Electronic Trany & GFCI Problem

[Dennis Alwon]

I am posting this for a friend of mine who is a licensed electrical contractor here in NC.

I got a call from him yesterday because he had removed a fixture in a house that is about 4 years old. In it's place he installed a brand new low voltage fixture with electronic transformer. Everything fine ----he then goes home.

Call comes in later that the fixture is tripping 3 GFI receptacles. They are all on different circuits then the light. For instance, the garage GFCI, the Kitchen and the bath GFCI's all tripped when the light is turned on.

Now the interesting part. He cannot duplicate the problem. The HO returns home and says it only happens when the fixture is off for 20 minutes or so. They wait 20 minutes and sure enough the GFCI trips when he throws the switch to the light. Reset the GFCI and there is no problem turning the switch on and off many times.

I had him put the original fixture back to see if it would do it again. No problem with the old fixture.

Now when my friend was there he could only get one GFCI to trip so I am not sure why 3 tripped for the HO. It may be it tripped one GFCI then the next time it tripped the next one and so on till the HO noticed the tripped receptacles. This I am not sure about.

The GFCI recep. that are tripping are not all on the same phases of the panel.

They will replace the fixture -- not sure but hoping this will solve the problem, but I am curious as to how it is possible for this to occur. Remember the light is on a different circuit not downstream of any GFCI.

I had thought that somehow there was a shared neutral but apparently that is not the case esp. since we have 4 circuits involved here.

Any theory explanations ?????

 

[ELA]

Sounds similar to this thread.

http://forums.mikeholt.com/showthread.php?t=89668

I have experienced GFCIs tripping due to noise. Noise can capacitively couple between legs of the service.

 

[quogueelectric]

low voltage xformers

low voltage xformers

Maybe inrush magnetizing current is causing voltage drop on the neutral enough to trip common neutral to gfciS??

 

[Dennis Alwon]

Maybe inrush magnetizing current is causing voltage drop on the neutral enough to trip common neutral to gfciS??

Apparently there is no common neutral. It is a totally separate circuit. My only thought was noise from the trany but I can't imagine how it can trip a gfi on another circuit. We will see if the new fixture solves the problem.

 

[iwire]

Seeing as I was not sleeping I actually spent some time trying to find any info I could about GFCIs and 'noise' or 'Harmonics'

The best I could find was an older NFPA digest saying that UL has not investigated how GFCIs and harmonics get along.

To a GFCI current is current and harmonic currents will show up on both circuit conductors running through the GFCI not just the neutral so the GFCI will see no imbalance.

 

[Dennis Alwon]

Seeing as I was not sleeping I actually spent some time trying to find any info I could about GFCIs and 'noise' or 'Harmonics'

The best I could find was an older NFPA digest saying that UL has not investigated how GFCIs and harmonics get along.

To a GFCI current is current and harmonic currents will show up on both circuit conductors running through the GFCI not just the neutral so the GFCI will see no imbalance.

Thank you for your efforts, I appreciate the homework you did. I had no idea where to look.

Well then something is causing it to trip that is related to the light. What it is may be a mystery. Maybe I can talk the EC to let me come over and trouble shoot. I love this stuff even if I don't understand the theory. I can usually find what's causing the problem but not understand why.

It's hard being half smart.:grin:

 

[iwire]

Maybe I can talk the EC to let me come over and trouble shoot. I love this stuff even if I don't understand the theory. I can usually find what's causing the problem but not understand why

It's an interesting problem and I would love to know the cause of this problem and the other thread about the Air Conditioner LCDI tripping when a light switch is thrown.

Keep us posted.

 

[iwire]

Maybe inrush magnetizing current is causing voltage drop on the neutral enough to trip common neutral to gfciS??

How would voltage drop trip a GFCI?

 

[quogueelectric]

gfci

gfci

How would voltage drop trip a GFCI?

over 30 yrs I have learned some basic properties of gfci that I dont nececarily have to completely understand but know that it will work. many of the jobs I am on are large and by design time is money and money is time. I have learned that when I have to splice into a live temp ckt that if I prep the wires and short the ground to neutral the gfci will trip if wired properly this will give me enough time to tie in my ckt before anyone even knows the ckt is off. My theory is that whatever trips the grci whether inductive resistive or capacitance current when you ground the neutral may do the same when a high inrush current drives up the voltage drop on a poorly grounded neutral. Remember that Ohms law applies to both the load and the source and you have to look at what point in the ckt you are exploring and how it effects both the source and the load. With the information given I can only see a high magnetizing current on the lv lights causing a brief surge/voltage drop on the neutral AND the hot conductors. Yet there is some sort of monitoring between the groung and neutral which is my gut feeling. Shoot me now or later????

 

[quogueelectric]

Gfci

Gfci

I am also a fan of the colorblind electrician can cause all kinds of mysterious problems with gfci devices. Usually only evident with high voltage lighting yet you can see the havock associated with swaping a grey with a green on a high voltage lighting ckt. Major problemo. The reason I am so aware of this is I myself am colorblind and see this on a regular basis. Some foreman or owners dont hear you when you tell them you are colorblind and everyone is different in thier degrees of colorblindness.

 

[CAPS]

Dennis,
Just curious, do the GFCI recepts. have a load drawing current when they trip, just have an appliance plugged in, or are they open circuited with nothing plugged in.

 

[Dennis Alwon]

Dennis,
Just curious, do the GFCI recepts. have a load drawing current when they trip, just have an appliance plugged in, or are they open circuited with nothing plugged in.

I would think they are open since the kitchen & bathrooms are 2 of the 3 circuits affected. The kitchen I would guess has nothing but the bathroom and the garage may have a night lite and something else plugged in, not sure. Like I said I have not even been to the job.

Curious as to why that info is helpful? What are you thinking, please share.

 

[quogueelectric]

This is such a freaky trip

This is such a freaky trip

Maybe someone sent a screw into a wire on a grab bar or something like that you say this is tied into a bathroom ckt ?? is every thing line to line or is anything fed from the load side of the gfcis??. There has got to be a solution there always is and it is usually simple.

 

[Dennis Alwon]

Maybe someone sent a screw into a wire on a grab bar or something like that you say this is tied into a bathroom ckt ?? is every thing line to line or is anything fed from the load side of the gfcis??. There has got to be a solution there always is and it is usually simple.

The bathroom, kitchen and garage GFCI's are all on separtate circuits from the problem lighting circuit. Therefore there is no downstream GFCI on the light circuit.

 

[Pierre C Belarge]

Dennis
I would get the info from your friend as to the make of the GFCI.
Also the make of the fixture and the info off the ballast. I would send all of that info to the manufacturer of the GFCI. They may already have someone else experience this type of situation and have an answer for you.
I am sure they would be as interested in this as we are.

 

[Dennis Alwon]

Dennis
I would get the info from your friend as to the make of the GFCI.
Also the make of the fixture and the info off the ballast. I would send all of that info to the manufacturer of the GFCI. They may already have someone else experience this type of situation and have an answer for you.
I am sure they would be as interested in this as we are.

I will try and get that info to them and you all as well.

 

[CAPS]

Dennis,
I had two IEEE articles to upload, but they were too big. Both about false tripping of GFCI's. They discussed switching transient surges, and capacitive coupling to the ground (10ma possible?). If the GFCI's are an older style they may not have proper filters and MOV placements to shield them from the 20-50kHz switching noise generated by the electronic ballast. My reason for asking about the loading of the GFCI was my thinking that a load may allow the switching traisient to be eliminated through the load rather than across the MOV in the GFCI.

The suggestion to contact the GCFI manuf. is good if they are new. Otherwise, maybe they need to be upgraded along with the light fixture.

Downward compatibility is rarely considered, they always want you to buy new. It keeps the market operating.

 

[Jack Maynard]

gfci--- tripping by ground to neutral connection

gfci--- tripping by ground to neutral connection

A gfci has an electronic comparator that senses the current on the hot and the neutral. When there is a diffference of more than somewhere between 3 and 6 milliamps the comparator trips the device. So when you cross the ground and neutral the current on the neutral is reduced, as some now travels on the ground, and the comparator sees that diffference--that is there is now more or less double current flow on the hot as compared to the neutral-- and the device (or gfci breaker) trips.

But here is the catch.... try this once when there is no load on the circuit downstream of the gfci. It won't trip. So don't count on the circuit being dead when you do this and are working remotely downstream on a gfci protected circuit. I've seen a couple guys blow up perfectly good strippers this way.

 

[Dennis Alwon]

Update

Update

Here is the latest update I have on this situation. Apparently the owner wasn't giving the EC all the info-- by that I mean he kept saying the GFCI's were all tripping. Apparently one would trip and the next time he turned the lights on another would trip and so on. When he finally noticed that a gfci was tripped he found all 3 tripped. That didn't seem to all trip at the same time.

My friend had changed 2 of the GFCI's that were tripping and the owner kept saying they were still tripping when in fact the one that was tripping was one of the old one.

To make a long story short--- he eventually changed all the GFCI (they were only 3 years old) to the new generation GFCI's and everything appears to be working. Either that or the HO called another electrician :grin: