CM15a: X10 devices keep switching to "OFF"

[SteveMSJ]

HS3/CM15A fails with 100% certainty within less than 2 hours, so it is effectively unusable

My experience is certainly different to yours. I have been running the CM15 plugin on HS3 windows since it was released and it has been running for some considerable period of time now with little problem. There were some issues with the first couple of betas. However, apart from the pretty rare strange log entry for a received command noted higher in this thread, it is as solid as it was on HS2 for me. I realise that's not much help to you.
I'm currently on 3.0.0.8.
I have a mixed X10 and ZWave setup and all new devices I install are ZWave. I suspect I am continually increasing the noise levels on my wiring as I add more and more little USB power supplies around the house for ZWave sensors of one sort or another. I'm sure my X10 setup which has been 99% reliable for years will eventually drown.

I'm generally finding HS3 pretty solid these days and I have come to like it more and more. My HS2 was rock solid and the migration to HS3 through various betas has been painful but I have come to appreciate the redesign, even CAPI.

Steve

[tarbat]

Yes, I'm using a CM15PRO with HS3-Pi without any problems. I've checked the log, and don't see any extraneous X10 messages that shouldn't be there. Although I do see several Z-Wave messages that shouldn't be there from devices not present on my setup!

Note that the CM15A plugin in HS3-Pi is v3.0.0.4.

[Pete]

Here I utilize CM11A's, TW-523's, WGL 800's and Jeff Volp's X-10 booster and dual phased TW-523 controller and booster. (along with Zigbee, Z-Wave, Insteon and UPB).

Jeff Volp's stuff is X-10 on steroids.

I have seen stray unexplainable poltergeist X-10 stuff.

Fer sure my playing around with the TED energy module caused a bunch of noise.

I do have a rack full of UPS's sitting next to the rack with the HS stuff and next to the controllers plugged into autonmous circuits today. The UPS's are work fine with X10 and not signal sucking that I can tell; well everything still works fine.

There was a time where I used the X-10 from the Insteon module (using both in Homeseer).

There too just watching the Insteon X-10 stuff on Homeseer I would get unexplicable X-10 signals. Not sure if I was seeing them or the source was the Insteon X-10.

I know when I removed the X-10 part of the Insteon PLM the extraneous signals went away.

The above said it doesn't explain why you didn't see this with Homeseer 2 but now seeing it with Homeseer 3.

You can if you want monitor your X-10 signals with the signal monitor (think I lost mine) watching your stuff with or without the CM15A plugged in. You can also monitor the serial port by mirroring the serial communications between Homeseer 3 and the CM15A using that free Microsoft serial port mirror monitoring program.

The HS2 X10 CM11 Plugin while working does show communications errors still today.

You can also use the WGL-W800 or the RFXCom stuff. I do not have issues today with the serially connected WGL-W800.

HS3/CM15A fails with 100% certainty within less than 2 hours, so it is effectively unusable

Monitor the console logging stuff and do a port mirror of the traffic between HS3 and the controller; you might see something there.

I do not have a CM15A though to test here.

A new Homeseer 3 lite RPi user is today testing his CM15A (tarbat- Christopher) with the Linux version of Homeseer 3 lite and I think that its been working fine for him. (sort of) See here:

http://board.homeseer.com/showthread.php?t=170455

Just guessing here....

BTW I have fond memories of working in Brussels, Belgium many years ago. Neato country you live in.

[WfromL]

OK Rupp, I appreciate your comments, but allow me to summarize my findings as follows:

• I triple checked I don't have a device sending M1
• I'm far away from neighbours, and I never saw any X10 signal that was not generated by my own devices
• Using other X10 receivers ( Marmitek, ActiVHome, RFXCOM ) I never see an M1 command,
• so yes, it is definitely not a device sending M1 ON/OFF.
• This is also confirmed by the fact that the X10 commands OFF are NOT executed by any devices, it is only the STATUS that changes in the HS3 software.

But is it noise? If so I still think there is something wrong with the CM15A plug in because

• the CM15A hardware has 99% reliability with HS2.
• HS3/CM11A is also 99% reliable
• HS3/CM15A fails with 100% certainty within less than 2 hours, so it is effectively unusable
  

My HS3 setup is essentially identical to my HS2 setup: same devices, same events, same plug ins (RFXCOM and CM15).
If it is indeed noise, that means the HS3/CM15A plug in software is much more sensitive softwareto noise than it should be. Is this even possible?

[Rupp]

The M code for X10 is 0 0 0 0 which is the most likely house code to be tripped by power line noise. HS should have received an M1 on or M1 off if a device was sending the code. If HS registers a M on or M Off this is again a very good indication that it's noise and not a device as the device would have sent the associated on, off, dim, etc along with the command.

[WfromL]

Tnx for looking at this Steve!
In fact I think this can easily go unnoticed. I only found out because I have an event that switches on my bathroom heating based on the status of a virtual X10 device. That device has to be "ON", but as the staus always changes to "OFF" I woke up with a cold bathroom and that's why I noticed.

[SteveMSJ]

CM15A DEBUG Received CM15 Address:M1,Command:socialite,7e,7b,f2,71,84

Out of interest I looked back through my log, with filtering set to only show CM15 commands, and I have a similar pair of commands received about 2 weeks ago.
Oct-05 23:29:38 CM15A DEBUG Received CM15 Address:M1, Command:socialite,d6,7b,ed,65,80
Oct-05 23:29:38 CM15A DEBUG Received CM15 Address:M1, Command:socialite,d6,7b,ed,65,80

I also don't have any devices on M1. I don't think it caused me any issues, although I probably wouldn't have noticed my X10 devices going to off at 11:30 on a Sunday night.

I have also in the past received some strange commands on A6, which I again don't have any devices set to:

Oct-10 08:23:48 CM15A DEBUG Received CM15 Address:A6, Command:devicestart
Oct-10 08:22:21 CM15A DEBUG Received CM15 Address:A6, Command:devicestop

I think these caused a device on O6 to turn Off which had me baffled at the time. I put this down to a glitch in my CM15 hardware as unplugging and plugging it back in whilst HS3 was shut down appeared to cure the issue, at least for a while.

Maybe these are coming from the plugin after all.

Steve

[WfromL]

Tnx Rupp for your prompt reply, but in my opinion unlikely to be noise, same X10 setup works perfectly with HS3/CM11A, and same CM15A interface has worked perfectly for many years on HS2 (and still does so at this moment).
In fact I can run both HS3/CM11 and HS2/CM15 perfectly on an old Win7 computer but my real problem is that I want to move my setup to a new WIN 8.1 computer and Win 8 doesn't have a CM11 driver, so HS3/CM15 is my only option
BTW, wouldn't noise actually also switch the devices, and not only change the status in the software? And why only switch to "OFF" and never to "ON"? This must be some bug hiding in the code.

[Rupp]

It's most likely noise on the powerline. Random noise can wreak havoc on an X10 installation.

[WfromL]

Using CM15A 3.0.0.8 / HS .94 on Win 8.1

• All x10 devices switched to status "OFF" seemingly at random, after a few minutes to at most 2 hours with nothing showing in the log
• Curiously the devices didn't actually switch to "OFF", it was only the device status that changed
• I (finally) switched to "developer" mode and found the switching occurred simultaneously with a log recording:
• CM15A DEBUG Received CM15 Address:M1, Command:socialite,7e,7b,f2,71,84
  
• I have no device with address M1, and I don't know what the "command socialite " means

Suggestions anyone?