AFCI Nuisance trip or a real issue

Status
Not open for further replies.
acrwc10

acrwc10

Master Code Professional
Location
CA
Occupation
Building inspector
The question of " why is the afci tripping " comes up on the forum often so I thought I would share a "nuisance tripping" that I traced out today.
A home owner who's remodel I wired about a year ago called and explained that the bedroom AFCI circuit had shut off multiple times over the last year. They would reset the breaker and it would be fine for a while. Until today, it would not reset. So I went through all the basic steps disconnecting all loads, opening up boxes etc. There were no obvious problems everything was wired correctly. Using an ohm meter there was no apparent problems or shorts. So I ran a megger test on the line and found a neutral to ground fault that only shows up when the voltage starts getting up in the 270 volt range.
The problem turned out to be when the carpenters put in some blocks for sheet rock backing they pinched an NM cable in the wall against its staple. This caused a small split , maybe 1/16th" , in the insulation on the neutral wire this bare spot would connect with the ground intermittently and trip the afci.
Had this of been a non-afci circuit it would have never shown up as a problem. If I had not used a megger on the circuit I would have never been able to locate the exact place in the line that had failed.
 
I wouldn't call that a nuisance trip. The AFCI was detecting a real fault. It was a fault that the system would be functional with, and a fault that a normal breaker would not detect. But it was a fault none the less.

Bravo for the proper fault finding, rather than simply blaming it on a faulty breaker.

-Jon
 
This is a good point you make that a lot of time the breaker gets blamed and off we go.
Not that long ago I had another job That I had installed an afci in a study and when the HO would turn on his computer it would trip the AFCI "sometimes". The problem with this one is the "utilization equipment" that was the cause of the problem. But there is no way the home owner was going to believe that 100% and you can't put a megger on a computer tower to prove it to the HO. So what I did was look at the approved blue print to see if the room was listed as a bedroom or a study. It was listed as a "study" so being under 2005 NEC I changed the breaker to a standard single pole. Still code compliant but I can't help wondering how many of these kind of problems we are going to see and what is the best way to head of the problem before it occurs. I also think that the problem with the computer is the power supply causing some sort of capacitance or in rush on start up that the afci sees as a fault.
 
acrwc10 said:
This is a good point you make that a lot of time the breaker gets blamed and off we go.
Not that long ago I had another job That I had installed an afci in a study and when the HO would turn on his computer it would trip the AFCI "sometimes". The problem with this one is the "utilization equipment" that was the cause of the problem. But there is no way the home owner was going to believe that 100% and you can't put a megger on a computer tower to prove it to the HO. So what I did was look at the approved blue print to see if the room was listed as a bedroom or a study. It was listed as a "study" so being under 2005 NEC I changed the breaker to a standard single pole. Still code compliant but I can't help wondering how many of these kind of problems we are going to see and what is the best way to head of the problem before it occurs. I also think that the problem with the computer is the power supply causing some sort of capacitance or in rush on start up that the afci sees as a fault.
Click to expand...




Kudos to you for being persistant and using equipment others are not familiar with to trace out your situation!!!:cool:
 
acrwc10 said:
This caused a small split , maybe 1/16th" , in the insulation on the neutral wire this bare spot would connect with the ground intermittently and trip the afci.
Click to expand...
So, arguably, a GFCI breaker, on that circuit, would have tripped intermittently, as well.

Your second example, the computer in the Study, brings up, perhaps, the more general case, in my opinion. A computer switching power supply will introduce a lot of short current draws, every half cycle, on the supplying branch circuit. The computer branch circuit load current is complex, full of high speed transients, unlike the smooth 60 Hz sine wave current of a simple incandescent lamp.

In a recent paper, Arc Fault Detection and Discrimination Methods, the author, Carlos Restrepo describes the theory that leads to the tripping of a Combination AFCI.

CombinationAFCIFaultCounter.jpg


The voltage and current waveform, that is "seen" at the AFCI breaker, is manipulated by the electronics of the breaker to create a number of "signals" that are then judged OK or not-OK. A "counter" circuit is incremented, as programmed, at each judgment, and if the counter total exceeds a set limit, the breaker mechanism is operated and the branch circuit is shut off.

There is a lot of room here for the design assumptions to not fit well with the wilds of the "real world".

I believe that as real world performance is returned to the manufacturer's design engineers, the programming in the breaker will be improved. However, the difficulty of field verification of true nuisance trips is going to be great, and will limit the feed back to the manufacturer.
 
any time a breaker trips its certainly a nuisance to me!
Click to expand...

Service call baby!!:cool:



If I had not used a megger on the circuit I would have never been able to locate the exact place in the line that had failed
Click to expand...


Tell me more?

The megger pinpoints the fault?

What type of megger. I need one.
 
220/221 said:
Service call baby!!:cool:






Tell me more?

The megger pinpoints the fault?

What type of megger. I need one.
Click to expand...

It is made with willow branches and it also finds bureid pipe:grin:

When I said that I found the exact location what I mean is that I could find the individual wire and run within the house without having to disconnect part of the circuit and then turn the breaker on to see if "that was the wire" . Remember this would not show up unless it had voltage at 120 volt or higher.
 
I could find the individual wire and run within the house without having to disconnect part of the circuit and then turn the breaker on to see if "that was the wire" .
Click to expand...

One more time.

s l o w l y this time.

Keep in mind that I don't own a megger and have never used one. I am assuming it induces voltage into the wire and reads.....something.
 
M e g g e r i s a t o o l t h a t c h e c k s t h e q u a l i t y o f t h e i n s u l a t i o n o f t h e w i r e b y i n d u c e i n g h i g h v o l t a g e o n t h e w i r e..Damn I hope that was slow enough it was driving me crazy..:grin: :D I will include a link for you..

http://www.megger.com/us/products/ProductDetails.php?ID=137&Description=

I am still laughing at my input sorry to have fun at your exspense it is serious question..you could locate the wire section but you can not locate the exact position with a megger..you would need some sort of wire tracer..
 
Last edited:
AFCI Nuisance trip or a real issue

-Actually, you can find the location within a few inches of an abrupt change in cable properties on an unpowered line using Time Domain Reflectometry. I don't think anyone makes one of these yet for house wiring, but the way things are going, they might soon.
-For this setup you need a reasonably fast scope and a special pulse generator, and they mostly use it to find coaxial cable faults.
-But, I don't know if Romex cable clamps and cabling kinks (both of which change the "characteristic impedance" of the cable) will give you a lot of false readings.
 
acrwc10 said:
If I had not used a megger on the circuit I would have never been able to locate the exact place in the line that had failed.
Click to expand...
Bravo.

About how long have you been carrying a megger on your truck?
 
langjahr@comcast.net said:
-Actually, you can find the location within a few inches of an abrupt change in cable properties on an unpowered line using Time Domain Reflectometry. I don't think anyone makes one of these yet for house wiring, but the way things are going, they might soon.
-For this setup you need a reasonably fast scope and a special pulse generator, and they mostly use it to find coaxial cable faults.
-But, I don't know if Romex cable clamps and cabling kinks (both of which change the "characteristic impedance" of the cable) will give you a lot of false readings.
Click to expand...

Megger makes some very nice TDR units - a graphical (as opposed to a numeric) would be very usefull for this type of fault finding IMO.
 
Last edited:
e57 said:
Megger makes some very nice TDR units a graphical (as opposed to a numeric) would be very usefull for this type of fault finding IMO.
Click to expand...
I'm curious if anyone has actually ever successfully used a TDR for NM cable?

I think you'd see a reflection at every snug staple and every 90 degree turn. Plus, we have this funny tendancy to feed through receptacles and cram wired back into boxes, so that would involve taking apart every single connection on the circuit... maybe even grounds too that were done with crimp sleeves. I think the general idea has potential, but I'm not sure the present batch of TDR's are quite suited to the task.

This has my gears spinning on how a longitudional stress test from a telco type kick meter might be useful in pinning down the location of this sort of fault.
 
Last edited:
mdshunk said:
Bravo.

About how long have you been carrying a megger on your truck?
Click to expand...


I must confess to the master, aka mdshunk , I finally went out and bought one after you had posted the link to " a stitch in time" a while back . It is not a supper deluxe model but a simple digital stile that goes to 1000volt and has volt and ohm meter. Simple yet perfect for what I need. Of course when I got it I went around test everything in the house garage and yard :)
 
Status
Not open for further replies.
Top