X10 Community Forum

🔌General Home Automation => Other Automation Systems => Powerline Control Reliability => Topic started by: JeffVolp on April 19, 2010, 11:11:10 AM

Title: Diagnosing a X10 Reliability Problem
Post by: JeffVolp on April 19, 2010, 11:11:10 AM
Well, I have to admit that after several years of virtually 100% reliability, even my X10 system started acting up last week.  One light on our most heavily loaded circuit started missing both its ON and OFF commands, and a couple of other lights furthest from the distribution panel started to misbehave too.  Most of the house was still running fine, so reliability was still in the high 90% range.

Checking signal levels with the XTBM showed the repeater was active with over 4Vpp of signal on the “problem” circuit.  Noise in all receptacles checked was 10mV or lower.  No new electrical devices had been added.

At first I suspected the prototype XTBR had somehow acted up, but swapping back my XTB-IIR didn't make any significant difference.  When the hall light failed to switch on, the XTBM logged Fxx ON, which means the unit code was not received.  My Ocelot log showed that light should have switched on.  I downloaded the program again anyway, but that didn't make any difference either.

Because I had been running with the XTBR rather than the XTB-IIR, I was again using a TW523 for my Ocelot powerline interface.  Hmm…

I went down to the distribution panel, unplugged the XTB-IIR, and monitored the incoming signal from the TW523.  It read only 120mVpp during the day with very few electrical loads on.  I suspect it is quite a bit lower at night when all the lighting loads are on, probably too low for the XTBR or XTB-IIR to receive reliably.

As a quick check, I swapped out the TW523 with another one, and the signal strength increased to 350mVpp.  The XTBR is reliably repeating that signal now, and I suspect all will be fine again tonight.

Since the TW523 can only manage 350mVpp “barefoot” at the distribution panel, there are some serious signal suckers here.  I've been able to ignore them because of the XTB and its successors.  The one weak spot is that the repeater needs to receive an acceptable signal for it to work properly.  Rather than searching for signal suckers, an alternate approach is to feed the TW523 through a XTB to guarantee a strong signal down at the distribution panel.  Of course, the digital link to the XTB-IIR is the best option.

It appears that my original TW523 was going soft.  When I get some time I'll open it up to try to diagnose what has degraded.

UPDATE:  Everything is working 100% again tonight.

Jeff
Title: Re: Diagnosing a X10 Reliability Problem
Post by: Brian H on April 19, 2010, 11:21:18 AM
Interesting find.

So we have one more thing to consider when an install that was OK for a long time starts getting flaky. A degrading X10 device.  ;)
Title: Re: Diagnosing a X10 Reliability Problem
Post by: dbemowsk on April 19, 2010, 08:02:57 PM
Thanks for the detail.  Helpful from me.