Print

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

[Brian H]

Not too clear to me.
Did you try it with only the 75W incandescent bulb and no transformer connected?

On but not Off is frequently the load making noise and stopping the X10 signals from getting through.

Another thought. Is the 75W bulb calmed down the bad interaction enough for it to be just stable but still making it difficult to control.

You may now have to start looking for power line noise makers and signal suckers.
Phase coupling between the two incoming power lines that are part of most split phase wiring. As X10 power line signals have a very poor time getting from one line to the other line. Though X10 is not unique in signal coupling between incoming lines being an issue.

Jeff has a great set of X10 Troubleshooting Tutorials.
http://jvde.us/x10_troubleshooting.htm

[CliffyJ]

The setup seems to still be working perfectly, with the 75w bulb and transformer (having a 4w load) in parallel.

My next test is to see how many watts (at 12v) I need to load the transformer with, before I can take off the 75w bulb.

[edit]

I steadily increased the number of 12v (multi-chip LED) bulbs fed form the transformer, while fiddling with load bulbs on the 120v side. Only with 75w (20 bulbs @ 12v) did things really stabilize. At that point, I could have a 120v load bulb on, switch it off, back on, breaker didn't trip and almost no hum.

I wonder, if I'd gotten a 100w transformer (vs a 200), would the performance have been better? Don't know.

Anyway, I plan to have more than 75w loading this transformer, and I'm glad that things seem to be ok with that. No issues with the palmpad either.
 

[Brian H]

Thank you for the update.

[bkenobi]

My motion sensor made by Heath-Zenith requires a 40W load.  I didn't know that when I originally installed it, so I was going mad trying to understand why it worked when controlling a bulb but not when I used a 4W mini-bulb as an indicator.  After talking to tech support, they advised using a 40W bulb and all is good.  I don't understand why these type of minimum wattage requirements aren't listed in the manual!   

[Brian H]

I looked at the XPDI3 Manual.
Says 40W-500W incandescent and 4A max inductive.
The 12V LED bulbs maybe a different requirement. As the load is being reflected back to the 120 volt primary from the 12 volt secondary.

[bkenobi]

If the module requires 40W+ but it's a dimmer, then how does that work when dimmed?  Doesn't that mean that there wouldn't be anywhere near 40W when dimmed to the lowest setting?  Obviously it must work, but I don't understand how 40W could be a minimum if it could be more like 2W (assuming there are 20 steps).

[dave w]

The forty watt bulb has a lower resistance than say a 25 watt bulb. The resistance of the load (or the current through the triac junction from anode 1 to anode 2) is what ensures saturation of the junction and full turn on ("latch") of the triac, not the gate drive. This allows the triac to maintain conduction even when the gate drive drops out.

A triac is two SCRs (thyristors) back to back, so the three paragraphs under the "Turn On" heading on page one of this link explains better than I can. 

http://www.nxp.com/documents/application_note/AN_GOLDEN_RULES.pdf

[CliffyJ]

One more experimental thing to point out: that with the 75w load on te transformer, the lowly LM465 seems to work just as well as the XPDI3. Go figure.

[Brian H]

Since an LM465 is not designed for an inductive load. Even with the 75 watt load on it.
Long term reliability is not guaranteed and it could have a short life.

[joe s.]

Be careful - I fried a LED (XMAS Bliss Lites) power supply with LM465 (my typical..."it must be OK, 'cause its working" approach).  And I wasn't even dimming it, ever - just on/off.  But that particular transformer hated the triac and folded within a week or so.  I use only appliance/relay types with them now.

[CliffyJ]

Thanks for the correction Brian and Joe, I'll not use the LM465 for this.

[dave w]

One more experimental thing to point out: that with the 75w load on te transformer, the lowly LM465 seems to work just as well as the XPDI3. Go figure.

The heavy resistance load of the 75W bulb helps reduce the phase shift normally caused by the large inductive load.

The eight amp transformer I was driving with a GE wall switch dimmer did not seem to be bothered by the waveform distortion of the triac dimmer. But years later I used a LM15A (http://www.amazon.com/X10-Socket-Rocket-Screw-In-Module-LM15A/dp/B000JK23VK) as a "solid state relay" to drive a couple of inductive 5V wall warts powering some flameless candles for accent lighting. The candles load was well below the capacity of each wall wart.

Although the LM15 can not be dimmed, just like a true "solid state relay",  it still had some turn on delay distortion. However everything "seemed" to work well.  The wall warts were out of plain sight and out of mind. A year later I pulled the wall warts out of their hiding place only to find the plastic cases had softened and melted all around the transformers and filter caps. The very tiny turn on delay distortion of the sine wave still caused the transformers to run hotter than designed and slowly met the cases.

I don't do that any more.

[CliffyJ]

Thanks for that further background Dave.

At the risk of  , one final question if that's ok.

The XPDI3 and LM465 (which I won't use) seem to control the dimmer & 75w led load in exactly the same way. That is, the steps seem the same, and at the last step there is a very slight hum in the transformer.

Question: how do I know the XPDI3 won't cause damage to the transformer (like the LM465 would) in the long run? Just take it on faith, that it's engineered for inductive loading?

Thanks,
Cliff

[Brian H]

The transformer is rated to be used on a dimmer and should be fine with the XPDI3. Unlike a wall wart that is not designed to be on a dimmer. Even at 100%.
I suspect the slight hum. Is the small off time when at the AC is crossing between the positive and negative part of the AC waveform.

[CliffyJ]

OK, thanks Brian.

BTW, the hum has decreased as I've applied more load to the 12v side of the transformer (I'm at 75w now). I expect that with the final design load of around 100w, the hum will be even less. But that's just a guess.

I'm afraid I don't have a frame of educated reference for the cause you described. But thanks for describing it anyway, I'm sure someone else will benefit.