined to try the brute-force method, i.e., putting a DSM on Shift-L4 to send repeated power toggles interspersed with short delays generated by the Pause protocol (that would get you up to seven power-toggles over a duration under your control) or change the Toadtog keymove to call Shift-L4 several times and put 13 consecutive toggles in a DSM on Shift-L4. That would get you 78 consecutive power-toggles - 6 each way in the Toadtog x 13 in the DSM. If necessary, you could nest that concept to get however many toggles you need to make it work.

Good luck,
Don

fichmana2003 · Post by **fichmana2003** » Fri Jan 16, 2004 6:02 pm

Actually, I've already acomplished that (and worked around the problem), but I don't like it... and it takes more memory.
I just send the code twice in a row (shift-L4 Shift-L4):
Phantom3 = toadtog - $02 $A8 $A8 $A8 $A8
where $A8=Shift-L4.
It works now, but I was still hoping for a more elegant solution

thanks anywah, DGG

Mark Pierson · Post by **Mark Pierson** » Fri Jan 16, 2004 6:22 pm

DGG wrote:When you hold down a key, the remote continues to repeat the last command - in this case a power toggle code - until released.

I'm not positive, but I don't think this applies in the LDKP protocol, since it can only determine a short or long press. You'd never be able to hold anything on the Short side of the command, and I believe it also ignores it on the Long side as well.

DGG · Post by **DGG** » Fri Jan 16, 2004 6:37 pm

You're right Mark, in the 8911 at least, the last keycode on a LPK is not repeated.

fichmana2003 · Post by **fichmana2003** » Fri Jan 16, 2004 6:57 pm

It is a mystery to me as to why a single code transmit doesn't work.
If I press shift-L4, even briefly, the TV's native on/off toggle works every time. Could it be that the protocol (Proton) repeats a keypress very quickly and the TV is dependant on it?