AFCI - Lightning trouble

[shockin]

I have a customer who has 3 AFCI breakers (C.H.-BR as I recall) installed in his panel. He claims that during the last few lightning storms one of the breakers (always the same) keeps tripping. We wired the house last year, so I am "assuming" that it has installed corectly and the grounding is properly done. Any thoughts? I thinking replace the breaker and see what happens.

 

[al hildenbrand]

Always the same AFCI breaker tripping. . .during lightning storms.

With lightning comes thunder, and, depending upon the intensity of the strike and its distance from the house, the thunder clap can offer a structure rattling impulse. I wouldn't rule out neutral - EGC contact.

 

[Kal-EL]

You could check the AFCI by swaping it with another and see which one tripps. If it still tripps the same circuit than there might be some form of damage to the insulation of one of the conductors.

 

[shockin]

Yes, it's always the master bedroom breaker.

The thunder theory / loose neutral is an interesting one. It's as plausable as anything else.

I have ruled out any unusual loads because it often happens at night. It can't be related to the phone/CATV either because there is a fiber optic line that comes all the way to each residence.

My only thought was either 1.)bad breaker? or 2.) surge realted to the lightning. I'm not sure how an AFCI functions or if that would affect it, and why it dosen't affect the other two but I reaching.

BTW thanks for the responce

 

[al hildenbrand]

I'm not sure how an AFCI functions or if that would affect it, and why it doesn't affect the other two. . .

An AFCI has several features. With respect to your problem, here is a list of the ones that relate, in my opinion.

Part of the circuitry behaves like a GFI with a trip of around 50 milliamps (not 5.0 as in a standard GFCI). To turn on and stay on, the AFCI circuitry has to see a load side neutral that has no contact with the EGC. Any accidental contact will trip the breaker.

Another part of the AFCI circuitry is a signal processor that is only turned on when it sees a signal that exceeds 50 to 70 amps. For the Branch / Feeder AFCI (that you would have installed in this Master Bed circuit a year ago) the signal processor then determines if the waveform is a good or bad arc. If the circuitry sees this "bad" high current spike in a half cycle, and then sees another within eight half cycles, the signal processor tells the breaker to trip.

Lightning creates an electromagnetic pulse (EMP) that can induce significant currents and voltages in conductors, even at a distance. It's easy to suspect this, but it will be extremely difficult to document.

This gets to the real issue surrounding AFCI troubleshooting, in my opinion. An AFCI has a number of things that cause it to trip, but, aside from the trip itself, the AFCI keeps secret what AFCI function actually initiated the trip.

As a result of this secrecy, the troubleshooting electrician must eliminate trip causes by showing what couldn't have caused the trip. One has to show that the branch circuit is, in fact, free of defects. Eliminating the branch circuit allows pointing a finger at the utilization equipment connected to the branch circuit at the outlets (lighting or receptacle or smokes), OR, at the breaker itself.

The first, least time involvement, method will be to replace the AFCI with another, of known condition, and wait.

 

[bphgravity]

I seiously doubt there is any relationship between the breaker tripping and lightning. It may appear that way by coincidence, but if one were to pay close attention to nearly any event, similar conclusions could be made about completely non-related phenomenon.

I am leaning toward a defective device. If that doesn'y hold up, I would check for cord-and-plug connected loads that may causing the problem, and if all else fails start looking at the wiring and connections within boxes and outlets.

 

[wwhitney]

Part of the circuitry behaves like a GFI with a trip of around 50 milliamps (not 5.0 as in a standard GFCI). To turn on and stay on, the AFCI circuitry has to see a load side neutral that has no contact with the EGC. Any accidental contact will trip the breaker.

Does an AFCI inject a small current into the circuit to detect EGC-neutral interconnection without any load, like a current GFCI does? Or will it only trip once there is a load on the circuit, like very old GFCIs?

Thanks, Wayne

 

[mdshunk]

Any thoughts? I thinking replace the breaker and see what happens.

You could try that, but CH makes famously good AFCI's. I would suspect a defective AFCI very last among the possibilities. I'd rather advise you to lift the hot, neutral, and ground of that circuit's cable in the panel and ring them out with a megger.

 

[al hildenbrand]

Does an AFCI inject a small current into the circuit to detect EGC-neutral interconnection without any load, like a current GFCI does? Or will it only trip once there is a load on the circuit, like very old GFCIs?

That's a great question.

So far, to my knowledge, manufacturers of AFCIs have stone walled the installers of their product, when presented requests for technical information on the operation of the as-manufactured-AFCI-circuitry.

I have not yet seen a manufacturer response to your question.

 

[mdshunk]

That's a great question.

And the answer is, no. The AFCI will ony trip on a hot to neutral short under load. A three light neon receptacle tester is not a sufficient enough load, but just about anything else is.

 

[BryKey]

And the answer is, no. The AFCI will ony trip on a hot to neutral short under load.

Not making waves but I have had to troubleshoot AFCI trips on several occasions and found on one it was neutral to ground contact that was the cause. The "installer" (an Electrician would not have done this) did nothing to the neutral in the ceiling fan box, no wire nut, and the end of the wire was against the side of the metal box. Once I placed a wirenut on the neutral there was no more tripping.

 

[tom baker]

This gets to the real issue surrounding AFCI troubleshooting, in my opinion. An AFCI has a number of things that cause it to trip, but, aside from the trip itself, the AFCI keeps secret what AFCI function actually initiated the trip.

.

The Siemens AFCI has LEDs for Arc Fault and Ground Fault.

 

[220/221]

I seiously doubt there is any relationship between the breaker tripping and lightning.

I don't know how ANY of this stuff works but after I read about Motorola walkie talkies tripping GFCI's and after I witnessed a neighborhood lightning strike 1/4 mile away kill several LV dimmers in my house, I don't doubt it a bit.

Electricity is funny. Not "ha ha" funny but wtf funny.

 

[nakulak]

I am of the opinion that anytime new electronics are added to any device (I repeat any device), it takes years before the device is fully functional and troubleshooted enough for ordinary use (I extend this theorem to software as well, as I am still happily using win2k and xp while others around me curse the new unproven, untested crap)

This quirk comes from all the unreliable crap I have tried to use in the past, from gfi's to electronic ballasts, etc etc. It continues to amaze me that manufacturers will even think of pushing unproven technology, and it amazes me even more that they want to charge exorbitent prices, and it baffles me that people will pay it. It would make boat loads more sense if they offered the latest technology at discount prices, while offering the proven technology at a higher price.

 

[e57]

I am of the opinion that anytime new electronics are added to any device (I repeat any device), it takes years before the device is fully functional and troubleshooted enough for ordinary use

Tend to agree - AFCI's the GFCI's of only a few years back, LED lighting, dimmable fluorescent ballasts and dimmers... The list could go on.

 

[al hildenbrand]

The Siemens AFCI has LEDs for Arc Fault and Ground Fault.

Thanks Tom, yes. The OP, however, is working on branch / feeder AFCI protected circuits. The branch /feeders don't give any indication of what operated the trip.

 

[weressl]

You could try that, but CH makes famously good AFCI's. I would suspect a defective AFCI very last among the possibilities. I'd rather advise you to lift the hot, neutral, and ground of that circuit's cable in the panel and ring them out with a megger.

AFCI's are fairly newly developed products. All Vendors were eager to jump on the bandwagon, realizing that there is some easy money to be made. So I would not discount that the product does have some flaws in it when it hits the street.

Since products became more complex and capable through the employment of microprocessors, it bought along a host of problems. On is the infamous software 'bug'.

This and the market pressure results in less well researched and prepared products to be out there. In the direct consumer market there is enough advocacy so the Manufacturer is compelled to disclose faults and initiate recalls. This happens to a lesser degree in the Commercial and Industrial market.

There were countless events and occasion where we be troubleshooting some phenomena, studying the manufacturers? instructions and troubleshooting guide forward and backward. Finally at the end of our rope we would call up the manufacturing engineers, only to be informed that they are aware of the problem, have no clue what is the cause, take two aspirins and call me in the morning.

I don't want to single out a manufacturer but this happened with companies we have been doing $10M's business.

 

[mdshunk]

AFCI's are fairly newly developed products.

Thank your for your treatise on AFCI's. I have to ask, though; Have you ever even held an AFCI in your hand? Sounds like a bunch of gibber-jabber from an old man trying to stay relevant. I could be wrong. You might not be old.

 

[weressl]

Thank your for your treatise on AFCI's. I have to ask, though; Have you ever even held an AFCI in your hand? Sounds like a bunch of gibber-jabber from an old man trying to stay relevant. I could be wrong. You might not be old.

Why don't you read your own footnote. Oh, 'the mote in thy neighbor...'

BTW I actively participated in the Eaton product development as user advisor.

 

[mdshunk]

....sounds like somebody needs a hug.