Flourescent kills X10

Certain loads create line noise that interfere with X10 signals. You might try a different brand bulb as that can make a significant difference. X10 is a 30+ year old technology that has always been a bit buggy. Newer technologies like Insteon, Zigbee, and UPB advertise that they don't suffer the same problems.

From:kviator snipped-for-privacy@gmail.com

These flourescent lights are high-end, 13 watt, quick start lights for under counter use. I'm not sure that replacing the bulbs would make much difference, but I may try as a last resort.

I guess I was looking for something like some sort a band-pass filter or something that I can put in the up-stream outlet that will bypass the flourescent light when on and pass the X10 signal down-stream to the next outlet.

Any other ideas? Maybe my macro idea is the only practical solution.

Thanks, Keith

BruceR wrote:

Yes, there are filters to solve the problem. Have a look at

or to see them. They have both plug in and hardwired versions. For the undercounter lights you could probably use the hardwired one at the j box. The fixtures you have contain circuitry that attenuates X10 signals. My reference to bulbs was based on the incorrect assumption that you were talking about the screw in compact flourescent bulbs. I too have the undercounter lights but was lucky enough to pick ones (purely by chance) that caused no problems.

From:kviator snipped-for-privacy@gmail.com

I'm not sure I understand what you are describing.

Light #1 works OK under control of an appliance module, turning on and off.

Light #2 works OK under control of an appliance module but only if light #1 is off.

Is this the situati>Arrrrgh. I finally got one flourescent light to work on an appliance

Keith,

Sorry you are frustrated with X10, but it CAN be a reliable system. It does take some planning and management to keep it reliable.

We use a bunch of compact fluorescents in ceiling cans to save energy. Most are controlled with Leviton 16293 relay X10 switches. To prevent potential problems, I included Leviton 6287 noise blocks in line between the switches and the ceiling cans. These filters are small enough to fit in the back of the switch boxes.

Unfortunately, the solution is not quite as easy with an appliance module. Years ago I ran into your problem when compact fluorescents first came out. The Lights of America CF bulbs had a filter capacitor directly across the power input. That killed any X10 signal on that circuit. I you are running into a similar problem.

The solution was fairly easy. I ordered some Miller 1000 uH Hash Chokes from Digikey and wired them in series with the bulb center contact. You can probably do the same with your under cabinet light. Just wire the choke in line between the "hot" input line and the circuitry. Be sure to well insulate the choke with shrink tubing to prevent any possible contact to the case.

We use X10 here for all lighting, sprinkler control, hot water recirculation, and exhaust fans. With hundreds of X10 commands sent daily, we see a missed command perhaps once a month. Not 100% reliable, but certainly better than 99%.

With some care, it should work fine for you too.

Jeff

Interesting idea Dave, but I don't think there is room. The Miller choke is about 1/2 inch diameter by 1 1/2 inches long. There may be smaller torrid chokes available. Something to consider is that the 1000 uH choke is only rated for 1A. That's ample for a CF bulb. But any modified appliance module would have to be identified so it isn't used for another purpose.

Along a similar line, I have modified floor and table lamps for screw-in compact fluorescents driven by an appliance module. I just snipped the hot lead inside the base of the lamp, wired a choke in series, and encased everything in shrink sleeving. But I used the 125 uH choke (3.5A) for these applications in case someone put in normal high wattage bulbs in the future.

Jeff

Gotta love those automatic spell checkers!

One would change our secretary's name from Denise to Dense. (She wasn't.)

Jeff

I took your admonitions about sizing the choke to handle the current to mean you wanted to avoid a torrid choke.

I guess I've never looked >Gotta love those automatic spell checkers!

Just to make sure that I have described the problem correctly, here's what I have.

Source -----------------------------|| end of line | | (AM1) (LM1) (F1) (I1)

, = Grounded Outlets , = Spliced-in additional grounded outlets

(AM1) = 3 pin appliance module set at A5 (LM1) = Lamp module set at A6

(F1) = 13 watt fluorescent light fixture (I1) = Incandescent light (just for testing)

has been modified to disable Local Control because it would not turn off and stay off (cut trace on IC pin 7). After that mod it blinked (F1) when in off state, so I modified it for current sense (sniped diode) and solved that problem.

I can control the (AM1)(F1) branch ok. If I have (AM1) in the OFF position, I can control the (LM1) circuit just fine. When the (AM1) circuit is in the ON position, (LM1) never gets any X10 commands because they are blocked by (F1).

I did write a ON macro that turned-on (LM1) first then (AM1) and an OFF macro that turned off (AM1) then (LM1) in that sequence and it worked.

I plan to replace the (LM1) module with another appliance module to run

2 more of the same exact kinds of lights after I get all of this sorted out. I was just using the lamp module for test.

I'm wondering what might happen when I add another appliance module in the (S2a) circuit and connect the fluorescent lights to it. Will I then loose control of the (S1a) circuit and be back to square 1?

Since all of these fluorescent lights will be used together, the macro solution is ok assuming that it continues to work when I add the second appliance module. A more permanent fix would be better though.

I just went out at lunch and bought another appliance module, so I'll be testing tonight.

Thanks for all of your help and suggestions.

Keith

Dave Houst> I took your admonitions about sizing the choke to handle the current to mean

If you refer to the schematic of the appliance module, you'll see that when the relay contacts are made (i.e. On), any line filter capacitor in the lamp is connected directly between line and neutral. Similarly, if the power supply in the fluorescent is generating noise and has no line filter capacitor this noise source is connected between line and neutral.

NOTE: The schematic has two mistakes. The -B in MCR100-B should be -8 and the device should be drawn as an SCR instead of as a triac.

Both Bruce and Robin think it's a line filter capacitor but I'm not sure there's enough evidence to conclude that it's not noise. You really need a scope to be sure or you need to look inside the fluorescent to see if it has a line filter cap that is of a size to cause problems. But, whichever is the case, Jeff's suggestion will probably cure it by blocking noise from the device or blocking the 120kHz X-10 signal from reaching the line filter capacitor.

The part I d>Just to make sure that I have described the problem correctly, here's >what I have.

It seems that if AM1 is on the noise is generated by the ballast blocking signal to LM1. A filter plugged into AM1 might do the job.

From:Dave Houston snipped-for-privacy@whocares.com

..... It sometimes works, but mostly it doesn't.

Radio Shack has a 100 uH RF choke rated at 2 amps. Will that work? It goes in series with the hot (typical black) line and not the neutral (white) or ground... right? Probably should do one for each circuit.

Thanks for all of your help, I'm going to lick this one way or the other! Keith

100µH gives 75 ohms impedance at 120kHz. That might be enough but with a 2 amp rating you'll need to put it in the fluorescent fixture rather than in the appliance module.

I'd put it in the neutral leg since X-10 uses the line as electronic ground. One in each leg would be good. A "typical" line filter uses a 0.1µF capacitor between line and neutral with 22mH (2200µH) chokes in each leg.

..... It sometimes works, but mostly it doesn't.