UV light + electric current = boom?

[Rive]

My idea about this case was that the filter circuit (cap and resistor in parallel with the dimmer) caused the sparks: especially the resistor (most likely some 1-2W cement type) which blown up after the cap started to conduct due the UV.

Prerequisite is that the cap should be some transparent, foil type and has reaction to UV.

I don't really know what kind of equipment do you have at hand. What I'm curious about is the behavior of (transparent) foil caps under UV light and high voltage (stay safe, please).

Ps.: a failure in the filter circuit would be also consistent with the fact that the stuff is still working.

 

[The Electrician]

My idea about this case was that the filter circuit (cap and resistor in parallel with the dimmer) caused the sparks: especially the resistor (most likely some 1-2W cement type) which blown up after the cap started to conduct due the UV.

Prerequisite is that the cap should be some transparent, foil type and has reaction to UV.

I don't really know what kind of equipment do you have at hand. What I'm curious about is the behavior of (transparent) foil caps under UV light and high voltage (stay safe, please).

Ps.: a failure in the filter circuit would be also consistent with the fact that the stuff is still working.

By "high voltage" do you mean anything other than the 120 VAC from the wall outlet?

 

[Rive]

By "high voltage" do you mean anything other than the 120 VAC from the wall outlet?

The best would be with 120VAC, but that's also the most dangerous, so no way I'll ask for that directly you know...

 

[The Electrician]

The best would be with 120VAC, but that's also the most dangerous, so no way I'll ask for that directly you know...

My EE degree plus decades of experience qualify me to do this test. I didn't bother saying it earlier because it didn't occur to me that anyone else would be likely to bother repeating the experiment, but I used a 1:1 isolation transformer for my own safety, and I don't recommend that anyone else do this.

Here is the stash of miscellaneous film capacitors that I've accumulated over the years. There aren't many transparent/translucent capacitors in there:

If there were an R/C series pair connected across the triac, I would suppose that a good designer would use an "X" capacitor for that purpose. I found a typical X capacitor which is also transparent:

You can see all the agency approvals on this capacitor (made by RIFA), but I have no way of knowing if the encapsulant is transparent to UV. I'm not going to create another video, so I will only report my results. I connected this capacitor across the gap in the line cord shown in the earlier video and played the UV from the flashlight on the capacitor from every angle. I got so close that the capacitor was touching the window in the end of the flashlight. There was no sign of any action from the light bulb.

I found two other transparent capacitors, a polystyrene cap and a polycarbonate cap:

I repeated the experiment with each of those, with the same negative result.

Later today I will try measuring the insulation resistance of these three capacitors with DC using my Fluke meter on conductance range. This can measure insulation resistance up to the gigaohm range, but with only a few volts applied. I will also apply about 50 VDC from a power supply and measure leakage current. Both these measurements will be performed with and without exposure to the UV.

 

[Rive]

I repeated the experiment with each of those, with the same negative result.

Thank you very much, that more or less settles it: it was also a dead end then.

 

[jim hardy]

If i could like a post multiple times i would like Electrician's 199X .

"One experiment is worth a thousand expert opinions"

I'm back to 'Open the fixture and look for smoke residue' .

old jim

 

[The Electrician]

If i could like a post multiple times i would like Electrician's 199X .

I'm back to 'Open the fixture and look for smoke residue' .

old jim

It's hard to imagine how very much UV could penetrate the box anyway, especially if there's only 5 mW from the pen to start with.

You're right on; a look inside is what's needed.

 

[darkwood]

So I guess we go back to the lying little lad scenario which with 30yrs experience in this trade was my first thought and most likely scenario; with a gullible father to boot

 

[The Electrician]

I followed up with a measurement of the 3 transparent/translucent capacitors' leakage as a measurement of insulation resistance, and as an indicator of internal photoelectric effects. With 50 volts DC applied to a given capacitor and with a DVM set to the microampere range in series, I shined the UV flashlight at very close range on the capacitor under test.

WIth the polystyrene and polycarbonate capacitors in circuit the leakage current without UV illumination was less than 10 nanoamperes. With UV illumination from the 750 milliwatt flashllight there was no detectable increase in leakage.

The RIFA X capacitor exhibited a leakage of about 270 nanoamperes without UV illumination, and showed no detectable increase with UV illumination.

I was hoping that there would be some detectable effect from the UV illumination, but no joy.

 

[Rive]

I was hoping that there would be some detectable effect from the UV illumination, but no joy.

Sigh... A single good experiment can massacre any amount of ideas
Thank you for following this up.

I know that some old potentiometers can produce fine graphite (or something like graphite) dust. Maybe that, on a surface? That equipment seemed to be quite old...
But it's just some afterthought. I agree with Jim Hardy that it's already hopeless without dissecting that box.

 

[Guineafowl]

Sigh... A single good experiment can massacre any amount of ideas
Thank you for following this up.

I know that some old potentiometers can produce fine graphite (or something like graphite) dust. Maybe that, on a surface? That equipment seemed to be quite old...
But it's just some afterthought. I agree with Jim Hardy that it's already hopeless without dissecting that box.

I’ve been watching this thread on and off and wondering why the box hasn’t been post-mortemed, with pictures. Lots of fruitful hypothesising and experimentation going on, but surely the answer lies within?! It’s a light switch, not Schroedinger’s box.

 

[jim hardy]

It’s a light switch, not Schroedinger’s box.

Laugh out loud on that one Guineafowl ! Dead cat in there?

Boring Anecdote
My old Dodge van refused to start one morning. Start relay on the firewall sounded weak.
I removed the starter relay, pried off the top and found a huge red ant squashed between the contacts.
Those big ants had a nest in the palm tree adjacent my driveway. I wondered what on Earth made that silly ant crawl up under the hood and into that relay.
Well ! Next morning same symptom exactly .
I figured no way could two ants be so crazy but sure enough i pulled another squashed ant out of the same contact.
Why they preferred the starter relay over the horn relay immediately adjacent only those ants knew.
Anyhow, at this rate in a few more days the little tabs on the cover that you must bend to get the top off would fatigue and break... so i decided to make the relay less palatable to them.
Sprayed the inside of the lid with Black Flag home insecticide and reinstalled. No more ant incursions..
old jim

.. .

 

[the_emi_guy]

I’ve been watching this thread on and off and wondering why the box hasn’t been post-mortemed, with pictures. Lots of fruitful hypothesising and experimentation going on, but surely the answer lies within?! It’s a light switch, not Schroedinger’s box.

If the OP is uncomfortable opening the box, an electrician should be called. I know if I had sparks flying out of my dimmers for unknown reasons I would not delay in fixing it or getting it fixed. I would be afraid that I would come home from work and find my house burned down.