Print

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

[SteveRF]

Dave L and others...

I guess I may be assuming too much concerning noise but I definitely have one computer, one UPS, One DVD and TV setup and the wife has a Quilting Machine, Embroidery and Monogramming machines (She has an in home Memory Quilt business) that seem to cause malfunctions depending on the time of day of operation.   We didn't realize the Sewing stuff was causing problems for quite a while because of the times of use.  Also have a passive coupler that may be an issue but have not noticed a complete phase failure so doubt that is the troublesome spot. 

I was mostly interested in Dave W solution of incorporating choke coils into a home built filter for testing....  I was hunting for the best solution in building a good filter because of the known noise sources... at least when the offending PC is "completely" removed from the power line the light modules in that area of the home suddenly start working again.

Now the WAF is suffering terribly. That is what happens when she picks up a remote and after several clicks... nothing happens...
Comments like "This is becoming annoying instead of enjoyable.

Guess I will order some choke coils and give it a shot after I happen upon the latest and greatest filter schematic.

Would appreciate and and all ideas and suggestions.

regards,
SteveRF

 

[JeffVolp]

Series choke coils should be at least 100uH to be effective.  I recommend 250uH to 1000uH.  Remember, they MUST be rated for the maximum current.  Except for low current devices, such as compact fluorescent lights, you may find that suitable chokes bought from major electrical suppliers are actually more expensive than buying the "Made in China" X10 filters on eBay.

Jeff

[SteveRF]

Jeff,
Thanks for the input...

OldTimer recommended a source for 22 uh chokes...
You are suggesting 250 to 1000 uhenry chokes...

So what is the formula for the noise freq we are trying to choke out here ??  Anyone know...
Is there a "proven" schematic here on the forum with "for sure" values for attenuation ??

Appreciate any and all suggestions

Many Thanks,
SteveRF

[JeffVolp]

22 uH has an inductive reactance of 16.6 ohms at the X10 frequency.  While better than nothing, that is too low to have a significant effect.

The formula is XL (ohms) = .754 x L (in uH)

Unless you are using resonant circuits, you want a series inductive reactance in the 100 to 1000 ohm range.  Your choice will be a tradeoff between current capacity, inductance, physical size, and cost.  If considering resonant circuits, the cost of parts will probably be higher than a real 5 amp X10 filter purchased through eBay.

Jeff

[Oldtimer]

22 uH has an inductive reactance of 16.6 ohms at the X10 frequency.  While better than nothing, that is too low to have a significant effect.

The formula is XL (ohms) = .754 x L (in uH)

Unless you are using resonant circuits, you want a series inductive reactance in the 100 to 1000 ohm range.  Your choice will be a tradeoff between current capacity, inductance, physical size, and cost.  If considering resonant circuits, the cost of parts will probably be higher than a real 5 amp X10 filter purchased through eBay.

Jeff

I have to respond to JeffVolp's comment since he's effectively saying that the filters I built using 22uH chokes can't possibly work when in fact they work extremely welll.  (Having said that I will admit that the one on the microwave has been marginal on occasion but I'll save that, and the solution, for another post.)

While Jeff's calculation of 16.6 Ohms for the 22uH choke is true it is out of context if he doesn't also state what the impedance is of the overall house wiring is.  I have no way of measuring it but I think that the impedance of  house wiring at the X10 frequency may be quite low.  If it is in the order of one or two ohms then the reduction provided by my two 22uH chokes may be in the order of (33.2/33.2+2))*100 or 94%.

I don't have an X10 meter so I can't measure noise reduction directly, but let me give an example of why I think the the house impedance is so low.  The branch circuit with our downstairs TiVo on it is only about 20 feet from the breaker.  When the TiVo was first installed  X10 modules about five feet from the TiVo stopped working but would still work on an outlet about 13 feet from the TiVo. Nothing else in the house was effected.

The wire wrapped around a steel bolt I mentioned in my "how to" post solved the problem for one module but not the other which is why I went to the double 22uH chokes, which work perfectly.  Note that the same double 22uH chokes on the upstairs TiVo buffer its noise from the CM15A on the SAME outlet so that it continues to work perfectly.  I would hardly call this kind of performance "better than nothing"!

As a practical matter I don't consider either my or Jeff's posts to be the last words on this matter so I hope some others chime in so we can all learn more about noise control.

[dave w]

Bill H and SteveRF

My ACT repeater is an old CR231. The CR234 is it's replacement.
As far as the line filters, I got them surplus from MECI in dayton. The model I use is under $3 per unit.

HOWEVER there are two caveats with this :
1. I do not see the same line filter on the MECI web site any more, although they have plenty of others (http://www.meci.com/index.php/cPath/396) 
and   
2. JeffVolp pointed out that the filter I am using has a .01mf across the input which is a good X10 attenuator, which is true. But the ACT CR231 cranks out enough signal that X10 attenuation from the home brew filters has only been a problem on one branch.

Each noise source tends to be unique. I have also used simple ferrite choke core filters from Radio Shark on the line cords of noise sources and quieted them down substantially.

For me the key has been using the X10 signal meter to locate my problem areas.

[JeffVolp]

I have to respond to JeffVolp's comment since he's effectively saying that the filters I built using 22uH chokes can't possibly work when in fact they work extremely welll.

Actually, I said:

While better than nothing, that is too low to have a significant effect.

That doesn't sound like "can't possibly work" to me.

X10 transmitters, such as the TW523 are spec'd to drive 5Vpp into a 5 ohm load.  So, 5 ohms is a HEAVY X10 load.  That should be what all the parallel loads on both phases looks like.  Most individual X10 loads are probably in the 100 ohm range, or higher.  So, adding 17 ohms in series with a noise source or signal sucker that might look like 13 ohms itself doesn't change things that much.  Yes, adding multiple inductors in series helps more, but why not start out with a higher value in the first place?  The 250uH Miller 5254 is rated for 2.5 amps, and costs less than 2 bucks each when you buy at least 10.

Anyway, the impedance of house wiring at 120KHz is not that low.  If it was, the signal wouldn't fall off that much as distance from the panel increases.  A typical wiring run might have 5 to 10 uH of inductance, depending on overall length.  Add to that distributed capacitance and resistance from other loads.

While developing the XTB and XTB-II, I made countless measurements of signal levels throughout this house.  I also created a model of a simple electrical distribution system that mirrors the attenuation I see here.  So I have a little more information than wrapping some wire around a bolt and saying IT WORKS!

Yes, 22uH is better than nothing, but not the best choice.  Now, if you want to build a resonant T-network with 120KHz traps and a high-frequency shunt using the 22uH inductors, that should work pretty well.  But, why not just buy one for 10 bucks?

Jeff

[Bill H]

Dave W--Tnx, for the info.

[gil shultz]

Good Evening,
Sounds to me like ground induced noise, I cannot be sure but you should consider checking it out.  Be sure you know what you are doing or hire an expert. A mistook could kill you.

Start with your ground rods; do you have two or more?  If not add them at least 8’ and connected with the proper size cable?

Are the ground connections solid in your main panel and sub panels if you have them?

Are your grounds (green or bare wires) separate from the neutral?  The only place they connect is in the main panel, not in the branch panels.

Are all the ground connections in your home connected together with the appropriate crimp connector (not twist on)?

Is the ground wire bonded with an appropriate screw to the metal boxes?

Are all the electrical connections in your facility tight.

Are the metal frames of the florescent lamps grounded?

Is the hot water heater properly bonded to the ground?

Is there an appropriate ground wire from the water line coming into the facility to the electrical panel and bridging the water meter if you have one?

Are all the switches etc including x10 crimped the ground wire?

If you have the meters removed or turn off everything and see if you have current in the cold water pipe coming into your home (use a clamp on).  I do and it is because I have a neighbor that has a bad ground, I see up to 12 amps, the actual voltage is in the millivolt range. Check it with everything operating normally also.

Please note this is just a starting point and not everything may be allowed in your area.  I went through this and much more and when finished the buzz was gone from the radio and the x10 worked.  The reason for doing this initially was because of a heinous power surge that destroyed most of my electrical system.  Yes I replaced a lot of wire.

Good Luck
Gil Shultz