Print

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

[Bill H]

I am finishing my basement and have temporarily added inexpensive florescent shop lights (standard 4-foot bulbs) as work lights (Lowe's); they will be replaced with dropped ceiling florescents eventually. When I added the lights, however, many X10 modules stopped working (from added noise on the line I assume). I am using an XTB and, when I added a phase coupler, it only coupled the noise to even more modules.

Question: Does anyone know if the higher quality florescents will work? Any suggestions on what to look for, how to choose for X10 use, etc.?

[JeffVolp]

Many of today's fluorescent lights use electronic ballasts.  They can be major noise producers.  I have received a number of emails regarding problems with those units.  While there may be "higher quality" units that do not produce noise, I suspect most of those units use similar electronic ballasts manufactured for the lowest possible cost.  Obviously, leaving out any line filter results in more profit for the manufacturer.

The one thing that seems to work for those units is to isolate them with a filter.  While I recommend the small Leviton 6287, some have also had success using the X10 XPPF as an in-line filter.  Since that is a low-pass filter, it isn't quite as effective as the 6287, which is a 120KHz notch filter.

The old fluorescent lights with magnetic ballasts do not cause a problem for X10.

Jeff

[Bill H]

Tnx, Jeff. Tried an ACT AF120 (15A Plug-in), but same results. It's a fairly large area, so florescents make sense...but not enough to give up X10. Maybe lots of well-placed candles....

[JeffVolp]

The question here is what frequency is the noise source, and what is the filter tuned to reject.  The AF120 is a notch filter.  The schematic I have shows it uses 7.2uH inductors and .22uF capacitors.  With those values, maximum attenuation is at 126KHz, which is a bit above the X10 frequency of 120KHz.  Depending on component tolerances and how sharp that notch is, it is possible the attenuation may not be that effective at 120KHz.  It is interesting that the Leviton 6287 is also tuned to the high side, but that filter has a larger value capacitor used as the leg of the "T".  So, it will provide somewhat more attenuation to high frequency signals.  The 5A X10 XPPF is a low-pass filter.  While its maximum attenuation is not as much as a notch filter, it is more effective attenuating broadband noise

I installed fluorescent lights in our utility spaces.  To avoid X10 problems, I chose lights with the older magnetic ballasts.  I did install a couple of the newer suspended ceiling fluorescent fixtures with electronic ballasts in the rec room.  While I haven't noticed any problem with X10 communication while they are on, noise from them does light up one bar on the ESM1.  I suspect that noise is far enough away from 120KHz that the receiving modules are able to reject it.

Jeff

[dave w]

FWIW
I don't have any E- ballasts to play with, but am wondering if wrapping the ballast feed wires around a heavy toroid would drop the noise. Radio Shark has an open frame toroid which they now charge a ridiculous amount for, but could used for testing.

[JeffVolp]

I don't have any E- ballasts to play with, but am wondering if wrapping the ballast feed wires around a heavy toroid would drop the noise.

That is an interesting concept, but I suspect the inductance would not be sufficient to have a significant effect at 120KHz.
 
If I ever get some free time, I'd like to run filter response tests on the XPPF, XPF, and 6287 to better understand their characteristics.

Jeff

[Bill H]

I thought the filtering was the same, only the wattage different on these units, so it never occurred to me to try the XPPF. I just did...and it worked beautifully. Thanks guys!!!

[Tony Olson]

Many of today's fluorescent lights use electronic ballasts.  They can be major noise producers.  I have received a number of emails regarding problems with those units.  While there may be "higher quality" units that do not produce noise, I suspect most of those units use similar electronic ballasts manufactured for the lowest possible cost.  Obviously, leaving out any line filter results in more profit for the manufacturer.

The one thing that seems to work for those units is to isolate them with a filter.  While I recommend the small Leviton 6287, some have also had success using the X10 XPPF as an in-line filter.  Since that is a low-pass filter, it isn't quite as effective as the 6287, which is a 120KHz notch filter.

The old fluorescent lights with magnetic ballasts do not cause a problem for X10.

Jeff

Sorry to bring back this thread, but I have a real old shop light (2 long bulbs) in our washroom. I have an XTB and passive device coupler installed but was thinking an in-line filter (for the florescent shop light) would quiet the noise down. But Jeff's comment in re: to the old fluorescent lights with magnetic ballasts not causing problems for X10 stopped me short <g>.
I was going to ask what filter (the Leviton 6287 et al) would be best. At times lights come on by themselves  when I'm using my treadmill (which I filter with a AF120). My X10 system is fairly responsive and has never worked better (closer but no no means  perfect).

Thanks

[JeffVolp]

Even a new fluorescent light with an electronic ballast would not generate noise when the light is off.  The problem reported by some folks is that once their fluorescent lights are turned on, X10 switches on that same circuit will not respond to commands.

When you say the lights come on by themselves, can I assume that means they are controlled by a X10 switch?  The fact that the lights switch on when your treadmill is running leads me to believe there might be a loose connection somewhere, and the vibration from the treadmill activates the local control feature.

It is also possible that the electrical noise generated by the treadmill is making it past the AF120 filter.  That is a notch filter.  Noise outside its rejection band can get by, and if strong enough might still effect a X10 switch.  I have seen noise morph one command into another, and that is a possibility here.

Jeff

[Tony Olson]

Yes Jeff the devices/lights that came when I was on the treadmill  were X10 Devices. I think your explanation of a noise spill over (my term) was no doubt due the type of filter the AF120 filter is.

Your XTB device has helped to make the occurrences of this behavior less and less and very manageable.   

I'm glad to leave my shop light as is. I'll save this thread for future reference (I make PDF's out of them).

Thanks again.

Tony Olson