Print

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

[Boiler]

azzar0,

I'm new to the site and was reading about your mysterious lights.  I hate to be a naysayer, but I'm betting you still have a noise source in the vicinity.  I say vicinity because it may not even be in your house. 

The changes you made should indeed improve the X10 signal level on your powerlines.  The level may now be just above that of the noise source.  You mentioned the behavior of your satellite reception when you turned on the lamps in your bedroom.

Oh, and get this: I also noticed that my satellite signal is pixelated when I turn the (x10) lights on in the bedroom. Turn the lights off - perfect picture!

 

Is your system operating properly now?  The X10 switches themselves should not have done this unless one of the triac drivers were oscillating.  If that were the case, it should still be happening (unless you replaced a switch).

Last year I tracked a similar intermittent noise problem down to a optical sensor in my post lamp.  The output stage of the sensor would go into oscillation (intermittently) and drive my X10 devices nuts.  Thankfully, it finally blew the breaker and I was able to put 2 and 2 together.  I've since purchased a X10 powerline tester and have a lot more hair on my head to show for it.  There is a site (can't remember the name) that offers a "loaner" ELK diagnostic unit to their patrons.  You may want to consider this if your problems return.

The problem could also be outside your home.  I have an audiophile friend at work who traced the noise in his high $ sound system to a compact florescent activated by an optical sensor (oscillating again).  This was a porch lamp three houses down (same transformer).

Best of luck

[nklght]

Do you have any repeaters, or transcievers?
My in-laws had a problem with the TV suddenly changing stations, and acting erractically.  The only thing that changed was that their nieghbor just signed up for the same company they had.  I changed their remote codes and the problem was solved.  X-10 is a little more difficult to troubleshoot, but it is possible a rogue signal from your nieghbors is getting into your system. 

[JeffVolp]

I had been watching this thread, but didn't have much more to contribute until now.  I was working on the XTB-IIR code yesterday, and had the return signal gain set at max while checking the automatic threshold control loop.  Then I turned on my noise generator - several compact fluorescent lights on an unfiltered circuit.  The noise level at the comparator input was incredible - dancing around up to several volts.  As the noise produced by each of the CFLs beats together, it produces peaks and valleys that can mimic X10 signals.  The noise was above my threshold limit during this test, and the XTB-IIR was recognizing occasional X10 commands.

So I don't scare anyone off, this was just high-gain test of the automatic threshold circuit.  The XTB-IIR would normally switch to a lower gain, pushing the noise down into the background.  But I wanted to bring this up here to point out that CFLs can simulate unwanted X10 commands.

Jeff

[Boiler]

Jeff,

I know this is a bit off thread but...

...The noise level at the comparator input was incredible - dancing around up to several volts.  As the noise produced by each of the CFLs beats together, it produces peaks and valleys that can mimic X10 signals.

This is just plain nasty.  Not only do we need to worry about the chopper frequency of an individual CFL, but the beat frequencies between multiple units?  This looks like it's a "new find" that's different from your CFL troubleshooting guide, correct?

• X10 & Compact Fluorescent Lights - 2007-02-24 @ 21:53:38
  Feedback Thread: Re: Use of compact florescent bulbs
  

Sorry but I have a host of questions:

• Were the CFL's X10 friendly or "known offenders"?
• Can you tell us the make/model of the CFL's used as well as the quantity?  Did you use multiple brands/models?
• Can a plug in filter reduce the beat frequency to tolerable levels?
• What is the typical chopper frequency for a CFL(sorry, I'm being a bit lazy here)?
• Why do individual frequencies vary so much that the "beat" winds up in the 120Khz range (or were these different manufacturers/models)?

Let me know if I'm reading this wrong, but this appears to be a new chapter in the "evils of CFL".

The Boiler


[TTA Edit: Added description to LINK; added FEEDBACK LINK; removed unnecessary {/LIST}{LIST} commands - just need one {LIST} and one {/LIST} per list. ]

[JeffVolp]

Not only do we need to worry about the chopper frequency of an individual CFL, but the beat frequencies between multiple units?  This looks like it's a "new find" that's different from your CFL troubleshooting guide, correct?

Actually, the photos in that tutorial were of the noise from those same bulbs.  I believe those TCP ESN18 were cheap bulbs offered by our power company about 5 years ago.  I can't comment whether others have identified them as known offenders, but they certainly are for me.  This is an unfiltered non-X10 circuit.  I'm sure a Leviton 6287 would significantly reduce the noise, but I have left them as is to use as a noise source for testing.

I don't think there is any "typical" chopper frequency.  Higher frequency usually means smaller magnetics, so the frequencies are probably going up as manufacturers push the circuitry into smaller bases.  These TCP bulbs have a much large base than the n:vision CFLs we bought at HomeDepot recently.

Ah, it isn't the beat frequency that winds up in the 120KHz region.  That is the carrier (chopper frequency).  The beat is in the 100Hz region so the noise sums together differently in successive X10 sample windows.  A 1% mismatch in the frequency determining components would produce a beat frequency in that range.

No, this isn't a new chapter - just a closer look at what is going on in one particular installation.  It explains how CFLs can cause X10 modules to apparently switch on or off by themselves.

FYI, after enabling the gain switch, and allowing the automatic threshold control to cover the full range, the XTB-IIR was able to recover a X10 signal just slightly stronger than the peak CFL noise without recognizing any false commands due to the noise.  So it is possible that more advanced X10 modules will be able to deal with the CFL issue without us having to put filters everywhere.

Jeff

[Brian H]

I have some DuraBright Brand CFLs that make noise right on the X10 powerline frequency. They are made by TCP.
A EDXO-23 23 Watt and an ES42 42 Watt! Yup aroud the same as a 150 Watt Light for the second one.

[Boiler]

Ah, it isn't the beat frequency that winds up in the 120KHz region.  That is the carrier (chopper frequency).  The beat is in the 100Hz region so the noise sums together differently in successive X10 sample windows.  A 1% mismatch in the frequency determining components would produce a beat frequency in that range.

Ok, I think I'm getting this.  If we filter the chopper frequency (120K carrier) the beat frequency will be ignored by X10.

Thanks for the explanation.

[steven r]

What someone needs to come up with is a screw in or drop in the socket solution that would go between the CFL bulb and the socket.
Is that possible? If not, how about CFL modules and switches with built in filtering.
Am I just wishing in dream land?

[JeffVolp]

What someone needs to come up with is a screw in or drop in the socket solution that would go between the CFL bulb and the socket.
Is that possible? If not, how about CFL modules and switches with built in filtering.

Given today's competitive CFL market, no manufacturer is going to spend a nickel on anything that isn't absolutely necessary.  So don't look for any help from the CFL manufacturers except as a byproduct of their natural evolution.

If the lampcord is accessable inside the lamp, perhaps in the base, it is possible to insert an inductor in series, and double insulate with shrink sleeving.  I have done that myself on several table lamps.

Many recommend running CFLs off appliance modules.  I don't like the KLACK, and use regular lamp modules to turn them ON and OFF (no dimming).  That has worked fine with all our CFLs for years.  The inductor inside the lamp module that suppresses the triac switching transient may help filter the noise generated by the CFLs.  Now that I think of it, I had to add those inductors back when I was using appliance modules...

Jeff

[JMac]

Jeff - What does the inductor look like and how do you wire it ?  Thanks in advance, and in behalf of all of us, we do appreciate the input and tips.

[JeffVolp]

Jeff - What does the inductor look like and how do you wire it?

Snipped directly from my CFL tutorial:

If you are handy, an inexpensive solution for many table lamps is to install a 1000uH Miller 5258-RC "Hash Choke" in series with the hot lead.  Double insulate it with shrink sleeving.  While the 1000uH inductor is rated 1 amp (100 watt bulb), I recommend the higher current 250uH 5254-RC if you will ever use a high-wattage incandescent bulb in that lamp.  Electronic supply houses such as Digikey and Mouser carry the chokes and suitable shrink sleeving.

The chokes are about 1/2" diameter and 1 1/2" long.  Just cut the "hot" lead (narrow plug prong), and solder it in series.  Be sure to split the zip cord back far enough so that you can slip on shrink sleeving.  A second layer over the entire cord gives a finished appearance if it is easy to free one of the cord ends.  You will need 5/8" shrink sleeving.

Jeff

[azzar0]

azzar0,

I'm new to the site and was reading about your mysterious lights.  I hate to be a naysayer, but I'm betting you still have a noise source in the vicinity.  I say vicinity because it may not even be in your house. 

The changes you made should indeed improve the X10 signal level on your powerlines.  The level may now be just above that of the noise source.  You mentioned the behavior of your satellite reception when you turned on the lamps in your bedroom.

Oh, and get this: I also noticed that my satellite signal is pixelated when I turn the (x10) lights on in the bedroom. Turn the lights off - perfect picture!

 

Is your system operating properly now?  The X10 switches themselves should not have done this unless one of the triac drivers were oscillating.  If that were the case, it should still be happening (unless you replaced a switch).

Last year I tracked a similar intermittent noise problem down to a optical sensor in my post lamp.  The output stage of the sensor would go into oscillation (intermittently) and drive my X10 devices nuts.  Thankfully, it finally blew the breaker and I was able to put 2 and 2 together.  I've since purchased a X10 powerline tester and have a lot more hair on my head to show for it.  There is a site (can't remember the name) that offers a "loaner" ELK diagnostic unit to their patrons.  You may want to consider this if your problems return.

The problem could also be outside your home.  I have an audiophile friend at work who traced the noise in his high $ sound system to a compact florescent activated by an optical sensor (oscillating again).  This was a porch lamp three houses down (same transformer).

Best of luck

Thanks for the reply, Boiler and everyone else.

It is very possible that there is noise generated from an outside source. I haven't had major problems lately, but I still can't say that my system is 100% working. Sometimes lights that are scheduled to come on don't come on (miss the timer) or they don't turn off. It is random though, I haven't been able to detect a pattern. Once in a while lights turn off and then right back on, as if the system got overloaded or something and then reset itself.

I still have the satellite issue. When I turn on the floor lamp the satellite signal goes down (only on some of the transponders though, not all of them). This could also be because the floor lamp uses a halogen bulb?

I don't have any CFLs in the house, but I am pretty sure that some of my neighbors do. I was thinking of installing a PZZ01 at the power box to eliminate outside noise sources, but I need to hire an electrician to do that and it can get expensive. It would be good though because I could also eliminate the dryer coupler I am currently using.

I was also using a repeater (model SR731 I think) which I discontinued because I have a feeling it may have contributed to the lights on/off by themselves issue. I don't understand why though - if anything the commands should repeat (hence the name of the device???), in which case I should have never seen such behavior. If anyone can explain why would a repeater cause the problems I experienced and related in the beginning of this thread, I would love to hear it.

In any event, I am leaning more and more towards the PZZ01 idea. What do you think? Is this likely to eliminate the possibility of the issue to come back? Is there anything else I can do instead of or in addition to?

Thanks again!

[dave w]

azzar0

Are the lights that come back ON after you turn them OFF, fluorescent?

[azzar0]

azzar0

Are the lights that come back ON after you turn them OFF, fluorescent?

Hi Dave W,

What happens is this:
(1) I turn the lights on (or a motion sensor triggers a macro that turns the lights on)
(2a) after a while the light turns itself off and comes right back on
OR
(2b) The light turns itself off

The amount of time between (1) and (2a) or (2b) varies, and a lot of times it doesn't happen at all. I did notice that the system is most likely to behave this way if there is a LOT of X10 traffic (motion sensors mostly).

This behavior is not reproduceable. I can't MAKE it do this, it just happens once in a while. I must say though, I haven't seen (2b) happen since I unplugged the repeater, I still experience (2a) once in a while though.

None of my lights are fluorescent, they are all incandescent (with the exception of a floor lamp that uses a halogen light bulb, which IS dimmable, however).

There was a time when certain lights (4 different unit codes) would turn themself off and stay off. They are also the main lights, so when they turned off everything was dark! Like I said, I haven't seen that happen since I took the repeater out.

Thanks,

Azzar0

[Brian H]

Are any of the lamps; with CFLs; on an appliance module? Some will turn back on from the CFLs electronics tickling the Local Control Circuit in an appliance module. I had to modify all of mine to disable the Local Controls so CFLs did not turn back on. If this is part of your setup. Let us know and we can point you to mods, that eliminate it.