AFCI to rescue?? maybe not??

Status
Not open for further replies.
Ponchik

Ponchik

Senior Member
Location
CA
Occupation
Electronologist
I found a nail from a trigger happy construction guy that had penetrated a 12-2 NM cable. The circuit worked fine for about 10 years. It started tripping the breaker and the client left it off for the last two years.

If this was on an AFCI circuit it would have prevented it to get to the burning stage.

View attachment 17283 View attachment 17284
 
If the NM had been stapled to code, at least 1.25" away from the stud face, it likely wouldnt have been perforated by the nail. and a fished-in wire may have moved whereas a rigidly mounted one cannot.
 
K8MHZ said:
How do you know that? I see no signs of arcing (melted metal) and what looks like a glowing fault where the staple turned into a heating element and slowly charred the wood.
Click to expand...

The staple was not at fault. The nail from the siding penetrated. It made contact between the hot and the EGC just enough to not cause a dead short, but good enough to heat up and eventually (because of heat) melt the conductors and kill the circuit down stream.
 
Classic case of pyrophoric carbonization. This is certainly something an AFCI is touted as being able to prevent, but can they in reality?
 
peter d said:
Classic case of pyrophoric carbonization. This is certainly something an AFCI is touted as being able to prevent, but can they in reality?
Click to expand...


If the hot is contacting the EGC just enough for current flow but not good enough contact for a dead short, won't that trip the AFCI?


I don't know this for sure, but doesn't the AFCI monitor the current flow between the hot & neutral and if there is a difference then it will trip. Similar to GFCI operation.
 
edward said:
If the hot is contacting the EGC just enough for current flow but not good enough contact for a dead short, won't that trip the AFCI?


I don't know this for sure, but doesn't the AFCI monitor the current flow between the hot & neutral and if there is a difference then it will trip. Similar to GFCI operation.
Click to expand...
Not all do. Some no longer have that function.
 
edward said:
If the hot is contacting the EGC just enough for current flow but not good enough contact for a dead short, won't that trip the AFCI?


I don't know this for sure, but doesn't the AFCI monitor the current flow between the hot & neutral and if there is a difference then it will trip. Similar to GFCI operation.
Click to expand...

The AFCI uses a microprocessor to compare current signatures with signatures stored in the device. It samples and compares, and makes a decision using the microprocessor. A GFCI looks for differences in the current of the hot and the neutral using a simple current transformer. Two different technologies. An arc will produce enough heat to melt an vaporize metal. The aftermath is easy to identify, the metal sometimes forms into little balls. If there is no evidence of melted metal, there was probably no arc, and an AFCI would have not tripped.
 
edward said:
If the hot is contacting the EGC just enough for current flow but not good enough contact for a dead short, won't that trip the AFCI?


I don't know this for sure, but doesn't the AFCI monitor the current flow between the hot & neutral and if there is a difference then it will trip. Similar to GFCI operation.
Click to expand...



There was current flowing into the wood itself, that's why it's charred around the nail. The GFPE setting on an AFCI is 30 milliamps, the current flowing into the wood turning it into charcoal is in the microamp range, so that would not pick it up. I doubt the AFCI circuitry would recognize it either.
 
ptonsparky said:
I thought they required a much higher current flow than what that process would need. IDK.
Click to expand...

Yes, far more. I really doubt an AFCI would have stopped this unless there was a decent H-G fault which would be picked up by the GFPE, but even a regular breaker would have picked that up as a dead short. In the case of the OP, it did turn into a dead short which tripped the breaker, which leads me to believe this started life as pyrophoric carbonization, which in turned heated up the nail to the point where it compromised the insulation on the cable and then finally burned away enough until it made contact with the ground or neutral, and then the dead short.
 
edward said:
If the hot is contacting the EGC just enough for current flow but not good enough contact for a dead short, won't that trip the AFCI?


I don't know this for sure, but doesn't the AFCI monitor the current flow between the hot & neutral and if there is a difference then it will trip. Similar to GFCI operation.
Click to expand...
They all used to have at least 30 mA of ground fault protection, not all do anymore.

ptonsparky said:
I thought they required a much higher current flow than what that process would need. IDK.
Click to expand...
The current does need to be above a certain threshold before it will look for known arc signatures. I don't know what that level may be. This is partly why some appliances give intermittent problems with AFCI's, if there is other loads on the circuit the threshold is reached easier then if the problem appliance is the only thing operating sometimes.

peter d said:
There was current flowing into the wood itself, that's why it's charred around the nail. The GFPE setting on an AFCI is 30 milliamps, the current flowing into the wood turning it into charcoal is in the microamp range, so that would not pick it up. I doubt the AFCI circuitry would recognize it either.
Click to expand...
You really think that wood is going to conduct any easily measurable current with only 120 volts being applied, I'd even ask the same for 277 volts as long as there wasn't a moisture problem present. When I say easily measurable I am talking about milliamps and higher, as it doesn't take all that expensive of a DMM to measure milliamps.
 
Stored Signatures?

Stored Signatures?

K8MHZ said:
The AFCI uses a microprocessor to compare current signatures with signatures stored in the device. It samples and compares, and makes a decision using the microprocessor.
Click to expand...

Hello K8MHZ,
You have peaked my interest with the "signatures stored in the device" statement?
Can you provide a source for that information?
When AFCIs were first were introduced that is what I theorized they might do but it seemed quite a sophisticated a task for the relative low cost of the devices.
 
kwired said:
You really think that wood is going to conduct any easily measurable current with only 120 volts being applied, I'd even ask the same for 277 volts as long as there wasn't a moisture problem present. When I say easily measurable I am talking about milliamps and higher, as it doesn't take all that expensive of a DMM to measure milliamps.
Click to expand...

Yes, absolutely. Pyrophoric carbonization is a proven fact. Again, we're talking about micro amps here, but that's enough given time to produce what happened in the OP's picture.


un250-66.jpg
 
Word of the day (at least for me) Pyrophoric carbonization. Never know.
:thumbsup::thumbsup:
 
edward said:
I found a nail from a trigger happy construction guy that had penetrated a 12-2 NM cable. The circuit worked fine for about 10 years. It started tripping the breaker and the client left it off for the last two years.

If this was on an AFCI circuit it would have prevented it to get to the burning stage.

View attachment 17283 View attachment 17284
Click to expand...



Thats pyroforic carbonization mostly. I guess it eventually wore down enough insulation eventually on the hot to contact the EGC.
 
kwired said:
You really think that wood is going to conduct any easily measurable current with only 120 volts being applied, I'd even ask the same for 277 volts as long as there wasn't a moisture problem present. When I say easily measurable I am talking about milliamps and higher, as it doesn't take all that expensive of a DMM to measure milliamps.
Click to expand...


Its microamps from what I've heard. Enough to heat the wood over the course of years if not decades ever so gradually.
 
Status
Not open for further replies.
Top