Print

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

[Charles Sullivan]

I agree with all the added data.
Will a True RMS meter be better?
Got a nice shiny Fluke at work the other day and maybe if I beg enough can borrow it. 

A True-RMS voltmeter and/or ammeter won't help with power measurement unless you know that the load is purely resistive, which is not the case for these modules.

The Kill-A-Watt meter mentioned by Take The Active measures power, but is not accurate enough at the low power dissipation of the X10 modules to be very definitive.  (It's rated at 0.2% but that's of the full scale of 1875 Watts.)  However for want of anything better at hand I plugged a few devices into it and looked at both the Watts scale and the VA (Volts x Amps) scale:
   Device         Watts    VA
   AM468           0          3
   AM14A          0           6
   LM465           0          3
   TM751           1        11

[Brian H]

Thanks for the added data. Scratch the new meter.

[JeffVolp]

I'm still looking for a way to disable the local control current as well as any other methods to reduce the power consumed by X10 modules.

The local sense resistor in a lamp or appliance module is 330K.  That is less than 0.05 watt, or about 1 watt hour a day.  At average utility rates, disabling that will save you less than a nickle a year.

Again, lamp and appliance modules do not consume anywhere near the power that you think.  They have capacitive power supplies.  Most of the current is out of phase with the voltage, and is not real power.

[steven r]

So bottom line is we can conclude that X10 modules don't add any significant amount to the electrical bill. Also it is likely that timed offs may save more electricity than the modules use.

Right?

[ArtClark]

Maybe I am missing something here, but the biggest power waste seems to be (Using the AM486 schematic) the 2.2K that is connected across the switch of the relay through a diode.  Assuming a lamp load (Which is at the lowest resistance cold), this would be a direct resistive load that ISN't capacitive, inductive, or in any way hard to read.

For a rough approximation, 110V (RMS) / 2.3K = 47.8 ma  (Approx.)      Therefore, wattage:  .0478 * 110 / 2  = 2.63 Watts !

Obviously, this is a LOT of power and assumes a 100 Ohm Load in the Off state.  Also this doesn't count ANY of the power that the actual module draws, which I assume is around .5 Watt.    All this really means is, the original estimate of 3.5 watts is right on with a real-world load attached and the module OFF. 

I haven't tried nor tested what the results would be to disconnect the circuit that leads from that 2.2k to the chip and then severly increasing the value of the 2.2K but I assume that would help.

Anyone have comments or want to prove me wrong, please do.  One should learn something new each day, and I am ready to be taught.  I can see nothing wrong with this check and Most loads are at least 50W or so.  A 50 Watt load would still draw 1.8 watts plus the .5 for the circuit or around 2.3 Watt total.  Removing this Off current would cut the module consumption greatly, though I can't see myself going through the work required, though with the cost of electricty going up so fast, I may try this sooner than I would have thought.

Input please.  I am sure there are poeple out there with lots more knowledge than I and I could use some reassurance.  I only got around to tuning modules recently, and what an eye-opener that was.  This could be even better.

TIA.

[JeffVolp]

So bottom line is we can conclude that X10 modules don't add any significant amount to the electrical bill. Also it likely that timed offs may save more electricity than the modules use.

Correct.

[Charles Sullivan]

FWIW, I plugged a 100 Watt lamp into an AM486 appliance module and that into the Kill-A-Watt meter.  I verified that the lamp is indeed On when the module is On.  With the module Off the readings are the same as reported earlier with no load connected, e.g., 0 Watts, 3 VA.
 
I repeated the test with one of the very old "BSR" AM486 appliance modules with the brown enclosure and codewheels on the sides.  I  got the same readings.

As mentioned earlier, the accuracy of the Kill-A-Watt is not high enough at these low levels to be definitive, so I tried a crude test to offset the bias in the meter:  A 4 Watt incandescent night light plugged directly into the Kill-A-Watt measure 2 Watts, 4 VA.  The same night light plugged into an "On" appliance module measured 3 Watts, 6 VA.

Jeff Volp has performed enough experimentation and measurements on X10 that I would have no hesitation in accepting his statement that the quiescent power dissipation by these modules is very low and not worth worrying about.

[JeffVolp]

UPDATE:  That 2.2K resistor on the schematic is WRONG.  It is actually 33K, so it dissipates just over .2 watt, or perhaps 2 cents a month in electricity.  What I thought was 2.2K in the brown BSR appliance module is actually the 220 limiting resistor in the power supply.  After tracing the wiring, I saw the sense resistor in the BSR module was actually 33K.  So I went back and measured the one in the newer module.  It is also 33K although it looks like red-red-red. END UPDATE

Just looking at the schematic, I agree with your numbers.  However, we must be missing something.  I ran the newer module open on my bench with a load connected but off.  I measured about 10C rise in the temperature of that resistor when it stabilized, so it can't be dissipating that much power.  I've burnt my fingers on enough devices over the years to know when they are pushing their limits.

I mentioned the 330K resistor because that is the part of the local sense circuit in a lamp module that does see the full line voltage when the module is off.  There is also that 100K across the triac to negative V.  While I'm not certain of its function, if it is used just for the local sense, eliminating it might save another penny a month on the electric bill.

If you wrap your hand around one of these X10 modules while off, you find they are barely above room temperature.  Compare that with a 4-watt nightlight.  Also, compare that with a lamp module when it is actually delivering 300 watts to a load.  The triac drop will be about a volt, so the module will be dissipating about 3 watts at that time.  Clearly these modules don't dissipate that much power when off.

[Brian H]

I see you too have observed the resistor shown on some schematics is incorrect. I have also seen 33K.

[ArtClark]

Thanks for the info.  I had to open my mouth (Keyboard?) after looking at the print, because it made no sense to me that it could be a 2.2K.  Without opening a unit and actually checking (Which I should have done.) I had to just go by the print.  A 33K  (Orange Orange Orange) would draw nothing to even bother with.  IF it HAD really been a 2.2k, the entire 2.x watts would have been disapated by it, causing it to have been a LARGE power resistor.  (Could have been used as a little heater, too.)

Doing a few checks of my own, (Using my customers test equipment.....), there isn't enough power being used by the whole module and sense current to really bother with.  (As you guys have been saying!)  The amount of cost is easily going to be offset by power savings in shutting One lamp off at the right time, etc.

You guys also, once again, proved the old addage "When in doubt, Check it Out".  I really should have opened and checked the unit before typing, but thats all history now.  As for module heat, I was surprised that I could actually notice it.  AFter a few hours, if you are careful, you can notice that the module is a couple of degrees warmer than the enviroment, maybe.  I cannot imagine the total dissapation to be more than .5 Watt max, to give off this little heat.  (Again, as everyone already said.)

Thanks for all the info, and for waking me up to remember to Never fully trust a print.  In my work I never do, but here at home the rules got a little slack.  Hopefully I won't get caught slacking again.