Print

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

[JeffVolp]

I agree fully with the unlikelihood of any X10 module “bursting into flame”.  I have seen the XPPF heat up and start to stink when loaded to the max.  And the beta CM11A could get stuck in a mode where it got rather warm.  But I have not seen ordinary X10 modules do anything other that get a little warm when controlling a heavy load.  That is a result of the roughly one watt per hundred watts of load that must be dissipated as waste heat.

Regarding using lamp modules for non-incandescent loads, this is more of a factor today since X10 added the ramp-on / ramp-off.  The lamp module is essentially a solid-state relay whose turn on time can be varied across each half cycle.  As long as it only turns on near the zero crossing, it can indeed control non-incandescent loads.  We have been using one like that to control a recirculation pump in our hot water system.  Since that is an old lamp module without the ramp-on / ramp-off, and is only controlled by ON and OFF commands, it has worked fine for years.  An appliance module would have produced an annoying Klack …. Klack every 15 minutes throughout the day.  And, I question how long its mechanism would have held up when cycling almost a hundred times a day.

The issue with the lamp module triac is that when dimmed at all, the switching occurs when there is substantial voltage across the device.  Since it switches on instantly, there will be a very rapid rise in voltage across the load.  That induces magnetic effects in some devices, and causes the low frequency buzz often heard in dimmed incandescent lights.  Some devices using transformerless power supplies can be stressed to the point of failure by the rapid rise time.

So, the bottom line is to only use dimmer modules to drive loads specifically designed to be used with dimmers.  When deviating from this rule, be sure you understand exactly what you are doing to prevent failure of either the module or the device it is powering.

And, like with any electrical device, it is always a good idea to keep them away from any particularly combustible objects.  Things do fail.  How many houses go up every year because of a short in holiday lighting?

Jeff

[dave w]

As long as it only turns on near the zero crossing, it can indeed control non-incandescent loads. 


So, the bottom line is to only use dimmer modules to drive loads specifically designed to be used with dimmers.  When deviating from this rule, be sure you understand exactly what you are doing to prevent failure of either the module or the device it is powering.

Jeff, a trifle disagreement with your first statement.
We use a lot of "accent lighting" in the home. One form is a series of "flameless" candles powered by two 3V DC transformer wall warts in the bedroom. I controlled these wall warts with two LM15A Socket Rockets, figuring they would be fine as the triac gating closely follows the sine, and dimming isn't possible.

When first trying the LM15A/wall wart combination I closely monitored the combination for problems, for a couple of days before tucking them behind a dresser and chest of draws. A couple of years later when one set of candles suddenly quit working, I paniced when I saw the wall wart with the plastic case completely deformed and melted down around the transformer which was now a beautiful smoked bronze color. I found the second wall wart to have the same symptoms, but it was still happily lighting it's candles. Two other identical wall warts, powering similar candles, but left on all the time, not switched with X10, were in like new condition. No evidence of any over heating.

Apparently the triac turn-on distortion of the LM15A was enough to cause the transformers to run slightly hotter than the plastic case could dissipate. I doubt there was ever enough heat developed to become incendiary. At least there was never any "burning tronics"  odor that we could detect. And for reasons I will not elaborate on, I am very sensitive to, and well practiced at identifying said odor's. 

I have replaced the two inductive wall warts with cheap cell phone chargers (over a year ago) and everthing seems happy.

But your sage advice of (paraphrasing): only use dimmer modules to drive loads specifically designed to be used with dimmers. When deviating from this rule, be sure you understand exactly what you are doing to prevent failure of either the module or the device it is powering."
Can not be over emphasized, to point of, even if you think it is OK, check it repeatedly for a while...

Finally, someone is going comment "Dave, only three post up this thread you rant about putting non-incandescent loads in a module labelled 'Incandescent Lights Only'  and now you confess you have done exactly that. How do you explain that?" 

u-h-h therapy(?).

[JeffVolp]

60Hz transformers can run fairly warm, especially if the manufacturer is cutting corners on wire turns and laminations to save a few pennies.  The LM15A triac probably doesn't switch on exactly symmetrically, so there could be some slight DC bias in the output waveform.  If the transformer is already running near its saturation limit (due to the forementioned reasons), a little bit of extra DC bias could have caused the core to saturate on one half of the AC waveform, resulting in higher than normal current.  Even if the saturation was not enough to make an audible hum, the plastic case would have made it hard for the transformer to dissipate the extra heat.  It would be interesting to test a well designed transformer from Stancor, Triad, Signal, or one of the other well-known brands to see if it exhibits a similar problem.  In fact, I think I'll do that myself out of curiosity.

Jeff

[dbemowsk]

I was running some of my outdoor pond lights with a lamp module using it simply as an on off switch, with no dimming.  The supply powering the lights was an AC corded plug in style transformer that had 2 - 24 volt receptacles for the lights.  This worked fine all last summer and part of this spring.  I had purchased a 120 LED light ring for my fountain and figured that a lamp module used in the same fashion as the other light would work fine.  This one had a 12 volt AC transformer wall wart style supply.  Within a day, I had lost the 2 transformers and the lamp modules.  Though the lamp module did not catch fire or appear to be overheated in any way, it did burn out both transformers primary coils along with the modules themselves. 

I have since rebuilt the 24 volt transformer and got a replacement wart for the LED light and put them both on appliance modules.  Lamp modules are not designed for inductive loads.

[Brian H]

Dan did you by chance verify they had transformers in them?
Some wall warts are real cheesy switching supplies now.

[JeffVolp]

I did some testing with both the LM15A and an ordinary lamp module (RCA version), and received some surprising results.  For a transformer I used a Triad 26.8VCT @1A F40X, so marginal transformer design should not be an issue.

With a 100-ohm load on the transformer, the transformer output voltage measured just over 30V, so that represented a 9W load.  A Kill-a-Watt read 9W, or 10VA.

With either the LM15A or RCA lamp module the waveform on the secondary was almost a pure sine wave with small glitches just after each zero crossing until the triac switched on.  The transformer emitted no hum, and there was no discernable heating of either unit after a few minutes of operation at that load.

Then I removed the load resistor to check the transformer by itself, and obtained very different results.  The waveform from the LM15A was still somewhat sinusoidal, but had severe distortion.  The transformer started to hum.  And the Kill-a-Watt reading jumped to about 150W.  I only left it in that configuration for a couple of seconds, so neither the LM15A nor the transformer had time to exhibit any significant heating.

The RCA lamp module acted differently.  With the purely inductive transformer load, the lamp module triac apparently did not remain switched on because the transformer output was just some high frequency ringing.  The Kill-a-Watt again displayed a high wattage reading, but I suspect that in both cases it was due to the corrupted waveform.

Note that for all testing with the lamp module, it was just switched on and off.  I used a RCA lamp module because it does not have the ramp on /off feature of the newer lamp modules.

Something to consider is that 120V variable speed power tools use essentially a dimmer designed to operate inductive loads.  X10, Leviton, and perhaps other manufacturers do make wall dimmers designed for inductive loads.  However, neither the LM15A nor the standard X10 lamp modules are designed to power inductive loads.

In summary, I totally agree that neither the LM15A nor a standard lamp module can be used for purely inductive loads.  However, when loaded properly, a good transformer can look almost resistive (as shown by the wattage and volt-amp reading being virtually the same).  As long as the asymmetry of the waveform does not cause core saturation, that configuration may be feasible.  The fact that our Grundfos hot water circulation pump motor has operated flawlessly for seven years gives some credibility to that hypothesis.

Jeff

[dbemowsk]

Dan did you by chance verify they had transformers in them?
Some wall warts are real cheesy switching supplies now.

Yes, these were definitely transformers.  Both transformers are AC output transformers, so I can't imagine that they would use a switching supply for that.  This is a picture of the 12 volt transformer.


The 24 volt one I had to take the case apart and found a green epoxied transformer in a brick with 4 wires coming out  I could actually see part of the transformer inside the epoxy.  I was able to get the epoxy loose from the plastic case and get it out.  I actually ended up replacing it with a 25 volt CT transformer from Radio Scrap and it works now.  It just ran 2 - 24 volt incandescent bulbs.  They just shine a little brighter now and I am guessing will not last quite as long, but hey, they work.

[Brian H]

Thanks for the information on it being an AC low voltage output.

[babybearjs]

anyone dumb enough NOT to read the warning label on this device should expect exactly what the get.... a CFL is NOT incandecant. Its INDUCTIVE! its concidered an APPLIANCE! PAY ATTENTION!

[sparkie951]

I think that I have found the fix for this... and it is very simple... There is a device called a "Triac". On the board it is at location MT1. Part number replacement is BTA10 fromn ST MicroElectronics and the parts are cheap. You can order them from Mouser Electronics for .99 cents each for a pack of 10 and real easy to replace.

[Brian H]

Replacing the triac isn't going to fix a users not following the warnings and using an improper load.  

Also many of the X10 Automation module owners. Will not have the technical skills to properly disassemble the module and replace the part.