Trying to decode Sassy Baby Mobile

[crust]

I have a sassy baby mobile, shown here http://sassybaby.com/sassybaby/finditem.cfm?itemid=189. I am trying to make a protocol upgrade for my 2117 Extender 3, but have not had much luck.

Device: Sassy Music Motion
Type of device: Baby Mobile
JP1 Remote model: 2117 Ext 3
JP1 user? Y
Still have original remote? Y
Checked the file section? Y
Checked the Pronto file section (at R/C)? Y
Partially working setup code? (no or setup code) N
Learning remote question? Y

I captured the signals and can replay them successfully from my remote, but IR was not able to decode them. In IR6, the text boxes in the middle of the screen vanished, but IR4 seems to work ok. I posted the file with the captured signal. My difficulty is that I am not able to figure out how to get from the 1s and 0s to the actual protocol upgrade hex. I do know how to find the 1s and 0s. Is there a tutorial for that step somewhere? Here is a link to the IR file with the captured signals:

http://www.hifi-remote.com/forums/dload ... le_id=2013

Thanks, baby and dad will appreciate it.

[johnsfine]

Does the original remote have only those four buttons?

Did you use a long press when learning?

Each of your learned signal is 5 frames long. Maybe because the signal needs to be 5 frames long, or maybe because that is just what happened when learning.

Each of the 5 frames is 5 bursts long. 5 bursts consists of 5 pulses with the 4 gaps between them plus a long gap between frames.

The pattern of the 5 pulses seems to be fixed (first long and the rest short). The difference between the four commands is which gap is long.

You could encode it in protocol builder with:
Lead-in = 1586, 562
One = 528, 562
Zero = 0, 524
Lead-out style 2
Lead-out off = 10868

Then you have 4 bits of OBC having the four possible values 7, 11, 13 or 14.

[crust]

Does the original remote have only those four buttons?

Yes, just 3 sounds and off.

Did you use a long press when learning?

Well during learning, i just held it down until the 2117 indicated that it had learned the signal. I did some analysis before my initial posting and realized that it repeated a couple of times, but due to the way it functions, the repeat probably makes no difference.

You could encode it in protocol builder with:
Lead-in = 1586, 562
One = 528, 562
Zero = 0, 524
Lead-out style 2
Lead-out off = 10868

Then you have 4 bits of OBC having the four possible values 7, 11, 13 or 14.

Ok, so this is the part that interests me, is there a reference or short howto that explains how you came up with some of this. The lead in and lead out, as well as 1 and 0 are easy, but the "style", and in other posts I have seen "MSB" as well. I have no idea how those are decided. I will give this a go. Thanks so much!

[johnsfine]

I made those numbers (7, 11, 13 and 14) assuming LSB, rather than MSB.

Last time I tried it, KM had some problems in custom upgrades that are MSB and fewer than 8 bits in the OBC. (This one is 4 bits).

With this few commands the choice of MSB vs. LSB is entirely arbitrary, so it's simpler to use LSB and not worry about whether KM still has that problem.

The "MSB" that DecodeIR generated for these signal (as part of incorrectly guessing that they are phase encoded) just meant to explain the sequence of bits it was displaying. It doesn't say anything about the actual protocol.

[crust]

Well, maybe i am either tired, or a fool, or both. I am using protocol builder v3.11 and I have enabled the macros. My problem is that i cannot figure out how to type in a protocol ID that the spreadsheet takes. I did use the howto, but it did not help. I type in 01FF, 1FF, 1FE, 0, none work. I also tried in C syntax and with a trailing h. What gives, how do you use this thing?

Thanks everyone