Vexing Electrical Problem

[ray hampton]

Finally some concrete data. I measured the resistance between the ground wire and black wire going to the light box in the kitchen and it has a little over 1 meg ohm of resistance. I'm not sure why that has given me the symptoms I've got when that switch is turned off though.

if your meter shows a reading between the hot black wire and the green ground wire THEN THEY must be touching at some point alone their path

[Mike Lassiter]

More information.

I reconnected everything except the pigtail in the bedroom switch box and turned the circuit breaker back on. I have 120 VAC on one wire in the box and nothing on the other two which confirms that box is the first box in the circuit.

Dennis I have been working on this with you (LOL) as I have been doing other things today. I am curious if you disconnected the power wire from the breaker and rechecked the circuit with the ohm meter if you would still get a ohm reading. Why I am thinking this is the breaker has a pigtail wire on it that goes to the buss bar on the neutral (or ground not sure on arc fault) which would electrically somehow tie them together inside the breaker I would think. I wonder if you are getting a reading from that and not a wiring issue. By disconnecting the power wire from the breaker you would eliminate the breaker from being a potentially high ohm electric conductor. If my thinking is right the breaker itself could be the source of the mega ohm reading you are seeing across the hot and ground wires. It likely shouldn't be a issue and has to do with how the arc fault is able to detect or operate.
I have worked on complicated wiring problems on computer controlled Allison transmissions and got weird ohm readings because of getting some continuity thru the circuit board in the ECU module when the harnesses were still connected to it. Regardless of anything being connected in the branch circuit you are working on, if the breaker is electrically connected to the power and ground internally it could give you a very high ohm reading. Even flipping the breaker off wouldn't prevent that I would assume as the electrical circuit would be on the load side of the breaker internally when flipping it off that connection I would think would still be there - just without power now.

You replaced the arc fault breaker as I recall - take a ohm reading across the pigtail wire and the terminal the circuit power feed wire attaches to and see if you get a similarly high ohm reading on it on that breaker. Will be interested in your reply.

[Von Bickley]

I have been trying to avoid that since it appears that I do have a problem Von.

I'll sit back and watch.... Let us know what you find.

[Dennis McDonaugh]

Dennis I have been working on this with you (LOL) as I have been doing other things today. I am curious if you disconnected the power wire from the breaker and rechecked the circuit with the ohm meter if you would still get a ohm reading. Why I am thinking this is the breaker has a pigtail wire on it that goes to the buss bar on the neutral (or ground not sure on arc fault) which would electrically somehow tie them together inside the breaker I would think. I wonder if you are getting a reading from that and not a wiring issue. By disconnecting the power wire from the breaker you would eliminate the breaker from being a potentially high ohm electric conductor. If my thinking is right the breaker itself could be the source of the mega ohm reading you are seeing across the hot and ground wires. It likely shouldn't be a issue and has to do with how the arc fault is able to detect or operate.
I have worked on complicated wiring problems on computer controlled Allison transmissions and got weird ohm readings because of getting some continuity thru the circuit board in the ECU module when the harnesses were still connected to it. Regardless of anything being connected in the branch circuit you are working on, if the breaker is electrically connected to the power and ground internally it could give you a very high ohm reading. Even flipping the breaker off wouldn't prevent that I would assume as the electrical circuit would be on the load side of the breaker internally when flipping it off that connection I would think would still be there - just without power now.

You replaced the arc fault breaker as I recall - take a ohm reading across the pigtail wire and the terminal the circuit power feed wire attaches to and see if you get a similarly high ohm reading on it on that breaker. Will be interested in your reply.

I'll check it tomorrow Mike.

[Dennis McDonaugh]

I'll sit back and watch.... Let us know what you find.

Von, I'm not discounting your advice, I'm just wary of replacing the AFCI with a regular breaker and masking a real problem.

I've got six other AFCI breakers in the box and don't have any issues with those.

[Tom M King]

In thinking about pulling a new wire, are the boxes in the wall metal, or plastic? I've dealt with a bunch of each, and can offer advice on how to do it, if you need the information.

[Pat Barry]

Hey Dennis, how old is the wiring? I don't recall - I don't think this was new construction, right?

[ray hampton]

Dennis , I got a GFCI in one bathroom and it keep tripping yesterday BUT seems to be okay today , this is the second time [I think ] and I believe that it acts up when a outside outlet gets wet

[Duane Meadows]

Not knowing what type meter was used and under what conditions, 1 Megohm of resistance would allow about 100 micro amps of leakage current. That should be well below what the AFCI should trip at.

Since the AFCI sees arcs as RF(radio frequency), I still think that especially with as much wire as that circuit seems to have, a high power CB radio going down the street would possible cause the AFCI to trip. It would also explain the highly intermittant aspect of the situation. If that is the case, Many ham radio operators can probably help with filtering the circuit, although that would make the AFCI less sensitive, period!

Else you have either a loose connection, or something like a staple thru the wiring somewhere. I think I would try a milliamp meter in series with the circuit with all loads off but the AFCI powered up. Any fluctuation over time could indicate a problem. Would of course require either a lot of meter watching or a recording meter! Then disconnect outlets from the end of the circuit toward the AFCI till the problem goes away!

Measuring resistance on a dead circuit, IMHO, is quite unlikely to locate the problem. Unless you are seeing like 10-50 thousand ohms or less!

Also it seems from my research into your problem, that neutral to Safety Ground contact can trip the AFCI.

Just my 40 years as an electronic tech thinking out load so to speak!

[Brian Elfert]

Dennis , I got a GFCI in one bathroom and it keep tripping yesterday BUT seems to be okay today , this is the second time [I think ] and I believe that it acts up when a outside outlet gets wet

Older houses often shared one GFCI receptacle for the bathroom and the outside outlets. My parent 's house has one GFCI receptacle for the entire house. It was apparently cheaper to run wire than to buy GFCI receptacles back in 1979. Try putting one of the newer outlet covers on the outside outlet that keep water out better.

[Rich Engelhardt]

Dennis , I got a GFCI in one bathroom and it keep tripping yesterday BUT seems to be okay today , this is the second time [I think ] and I believe that it acts up when a outside outlet gets wet

We have a GFCI @ one of our rentals that we've replaced three times now in a little under two years.
We've use a Cooper, a Leviton and a Harbor Freight & they all three do the same thing.
The GFCI will function fine as long as it doesn't get tripped.
Once it get's tripped though, it's all over for it. It becomes "flakey" and unreliable.
Sometimes it goes days or weeks between tripping and sometimes it trips right away as soon as a load is plugged into it.

We replace it & it works fine, up until the point the tenant (or the tenant's aunt that comes over to watch her daughter) absent mindedly plugs her counter top oven into it.

From the reading and research I've done on GFCI outlets, that seems to be fairly common/normal.
Newer GFCI's I'm told, also have a lifespan of only a certain number of years before they get too sensitive.

BTW - I don't know if arc fault outlets are like that or not or if the QC is any better on them than it is on GFCI outlets - but - we can usually count on one bad outlet out of a couple of boxes of 6.
We buy six packs of GFCI outlets when they go on sale anywhere. Sometimes it's Cooper, sometimes Leviton and sometimes Harbor Freight.
It doesn't seem to matter what brand they are.

[Dennis McDonaugh]

Sorry for the delay, been working on other things. This is new construction, just about 11 months old and I called in a warranty repair. I have been reluctant to do this because we built the house and the electrical contractor just installed the breaker panels and wiring to the boxes, switches and outlets. Waiting for a reply, but I know there is going to be some finger pointing, but I am stuck as this is not as simple as I thought it would be.

[Dennis McDonaugh]

In thinking about pulling a new wire, are the boxes in the wall metal, or plastic? I've dealt with a bunch of each, and can offer advice on how to do it, if you need the information.

I'm thinking this is going to be the fix, but it's not going to be easy because there is a floor in the attic above the laundry room (location of the breaker) and hall. All the interior walls are also filled with insulation.

[Dennis McDonaugh]

Not knowing what type meter was used and under what conditions, 1 Megohm of resistance would allow about 100 micro amps of leakage current. That should be well below what the AFCI should trip at.

Since the AFCI sees arcs as RF(radio frequency), I still think that especially with as much wire as that circuit seems to have, a high power CB radio going down the street would possible cause the AFCI to trip. It would also explain the highly intermittant aspect of the situation. If that is the case, Many ham radio operators can probably help with filtering the circuit, although that would make the AFCI less sensitive, period!

Else you have either a loose connection, or something like a staple thru the wiring somewhere. I think I would try a milliamp meter in series with the circuit with all loads off but the AFCI powered up. Any fluctuation over time could indicate a problem. Would of course require either a lot of meter watching or a recording meter! Then disconnect outlets from the end of the circuit toward the AFCI till the problem goes away!

Measuring resistance on a dead circuit, IMHO, is quite unlikely to locate the problem. Unless you are seeing like 10-50 thousand ohms or less!

Also it seems from my research into your problem, that neutral to Safety Ground contact can trip the AFCI.

Just my 40 years as an electronic tech thinking out load so to speak!

I am getting such flaky readings--I seem to have some continuity between hot and ground at several different places, even with wiring disconnected, that I suspected the meter was wrong. I have about 100' of 12 gauge wire left over from the garage wiring and I ohmed that out just to be sure. The meter reads infinite between each wire and short form one end to the other so now I'm sure its functioning properly.

[Dennis McDonaugh]

I disconnected the hot and neutral at the breaker and took apart the pigtail in the bedroom light switch. That leaves me a wire, disconnected at either end, from the light box to the breaker box. I have some continuity between the ground and hot. I also have some continuity between the hot and ground going the other way. I can go down the line disconnecting switches, outlets and lights and confirm the continuity, the only thing that changes is the reading.

I have strange behavior when I reconnect everything.

The breaker blows immediately if I switch either the fan or light in the bedroom on. This was the original problem.
The hall lights, dining room lights, foyer lights, and kitchen lights all can trip the breaker, but not immediately. This is weird. I flipped the hall light on and it tripped the breaker. I reset the breaker and the hall light didn't trip the breaker the next time I turned it on, but the foyer light did. I reset the breaker, turned the foyer light on and it didn't trip the breaker so I tried the hall and kitchen and it didn't trip the breaker either so I tried the dining room light and it did trip the breaker. It just goes 'round and 'round in a seemingly random pattern. Reset the breaker and one of the light switches will cause it to trip, but there is no telling which one will do it at any time. Also, if I reset the breaker and do nothing it will trip after a random period of time.

I quit.