METechs Drapery Control

kkl · Post by **kkl** » Tue Aug 16, 2011 10:54 am

This probably won't help, but I have a TIRA IR receiver/transmitter and captured the codes from the METechs remote using the TIRA protocol analyzer (http://www.home-electro.com/jp1.php). It can also utilize decodeir.dll, but that appears to be the same info as IR.

Edit: From description - "As of Feb 8, 2003 Protocol Analyzer supports the following protocols: NEC (one and two device word versions), RC5, JVC, and JVC1. For unsupported protocols it prints uprocessed timing of the signal. All times are in microseconds, odd fields are duration of bursts, even fields are duration of pauses."

Code: Select all


C:\Users\default.M7650N\Downloads\ProtAn>pa.exe com
IR Protocol Analyzer, ver. 0.3a (c) 2003 Home Elect
Using decodeIR!!
=========================================

=========================================

Key 1
Uknown Protocol, printing unprocessed data:
424     1256    408     1256    416     1256    416     1256    1248    424
408     1256    1256    416     416     1256    408     1256    416     1256
416     1248    416
Key 2
Uknown Protocol, printing unprocessed data:
424     1256    408     1256    416     1256    416     1256    1248    416
416     1256    416     1256    1248    424     408     1256    416     1256
416     1256    408
Key 3
Uknown Protocol, printing unprocessed data:
424     1256    408     1256    416     1256    416     1256    1248    416
416     1256    416     1256    416     1248    1256    416     416     1256
408     1256    416
Key 4
Uknown Protocol, printing unprocessed data:
424     1256    408     1256    416     1256    416     1256    1248    416
416     1256    416     1256    416     1248    416     1256    1248    424
416     1248    416
Key 5
Uknown Protocol, printing unprocessed data:
424     1256    408     1256    416     1256    416     1256    1248    424
408     1256    416     1256    416     1248    416     1256    416     1256
1248    416     416
Key 6
Uknown Protocol, printing unprocessed data:
424     1256    408     1256    416     1256    416     1256    1248    416
416     1256    416     1256    408     1256    416     1256    416     1256
408     1256    1256
Key 7
Uknown Protocol, printing unprocessed data:
424     1256    408     1256    416     1256    416     1256    408     1256
1256    416     1248    424     416     1248    416     1256    416     1256
416     1248    416
Key 8
Uknown Protocol, printing unprocessed data:
424     1256    408     1256    416     1256    416     1256    408     1256
1256    416     416     1256    1248    424     408     1256    416     1256
416     1248    416
Key <>
Uknown Protocol, printing unprocessed data:
424     1256    408     1256    416     1256    1248    424     416     1256
408     1256    416     1256    1248    424     408     1256    416     1256
416     1248    416
Key ><
Uknown Protocol, printing unprocessed data:
424     1256    408     1256    416     1256    1248    424     416     1248
416     1256    416     1256    408     1256    1256    416     416     1256
416     1248    416
Key Stop (Left)
Uknown Protocol, printing unprocessed data:
424     1256    408     1256    416     1256    1248    424     416     1248
416     1256    1248    424     416     1256    408     1256    416     1256
416     1256    408
Key Stop (Right)
Uknown Protocol, printing unprocessed data:
424     1256    408     1256    416     1256    1248    424     416     1248
416     1256    1248    424     416     1248    416     1256    416     1256
408     1256    416

Post by **The Robman** » Tue Aug 16, 2011 11:10 am

For any other device, these learns would be considered identical. The format is exactly the same and now the carrier frequencies are exactly the same, all that is left is some VERY minor differences in the burst times, which no other device on the planet would care about!

I have tweaked the burst times in this version just a little, so hopefully now they will be identical to the TV learns.

https://www.hifi-remote.com/forums/dload ... le_id=9835

If they still don't work, and the learns come back as identical as I expect them to, the only thing left will be the duty cycle, but if you've tried testing the remote up close, that's unlikely to be the fix.

Just in case you're interested, here are the burst times from the good learns:

+420 -1272
+1260 -432

and here are the ones from the bad learns:

+420 -1268
+1260 -412

Post by **vickyg2003** » Tue Aug 16, 2011 11:24 am

The Robman wrote:For any other device, these learns would be considered identical. The format is exactly the same and now the carrier frequencies are exactly the same, all that is left is some VERY minor differences in the burst times, which no other device on the planet would care about!

Amen!

kkl · Post by **kkl** » Tue Aug 16, 2011 11:24 am

Woohoo!! You've done it. Appears to work perfectly. It's a PITA to check the number keys because I have to climb up on a ladder and change the DIP switches one at a time and then power cycle, but they all seem to work too. It works from across the room also. Great job!

Post by **The Robman** » Tue Aug 16, 2011 11:41 am

Great, glad we were able to put this one to bed.

kkl · Post by **kkl** » Tue Aug 16, 2011 11:46 am

How do I get this into RM/KM for posting, or are you planning to do that?

Post by **The Robman** » Tue Aug 16, 2011 11:52 am

I'm doing that now.

KM upgrade file here:
https://www.hifi-remote.com/forums/dload ... le_id=9843

And just FYI, the PB file is here:
https://www.hifi-remote.com/forums/dload ... le_id=9844

Post by **The Robman** » Tue Aug 16, 2011 12:10 pm

kkl wrote:This probably won't help, but I have a TIRA IR receiver/transmitter and captured the codes from the METechs remote using the TIRA protocol analyzer (http://www.home-electro.com/jp1.php). It can also utilize decodeir.dll, but that appears to be the same info as IR.

I'm curious, why did you get the TIRA, do you have an interest in what IR signals look like? A better idea might have been to get an IR Widget ($29 +$5 s/h) from Tommy Tyler, but it looks like the TIRA did a good job of capturing the signals.

kkl · Post by **kkl** » Tue Aug 16, 2011 1:31 pm

It's been so long, I can't even remember why I bought it. I think it was for HTPC control, before the Windows MCE became popular. Everytime I've started to even try to understand IR codes, my head starts to hurt. I do better with fuzzy logic than with binary.

Funny you should ask about the TIRA, because it's the reason I started looking for a JP1 protocol. I want to use EventGhost on my PC with TIRA to control the drapery controller. However, it doesn't seem to work. I wonder if the TIRA is not accurate enough for the specificity you found that the METechs requires.

Could you look at this code briefly and tell me if anything jumps out at you? Here's what EventGhost gets from TIRA for the OPEN button on the original remote.

Code: Select all

00 00 00 00 08 00 00 00 35 00 00 00 9D 00 00 00 ED 10 00 00 08 2B 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 34 88 27 05 00 00 00 00 00 00 00 00 02 00 00 00 0C 00 00 00 84 67 00 00 00 00 00 00 5F 00 00 00 FF FF FF FF 00 01 00 01 00 01 01 00 00 01 00 01 00 01 01 00 00 01 00 01 00 01 00 02 00 01 00 01 00 01 01 00 00 01 00 01 00 01 01 00 00 01 00 01 00 01 00 03 00 01 00 01 00 01 01 00 00 01 00 01 00 01 01 00 00 01 00 01 00 01 00 02 00 01 00 01 00 01 00 01 00 01 00 01 00 01 00 01 00 01 00 01 00 01 00

Here's the JP1 (15-2117) code. Only a few digits different.

Code: Select all

00 00 00 00 08 00 00 00 35 00 00 00 9D 00 00 00 ED 10 00 00 15 2B 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 34 88 27 05 00 00 00 00 00 00 00 00 02 00 00 00 0C 00 00 00 84 67 00 00 00 00 00 00 5F 00 00 00 FF FF FF FF 00 01 00 01 00 01 01 00 00 01 00 01 00 01 01 00 00 01 00 01 00 01 00 02 00 01 00 01 00 01 01 00 00 01 00 01 00 01 01 00 00 01 00 01 00 01 00 03 00 01 00 01 00 01 01 00 00 01 00 01 00 01 01 00 00 01 00 01 00 01 00 02 00 01 00 01 00 01 01 00 00 01 00 01 00 01 01 00 00 01 00 01 00 01 00

Do you see any changes that could be made that matches the code you finally used? If you don't have time or interest, I completely understand.

Post by **The Robman** » Tue Aug 16, 2011 3:17 pm

I don't know how to read the EventGhost/TIRA code so I don't know what effect the differences would have. Do you know how to read it, or is it documented somewhere.

Let's say that I did know how to read it, what would you do with that knowledge? In other words, do you have a way to tweak the signal like how I did with the JP1 signals (both the learned and the upgrade).

kkl · Post by **kkl** » Tue Aug 16, 2011 4:46 pm

The Robman wrote:I don't know how to read the EventGhost/TIRA code so I don't know what effect the differences would have. Do you know how to read it, or is it documented somewhere.

Let's say that I did know how to read it, what would you do with that knowledge? In other words, do you have a way to tweak the signal like how I did with the JP1 signals (both the learned and the upgrade).

I was afraid you might say that. Too much to ask for some sort of universal IR language. I found (Edit: Corrected Link)this document which I believes describes the format in the Timing section of page 1.

There is a way to edit the signal in EventGhost with the TIRA Plug-in. It learns the key, and then you can transmit that signal. The sample code I provided is in the format that can be edited. The code for the <> (OPEN) command above should somehow correspond to this code (I think), captured by TIRA's protocol analyzer:

Code: Select all

Key <>
Uknown Protocol, printing unprocessed data:
424     1256    408     1256    416     1256    1248    424     416     1256
408     1256    416     1256    1248    424     408     1256    416     1256
416     1248    416

kkl · Post by **kkl** » Thu Aug 18, 2011 3:19 pm

Since interpreting the TIRA codes doesn't appear to be feasible, I thought I'd try my MCE IR receiver/blaster. EventGhost with this device uses IR codes in the Pronto format. I used the "Export Pronto" function in IR.exe for the learned keys, but they don't appear to work. I downloaded the MakeHex program, but it appears to require an *.irp file for each protocol.

Questions:
1. How would I get an *.irp file for this new super-sensitive F12 protocol?
2. Are Pronto codes robust enough to maintain high fidelity to the source protocol?

Thanks.

Post by **The Robman** » Thu Aug 18, 2011 4:04 pm

The pronto code generated by IR.exe looks good to me, but as we already know, you have the fussiest drapes on the planet! So, what you should try to do is learn the signals generated by the MCE and see what they look like. We can't tweak the pronto hex until we know whether we need to teak the times UP or DOWN.

kkl · Post by **kkl** » Thu Aug 18, 2011 9:56 pm

I got it working! It's been years since I used the MCE IR blaster. I mistakenly thought that there was a blaster in the main unit. Once I plugged in an IR emitter and used that, it worked!

For the sake of completeness, I'll post the Pronto codes here:

OPEN
0000 0073 0000 0030 0010 002D 0010 002B 0010 002D 002F 000E 0010 002B 0010 002D 0010 002D 002D 000E 0010 002D 0010 002B 0010 002D 0010 04E8 0010 002D 0010 002D 0010 002B 002F 000E 0010 002D 000E 002D 0010 002D 002F 000D 0010 002D 0010 002D 0010 002B 0010 0C7D 0010 002D 0010 002B 0010 002D 002F 000E 000E 002D 0010 002D 0010 002B 002F 000E 0010 002D 0010 002B 0010 002D 0010 04E8 0010 002D 0010 002D 000E 002D 0010 002D 0010 002D 000E 002D 0010 002D 0010 002B 0010 002D 0010 002D 0010 002B 0010 0E14

CLOSE
0000 0073 0000 0030 0010 002D 0010 002B 0010 002D 002F 000E 0010 002B 0010 002D 0010 002D 000E 002D 002F 000E 0010 002B 0010 002D 0010 04E8 0010 002D 0010 002D 0010 002B 002F 000E 0010 002D 000E 002D 0010 002D 0010 002D 002D 000E 0010 002D 0010 002B 0010 0C7D 0010 002D 0010 002B 0010 002D 002F 000E 000E 002D 0010 002D 0010 002B 0010 002D 002F 000E 0010 002B 0010 002D 0010 04E8 0010 002D 0010 002D 0010 002B 0010 002D 0010 002D 000E 002D 0010 002D 0010 002B 0010 002D 0010 002D 0010 002B 0010 0E14

STOP
0000 0072 0000 0030 0010 002C 0010 002D 0010 002D 002D 000F 0010 002D 0010 002D 002D 000F 0010 002D 0010 002D 000F 002D 0010 002D 0010 04F3 0010 002D 0010 002C 0010 002D 002F 000F 0010 002C 0010 002D 002F 000F 000F 002D 0010 002D 0010 002C 0010 002D 0010 0C99 0010 002C 0010 002D 0010 002D 002D 000F 0010 002D 0010 002D 002D 000F 0010 002D 0010 002C 0010 002D 0010 002D 0010 04F3 0010 002D 0010 002C 0010 002D 002F 000F 000F 002D 0010 002D 0010 002C 0010 002D 0010 002D 0010 002C 0010 002D 0010 0E34

8
0000 0072 0000 0018 0010 002D 0010 002C 0010 002D 0010 002D 000F 002D 002F 000F 0010 002C 002F 000F 0010 002D 0010 002C 0010 002D 0010 04F3 0010 002D 0010 002D 0010 002C 0010 002D 0010 002D 002D 000F 0010 002D 002D 000F 0010 002D 0010 002D 0010 002C 0010 0E34

I looked through several dozen posts for an answer to the RM/KM to *.irp question and didn't find one. I'm assuming that there's no direct method, that you have to understand the "behind the scenes" workings.

Robman, I truly appreciate your help with this.

3FG · Post by **3FG** » Thu Aug 18, 2011 11:33 pm

So the duration of the 1 bits is slightly shorter than standard F12. But I'm curious: The format of the open/close/stop signals is not standard F12. For example, Open is function 8.1 sent as a pair followed by the signal in a second pair. But the 4th signal is device 0.0, function 0. This is not a legal F12 signal, because it imples that none of the 3 groups of buttons was pressed.

Do you know if this 8..8......8..0 format is required for this non-digit buttons?

From DecodeIR.html in version 2.43 of DecodeIR:
IRP notation: (Toshiba specification) K=(D:3,H:1,A:1,B:2,F:6) {37.9k,422}<1,-3|3,-1>(K,-34,K) for A=1 or B=1
{37.9k,422}<1,-3|3,-1>(K,-34,K,-88,K,-34,K)+ for H=1. Exactly one of H, A, or B can have a value of 1. If H=1 the signal can be sent repeatedly, and F can take any 6 bit value.
If A or B=1, the signal is sent once only per button press, and only a single bit of F can be non-zero.

IRP notation: (JP1) K=(D:3,H:1,F:8) {37.9k,422}<1,-3|3,-1>(K,-34,K) for H=0.
{37.9k,422}<1,-3|3,-1>(K,-34,K,-88,K,-34,K)+ for H=1.
A and B are subsumed into F, and the value of H is computed in the executor. H=A^B.
DecodeIR reports H as the subdevice. This is useful when making a Pronto Hex file, or other description based on durations. Remotes with executors (e.g. UEI remotes) normally compute the value of H in the executor, and the "subdevice" is not needed as a parameter.