X-10, GFCI, outdoor lights & problems

Hello,

I have searched the archives, and I have seen anecdotal issues. Here is my situation:

Today I installed 5 outdoor 120v light fixtures (porch lights). Two are fed from an existing wall switch that I replaced with an X-10 switch. The remaining 3 lights are on a new extension of a circuit downstream of a brand new GFCI.

All 5 lights are on the same code (I-9). When I send the "on" code from inside (from a PHC02 maxi controller), the on and off are somewhat sporadic. The non-GFCI circuit responds as directed and the downstream X-10 reciever does not alway "hear" the instruction.

If I plug the transmitter in downstream of the GFCI it all seems fine.

Is there a GFCI/X-10 signal issue? If so, how do I work around? I could change the wiring so that the 3 lights downstream of the GFCI are fed from a different non-GFCI circuit.

What do you think?

Reply to
neergqj
Loading thread data ...

GFCI brekers can cause problem. I tried both an appliance and lamp switch on a GFCI breaker protected pool light and the breaker would pop everytime. I used a relay controlled by a modified appliance module and it worked fine.

Reply to
brobin

My issue is not the GFCI or breaker, but the erratic nature of the X-10 units. It seems that they will (usually) both turn on OK, but the X-10 controller downstream from the GFCI won't turn off unless I plug the X-10 controller in downstream of the GFCI. The the other X-10 controller can't hearthe signal.

I guess I will re-wire the second outdoor circuit so it is not on the GFCI circuit.

Reply to
neergqj

Several thoughts:

1) As I understand it, if you move your transmitter close to the X10 switch, it works better. There may be a signal sucker somewhere on that branch circuit that is reducing the signal level. Computers, electronic equipment, power strips with surge protectors are all culprits.

2) The signal path may be too long for reliable X10 operation. We had a long run out to a yard light at our last house that sometimes would miss X10 commands. The problem went away after I added a Leviton 6201 coupler/repeater to boost the signal level.

3) While I have used X10 modules downstream of GFCIs with no problems myself, that doesn't mean that all GFCIs are the same. If the design of your GFCI includes an internal line-to-line capacitor for noise suppression, it could also attenuate X10 signals.
Reply to
Jeff

What kind of lamps are you trying to control? If they are compact fluorescents you may be experiencing the well known problem many of these units have in terms of putting noise on the powerline when they are operating. They'll turn on OK, but their ability to both attenuate X-10 signals as well a generate noise on the powerline prevents them from "hearing" an OFF command once they have been turned on. If they are CF's, try incandescents in the same circuit before you rewire. If those work, you may simple need to find a different make of CF bulb that's not so X-10 unfriendly.

I'd try very hard to look for a solution that doesn't bypass GFCI protection to something mounted outside the house or near water or rain.

-- Bobby G

Reply to
Robert Green

Aha!

Yes I am trying to control compact flouescant bulbs. Five to be exact. Since they are so economical, is there an X-10 workaround?

Reply to
neergqj

About half the bulbs we control with X10 are compact fluorescents. It just takes some special care. We use Leviton X10 switches that require a neutral to guarantee a good signal level. And then I used the small wire-in filters in all compact fluorescent circuits to prevent noise or signal loading from corruptiing the X10 signal.

Just take a few precautions, and it should work fine.

Reply to
Jeff

The simplest work around is to try different CF bulbs - they vary greatly in X-10 friendliness. I bought two blister packs a month or so ago of 60 and

100W equivalent bulbs. Both by the same manufacturer (GE). The 60's didn't generate line noise nor did they attentuate the signal, but the 100's did.

I would advise once you find a bulb that works, buy a 5 or 10 year supply of the same lot number. There's no guarantee that the next set from the same manufacturer won't foul up X-10 communication.

Jeff's advice is also good - it will prevent someone from changing to the wrong bulb in the future and messing up the X-10. It's a little more expensive to implement than finding non-interfering bulbs. If I were in your shoes I'd spend the $50 on an ESM1 meter rather than "heavying up" your outside circuits.

I would say that almost 1/3 of new electronics that I buy has some sort of X-10 problem. )-: The lastest was a Norcent DVD player that knocked out a whole circuit with a 2V drop in the X-10 signal. I'm relocating a lot of such equipment to the same place where I can put them on a power tap that's plugged into a X-10 filter.

If you're going to use more X-10 (and almost everyone does expand their setup) a meter will save you so much time and frustration you will be amazed.

-- Bobby G.

Reply to
Robert Green

Good point. If it's the GFCI that's causing the problem, you may have to look for another one since filtering it means you can't control any loads beyond that point. It's another reason why even a cheap X-10 meter is so important to have if you're serious about X-10. Once you can actually

*locate* signal suckers and noise generators, cleaning up an X-10 network becomes ten times easier.

-- Bobby G.

Reply to
Robert Green

Thanks for the feedback. Suppose I get an ESM1, how involved is the diagosis?

Reply to
neergqj

Quite simple. You may need a screw-in adapter (an outlet that screws into a lamp base) and an extension cord to use the ESM1 on an outdoor lamp circuit. The idea is to plug the ESM1 as close to the offending circuit as possible. If the X-10 signal it attenuated, not many bars will light up on the LED display. If there's noise present, bars will light up, but the "X-10 Good" LED will not. When plugged in right next to a X-10 transmitter almost all the bars will light up (depending on which transmitter, of course - they output substantially different levels with the CM11A being one of the weakest). From there, the signal will get weaker and weaker the further you go from the transmitter. Readings near signal suckers will be dramatically lower. Readings near noisemakers will show constant LED bargraph activity without the "X-10 Good" LED lit.

The idea is to locate and filter devices that produce noise or attenuate the signal. If you get a long enough extension cord, you can take the ESM1 to the electrical panel and shut off each breaker in turn until you find one that either kills the noise or increases the signal level. If you're into high tech, you've doubtless got at least a few X-10 unfriendly devices. It's becoming a more and more annoying problem as there's really no way other than plugging it into the home wiring whether something's going to affect the X-10 signal.

-- Bobby G.

Reply to
Robert Green

OK, I think I have this straight. The CF bulbs I am using are generating noise when they are on. That's why the lights go on fine, but then act up. The curious part is that one circuit seems to respond OK, and the other not. It is only when I place the transmitter downstream of the GFCI that both circuits respond as intended.

I suppose it could be luck of the draw on the 2 sets of CF units, as one set is 75W equivalent andthe other is 100W equivalent, purchased at different times.

I have googled a fair amount of anecdotal evidence of X-10/CF problems but I haven't seen as much on X-10/GFCI problems and cures.

I am not sure that if I install noise filters (Leviton 6287) on each of the circuits that it will fix the problem. I guess I can try...

Reply to
neergqj

Those are the filters I'm using. We have different brands and sizes of compact fluorescents on those circuits, and have no problems at all.

We also have X10 modules plugged in downstream of Leviton GFCI's, and have had no problems with those.

As I said in an earlier post, it could be the length of your circuit. With a switch that doesn't require a common connection, the signal path must flow out to the load and back. Adding the filter will place the signal return path at that point.

Reply to
Jeff

Out of curiosity, how did you identify this noise? Did it show up with your ESM1?

Reply to
Dave Houston

Do the cheapest, easiest tests first: Just use garden variety incandescent bulbs to test. If THEY work, and turn on and off with no problem, then it's probably NOT the GFCI but the CF bulbs themselves or their interaction with the GFCI that's giving you trouble.

If that didn't clear up the problem, I would then kill all the breakers at the main panel except the one you're using to see if there's a serious signal sucker somewhere on the electrical wiring. I've pumped X-10 through hundred foot extension cords without serious measurable signal loss. But put one serious signal sucker on the line and you can knock a whole house into trouble.

A meter is essential for finding noise and attenuation in the system. Otherwise you're going to spend a lot of time wondering what component is screwing things up. It's a rite of passage with X-10. Most people having to chase their tails for a awhile trying to solve mystery X-10 problems before they can justify the cost of a meter but it's really a must have.

My attitude about X-10 reliability changed after getting the ESM1 and finally being able to find out which devices were interfering with X-10 and how much. I pulled out 5 power strips, two PC's, two UPS's, some CF bulbs, some 48" fixtures, a VCR, a laser printer, a laptop PS, an external HD power supply and a few other things that caused anywhere from a .3 volt to a 2 volt drop of the X-10 signal. A few fluorescent fixture put noise on the line as well, and the ESM1 measured those at .1 volt.

-- Bobby G.

Reply to
Robert Green

Yes. I thought there was something wrong with the meter at first. It would display nothing but the "ON" LED for a while and then the first bar of the LED display would light up and stay lit, but not the "X-10" good signal. Using a plug-in Leviton filter stopped it. Since the lamp also attenuated legit X-10 signals, it's hard to estimate how much the noise effected signal propagation.

-- Bobby G.

Reply to
Robert Green

Thanks. I've never found any noise sources that show up on the ESM1. I have generated some on my own to test the ESM1 but it's good to know it can detect noisy switchmode power supplies.

I am still curious though about how a device that obviously filters ~120kHz still outputs noise in that vicinity. Their filter must not be all that effective.

Reply to
Dave Houston

Cabling-Design.com Forums website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.