Elan V883

jetskier · Post by **jetskier** » Wed May 26, 2010 3:48 pm

Here's a new one I can't get decoded. It appears to follow something resembling the NEC2 format, but there's a burst pair between the complement. DecodeIR can't do a thing with it, but ExchangeIR produced this IRP notation.

{40.2k,397,msb}<1,-1|1,-2|2,-1|2,-2>(3,-2,A:66,1,-14.0m)+{A=$00011554700015554}; Alt leadout form: ^50m

0000 0067 0000 0023 0030 0020 0010 0010 0010 0010 0010 0010 0010 0010 0010 0010 0010 0010 0010 0020 0010 0010 0010 0020 0010 0020 0010 0020 0010 0020 0010 0020 0010 0020 0010 0010 0010 0020 0020 0020 0010 0010 0010 0010 0010 0010 0010 0010 0010 0010 0010 0010 0010 0010 0010 0020 0010 0020 0010 0020 0010 0020 0010 0020 0010 0020 0010 0020 0010 0020 0010 0010 0010 0233

{40.2k,397,msb}<1,-1|1,-2|2,-1|2,-2>(3,-2,A:66,1,-14.0m)+{A=$00011554700045551}; Alt leadout form: ^50m

0000 0067 0000 0023 0030 0020 0010 0010 0010 0010 0010 0010 0010 0010 0010 0010 0010 0010 0010 0020 0010 0010 0010 0020 0010 0020 0010 0020 0010 0020 0010 0020 0010 0020 0010 0010 0010 0020 0020 0020 0010 0010 0010 0010 0010 0010 0010 0010 0010 0010 0010 0010 0010 0020 0010 0010 0010 0020 0010 0020 0010 0020 0010 0020 0010 0020 0010 0020 0010 0010 0010 0020 0010 0233

{40.2k,397,msb}<1,-1|1,-2|2,-1|2,-2>(3,-2,A:66,1,-14.0m)+{A=$00011554700045551}; Alt leadout form: ^50m

0000 0067 0000 0023 0030 0020 0010 0010 0010 0010 0010 0010 0010 0010 0010 0010 0010 0010 0010 0020 0010 0010 0010 0020 0010 0020 0010 0020 0010 0020 0010 0020 0010 0020 0010 0010 0010 0020 0020 0020 0010 0010 0010 0010 0010 0010 0010 0010 0010 0010 0010 0010 0010 0020 0010 0010 0010 0020 0010 0020 0010 0020 0010 0020 0010 0020 0010 0020 0010 0010 0010 0020 0010 0233

Post by **The Robman** » Wed May 26, 2010 7:12 pm

I imported these codes into IR and then converted them to binary (using 0= +400 -400 and 1= +400 -800) and got the following:

+1200 -800 00000010 11111101 +800 -800 00000001 11111110 +400 -14000
+1200 -800 00000010 11111101 +800 -800 00000010 11111101 +400 -14000
+1200 -800 00000010 11111101 +800 -800 00000010 11111101 +400 -14000

So, I do see the extra "+800 -800" pair that you were talking about. We call this a "mid frame burst pair". Unfortunately, I don't think there's any way to generate that particular mid-frame burst pair using the IR engine, so we would have to generate the signal by hand (ie, we'd need to write the assembler).

xnappo · Post by **xnappo** » Wed May 26, 2010 7:25 pm

The Robman wrote:I imported these codes into IR and then converted them to binary (using 0= +400 -400 and 1= +400 -800) and got the following:

+1200 -800 00000010 11111101 +800 -800 00000001 11111110 +400 -14000
+1200 -800 00000010 11111101 +800 -800 00000010 11111101 +400 -14000
+1200 -800 00000010 11111101 +800 -800 00000010 11111101 +400 -14000

So, I do see the extra "+800 -800" pair that you were talking about. We call this a "mid frame burst pair". Unfortunately, I don't think there's any way to generate that particular mid-frame burst pair using the IR engine, so we would have to generate the signal by hand (ie, we'd need to write the assembler).

Just curious - has it ever been the case that we have found an error in the protocol generation of the remote we are trying to duplicate? I mean - maybe it is a mistake and isn't really needed?

xnappo

Post by **The Robman** » Wed May 26, 2010 8:27 pm

No, I don't think we've ever found an error, but sometimes we find that the signal doesn't need to be as exact as the learns might imply.

I can never remember the two ways that we can generate mid frame burst pairs, so I need to document it and post it somewhere, but if I recall, one way just repeats the leadin pair and the other way combines two partial pairs (but I forget which).

jetskier · Post by **jetskier** » Wed May 26, 2010 9:26 pm

Yeah, that's what I came up with too when I broke it down last time I messed with it in November. This was my first attempt at learning decodes. Like you said, there's not a lot of documentation on the mid frame bursts that I could find when I played with these codes. I tried PB, but I'm not an expert there. I couldn't figure out the midframe burst settings.

I don't actually have the piece of equipment to test the signal (backed out on the purchase). I did try to build an upgrade in the process. It looks like an interest protocol. If work picks up again, I may end up getting the video switcher after all. And I know you guys like challenges.

Post by **The Robman** » Wed May 26, 2010 9:32 pm

Do you have a complete set of learns?

jetskier · Post by **jetskier** » Wed May 26, 2010 9:36 pm

Just the Pronto Hex of the 64 or so codes (8X8 matrix)

Post by **The Robman** » Thu May 27, 2010 6:24 am

Could you post them please. Or better yet, use the new IR functionality to add them all to an IR file and post that.

jetskier · Post by **jetskier** » Thu May 27, 2010 7:58 am

Is there a batch processor in the new IR to import more than one at a time?

Post by **The Robman** » Thu May 27, 2010 9:17 am

Nope.

jetskier · Post by **jetskier** » Thu May 27, 2010 11:34 am

I just discovered there are 256 codes for this device. Apparently you can stack 8 component inputs to 32 component outputs if you have 4 units.

I built an IR file with the first two outputs and the associated 8 inputs (16 total learns from the CCF).

I found the manufacturers file that has a format more processed than a CCF. It has the burst pairs listed first and then how the pairs are sent out. It's in a format more usable to us (binary type). I'm sure once we establish the executor, the rest of the 240 codes can easily be figured by a simple binary to hex conversion.

here's what I figured...

~0000OUTPUT1 INPUT1$
:00000920000092 <---Burst Pair 0 timings (00 below)
:00000920000184 <---Burst Pair 1 timings (01 below)
:00002760000184 <---Lead IN Burst Pair (02 below)
:00001840000184 <---Mid Frame Burst Pair (03 below)
:00000920003226 <---Lead Out Burst Pair (04 below)
^
:00460000 Frequency ~ 40.2kHz
^
:02 <---Lead IN Burst Pair
:00 :00 :00 :00 :00 :00 :01 :00 data
:01 :01 :01 :01 :01 :01 :00 :01 data complement
:03 <---Mid Frame Burst Pair
:00 :00 :00 :00 :00 :00 :00 :00 repeated
:01 :01 :01 :01 :01 :01 :01 :01
:04 <---Lead Out Burst Pair

The italicized comments are my notes.

Post by **The Robman** » Thu May 27, 2010 2:21 pm

I think I've figured out how to format this sort of mid-frame burst and I've created a test upgrade based on that theory.

https://www.hifi-remote.com/forums/dload ... le_id=8467

jetskier · Post by **jetskier** » Thu May 27, 2010 2:36 pm

Thanks Rob.

How does it change for the HCS08 processors? I'm sure the entry points change for the mid frame burst execution on the HCS08. That's where I got stuck earlier.

Post by **The Robman** » Thu May 27, 2010 3:59 pm

I've just updated the files to include the HCS08 protocol code too.

Keep in mind that this is all experimental, I don't have any proof that it will work yet.

jetskier · Post by **jetskier** » Thu May 27, 2010 4:05 pm

It's a start though.