Wrong X10 devices triggered

[sponnath]

I have recently found that the wrong X10 devices are getting activated. For example, motion in the master bath could turn on the hallway bath light or exhaust fan. It will happen repeatedly but can eventually correct to change to another device. Seems like I started to notice this around 2.0.19. Now I am running HS Pro 2.4.0.1, T103 with latest code. I used Doomotion to run scripts that use ExecX10 commands.

Anybody seen anything like this?

[Rupp]

What shows in the HomeSeer log file? The correct code ?

[sponnath]

So far, I haven't been able to track it to anything that looks odd in the log. Been working a little too much to spend time on this a lot. I did get all my software up to date this weekend. I have still seen it occur. I do have debug logging turned on for the T103.

I need to note the time of the event in my phone and then I can get back to the mdb that holds my log files. I dump them in 5 mins to an mdb.

I am off call next week so maybe I can get more time on the logs.

I was just wondering if anybody saw anything like this. I have to solve this. It's a WAF killer when the bathroom lights go off whil she's taking a shower.

I don't think it's doomotion triggering the wrong CAC code. My hall bath does not normally get the lights turned on by doomotion. So when I move in the master bath, I can see the hall bath light or exhaust fan turn on, sometimes. But it's not consistent like a programming error would be.

[outbackrob]

What housecodes are you using for the devices that are giving you problems?

I occasionally get similar issues with housecodes L and D. Triggering L will cause the corresponding D device to trigger. USUALLY both the L and the D device will trigger so the desired action occurs, but additionally the D device is triggered. It so far has been only L -> D so I no longer use D for any WAF critical devices.

An interesting discovery; for our house, this L-> D issue only happened in the mornings and the evenings. Never saw it happen between 10AM and 5PM.

I would suggest reswizzling your housecodes a bit to see if your phenomenon is housecode specific.

Rob

[Gogs]

Beware of using house codes A1 and A2 as motion sensors default to this when the battery gets low, unless the have an Eprom fitted.

[sponnath]

What housecodes are you using for the devices that are giving you problems?

I occasionally get similar issues with housecodes L and D. Triggering L will cause the corresponding D device to trigger. USUALLY both the L and the D device will trigger so the desired action occurs, but additionally the D device is triggered. It so far has been only L -> D so I no longer use D for any WAF critical devices.

An interesting discovery; for our house, this L-> D issue only happened in the mornings and the evenings. Never saw it happen between 10AM and 5PM.

I would suggest reswizzling your housecodes a bit to see if your phenomenon is housecode specific.

Rob

Interesting discovery you made.

So far I have noticed issues where the codes are close. So the hall bath light is I11 and it's tripped when I12 is set to on which is the master bath. Also, the hall batch fan can be set on and it's I9.
What's also interesting is that you find time windows for these things. I seem to notice the bad behaviour in the morning and not so much at night.

Thanks for this as it helps me think about patterns.

What's odd is that I didn't see any of this behavior until recently. Lights tripped very reliably, virtually 100%.

[outbackrob]

The Grinch hates this too!

What's odd is that I didn't see any of this behavior until recently. Lights tripped very reliably, virtually 100%.

Hmmm. Then I bet the problem is noise. You could swap a well-functioning motion detector with the one in the master bath to rule out a bad motion detector. If it still happens, I bet you have a new or moved wall wart transformer somewhere that is noisy. ....do you guys plug in anything in the mornings? Coffee pot?

Another noisemaker could be your HA machine's power supply. The older they get, it seems the noisier they get.

If you don't have any filters on anything, invest in a few and put them on anything with a motor or a power supply. Automated Outlet ususally has the best prices on them. (DON'T put one on a microwave! It will fry it and stink up the house! Don't ask me how I know...)

The only other thing I can think of is distance. How far away are your wireless motion detectors from your receiver? (What receiver are you using?) You could be on the very edge of your reception range.

In my 25 year experience with X10, 99.999% of the time when something that has been working suddenly stops working or working consistantly, the problem is noise.

[buvens]

Battery operated controllers, motion sensors, keypads etc, are where most of the issues I've seen as the batterries go down.
The other thing is what was suggested, filters. Especially on UPS's or regulators.
My house had gotten down to probably 60% reliability but needed to use sprinkler controller this summer. Bought some of the small filters and a couple of the larger filters. Used them on the wiring closet and UPS's and back to as close to 100% as ever.

Good Luck...Sonny

[JeffCharger]

Interesting discovery you made.

So far I have noticed issues where the codes are close. So the hall bath light is I11 and it's tripped when I12 is set to on which is the master bath. Also, the hall batch fan can be set on and it's I9.
What's also interesting is that you find time windows for these things. I seem to notice the bad behaviour in the morning and not so much at night.

Thanks for this as it helps me think about patterns.

What's odd is that I didn't see any of this behavior until recently. Lights tripped very reliably, virtually 100%.

Could this be related to the fact that motion sensors sometimes use 2 codes........one for daylight sensor?

[Uncle Michael]

You may need to buy/rent/borrow an X10 signal meter/monitor to help you track down what is going on.

[sponnath]

Well I dug out my meter. I have the AT004 from ACT. In the kitchen where the under cabinet lights are not coming on, I have >5 signal and at most 145mv of noise. I did a trace on the codes and found something odd but repeated each time, it did not seem like noise. The code is F5 and I get the following across the meter. F5 F5, Hpsd0 Hpsd0. I expected ON ON, not Hpsd0. It started me wondering if that is some kind of extended code sent by Homeseer for a certain device type. This is very repeatable so I have a hard time thinking noise. I am using Doomotion to execute the ExecX10 command. That command seems to take into consideration the current state of the device to determine what it will send.
I also noticed a light up in the bedroom that did not turn on when motion was tripped. I tripped the motion sensor 4 times and no X10 codes came across the meter. So, either Doomotion did not function or the ExecX10 thought the light was already on. Gotta check that one tonight in the logs.
Again, upstairs on those circuits, my noise was running about 125mv, not too bad.
I will do some extra checking around tonight to see if anything new got plugged in around the house that is not on filters. I use some old X10 pro filter, smartlinc filters, and ACT large filters.
Thanks for all the tips.

[randycboone]

I have recently found that the wrong X10 devices are getting activated. For example, motion in the master bath could turn on the hallway bath light or exhaust fan. It will happen repeatedly but can eventually correct to change to another device. Seems like I started to notice this around 2.0.19. Now I am running HS Pro 2.4.0.1, T103 with latest code. I used Doomotion to run scripts that use ExecX10 commands.

Anybody seen anything like this?

I have a similar problem where the hall and living room lights seem to be going off together, and where they are supposed to be triggered by separate motion sensors on separate codes. Here's what I think is happening with my set up.

• Hall motion detector (M10) triggers OFF
• House and Unit Code is received but the OFF command is not
• Hall light stays on
• Living Room motion detector (M12) triggers OFF
• House and Unit Code is received followed by the OFF (ie. M12 + M OFF)
• Hall light goes off with the Living Room light

I'm thinking the system is seeing M10+12 followed by the M OFF, but I have no proof.

Does this make any sense?