X10 Community Forum
💬General Category => General Discussion => Topic started by: gga on December 09, 2011, 04:32:40 PM
-
I'm new to this site so I hope this is the proper place to ask for help? I've used X-10 for many years (maybe 15 or more) and had very few problems. I just use it to turn lights on and off. I have 16 or so light groups with half on house code A and half on B. Just in the last two weeks I find various lights turning on by themselves. It seems random and can be on either code A or B. I can't think of any new electrical equipment that I have installed. I tried using some different house codes including C,D,E and F. The problem remained with C,E and F. For several days D had no problems so I was about to switch everything to D but last night the lights on D2, D3, D4 & D5 all came on. Another clue? is that no lights have ever turned off on their own. I have also tried unplugging all my "sending" devices (maxi control, mini control and RF control) but that had no effect. Is it likely I'm getting outside electrical noise? Would some kind of filter help?
-
Has your electric utility recently installed smart meters?
-
Yes. Just a few weeks ago. Is that causing the problem? If so how I can block it out?
-
Long 12 page thread on this subject.
May not be able to filter it out.
http://forums.x10.com/index.php?PHPSESSID=02e0hasuc9fmt0vl0140f32s90&topic=22508.0
-
Yes. Just a few weeks ago. Is that causing the problem? If so how I can block it out?
At this time it is still unclear. There have been several reports of problems after smart meters were installed but not all have been traced to the meters and I'm still looking for a smoking gun, especially for the random turn-ons like you report. I think it's time to try to build a database of meter suppliers (and utility company), of specific problems (i.e. random ons, offs and/or blocked commands) and of the model numbers of modules and switches affected.
For instance, has anyone seen this phenomena with a Smarthome switch/module (whether X10 only or Insteon/X10) or is it limited to those of X10 manufacture?
Knightrider has set up a topic where those with similar issues can report in a standardized way so we can start tracking these. See...
- http://forums.x10.com/index.php?topic=25769.0
-
I checked my new Smart Meter installed by the utility company. The brand in Echelon model 83021-21AA. Went to their website and saw description of a "smart grid" which includes powerline communication so I'm assuming that's the problem. Called Duke energy and they said someone would call me back on Monday to discuss the problem.
-
Would you mind going to the link in my message above and filling in the data requested. So far we are seeing anecdotal reports and it would be helpful to start building a bit more rigorous structure for tracking purposes.
Where are you located?
Do you have any Insteon/X10 modules?
If anyone in the Cincinnati area is having this issue with Duke, I can loan you an Insteon/X10 module for testing a theory. I'd like to see what happens to it when set to the same X10 address as an X10-made device that is exhibiting random behavior.
-
I wonder if there is any power line command monitoring device that can actually record all the things going on on the power line.
-
I wonder if there is any power line command monitoring device that can actually record all the things going on on the power line.
Yes and no. An oscilloscope through the proper coupling may show it but it depends on the nature of the event causing the problems. Spikes of very short duration, which are known problems for most X10 devices, can be very hard to capture because there's no simple way to trigger the 'scope and recording all the time will quickly fill up memory on a storage type scope.
OTOH, the periodic bursts of 75kHz that Jeff Volp thinks are causing this should be fairly easy to capture and, as soon as I get some time to spend on it, I'll try to set up my Protek 220 'scope card to see if they are present on my line which has an Echelon meter of the type under suspicion.
-
There is another thread with a poll for people with smart meters. None of the ones listed there seem to cause any problems, but none are Echelon. The fellow I am working with also has an Echelon smart meter. It looks like Echelon may be the common denominator for this problem. Echelon is the company that makes those high power transceivers for powerline communication.
Jeff
-
There is another thread with a poll for people with smart meters.
Different from the one above? Have a link?
I believe there have been a few reports of problems with non-Echelon meters.
-
There is another thread with a poll for people with smart meters.
Different from the one above? Have a link?
No, that is the one. I was referring to different from the discussion here.
Jeff
-
Where are you located?
Do you have any Insteon/X10 modules?
If anyone in the Cincinnati area is having this issue with Duke, I can loan you an Insteon/X10 module for testing a theory. I'd like to see what happens to it when set to the same X10 address as an X10-made device that is exhibiting random behavior.
West Chester, Ohio
I have no Insteon Modules.
I'm in the Cincinnati area, using Duke and I'd be happen to try one.
-
Do you have any Insteon/X10 modules?
... I'd like to see what happens to it when set to the same X10 address as an X10-made device that is exhibiting random behavior.
I can tell you what I experienced.
When my neighbor's noisy CFL shut down my system, ONLY my Insteon modules stopped responding to the X10 commands. My "true" X10 modules still worked.
I remember trying Insteon switch-to-switch control, but I don't remember the results.
-
When my neighbor's noisy CFL shut down my system, ONLY my Insteon modules stopped responding to the X10 commands.
That's a different phenomenon - the Insteon module sensitivity is much greater (and the bandwidth is narrower). But I'm only interested in the random turn-ons, not blocked signals.
-
I plugged my ELK ESM1 in so I could test some switching power supplies sent me by the distributor I'm working with for the ZarduinoTM project.
I left the ESM1 plugged in and just noticed its LED bargraph lights one bar briefly in a rather random pattern. It will flicker on/off with a few seconds interval, then there might be an interval of several seconds before it repeats. A little later, it was showing a near constant one bar and occasionally flickering a second and with the X10 Good LED lighting sometimes. IIRC, Paul Beam told me the bandwidth of the ESM1 was ~!75-150kHz. It doesn't do anything sophisticated to decide good/bad but only looks for a startcode followed by the correct and equal number of 1 and 0 bits.
I unplugged both transceivers so that eliminates a stuck button on a remote. Unless my mini-timer is free lancing there are no other X10 sources.
I have a Parallax USB oscilloscope that I can setup fairly quickly that should be able to see and capture whatever is causing it.
NEVER MIND! It dawned on me that the ESM1 was on the same powerstrip as a Sony NIMH battery charger. When I unplugged the charger, the ESM1 registers nothing.
-
I, too, am in West Chester OH (Duke Energy) and have been having the exact same problem as gga with our X10 lights since the new smart meter was installed on our house. Duke was here twice today and are now referring it to their engineering department. The second Duke rep just left and told us that he believes the problem is the new smart meter. Will post Duke engineering findings after they contact us.
-
When my neighbor's noisy CFL shut down my system, ONLY my Insteon modules stopped responding to the X10 commands.
That's a different phenomenon - the Insteon module sensitivity is much greater (and the bandwidth is narrower). But I'm only interested in the random turn-ons, not blocked signals.
Ok. I didn't know if it was relevant information, but I figured I'd contribute, anyway.
-
I, too, am in West Chester OH (Duke Energy) and have been having the exact same problem as gga with our X10 lights since the new smart meter was installed on our house. Duke was here twice today and are now referring it to their engineering department. The second Duke rep just left and told us that he believes the problem is the new smart meter. Will post Duke engineering findings after they contact us.
Do they think it is specific to YOUR new smart meter (meaning replacing it might solve the issue), or something symptomatic to that particular model?
Can they install a "dumb" meter for testing purposes?
-
Noam - the Duke rep talked like it was a problem with smart meters in general, not our specific one. Because part of the functionality of a smart meter is to receive/send wireless communication to the node located in the electrical box at the street (we have underground wiring) and that could be causing an interference with our wireless automation. The first Duke rep told us that the new smart meter technology gives Duke more control over individual electric lines. The example he gave us was that if we would stop paying our bill, Duke could "flip a switch" in North Carolina and shut off our service. Now having said that, (and please, I am not crazy!) today was the first morning in over a month that none of our X10 lights were on when we awoke. All the X10 controlled lights were still off. I know, it sounds crazy to me too. I will continue to update this post.
-
Maybe they "flipped a switch" and turned off the remote signalling to your meter or something.
-
Well, I spoke too soon. Both bank of x10 lights just turned on by themselves. :-[
-
acce06t: Have you heard back from Duke engineering yet? I have called Duke several times and they told me someone from the "smart meter" engineering group would contact me but so far they haven't. My lights still continue to turn on at seemingly random times throughout the day and night.
-
I have been working with a customer in the Duke service area to develop a fix for the smart meter problem. What appears to be working fairly well are clamp-on ferrite inductors placed over the two heavy service entrance conductors in the distribution panel, and low-impedance notch filters on each phase to severely load down the incoming smart-meter signal.
Jeff
-
Today I was finally able to talk to an engineer from Duke. He said that Echelon (the company that makes the Smart-meter) is working on a filter to try to solve the X10 problem. He expects to get one in the next couple of weeks and he will try on my system as a test case.
-
We have a solution that involves adding clamshell ferrite filters that snap over the incoming service entrance cables and a shunt filter across the 240V inside the home to attenuate the signal that leaks through the ferrite filters. You can read more about it here:
http://forums.x10.com/index.php?topic=26056.0
Jeff
-
GGA, Sorry for not replying sooner but we are wintering in Florida. I will deal with this problem after we get back in April. Keep me posted about the filter and if it is successful. Good Luck!
-
Talked to the Duke engineer today. It's been several weeks and still no filter. He said Echelon was working on it but apparently haven't solved it yet? He suggested I might want to be removed from the Smartmeter grid. If they solve the problem later then I can always be turned back on. So that's what I'm going to do. He said the "turning off" process itself will take about 2 weeks. They have to come to your house to verify that you have X10 equipment, etc.
To acce06t: what did you do about your X10 stuff when you left town. Were you able to unplug everything? If you have any X10 wall switches did you just turn off the circuit breaker they were on? I've been away a few weekends since this all started and I typically come home to find the house all lit up.
-
If you have any X10 wall switches did you just turn off the circuit breaker they were on?
You could use the "safety" ON-OFF slider switch which disconnects power to the switch. It is designed in as a safety measure when changing bulbs.
-
Talked to the Duke engineer today. It's been several weeks and still no filter. He said Echelon was working on it but apparently haven't solved it yet?
I have also been in contact with Duke. There is a solution that is working well at one home. I am offering that as a kit, and one of the Duke engineers said they would order a kit from me, but they have not done so yet.
The kit contains 4 special low-frequency clamshell ferrite filters to snap over the "hot" service entrance cables, and a XPNR to go across a 240V circuit to further attenuate the smart meter signal that makes it through the ferrite filters.
Jeff
-
Duke turned off my Smartmeter hookup 3 or 4 days ago and since then I've had NO random events where my lights turned on. So for me it appears the problem is fixed.
-
And more conformation that Smart Meters play havoc with X10 protocol. Thanks for keeping the forum up to date.
-
My HAM radio turns on every X10 device i pass by. Especially on 222mhz
-
GGA, glad to hear (I guess!) that Duke has taken you off the Smartmeter until they come up with a solution. While we are gone, we removed the light bulbs from all the lights that are x10. We will be back in town the end of March and plan to contact Duke again. Thanks for your information.
-
We have a solution that is working very well to eliminate the Duke Smart Meter problem at one home in the Cincinnati area. It involves installing several low-frequency clamshell ferrite filters over the "hot" service entrance cables inside the distribution panel, and then installing a XPNR or custom filter to attenuate the remaining signal.
The smart meter rejection kit is available on my website. Duke expressed an interest in it about a month ago, but as yet they have not purchased even one kit.
Jeff
-
The smart meter rejection kit is available on my website. Duke expressed an interest in it about a month ago, but as yet they have not purchased even one kit.
Well, since X10 was here long before Eschelon and Smartmeters, Duke should be buying them from you in bulk to supply to X10 homes. It sounds like Duke is being cooperative by turning off the meter communications for complaining X10 customers, but I wonder how long that will last(?).
-
Update. Duke has finally turned off the smartmeter transmitter and my problem is solved. My X10 lights have not randomly turned on or off since. Duke had to notify my neighbors first because they had to turn off the smartmeter transmitter at the main box that transmitted to them also. For some reason, Duke could not just do something to my meter. So, unfortunately, my problem affected the entire neightborhood. Talked to a couple neighbors about it, consensus was -- they could care less! Life is good again.
-
Update. Duke has finally turned off the smartmeter transmitter and my problem is solved. My X10 lights have not randomly turned on or off since. Duke had to notify my neighbors first because they had to turn off the smartmeter transmitter at the main box that transmitted to them also. For some reason, Duke could not just do something to my meter. So, unfortunately, my problem affected the entire neightborhood. Talked to a couple neighbors about it, consensus was -- they could care less! Life is good again.
That's really good to hear.
If Duke REALLY wants to use the SmartMeter functions for your house (and the neighbors around you), they could always look into a better model of SmartMeter, one that doesn't mess up Home Automation systems.
-
I don't thing a "a better model of SmartMeter, one that doesn't mess up Home Automation systems" will ever be made. BGE in Baltimore has an "opt out" for those of us who don't want to have to put out $$$ to keep our X10 working properly or for other reasons. I've already put my name in the "no smart meter" list.
-
There are a multitude of SmartMeters that do not put anything on the powerline, even including the Echelon meters chosen by Duke which offer multiple methods for communicating with the data aggregators which then send the data to the power company. In some areas, Duke chose to use the powerline between the meter and aggregator.
And, I suspect, it's only one area around Cincinnati where Duke has gone this way. I live a few miles away in a Northern Kentucky city (Fort Mitchell) which Duke used as a test area for the SmartMeters. In 5 years or so I've seen no issues with X10 although that may partly be due to the fact that my switches were made by Smarthome.
I also suspect the fundamental problem is with X10's designs rather than with the meters. Only certain X10 switches seem to be affected and they are models that have a lengthy (pre-SmartMeter) history of being affected by spikes and other noise on the powerline. In fact, two of the Cincinnati area victims tried Smarthome-made modules on the same addresses without seeing any random events.
-
I don't thing a "a better model of SmartMeter, one that doesn't mess up Home Automation systems" will ever be made. BGE in Baltimore has an "opt out" for those of us who don't want to have to put out $$$ to keep our X10 working properly or for other reasons. I've already put my name in the "no smart meter" list.
From what I've read, the Landis+Gyr Focus AXR meter that Pepco (boo!) is supposed to be installing on Monday doesn't communicate over the powerline at all. It uses a wireless transmission network that is completely independent of the power lines.
Since last weekend's storm, Pepco had a really hard time keeping track of which customers were still without power. As they restored power to a large feeder, their system assumed that ALL customers served by that feeder were restored, and those who still did not have power had to call again to report their outage. It happened to us AT LEAST 3 times in the 104 hours we were without power. If having a SmartMeter will help the utility know when my power goes out, and let them know when they REALLY have restored it, I'm all for it.
Even if YOU don't get a SmartMeter on your house, if BGE installs them for your neighbors, and if they decide to use powerline communications for those, it STILL could mess up your X10 system.
-
I also suspect the fundamental problem is with X10's designs rather than with the meters. Only certain X10 switches seem to be affected and they are models that have a lengthy (pre-SmartMeter) history of being affected by spikes and other noise on the powerline.
The Echelon smart meter that we tested uses very powerful signals just below the X10 bandpass. X10 modules do not have a sharp enough input filter to reject those signals. If they did, tuning of the transmitted signals would be more critical. (I measured an old Sears Maxi Controller at 130KHz, and it was still controlling modules just fine.)
Jeff
-
The Echelon smart meter that we tested uses very powerful signals just below the X10 bandpass. X10 modules do not have a sharp enough input filter to reject those signals. If they did, tuning of the transmitted signals would be more critical. (I measured an old Sears Maxi Controller at 130KHz, and it was still controlling modules just fine.)
I had intended to edit my post to include X10's wide pass-band as another contributing factor but got involved with another task. I once tested several modules and found most responded to 75-150kHz and, IIRC, the designer of the ESM1 meter told me he found about the same.
Even the Echelon meters can use ZigBee between meter and aggregator. It was just a choice made by Duke and, judging from lack of reports from other areas within Duke's widespread territory, only in Cincinnati. And, one other factor is that when these meters sense noise in their pass-band, they crank up the volume and a very high amplitude signal might have the same effects as an inductive spike.