AHP used to work fine - now it only controls one module

[Noam]

You could always try turning off one breaker at a time, to isolate the noise.
Perhaps some device you have was affected by the power outage, and is now creating more noise than it did before.
I had the same thing happen (though not from a power outage) with a CFL that was going bad. Prevented ALL X10 communications in my home.

[drfthorse]

Everyone,
  My four XPPF filters arrived today.  I installed only one on my PC UPS.  All modules seem to work as they should from the AHP interface.  I will install the other three on the entertainment center and other "tech" areas in the house. 

Thanks to everyone for their input!  Now I get to install more switches and do some more programming!

Problem solved - case closed!

[eagle]

I will buy at least 6 of these.

I have to be mindful of the load though.

Here's a link to a scary experience by an owner where one of the filters caught fire!

http://www.remotecentral.com/cgi-bin/mboard/rc-x10/thread.cgi?4621

Yikes!

r,
eagle

[Brian H]

Yes I would not run an XPPF close to its five amp rating. They have a tendency to get warm. I tested one with 4.85 amps of resistive load. 600 watts of incandescent light bulbs. Ran it for thirty minutes. The filter got very warm and started to smell. Disassembly showed the plastic bobins the coils where wound on had started to melt.

Now in the link provided for the overheating.
If the user was using older inductive ballast load lights. This causes the current to be out of phase with the voltage and the current can be much higher than many meters measure correctly.
I am not familiar with the new high effency lights but their electronic ballasts may have current spikes that could be hard to read on most meters.

[JeffVolp]

If the user was using older inductive ballast load lights. This causes the current to be out of phase with the voltage and the current can be much higher than many meters measure correctly.

Yes, reactive loads are a problem.  The XTB includes a low-current low-pass filter similar to the XPPF to feed power to its X10 Input receptacle.  Specifications on the inductors say they will handle .2A, or 24 watts at 120V.  Since a typical X10 transmitter consums about 2 watts, one would expect that rating to be very conservative.  However, simulations show the out of phase power used by the transformerless supplies in most X10 transmitters is much higher than the actual power consumed, which is why I recommend that no more than two X10 transmitters be connected the X10 Input receptacle on the XTB.  (The inductors on the XTB-IIR are larger, and that can handle about twice as much current.)

Jeff

[drfthorse]

Do most of you not worry about a filter on the entertainment center then?  A stereo reciever and a large TV puts you well above the 800W 5A range for this filter.

Should I only/mostly be concered about computers?  Or is it always a trial and error thing?

[Brian H]

That is what the Smarthome 1626-10 10 Amp FilterLinc and ACT AF120 15 amp filters are used for.

Computers, UPS units and many cell phone chargers are high on the list of things to filter.

Entertainment devices can also make noise or absorb signals.

I even have found a rechargeable flashlight I have absorbs X10 signals.

[drfthorse]

Per everyone's advise I was monitoring the watts running through my newly purchased XPPF.  Tonight I turned on an additional computer increasing the overall load totalling close to 350 watts (digital readout on the UPS).  Sure enough the XPPF which was already warm got hot and there was a burning smell of basically melting plastic and electronics. 

I will echo previous posts and say that I would be very concerned about these products with a medium load on them.  Mine was melting at 3 amps even though they are rated for 5 amps.  I will now be purchasing some AF120s and not messing around.

[JeffVolp]

I suspect the XPPF can handle a 5 amp resistive load.  The problem is that the devices we need to filter do not look like resistive loads.

Power (and heat) is a function of the square of the current.  So if current is drawn over a relatively small conduction angle, such as near the peak of the AC waveform, the average current may be 5A, but the peak current (and resulting heating) will be much higher.

Take for example a 50% conduction angle with a 5A average load.  The actual current during the portion of the cycle where power is consumed would be 10A, resulting in twice the heating relative to a resistive load with a 100% conduction angle.  ((2I)^2) X 0.5  versus (I^2) x 1  So, the heating of the filter would be the same as for a 10A resistive load.

Unfortunately, most people assume a 5A current limit means the same for all loads.

Jeff