Multiple arc fault breakers tripping

[CSElectricCO]

A customer has an existing 200 amp Eaton BR panel that I added a 125 amp sub panel to. The sub panel is drawing about 75 amps. Problem is, ever since the sub has been installed, it will randomly cause one of the existing arc fault breakers to trip. After one trips, most of the other existing arc faults start to trip one by one. Never ran into this before. Could it be a heat issue or a problem with these type of breakers? Not overly hot in there...

 

[Buck Parrish]

What does the customer do to get the power back on to those circuits?
Did you see it?
Is it possible the customer is embellishing what's actually happening ?

 

[ptonsparky]

Time to go back and see what is actually happening. It may not be your fault, but you’re getting blamed.

Did you touch the existing AFCI circuits at all, in any way shape or form?

 

[CSElectricCO]

I did see it happen for myself. To reset the breakers, first you have to shut the sub panel breaker down. That's the only way they will hold for any length of time. Then eventually one will trip followed by multiple others. All the arc faults are next to each other, 10 in total

 

[CSElectricCO]

I did not mess with any existing circuits

 

[ptonsparky]

Being next to each other doesn’t matter unless you have the neutral from one going to the other and that to the other wrong one, etc.

Did you move the the AFCIs to your sub?

Add: ok I see you didn’t mess with the existing

 

[CSElectricCO]

The breaker that trips first is always the same one so I'll try replacing it and see what happens. Just hoping somebody else had seen this before or is this just my luck?

 

[GoldDigger]

The breaker that trips first is always the same one so I'll try replacing it and see what happens. Just hoping somebody else had seen this before or is this just my luck?

We have seen descriptions where a single circuit, not necessarily even on an AFCI breaker, will generate a waveform that matches a series arc signature and leaks that signal back through the panel bus to the input side of some AFCI breakers which will recognize it as a series arc.
But the AFCI breakers that see the (phoney!) arc signature will only trip if they are carrying more than about 7 amps through their own circuit at the time.
This can cause different seemingly unrelated breakers to trip whenever their normal load current goes above that current threshold.

 

[CSElectricCO]

This seems to be the case. By just replacing that breaker, the problem hasn't happened again. Very interesting. The arc fault breakers that were tripping after the first one tripped weren't carrying much current though. Maybe 2 amps max on one of them but that's it. At least I'll know for next time. Thank you!

 

[retirede]

My guess: one of the new loads fed by the sub was generating noise that the afci saw and interpreted as a fault. (As described in post 8) New afci has better signature identification logic and is unaffected by the noise???

 

[GoldDigger]

My guess: one of the new loads fed by the sub was generating noise that the afci saw and interpreted as a fault. (As described in post 8) New afci has better signature identification logic and is unaffected by the noise???

But how could that trip the other AFCIs with no current flow? And why did replacing the first breaker fix it?

 

[ActionDave]

And how do we know there is any arc detection left after so many changes in the breakers arc detection.

 

[retirede]

But how could that trip the other AFCIs with no current flow? And why did replacing the first breaker fix it?

I don't think he stated that the other AFCIs had no current flow.

Guessing again (this is far-fetched): The circuit of the first to trip muted the noise that the other breakers were seeing, but when it tripped and that circuit opened, the noise seen by the next breaker increased and it tripped. This continued, tripping one by one.
The new breaker installed in the first position is not susceptible to the noise, so the sequence never starts.
If I'm right, simply switching off the new breaker should start the ball rolling again with the subsequent breakers.

I admit this is way out there, but it's the best I got.

 

[GoldDigger]

An interesting hypothesis.

 

[growler]

My guess: one of the new loads fed by the sub was generating noise that the afci saw and interpreted as a fault. (As described in post 8) New afci has better signature identification logic and is unaffected by the noise???

I don't think he stated that the other AFCIs had no current flow.

Guessing again (this is far-fetched): The circuit of the first to trip muted the noise that the other breakers were seeing, but when it tripped and that circuit opened, the noise seen by the next breaker increased and it tripped. This continued, tripping one by one.
The new breaker installed in the first position is not susceptible to the noise, so the sequence never starts.
If I'm right, simply switching off the new breaker should start the ball rolling again with the subsequent breakers.

I admit this is way out there, but it's the best I got.

Seems logical to me and easy enough to test.

You could also slut of circuits in the sub panel one by one to determine what loads provide the noise in the first place.

What I don't get is if he didn't add any circuits to the sub panel then why didn't they have this problem before? Was it just bad timming to be there when things started to go bad?

I saw that happen to a security company. They were working at a house when the AFCI started to trip. They didn't have a thing to do with it. Bad cable box or a bad breaker ( switching out the breaker fixed the problem).

 

[retirede]

What I don't get is if he didn't add any circuits to the sub panel then why didn't they have this problem before? Was it just bad timming to be there when things started to go bad?

The way I read the OP, it was a newly added sub panel (not a replacement of existing sub panel). If so, probably new load(s). But I may be misinterpreting....