JP1 Remotes

Hey Rob,

You ought to sell modified URC-6131's with fixed button and the X10 RF mod (once I provide the design for you). After all, the tough part is opening the case without damage.

The RF hookup will merely require wiring to power, ground, and to the driver pin for the IR LED. Additionally, to save power, the keychain remote will need two disconnections or trace cuts to turn off its LED and control chip. I downloaded some photos and schematics from the FCC, if you are interested. (Unfortunately, the schematic download for the remote I wanted to use, turned out to be for a totally different system. It is similar to two others that were available, however).

Fred

fpreston wrote:Hey Rob, You ought to sell modified URC-6131's with fixed button and the X10 RF mod (once I provide the design for you). After all, the tough part is opening the case without damage.

I've thought about doing the key mod, but it's quite fiddly so it would add another 15 minutes or more to each mod and I just can't see people being willing to pay any extra for a remote with that mod, so I've never done it. It would increase the cost of the remote from $23 to about $35.

Now the X10 mod is another story, if I can get the keychain remotes cheap enough, I can certainly see people being willing to pay for a JP1 remote that sends X10 RF signals.

FYI. To search the FCC database, I used

https://gullfoss2.fcc.gov/prod/oet/cf/e ... Search.cfm

for Grantee: B4S (meaning X10)

I have my URC-8820 ready to install my X10 RF transmitter, but my son rearranged and my soldering station is temporarily missing. Income tax time is also hampering progress.

First, the URC-8820 style case has ribs about every 1.5 inches, so one must remove a portion of a rib to make room for the keychain remote install.

Second, the URC-8820 uses an inverter driver for the IR LED, so the drive pin cannot be used to modulate the RF transmitter. The drive at the IR LED is the correct polarity, but is not the full 3 volts. It is limited by the on voltage drop of the IR LED. Fortunately, this is typically 2 volts for LEDs, and this should be enough to modulate the RF transmitter.

So, I believe I am set to get it working after I complete soldering the 3 needed wires, and complete the minor mod to the keychain remote (cutting one trace). The latest documentation is at https://www.hifi-remote.com/forums/dload ... le_id=4381

Fred

I'm looking forward to hearing if you get it to work.

I located my soldering station and completed the mod wiring. Unfortunately, it does not appear to be functioning as anticipated. I do not know whether this is due to:

1. Something incorrect with the X10 device loaded into my URC-8820 ?
2. Inappropriate drive for the X10 RF modulator ?
3. Issues regarding not sending the house code ?
4. A non-obvious problem with my wiring mod workmanship ?
5. Something else TBD ?

I tried both the X10 and the Hacked X10 devices (although I had to change the Hacked device to use the X10 protocol instead of the custom one it included).

I will have to do further diagnostics, probably at a friend's house (who has full RF test gear). Any ideas?

Fred

PS: Photobook file update at https://www.hifi-remote.com/forums/dload ... le_id=4381

fpreston wrote: 3. Issues regarding not sending the house code ?

I will have to do further diagnostics, probably at a friend's house (who has full RF test gear). Any ideas?

Fred

Problem is due to no house code being sent. All OFA remotes' x10 functions are designed to send IR to an IR543, which supplies the house code.

All x10's own rf remotes, incl their universal ( audio/video) remotes have a method to enter the housecode, either thro' the wheel, or digitally keyed. Such is not the case with OFA's.

Just transmitting OFA's x10 commands via RF will never work 'cos of the missing housecode, unless one can configure the RF receiver to supply the housecode.

I still don't know if other problems exist with my RF mod, but your confirmation of the "device" not transmitting house codes makes sense. The advanced IR543, however, is supposed to allow the remote to control the house code. I tried the hacked "device" (which includes the house codes) with attempting to manually transmit house codes, but this didn't work. Perhaps I needed to merge the codes instead, so that they make a continuous string.

Any further thoughts.

Fred

PS: A friend is building me one of Tommy's universal RF detectors so that I can see what is actually being transmitted.

I also tried the X10 RF Device "X10dream.rmdu" last night. This is apparently supposed to send the entire string with house code, but did not work. This may indicate that something else is wrong. (It did give a list of Java errors when switching to the URC-8820, though). Even if it worked, it is a little inconvenient in that you have to make up a separate device for each 8 control addresses (2 for each house code to cover the full 16 channels).

I don't quite understand the use of NEC1 Device numbers versus having a protocol that can output all commands, versus the "X10" protocol. Can anyone shed some light on this?

Fred

I thought I remembered reading somewhere that the X10 RF signals actually used the standard NEC protocol (NEC2 actually), so I did a quick search and found this...

http://www.printcapture.com/files/X10_RF_Receiver.pdf

Assuming that your mod is all good, you would need a new upgrade to work the RF puck. The data for all the buttons is included in the PDF so if I get a chance later, I'll put one together for you.

I have created an upgrade based on the data in that PDF, which can be found here...
https://www.hifi-remote.com/forums/dload ... le_id=4604

X10 uses a different device code for each bank within each house code. Bank1 works for units 1 thru 8 and bank2 works for units 9 thru 16.

The house/bank device codes are as follows:
A1 = 06, A2 = 38
B1 = 14, B2 = 46
C1 = 02, C2 = 34
D1 = 10, D2 = 42
E1 = 01, E2 = 33
F1 = 09, F2 = 41
G1 = 05, G2 = 37
H1 = 13, H2 = 45
I1 = 07, I2 = 39
J1 = 15, J2 = 47
K1 = 03, K2 = 35
L1 = 11, L2 = 43
M1 = 00, M2 = 32
N1 = 08, N2 = 40
O1 = 04, O2 = 36
P1 = 12, P2 = 44

Rob, I've written a couple of VB programs for PC and PocketPC to control a firecracker via serial. My codes match the table in the document you referenced. But I don't quite follow yours. For example, the first byte for A1 thru A8 would be hex 60, A9 thru A16 hex 64 (all upper banks are 4 greater than the lower bank), which matches the pdf file.

FWIW, one thing missing from the tables that I had a really hard time finding was All On/All Off. Byte 2 for All Units Off is hex 80, and All On is hex 90 (Bright is hex 88 and Dim is hex 98).

The PDF defines a logical "1" as having an OFF time of 2.25 msec and a logical "0" as having an OFF time of 1.125 msec. Therefore, the binary in the tables in not complimented. The UEI executor for NEC *is* complimented, so if you simply converted the X10 binary to hex, it would be the complement of the hex needed for the UEI executor.

Furthermore, the NEC protocol is "LSB", which means you would need to read the binary backwards in order to convert it to decimal.

The device code binary for A1 is listed as "01100000", which as you correctly point out, equates to a hex code of "60".

To get the decimal device code from that binary, you would first have to read it backwards as "00000110", which is decimal 6.

When you select NEC2 and enter a device code of 6 in KM, you'll see that the hex code generated is "9F". "9F" equates to binary "10011111", which is the complement of "01100000".

Thanks for the additional 2 codes (ie, All units On/All units Off). I have converted your hex codes to OBCs and added them to the KM file.

Thanks for the explanation. I've got my soldering iron ready for when fpreston reports success (hopefully).

Here's another page that has alot of info on the X10 RF signal...
http://www.nomad.ee/micros/x10faq.html#section3

The binary listed there is the reverse of the binary listed in the PDF, which is not uncommon.

I also found an archive of a page that details a different mod for the X10 keychain remote, so I'm posting that just for reference purposes...

http://www.geocities.com/ido_bartana/modifying_RF_Remote.htm

I found another forum where someone posted a snippet of the binary, which matches the binary in the RDF and confirms that I'm reading it correctly:

http://www.integratorsforum.com/archive ... -3950.html