Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[JeffVolp]

Here are a couple snippets from an email I just received:

I have been using an X10 system in my home for around 20 years with very few problems until recently.  Suddenly I have several lights that come on and go off at random.  I have tried changing house codes, replacing, switches and disconnecting command modules with no improvement."

...  "I find it strange that the random events first showed up when my power company installed a digital power meter. replacing the old style meter.  Even if the problem is the new meter, complaining to Edison Electric is like pounding sand."

Jeff

[Backward Engineering]

Those snippets on Jeff's post yesterday are from the email I sent Jeff.  After thinking about my problem more, I tend to think that the new smart meter may have less inductance than the old meter and thereby allowing X10 control signals to arrive from a neighbor's house.  The old meter was an induction motor technology where the motor windings provide a flux field to drive an aluminum disk.  I have ordered an XTBM Kit from Jeff and as soon as I finish the construction of the XTBM, I will continue to investigate the problem.  My meter is mounted in the same box as the breakers and I will not be able to install one of the whole house filter on the neutral between the meter and the breakers.  I have some other thoughts as to how I might install a filter on the output of the breakers that go to the lighting and receptacle circuits.  I will test my filter idea once I have the XTBM working.  My idea involves using a ferrite  core with a tertiary winding and capacitor tuned to 120Khz.  I will use a fairly large core that will allow me to pass multiple circuits through the center of the core.  I hope it helps.

[Dan Lawrence]

As to X10 control signals to arrive from a neighbor's house, doubt that.  X10 is NOT mass marketed and never was.  I got turned on to X10 from a close friend in the middle 1980's and to this day, NOBODY else on my street (or my transformer) has X10.  Where I got my X10 start is long gone.   If you read my earlier post, I live in Baltimore and BGE (our utility) is pushing Smart Meters, but both my gas and electric meters are inside my house, so I'm not letting them in if they want to install a Smart Meter. 

[Backward Engineering]

My smart meter is an OpenWay by Itron.  They use 902Mhz at up to 1 watt for the communications.  902Mhz should not be a problem for the X10 system, however, if they modulate the carrier with 120Khz at times, and any X10 device rectifies the signal, random interference may result.  I found that even after removing all of my command modules, I still had random on/off events occurring, mostly in the evenings.  I don't use any RF remotes in my X10 system.  I also noticed that the modules which had the random events were the ones closest to the power panel (smart meter).  I'm considering installing an RF filter at one of the problem X10 modules to see if it will help.

[dave w]

  I found that even after removing all of my command modules, I still had random on/off events occurring, mostly in the evenings.  I don't use any RF remotes in my X10 system. 

So you are saying; at that point there was nothing in your house that could send an X10 code? Does sorta point to the meter or out side signals creeping in. If from meter that will be yet another nail in X10 coffin. Let us know after assembling the "saa-weet"  XTBM.

[dhouston]

In my case I'm still leaning towards RF from the lady in the apartment above mine.

But, I have a 1MHz-3GHz frequency counter, that I'll try as soon as I can find a power supply. It may not have enough sensitivity. As I recall it could only pick up an X-10 remote when within 1/2 inch.

[Backward Engineering]

I don't think you will ever see an RF signal at this low of a level with a frequency counter.  You would need a communications receiver tuned to the X10 receiver frequency and a way to capture and record the signal.  The signal levels for the power level allowed are down around several micro-volts/meter.  I'm assuming you have an X10 RF remote, or are you thinking that the RF signal is strong enough to somehow effect your wired modules?

[dhouston]

The one thing most, if not all, of the smart meters have in common is ZigBee @ 915MHz. (Itron also uses ZigBee.) Older X-10 RF transceivers use superregenerative receivers @ 310MHz. 915 / 3 = 305 so it would not surprise me to find things like an RR501 or older TM751 (both of which I'm using) being affected by a nearby 915MHz signal. I've seen other, non-X10 superregenerative receivers interfere when within 6' of each other.

• http://davehouston.org/modules.htm

Newer X-10 transceivers (more recent TM751s, CM15A) use a single chip superheterodyne receiver (even newer RR501s still use superregenerative) and I don't know how well they deal with third harmonics. X-10 uses a coil with a tuning slug for the oscillator on these so they may also be susceptible to a local 3rd harmonic with the tuning coil acting as antenna.

• http://www.allparts.co.kr/pdffile/pdffile_001/3274.pdf

[Backward Engineering]

My smart meter is an OpenWay by Itron and the radio frequencies used are defined in this document:
http://www.sdge.com/documents/smartmeter/OpenWay%20RF%20FAQ.pdf
My system does not have a radio frequency remote control, yet I have the random on/off event problem.
I did not have a problem with non-responsive modules.  I can see the possibility of the 900Mhz smart meter producing random events if you are using an RF remote, however I can not find an explanation for any non-responsive module problems being caused by the 900Mhz smart meter communications unless the non-response problem is also random.  The smart meter should not be transmitting continuously.  I believe that the X10 system signal might be attenuated by the electronics in the meter or a reduction of inductance between the source and load side of the smart meter compared to the old meters.
Or simply stated, the smart meters may be signal suckers.  It also occurred to me that the power companies are pushing residential solar power systems.  These systems use static converters to convert the solar panel direct current to 60Hz AC.  These are switch mode power converters and are often the source of harmonics.  If solar converters combined with the new smart meters are contributing to the noise problem, we should see a noticeable difference between night and day.

[dhouston]

... the radio frequencies used are defined in this document:

The referenced document says 902-928MHz ISM band with frequency hopping (FHSS) and 2.4GHz. However, another press release said Itron was certified by the ZigBee alliance and ZigBee Certification requires Direct Sequence Spread Spectrum  (DSSS) instead of frequency hopping. Checking the FCC IDs from the document gives 927.45MHz for all but one model which is listed as 921.8MHz. 927.45/3=309.15

• https://www.itron.com/newsAndEvents/Pages/OpenWay-by-Itron-Earns-ZigBee-Smart-Energy-Certification.aspx

Momentary brownouts and/or spikes will cause the chips used by X-10 to reset, resulting in random on/off events. Since these meters also can shed loads, it may be they are putting a bit of hash on the powerlines but this might be really, really hard to document. I had an LM14A that would turn off nearly every time a 60 year old fluorescent in one bathroom was turned off. I could see nothing on the powerline because it was so brief but replacing the switch (in which the arcing was visible) with an Insteon Icon model (used in X10 mode) cured the problem.

[dhouston]

One more bit of data. I did a search using "smart meter" of the Smarthome Insteon forum without getting a single hit. That, along with my ESM1, makes me doubt it's powerline noise causing my intermittently non-responsive transceiver problem.

Some of these meters may have filters to block their powerline signals from reaching the residence side and those might be attenuating X-10 signals.

[JeffVolp]

Some of these meters may have filters to block their powerline signals from reaching the residence side and those might be attenuating X-10 signals.

One might think that, but one of my customers had very strong signals, but still could not get his X10 system to work reliably after a smart meter was installed.  And low signal levels should result in devices not responding at all.  Some of the reports have been random actuations after a smart meter installation - particularly during the middle of the night.

Jeff

[dhouston]

That is more evidence that the random on/offs may be from spikes or brownouts affecting X-10's chips. I've never heard of similar issues with Insteon so that could also explain why Insteon seems to have no issues with the meters. Here's a thread from a Smarthome forum that may point us in the right direction - it mentions pulses from some meters used for measuring usage - these might be the culprits.

• http://www.smarthome.com/forum/topic.asp?TOPIC_ID=5507

[Backward Engineering]

Momentary brownouts and/or spikes will cause the chips used by X-10 to reset, resulting in random on/off events. Since these meters also can shed loads, it may be they are putting a bit of hash on the powerlines but this might be really, really hard to document. I had an LM14A that would turn off nearly every time a 60 year old fluorescent in one bathroom was turned off. I could see nothing on the powerline because it was so brief but replacing the switch (in which the arcing was visible) with an Insteon Icon model (used in X10 mode) cured the problem.
Posted on: October 27, 2011, 10:08:56 PM
Posted by: Backward Engineering

OK, I have the equipment to instrument the power line but I don't know when I will have the time to set it up.  I have a Astro-Med 16 channel chart recorder with high speed capture to memory.  I can set it up to monitor line voltage on one channel and the output on the load side of a light switch module on a second channel.  I can program it to trigger on a change in the status on the load side.  This recorder can capture events as short as a few microseconds and display both pre and post event analog data.  I also have a digital storage oscilloscope with a printer interface that I might also try.  I received my XTBM kit today and I will be focusing on assembling the XTBM initially.  I always thought that when I retired I would have as much time as I wanted to work on these things.  I now see that I was wrong.  I was working in power conversion engineering before retiring.

[Backward Engineering]

If these smart meters can shed loads, it would be interesting to learn what technology is used to switch the 100-200 amp load.  Many solid state power devices will generate waveform distortion at zero crossing, right where X10 communicates.  Do the smart meters use solid state devices to load shed?  Do they use IGBTs?  If IGBTs are used and the load shed is reconnected at any time during the cycle other than near zero crossing, the result can be very high di/dt which can be destructive to some types of loads.  I guess I need to setup that chart recorder to get some answers.