Barthelmess Christmas tree lights

ledan · Post by **ledan** » Sat Nov 20, 2004 11:25 pm

Hi. I cant get information from IR learned signal to make an upgrade device with. Could you help. thanks

Barthelmess-Christmas-tree-lights.txt

Capn Trips · Post by **Capn Trips** » Sun Nov 21, 2004 7:12 am

Try reading this first

jon_armstrong · Post by **jon_armstrong** » Sun Nov 21, 2004 7:38 am

From the ten comands it looks like:

{37.6,540}<1,-3|3,-1>(D:15,F:8,P:1,1,-31)+

The last nine bits vary and the final bit appears to be a parity bit with those ten commands. These are what I decode in hex:

50 1
80 2
18 3
60 4
00 5
06 6
07 7
19 8
81 9
61 0

What is the equipment that these commands control? You might as well learn all the other commands and post them in the diagnostics section. It may take more than one file depending on the number of commands.

ledan · Post by **ledan** » Sun Nov 21, 2004 8:36 am

I'm not sure how to read information above.

1. Device: Christmas tree light
2. Type of device: Barthelmess Christmas tree light
3. Remote model: URC-8811
4. JP1 user? YES
5. Still have original remote? YES
6. Checked Yahoo file section? YES
7. Checked Pronto file section (at R/C)? YES

I want to make an upgrade device but I need protocl and EFC command. How can I get some information to read my TXT file and get protocol and EFC code.

Thanks

Post by **The Robman** » Sun Nov 21, 2004 10:58 am

Ledan,
Are those 10 buttons the only buttons on the Xmas tree remote? If there are more buttons, could you please learn all of them. It might take several passed to get them all, but that's OK. We already have the first 10, so we don't need those again. Also, please mention what each button really is so we can label the upgrade appropiately.

The reason we need more buttons is this, fromthe 10 codes posted, it appears that the last bit of the signal is a repeat of the 2nd to last bit, but we would need to see all the signals to know if this holds true for them all.

Also, please don't post IR files in these threads, please post the file to the Diagnosis Area and then post a link, just like I did in your first post.

Post by **The Robman** » Sun Nov 21, 2004 12:06 pm

OK, unless we get any information to the contrary, I have built an upgrade based on the assumption that the last bit is always a repeat of the previous bit. Here it is...

Upgrade Code 0 = 37 D0 (Video Acc/2000) Barthelmess Christmas tree light (KM v8.30)
FF 00 81 AA AA 79 F5 FE E7 F9 FF 9F 1F 67 7E
End

Upgrade protocol 0 = 00 FF (S3C8+) Barthelmess Christmas tree light (PB v3.10)
3F 95 21 8B 15 8A 41 08 08 01 0E 03 0C 03 20 00
FA 20 9E FF FF FF FF 07 FF 01 08 05 10 C0 C0 06
20 11 8D 01 46
End

ledan · Post by **ledan** » Sun Nov 21, 2004 1:03 pm

Wow. I dont know how you do that but I need information or practice.
The original remote has ten button. Here's a description.
COLOR LIGHT:
1:ON
2:OFF
3:TWINKLE/FLASH
4:ALL FADE
5:SEQUENTIAL
6:SINGLE FADE
CLEAR LIGHT:
7:TWINKLE/FLASH
8:ALL FADE
9:OFF
10:ON

I have tried the device and protocol upgrade you gave me and it's partly working. Here is what is working.
COLOR LIGHT:
1:ON
2:Seem to do same as 5
3:TWINKLE/FLASH
4:ALL FADE
5:SEQUENTIAL
6:SINGLE FADE
CLEAR LIGHT:
Nothing is working.

Thanks for all your help.

Post by **The Robman** » Sun Nov 21, 2004 1:43 pm

OK, silly mistake. I did a rotate left when I should have done a rotate right. This protocol should fix it...

Upgrade protocol 0 = 00 FF (S3C8+) Barthelmess Christmas tree light (PB v3.10)
3F 95 21 8B 15 8A 41 08 08 01 0E 03 0C 03 20 00
FA 20 9E FF FF FF FF 07 FF 01 08 05 C0 C0 C0 06
20 11 8D 01 46
End

And here's an upgrade with the buttons laid out in a more logical manner:

Upgrade Code 0 = 37 D0 (Video Acc/2000) Barthelmess Christmas tree light (KM v8.30)
FF 00 68 A8 81 AA AA F5 FE F9 79 7E 9F E7 1F FF
67
End

Here's the RM file, in case you want to re-arrange the buttons...
http://www.hifi-remote.com/forums/dload ... e_id=26795

And just for reference, here's the PB file...
http://www.hifi-remote.com/forums/dload ... le_id=1388

ledan · Post by **ledan** » Sun Nov 21, 2004 4:19 pm

Everything is working perfect !!

Thanks for your help.

Is there any information that I can read to do what you've done from my file ?

Post by **The Robman** » Sun Nov 21, 2004 4:34 pm

ledan wrote:Is there any information that I can read to do what you've done from my file ?

Could you re-phrase that, I don't quite follow.

ledan · Post by **ledan** » Sun Nov 21, 2004 6:14 pm

Sorry for my english...

I want to know if it exist documentation on how to make device and protocol upgrade from a file containing learned code like I post here because I really dont know how you have done that.

Post by **The Robman** » Sun Nov 21, 2004 6:43 pm

Oh I see what you mean, you want to know how I did what I did. First I copied all of the timing info from your learned signals into a text file. Then I looked at the data to see if I could spot the logical burst pairs, in this case the pairs are "+540 -1600" and "+1600 -540". Then I used the Protocol Builder spreadsheet to create the basic protocol.

There is a help document that describes how to use Protocol Builder here:
http://www.hifi-remote.com/files/help/P ... uilder.zip
http://www.hifi-remote.com/wiki/index.p ... ol_Builder

And you can download PB itself from here:
http://www.hifi-remote.com/forums/dload ... le_id=3591

However, PB has changed since I wrote that doc as it now includes an assembler, which is a very cool addition.

Your signals have 15 bits of fixed data and 9 bits of variable data. Here's the binary pattern...

101010101010101 011110011
101010101010101 111101011
101010101010101 111111100
101010101010101 111001111
101010101010101 111110011
101010101010101 111111111
101010101010101 100111111
101010101010101 000111111
101010101010101 011001111
101010101010101 011111100

Normally, we like to see 8 bits of variable data as this fits nicely into 1 byte, but looking at the above patterns you'll notice that the last bit is always a repeat of the previous bit, so I programmed the upgrade to use 8 variable bits and to generate the final bit itself.

In PB, I set the code to use 2 device code bytes, one of 8 bits and another of 7 bits. I then set it to use 2 variable bytes, one of 8 bits and another of just 1 bit.

Then, in the new Assembler sheet, I used the "Load Generated Code" button to convert what I had done so far into assembler. I then converted the format from 2-2 (ie, 2 device codes, 2 command codes) to 2-1 by editing the 3rd byte from 22 to 21.

I then added logic in assembler to do the following:

1. Copy the variable byte to a scratch register (LD RC0,R05)
2. Rotate the scratch register right in the CARRY bit (RRC RC0)
3. Rotate the carry bit into the 2nd command byte (RRC R06)
4. Increase the number of command bytes from 1 to 2 (INC R11)
5. Jump to the IR engine to send the signal (JP $0146)

You can look at what I did either by opening the PB file I created using the PB spreadsheet, or just by looking at the disassembled version of the protocol in IR.exe.

As you can see, this is a little bit above what we can expect a beginner, or even a mid-level JP1 user to be able to do, that's why the experts are usually more than happy to jump in and take care of this sort of stuff for you.