LED lighting AFCI and ferrite chokers

Status
Not open for further replies.
C

Cutima

Member
Location
Philadelphia
Occupation
Teacher
Ive been trying to troubleshoot my constantly tripping CAGCI arc fault breakers (Siemens) with two electricians who can’t figure out the problem. (I Have new wiring and a new panel in the house, along with tripping on circuits that are literally one foot from the panel, so the wiring isn’t the issue. I am using 3 different generation Siemans breakers. All have tripped at some point.)

Through this forum, I think I’ve narrowed the problem to LED lights causing some type of interference in the panel that the arc faults are picking up on, even though they are on separate circuits. My arc faults only trip with a heavy load applied to them and with my LED lights on (which are different circuits).

My question is what is the current solution for this, outside of getting rid of LEDs throughout the house. I’ve read about Ferrite chokers, but heard they don’t work with switches that have dinners (almost all of mine do). Is there a brand of LEDs that work best for this scenario. Do can versus cableas LED matter? Any help is appreciated.
 
Cutima said:
Ive been trying to troubleshoot my constantly tripping CAGCI arc fault breakers (Siemens) with two electricians who can’t figure out the problem.
Click to expand...

Why don't you put the blame where it belongs? You're going crazy with LEDs and even considering replacing them. How about replacing those useless breakers with regular ones?

-Hal
 
They aren't smart dimmers. Lutron brand.

I'd love to use put the regular breakers, but my house is under warranty and I'm trying to get this all squared away now before I try to sell the house in a few years and the inspector tells me this is my problem. Unfortunately, these breakers are here to stay and only getting more ingrained in the code.
 
It's anecdotal as I've had no personal experience with the chokers but have had a few folks say they helped. Fairly inexpensive experiment
 
House is under warranty? Make it their worst headache. This is why we have this useless junk because people don't fight back. Many actually think they work and save lives. Don't be fooled! They only exist to make money for the manufacturers.

-Hal
 
hbiss said:
It's anecdotal as I've had no personal experience with the chokers but have had a few folks say they helped. Fairly inexpensive experiment
Click to expand...
I guess it's worth a shot. I came across this paper below from Lutron about their dimmers. But, if I'm reading this correctly, this would only matter if I had non LED bulbs that approached 1000 watts. My LED lights each are about 12 watts even though there are about 8 on a circuit.

 
The breakers are looking for a current waveform that it considers an 'arc signature'.

The first is you really confirm that this is a false trip. A slightly loose screw connection can cause enough of an arc signature to trip a breaker, and this is a real arcing situation detected by the breaker. Confirming this is a job for the electrician.

If you have reliable false trip data, eg a log of trip events listing what is on, demonstrating that the breaker only trips when the LEDs are on, and the same high loads (in the same receptacle) function without tripping when the LEDs are off...then IMHO the thing to do is to have the contractor push back with the breaker manufacturer. Very often the manufacturer will update the software in response to false tripping.

Jon
 
Thanks for your response. Just to be certain the LEDs and the tripping breakers are on different circuits. The LED circuit actually never trips, so I doubt it's anything loose with the wiring. It's 5 other circuits, all Arc Fault, all with appliances of a decent load (fridge, coffee pot, air fryer, microwave) on them that trip with the LEDs on.

I think we're going to try with a call to Seimens next week and see if they're any help, but I'm afraid they might default to the LED manufacturers or even the dimmer manufacturers.
 
The heavily loaded circuits _never_ trip with the LEDs off?

Is there a possibility that the tripping is time of day dependent rather than LED dependent, and just happening at a time of day that the LEDs were usually on?

The LED circuits have been checked for tight connections?
 
winnie said:
The heavily loaded circuits _never_ trip with the LEDs off?

Is there a possibility that the tripping is time of day dependent rather than LED dependent, and just happening at a time of day that the LEDs were usually on?

The LED circuits have been checked for tight connections?
Click to expand...
Not the first time someone has brought up AFCI trips that are triggered by some other circuit here.

I'm one of those not really for AFCI's myself but kind of always wondered if using ferrite choke essentially reduces or eliminates what AFCI is supposedly trying to detect. These are win for the manufacturers, a pain for end user and a loss for the installers* and in the end possibly don't really accomplish much.

*Loss because they are the ones stuck between owner complaints and code requiring them. When an owner has a problem you often have to figure things out at your own costs. Nobody is reimbursing you. Sure a manufacturer may at times give you updated replacements, but won't pay you anything for your efforts.
 
winnie said:
The breakers are looking for a current waveform that it considers an 'arc signature'.

The first is you really confirm that this is a false trip. A slightly loose screw connection can cause enough of an arc signature to trip a breaker, and this is a real arcing situation detected by the breaker. Confirming this is a job for the electrician.

If you have reliable false trip data, eg a log of trip events listing what is on, demonstrating that the breaker only trips when the LEDs are on, and the same high loads (in the same receptacle) function without tripping when the LEDs are off...then IMHO the thing to do is to have the contractor push back with the breaker manufacturer. Very often the manufacturer will update the software in response to false tripping.

Jon
Click to expand...
There was a video of an engineer showing him just touching/striking the screw terminal of a receptacle with the branch wires with a space heater plugged in. The receptacle was fed from an AFCI breaker and it never tripped. He did that with different brands of AFCIs. This would negate the "loose connection" theory.

I tried to find the video but couldn't. It was posted here before but can't find that either.

Anyone have a link to that video?
 
Cutima said:
... My question is what is the current solution for this, outside of getting rid of LEDs throughout the house. I’ve read about Ferrite chokers, but heard they don’t work with switches that have dinners (almost all of mine do).
Click to expand...

Do the LED llights on circuits with regular switches cause the AFCIs to trip, or is the tripping only when the dimmers are used?

If the tripping is only when usings dimmers, then you might try a dimmer that can provide "reverse phase" operation instead of the more common "forward phase". Legrand Tru-Umiversal is such a dimmer. The reason I mention this is that LED lights often present capacitance on their supply input, and the fast rising edge of the voltage from forward phase dimmers can cause current spikes with capacitive loads. And maybe such spikes are causing the tripping. This is just a hunch but perhaps worth a try.

leds-deepdiveindimming-02-tcm47-2138820.jpg
 
Little Bill said:
There was a video of an engineer showing him just touching/striking the screw terminal of a receptacle with the branch wires with a space heater plugged in. The receptacle was fed from an AFCI breaker and it never tripped. He did that with different brands of AFCIs. This would negate the "loose connection" theory.
Click to expand...

I don't have the link but have seen the video.

This doesn't negate the loose connection possibly, but does limit the sensitivity threshold.

The problem with AFCIs is that 'normal' operation and 'arcing' operation current waveforms are both very broad ranges, and IMHO overlapping. So you will always have a tradeoff between false positives and false negatives.

IMHO you will either have AFCIs that trip on normal operation or that fail to detect real fault arcing. But that doesn't negate the reality that they detect most real arcing, and doesn't negate that the OP has to rule out real arcing.

Jon
 
I have heard short runs on an afci will cause nuisance tripping. Have your electrician try adding a couple of feet of wire in the panel before it goes to the breaker.
 
winnie said:
The heavily loaded circuits _never_ trip with the LEDs off?

Is there a possibility that the tripping is time of day dependent rather than LED dependent, and just happening at a time of day that the LEDs were usually on?

The LED circuits have been checked for tight connections?
Click to expand...
I've been keeping my LEDs off in my house for 6 days now and no trips. I've never gone more than 2 days without a trip before. It's definitely not time dependent. Happens at all hours. I haven't ruled out LED circuit connections. I should also add I have flickering of lights on a ceiling fan with normal bulbs, which is on the same circuit as one of my LEDs. It doesn't cause anything to trip, but the flickering is odd, considering it doesn't coincide with the AC turning on or any other large load starting in the house.

synchro said:
Do the LED llights on circuits with regular switches cause the AFCIs to trip, or is the tripping only when the dimmers are used?

If the tripping is only when usings dimmers, then you might try a dimmer that can provide "reverse phase" operation instead of the more common "forward phase". Legrand Tru-Umiversal is such a dimmer. The reason I mention this is that LED lights often present capacitance on their supply input, and the fast rising edge of the voltage from forward phase dimmers can cause current spikes with capacitive loads. And maybe such spikes are causing the tripping. This is just a hunch but perhaps worth a try.

View attachment 2567285
Click to expand...
This is an interesting theory. Today, I'm going to start using the LEDs in the house that don't have dimmers attached to them and see if they hold. I know Lutron has had issues in the past with their dimmers replicating the arc. I'm going to "run my experiment" with the dimmers over the next couple of days and see if that helps. Thanks for the suggestion
winnie said:
I don't have the link but have seen the video.

This doesn't negate the loose connection possibly, but does limit the sensitivity threshold.

The problem with AFCIs is that 'normal' operation and 'arcing' operation current waveforms are both very broad ranges, and IMHO overlapping. So you will always have a tradeoff between false positives and false negatives.

IMHO you will either have AFCIs that trip on normal operation or that fail to detect real fault arcing. But that doesn't negate the reality that they detect most real arcing, and doesn't negate that the OP has to rule out real arcing.

Jon
Click to expand...
My electrician hasn't checked every wire yet, but he did change the panel and add a new circuit breaker to an empty spot on the panel and run a a wire with a GFCI receptacle right outside the panel. That receptacle is one of them that trips pretty often and it's got about a foot and a half of brand new wire to it and brand new everything to it. Nothing loose or old there. Wish that were the case, but I still have to have him look at the LED light switch because a ceiling fan on the same switch often flickers. Thanks for the suggestions
 
So a gfi receptacle is tripping too? Have your electrician check for voltage between the neutral and ground at that panel, then check resistance between neutral and ground. I have a feeling you may have a grounding/bonding issue, along with a fault somewhere.
 
The flickering of lights and the tripping of both a GFCI receptacle and AFCI breakers suggests that you might have a poor connection to the service neutral in the panel. That could allow the nominally 120V on branch circuits to vary up and down as the loads on the panel change. If the voltage gets high enough, even intermittently, it could trip AFCI or GFCI devices.

I suggest having your electrician check the line to neutral voltages when a substantial 120V load is applied. With a poor connection to the service neutral, when the loading on one of the phases causes its L-N voltage to drop, the voltage on the other phase will increase correspondingly.

When the panel was replaced the service neutral connections had to be re-established, and so that would have provided an opportunity for a problematic connection.
 
The resistance of a poor connection to the service neutral would also allow the noisy and distorted current from dimmers, etc. to induce a noise voltage on the neutral bus in the panel. That noise voltage could then modulate the currents feeding loads on AFCI breakers in the panel, which could then trip trip them if the load current exceeds the 5A threshold spec for series arc detection.
 
Status
Not open for further replies.
Top