Raspberry Pi B+ & Homeseer Pi X-10 (CM11a) not working

[JWD]

Thanks very much Barry. Was that consistently better on the CM11A over several tests?
I have a CM11A and will try mine but from previous experience I know the power output of the TI103 X10 signal reaches my devices much more reliably than the CM11A. I also remember the CM11A randomly getting itself in a muddled state.
So it would appear that something about the Zee/HS3 TI103 driver isn't right.
I will add this to the Bugzilla ticket and hope someone at HST looks at it soon!

[lhfarm]

I'll do a few more tests later. I only have a few devices on the Zee. I it is my garage and was going to be my way of expanding my Z-wave out there. So not a robust system at all.

I also notice that there was a slight delay when I used the Zee to turn off the lights, vs the HS3/TI103. So that may be a function of the RPI too.

Barry

[JWD]

Now I have done some more controlled timed tests comparing CM11A and TI103 Zee response times.
Test setup is HR10 X10 RF remote control, receivers are TM13s (2x) and then either TI103 or CM11A into a USB Serial adapter into the Zee.
The X10 received signal just triggers a simple Speak action from the Zee.

With the TI103 I'm averaging 7-8 seconds from HR10 button press to hearing the Speak. With the CM11A the average is about 2 seconds and much more acceptable.
However, (and this is part of the reason I moved to the TI103 on HS2) the CM11A doesn't appear to be as good at detecting the inbound X10 signal as the TI103 so although I get the faster response it is a bit more hit and miss with the CM11A. I have found before that the TI103 is more reliable at sending / or sends stronger signals out to devices.
With Barry's evidence this seems quite conclusive about the Zee/HS3 TI103 handling having a performance problem. It would be good to have an HST comment / fix for this.

[sparkman]

I don't have any personal experience with the TI103, but I understand the baud rate is configurable on it. I believe the CM11a's baud rate is 4800bps. If there's a baud rate mis-match, it could perhaps explain the delays.

Cheers
Al

[JWD]

Al, selecting the TI103 X10 interface in HS3 gives 2 parameter choices - 9600 (default) or 19200 baud. 9600 is what I have always used with HS2.
The Other parameter is a polling interval which defaults to 4000 (4 seconds) with a warning that setting too low can cause unexpected results. I have tried lowering this but I noticed no change to response times - I might try a more methodical test. One assumes that the 4000 default in the software was chosen for a reason originally.

[Rupp]

Al, selecting the TI103 X10 interface in HS3 gives 2 parameter choices - 9600 (default) or 19200 baud. 9600 is what I have always used with HS2.
The Other parameter is a polling interval which defaults to 4000 (4 seconds) with a warning that setting too low can cause unexpected results. I have tried lowering this but I noticed no change to response times - I might try a more methodical test. One assumes that the 4000 default in the software was chosen for a reason originally.

9600 is the proper rate.

[travisdh]

I am having the same issues with the TI103 on the linux X10 plugin, my logs say this and I can't even get it to connect:

Mar-05 1:44:16 PM ACT TI103 Cannot open com port: Invalid argument
Mar-05 1:44:16 PM Info TI103 Plug-in, Version 3.0.35
Mar-05 1:44:16 PM ACT TI103 Calling initIO in TI103 with port:/dev/ttyS4
Mar-05 1:44:16 PM Info Plugin X10 has connected. IP:127.0.0.1:55068

Baud rates and so on are fine from what I can see.