Protocol for homemade remote

[Chester]

I have a URC-8811 using extender 3.
I would like to know how to make an upgrade protocol for a remote control I
made using the Motorola Enc./Decode pair "MC145026 MC145207"
I used 5-bit address and 4 bit data the carrier is 40K. The data sheet can be found @ http://www.alldatasheet.co.kr/datasheet ... 45027.html
I have used the learn feature and have a working IR (5.15) file witch is @
http://www.hifi-remote.com/forums/dload ... le_id=1250
Looked at PB but ???
Any help would be appreciated

[The Robman]

The info that you need for PB is as follows:

freq: 46.783
1 = +318 -2090
0 = +2114 -290
leadin = +2114 -290
leadout = 16550

I whipped up a quick protocol using PB and an upgrade using KM, see if these work;

Upgrade Code 0 = 0F D0 (TV/2000) (KM v8.31)
FF 00 80 A1 38 BF EA BA FA EE EF FE EB BB FB AA
FF
End

Upgrade protocol 0 = 01 FF (S3C8+) PB v3.11
36 71 11 8B 12 85 45 08 08 00 9F 04 01 04 21 00
7D 20 53 04 21 00 7D 8D 01 46
End

[Chester]

WoW That was quick I will try it right away and let you know
Thank You very much

[johnsfine]

Edit: While I was typing slowly, Rob answered. If his answer works, then I misunderstood that pdf. Otherwise ...

Are you using it in binary or trinary mode? (I haven't looked at your learned signals yet. Those might tell me).

The data sheet indicates two physical bits are sent per bit. In binary mode, a 1 is sent as 11 and a 0 as 00. In trinary mode those two still apply and the third value is sent as 10.

For binary mode, I think there is a bit doubling option in PB that will do the 1->11 and 0->00 translation.

For trinary mode, I can't think of a decent choice other than treating it as 18 bit binary instead of 9 bit trinary.

There seems to be no lead in and no On portion to the lead out, so a quick look at the learned data ought to show the on/off durations of "0" and "1" as well as the off duration of the leadout. That's all you need for PB.

[Chester]

IT Works GREAT
Thanks again

[johnsfine]

I looked at your learned data. By the Motorola spec it is trinary. I notice the first half bit was always 1, meaning the first trinary digit was either high or open. The last half bit was always 0, meaning the last trinary digit was either 0 or open.

If those things would be consistent, then you can (as I assume Rob did) bury the first half bit in lead-in and the last half bit in lead-out and have 16 bits in between, which I assume Rob did as 8 device bits and 8 obc bits.

[The Robman]

I didn't even look at the data sheet, I just looked at the learned data. I did indeed bury the first bit in the leadin. I had no way of knowing if there was an additional bit buried in the leadout pair as the ON time was the same for all of the learned signals. If the ON time was alternated between the two burst pair ON times, I would have had to extract it from the leadout pair.

I also checked my documentation for official UEI protocols and didn't see any that use this frequency.

[Chester]

Well I built this about 15 years ago and I am looking at the encoder board
A1 A3 A4 ARE "HIGH" A2 A5 "OPEN" If that clarifies anything.

I backtracked the upgrade protocol by importing into PB and now understand
quite a bit more.
However the device upgrade fixed data is 38 I don't know where this came from could you explain how that was determined please.

[johnsfine]

I just grabed the first five trinary bits from one of your signals:

+2114 -290 +2114 -290 +2114 -290 +318 -2090 +318 -2090 +318 -2090 +2114 -290 +2114 -290 +2114 -290 +318 -2090

I read that as high, open, low, high, open. That doesn't match what you claimed, which I assume meant high, open, high, open, high.

By Rob's encoding rules, I read it as 0 0 0 1 1 1 0 0 0 1

The leading 0 is treated as lead-in and the trailing 1 becomes the top bit of the variable data. That leaves 8 bits of fixed data 00111000. If you convert that to hex you get 38.

[Chester]

Yes I read your post got the multimeter out and rung the address pins out and got the same result as you did from the file. That only makes sense!
A3 is connected to a spst dip switch that allows me to change the address so by looking, it goes to a pull up resistor. But did not see the switch is closed to ground.

So if I want to change the address how can I mod the upgrade for that.

[johnsfine]

So if I want to change the address how can I mod the upgrade for that.

If you changed A1 to low, the whole upgrade design breaks down. But other than that, it would be pretty easy.

First convert the five trinary digits to ten bits using the inverse of what the PDF says. (Inverse because Rob needed to switch 0 and 1 to get your last half bit merged into the lead-out). So the rule is:

high = 00
open = 01
low = 10

String those ten bits together. Then use just the middle eight. Convert those from binary to hex. Then you have your fixed data.

If limiting A1 to High or Open is a problem, or limiting the last data bit to LOw or Open is a problem, you could redo the protocol in PB to have no lead-in and no On part of lead-out, and to have ten device bits and eight function bits. Then there would be two fixed data bytes.

[Chester]

Ok I think I understand and I will not need to change the A1 state
I still have one decoder chip laying around but have not found a use for it yet
and ther are better ways now anyway.
Thanks for yor help. Now I have a starting point to play.