Line Noise Interference Question

What you have described is a perfect example of an X-10 sink. The power supply in your new computer has some large and effective capacitors across the AC line which are shorting out the X-10 signal. This is a common problem with switching power supplies as used in computers, televisions, and an increasing amount of consumer equipment.

As a partial fix you can try plugging the computer in a different outlet, hopefully a different circuit. It may improve performance a bit. The 100% fix (for this specific sink) Is to separate it from the outlet and your house wiring with an X-10 "filter". For example:

Geo.

well, i just got a filter to address my ws12 switch problems, and somehow I managed to get a bum unit myself, so it does happen apparently...I got the smarthome branded 10amp filter, and when I plug a module into either the filtered or unfiltered outlet, i can still control the module. Needless to say i'm pretty irritated at getting a unit that never should have made it out of quality control. Have not contacted smarthome yet to see what they're going to do about it (after business hours) but they better have a free shipping label for me.

Yours is the opposite of a noise problem. The power supply has a capacitor across the mains that acts as a near-short to the 120kHz X-10 signal. It's intended to block noise from the power supply from reaching the powerline but it's also eating your X-10 signals.

I'm surprised that the filter isn't helping. It should block the 120kHz signals from reaching the capacitor. Which model filter are you using? The filter should be plugged into the wall outlet and the computer plugged into the filter.

Two things to try: Double up the filters, thatis plug one filter into another and see if that helps. If not, an isolation transformer should do the trick or you can try a different power supply in the PC.

From:Dimbo Spams cuervojose(remove)@jps.net

It may be the powerstrip that is the signal eater. How do you have the filter, powerstrip and PC connected?

If it's wall plug -> powerstrip -> filter -> PC and you have to unplug the powerstrip to eliminate the problem, the powerstrip is the likely evil doer.

Try wall plug -> filter -> powerstrip -> PC.

If that doesn't fix things then you may have a defective filter but I think the chances of that are very low. You can test it using wall plug -> filter

-> lamp (or appliance) module. If you can control the module, the filter is defective.

If it is the powerstrip, you can open it up and remove the capacitor.

That one is fine but the power strip is likely part of the problem. The power strip should be plugged into the filter and the filter to the wall. Sometimes two filters will solve a particularly bad problem. You can use the 5 amp filters instead as it's unlikely that your PC and associated equipment will draw more than that.

From:Dimbo Spams cuervojose(remove)@jps.net

You may be right. Doesn't hurt to try though. I've heard it sometimes works but have never needed to try it. Sorry my last post was redundant to what you had already said. Should have read through the thread first.

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

The ACT filter he's using is a 30:1 bandstop filter. It blocks the passage of 29/30 of the signal, preventing it from getting sucked down the drain. Since only 1/30 of the signal gets through the filter, adding another filter will only help ACT's bottom line.

OTOH, when the problem is noise, adding a second filter may sometimes be necessary. It depends on the efficiency of the filter.

It's unfortunate that ACT uses the term "attenuate" >You may be right. Doesn't hurt to try though. I've heard it sometimes

It depends on the filter. None are 100% efficient so one might allow enough signal through to operate a module. Are there any published specs for the filter you have?

I was a little too quick with that suggestion. I can't find a spec on the X-10 web site but, from memory, the older units need 100mV signal and newer ones only 50mV. I'm not sure whether that is peak-to-peak or just peak.

With a 5V signal level even the 30:1 filter will allow 167mV through and many transmitters have a 10Vpp signal level.

this is the filter i got:

anyway I got an RMA on it (no questions asked), so hopefully it is indeed non-functional. (before trying it with a module, I was trying to use it to filter the PC that i had established as the source of my problems, and was noticing no effect whatsoever). Otherwise I guess i get to start trying to relocate my PC or something.

I noticed that when I put an ACT 5A filter between a signal sucker and the line that a maxicontroller could still reach and control an appliance module located very closer to the other side of the filter. I also noticed very odd results combining a surge protector with a Leviton filter.

I confess to not having read up as much as I would like on filter design, but perhaps you can give me some guidance. Is it possible for two "signal suckers" to interact in unexpected ways? I plugged two surge strips that gave me X-10 troubles and used the ESM-1 to measure a signal at the end of both of the units. It was higher at the end then it was in the middle. Is that even possible?

Since I can't get the level of precision I want from the ESM-1, I decided to bite the bullet and buy the Monterey unit. The Lynx wasn't very useful to me in making detailed analyses of the "power picture" at my house. The ACT unit was just too damn big. From what I read, aside from the BSC problem, it's going to give me a signal and noise readout that promises to be useful in both bug hunting and testing.

What I want to do is design a test bench that can not only get some sort of "device profile" for all my 110VAC equipment. I was going to set up the meters (I want to make sure the ESM-1 and the Monterey agree) on a plain vanilla power strip plugged into an ACT 15A filter. Is that a good idea? I'm worried that devices might react differently in the test bench than they might in the real world. What else should I be measuring to try to eliminate variables? I want to know which devices present obstacles to X-10 signal propagation but I don't want to waste time recording worthless readings because I didn't think it through first.

Thanks!

-- Bobby G.

It may be that the combination is creating a resonant circuit. That's why I suggested to the OP that he remove the powerstrip and test with just the filter. It's always best to start with the simplest possible setup.

Without knowing exactly what components are used in each filter, it's hard to say what interactions are possible.

Ground loops might cause erroneous results.

I didn't think the Lynx-10 PLC would be very useful as a test instrument. I also hate to see anyone spend so much money for the Monterey. The Monterey maxes out at 4.096V so, while being able to measure =might in the real world. What else should I be measuring to try to

I've found that plugging everything into one powerstrip makes for resonably reliable benchmarks. That's how I measured the outputs of the various transmitters at

The filter will provide further isolation but you probably don't need it unless there are signal suckers nearby. You can verify this by taking readings with and without the filter in place.

It really is a shame that the web site that had all those papers on powerline communications was taken down. There were some really informative studies there that, if nothing else, illuminated the complexity of using the powerline as a communications medium. The powerline acts like a transmission line with a characteristic impedance that varies from ~2-80 ohms. Like any transmission line, there can be reflections, nulls and peaks.

all

informative

transmission

I decided I wasn't as concerned about the Monterey's lack of max voltage indication as I was that the ESM-1 and the Monterey agreed. In fact, it was you message about how the ESM-1 was calibrated at the factory to agree with the Monterey that made finally decide to buy one. That turns out to be a plus for me since one meter can serve as a sanity check for the other in the face of bizarre readings. As far as profiling the transmitters that overwhelm the Monterey, I can read them through a filter I'll keep offline as a "reference attenuator" that shifts the signal into the range the Monterey and ESM-1 are set to measure. Not ideal, and certainly not useful to compare my readings to anyone else's, but it shouldn't affect my goal of having baseline numbers for every device. I wonder if the Monterey has a similar potentiometer to the ESM-1 for adjusting the scale?

This has got to be the most heavily researched thing I ever bought! )-: I've read lots of previous threads about the unit, particularly between you and John Gavin (sp?) where you discussed decay rates of the CM11A signal and methods of measuring the signal quality (where on the AC cycle samples are taken, how many, etc.). I struggled mightily with the Monterey over the ACT that Bruce R. likes and decided as long as neither has an output port, the smaller size of the Monterey makes it worth the premium.

While an oscilloscope would be the best way to look at an X-10 signal, on a practical level, I need something I can tote to the places I have having problems. I forsee getting the USB scope you cited a while back, but right now I need real world tools that help me diagnose the few intermittent problems that still plague me.

As for a test bench, it's going to be hard to keep it isolated from other X-10 transmitters, which I learned through you and Google all attenuate (is that the right word in this context?) the X-10 signal slightly. I've got

I thought about plugging the test strips into two 100' 14AWG extension cords I have just to put some distance between the test bench and the rest of the house, but your suggestion of measuring with and without a filter will probably tell me (with the Monterey or a scope, at least) whether there's a signal sucker on that circuit leg. If the measurements are too far out of synch, I might go back to the 200' of cord idea.

Probably the best way to test would be to shut down all breakers in the house except for the one I was using to test, but the last time I did that, I got into BIG trouble for blowing up the VCR's programming. (Actually, I did it twice. When I reset the programming, I inadvertently set the time for AM instead of PM. Oddly enough, the unit seemed to be coming on at the right times, but in reality, they were out of phase by 12 hours).

Since the great breaker box fiasco, SWMBO has put a limit on the number and type of "research projects" I can engage in at any one time. I have to apply for permission to shut off the panel since I mucked up the recording of a "Series Finale." The currently approved project is to get the HAI Omni LT function as a basic alarm system, capable of calling a series of numbers when an alarm event occurs. I have one week, starting tomorrow, to get it done! (So, why am I here in CHA, you ask?)

Wasn't someone in CHA measuring their home's impedance at the breaker box?

I saw an article that talked a bit about powerline impedance here:

Protocol to Avoid Noise in Power Line Networks engrwww.usask.ca/research/ ee/faculty/ded632/files/Powerline2005.pdf

And there's an enormous powerline FAQ at:

Thanks again for you insight, Dave. I'll be sure to post when "Monty" arrives. My main purpose in buying it is to get very accurate readings of the X-10 signal strength at multiple points throughout the house. There are just too many "semi signal suckers" that seem to cause a degradation of the signal, but it's just too small a voltage for the ESM-1 to see it (and I've tried videotaping the display and playing it back in slow motion to see if I could see a difference in the # of bars lit per frame). That's when I decided that despite its shortcomings, the Monterey is going to give me the degree of precision I need as admirably as it's going to unfill my wallet. :-( It really is just a little too expensive to make me feel really comfortable with it, but my vendor assures me they'll take it back if I don't like it. We'll soon see!

-- Bobby G.