Arc Fault nuisance tripping

[iwire]

I thought that someone said that an arc-fault worked like a Gfci and that is not true .

Well it is true....sort of.

All AFCIs have circuitry looking for arcing faults.

However all AFCIs also incorporate circuity that mimics a GFCI at a higher level trip level.

Any of the wiring issues that cause a GFCI to trip will also cause an AFCI to trip.

 

[don_resqcapt19]

John,

I thought that someone said that an arc-fault worked like a Gfci and that is not true and I said ground to neutral, ...

A ground to neutral fault is the most common installation error that causes an AFCI to trip. This fault puts some of the neutral current on the grounding conductor and will cause the GFP part of the AFCI to open the circuit.

Actually you could have a hot to nail arc and you're not going to find that with a continuity tester.

For the hot to arc to a nail the nail would have to be grounded. The branch circuit/ feeder type AFCI (the most common one at this time) won't even start looking at the arc characteristics until the current exceeds 75 amps. I would expect in the case of the arc to the grounded nail the GFP part of the AFCI would open the circuit. As a matter of fact it is my opinion that in a very large percentage of the AFCI trips will be by the GFP part of the circuit and not the fancy AFCI arc detection circuit. About the only time I see the arc detection part of the AFCI open the circuit is when there is a hot to neutral arcing fault...in other words a parallel arcing fault...the only type of arcing fault that the branch circuit/feeder type of AFCI is designed to detect.
Don

 

[Greg Avakian]

Here's an interesting problem; maybe I don't get how the load on AFCIs work...
In a 20A bedroom circuit that has 6 receptacles, I can plug in my GFI tester and that load doesn't trip the AFCI breaker. The customer can plug in his clock radio and that doesn't trip the AFCI.
But if we plug in either of 2 different lamps, the AFCI breaker trips.
I have tried various combinations of having a lamp on when I flip the tripped breaker: breaker immediately trips. Then turning the lamp off, flipping the tripped breaker: breaker comes on ...and then trips as soon as I turn on the lamp. I've checked the bulbs in both lamps and even tried a flourescent bulb just for fun.

Why would the AFCI breaker work with the clock radio and not a lamp??? Does the type of load make a difference???

This is my first post and I did do a quick search before posting, so forgive me if this type of issue was discussed elsewhere and I missed it.

Thanks in advance for what I'm hoping will be an enlightening discussion...


Greg

 

[George Stolz]

Did you read this thread? You likely have a neutral touching a ground.

 

[tom baker]

95% or more AFCI trips are caused by neutral to ground trips. turn every thing off on the circuit. At the far end, turn on a 100 watt lamp. if the AFCI trips you have a neutral to ground connection in a box.
And of course the AFCI could be doing its job. You may have an future arc fault.

 

[Greg Avakian]

At the far end, turn on a 100 watt lamp.

Thank you.

My question isn't about how to trouble shoot, but rather about the theory of how much load the AFCI needs to read before tripping.

So, generally speaking, I should not expect an AFCI to properly function unless the load is approximately 1 amp or more?

My inspector thinks it's more likely that the breaker is at fault (haha) since the AFCI "should" function properly at any load (ie the load from the alarm clock does not trip the AFCI).

The reason I'm surprised is that I once trouble shoot 2 AFCI circuits -it turned out the electrician had shared a neutral. The breakers came on, but ANY load-including the tiny load from my tester- tripped the AFCI breakers.

(Solution was a DP AFCI breaker -$120 and a 1 week wait)

 

[George Stolz]

Greg, welcome to the forum.

The AFCI comes with ground-fault protection of equipment that is similar to a GFCI but with a higher threshold. It will trip if 50 mA of current is unaccounted for, unlike the GFCI which trips at 5 mA. (Edit to add: The current has to be flowing for it to be unaccounted for, so loads have to be connected in order to cause the GFPE or a GFCI to trip.)

It does not look for arcing faults at anything below 75A, so usually if there is a nuisance trip, it is due to the GFPE inside the AFCI breaker.

This was talked about above, which is why I responded a bit confused as to whether you'd read this thread when you posted.

Also, check out the FAQ on this if you'd like.

 

[kbsparky]

Here's an interesting problem; maybe I don't get how the load on AFCIs work...
In a 20A bedroom circuit that has 6 receptacles, I can plug in my GFI tester and that load doesn't trip the AFCI breaker. The customer can plug in his clock radio and that doesn't trip the AFCI.
But if we plug in either of 2 different lamps, the AFCI breaker trips.
I have tried various combinations of having a lamp on when I flip the tripped breaker: breaker immediately trips. Then turning the lamp off, flipping the tripped breaker: breaker comes on ...and then trips as soon as I turn on the lamp. I've checked the bulbs in both lamps and even tried a flourescent bulb just for fun.

Why would the AFCI breaker work with the clock radio and not a lamp??? Does the type of load make a difference???....Greg

I've found that those symptoms will be present IF the neutral conductor from the circuit is mistakenly placed on the neutral bar, instead of the breaker terminal.

Another mishap I've encountered is adjacent AFI breakers have their circuits crossed, with the respective neutral wires on the wrong breaker. :roll:

 

[kbsparky]

WHILE we are on this subject, let me relate a service call from last week:

Customer complained that every time they ran the vacuum cleaner in the master BR, the breaker tripped. No problems in the other bedrooms.

So, when I took a look, they plugged in the vacuum, switched it on, and sure enough it tripped out.

I replaced the offending breaker, and now it works just fine.

It is possible that those AFI breakers get a little too trigger happy.

Another observation I've made about these is they tend to go downhill once they start acting up: The trip threshold is less and less until they can't hold the circuit on without any loads switched on as well. Replacing the breaker has always solved this problem.

I still don't believe that the manufacturers have worked out all the bugs of this critter. It seems that the public is playing the part of guinea pig, with the NEC CMP's cooperation.

 

[Greg Avakian]

Trouble shot today.

First, the clock is 6.6 watts - over the 5mA threshold. Grrr. But then I realized that the rating probably included the alarm going off, so it wasn't using that much juice when it was just plugged in. That was a good learning experience.

The result was what I kinda expected; "Somewhere" in a receptacle box there was a ground-neutral short.

I draw all my plans with each circuit labeled and each location on the circuit labeled so I know the order of each of the receptacles on the circuit. So trouble-shooting took about 10-15 mins. to find the short location. Once I separated the neutral wires from the receptacle, the short disappeared though. I tried to find a nick in the insulation or some damage and couldn't find anything. I assume that the ground folded up and hit a neutral screw on the receptacle when originally installed. This was especially annoying since I always tape my receptacles for safety's sake.

 

[George Stolz]

Good to hear you figured it out.

First, the clock is 6.6 watts - over the 5mA threshold.

Bear in mind, with a bootleg neutral (EGC touching neutral), that current would be split over the two paths, more or less equally given their proportional resistance.

So, with the .05A imbalance trip setting on the breaker, and the .055 load splitting between the EGC and the neutral, then the imbalance would likely only be .0275A, which is well below the GFPE of the breaker.

 

[tallguy]

Another mishap I've encountered is adjacent AFI breakers have their circuits crossed, with the respective neutral wires on the wrong breaker. :roll:

Been there, done that with a GFCI... Of course I started at the far end of circuit and worked my way back towards the panel. :roll: I think I'll go the other way around next time.

 

[Greg Avakian]

So, with the .05A imbalance trip setting on the breaker, and the .055 load splitting between the EGC and the neutral, then the imbalance would likely only be .0275A, which is well below the GFPE of the breaker.

I R NOT SO SMART.

Thanks for pointing that out. Certainly answers my question then!

 

[JohnJ0906]

WHILE we are on this subject, let me relate a service call from last week:

Customer complained that every time they ran the vacuum cleaner in the master BR, the breaker tripped. No problems in the other bedrooms.

So, when I took a look, they plugged in the vacuum, switched it on, and sure enough it tripped out.

I replaced the offending breaker, and now it works just fine.

It is possible that those AFI breakers get a little too trigger happy.

Another observation I've made about these is they tend to go downhill once they start acting up: The trip threshold is less and less until they can't hold the circuit on without any loads switched on as well. Replacing the breaker has always solved this problem.

I still don't believe that the manufacturers have worked out all the bugs of this critter. It seems that the public is playing the part of guinea pig, with the NEC CMP's cooperation.

Agreed. And they are aware of it too. I had an AFCI situation with vacuums tripping the circuit, but nothing else. Square D sent this response -

Dear John,
Thank you for using the Schneider Electric web site. There are some models of vacuum cleaners that have problems with AFI circuits. Most are 2-speed models that are started out in high speed. We have found that starting in low speed does not trip the breaker. On single speed vacuums start out the vacuum in high carpet setting to eliminate startup surge.
Please feel free to contact us if we can assist you further.
Thank you,
xxx xxxxxxx
Product Support Specialist

 

[don_resqcapt19]

There are some models of vacuum cleaners that have problems with AFI circuits.

Just another example of "spin" by the AFCI manufacturers. Its not the vacuum that has a problem....its the AFCI design.
Don

 

[cschmid]

I have seen allot of math on the AFCI can someone link me with the IEEE standard on AFCI's not the afci safety site.. I want to see some of the design carteristics...So far have had no problems with AFCI's...just looking for more in-depth info on AFCI's...I have learned a little so far from the moderators here but if George is correct is not a 75a spark good enough to start a fire in the wall? and is that not what these are trying to protect arc faults? I thought a GFCI's were for current imbalance's and AFCI's were for arcing? but from what I read in this thread they are both rolled into one? I think I need more information.

The requirements for AFCIs can be found in 210.12:


Quote:
210.12 Arc-Fault Circuit-Interrupter Protection.
(A) Definition: Arc-Fault Circuit Interrupter. An arc-fault circuit interrupter is a device intended to provide protection from the effects of arc faults by recognizing characteristics unique to arcing and by functioning to de-energize the circuit when an arc fault is detected.
(B) Dwelling Unit Bedrooms. All 120-volt, single phase, 15- and 20-ampere branch circuits supplying outlets installed in dwelling unit bedrooms shall be protected by a listed arc-fault circuit interrupter, combination type installed to provide protection of the branch circuit.

Branch/feeder AFCIs shall be permitted to be used to meet the requirements of 210.12(B) until January 1, 2008.

FPN: For information on types of arc-fault circuit interrupters, see UL 1699-1999, Standard for Arc-Fault Circuit Interrupters.

Exception: The location of the arc-fault circuit interrupter shall be permitted to be at other than the origination of the branch circuit in compliance with (a) and (b):
(a) The arc-fault circuit interrupter installed within 1.8 m (6 ft) of the branch circuit overcurrent device as measured along the branch circuit conductors.
(b) The circuit conductors between the branch circuit overcurrent device and the arc-fault circuit interrupter shall be installed in a metal raceway or a cable with a metallic sheath.

 

[don_resqcapt19]

The branch circuit/feeder type AFCIs (the ones you have been using) do not look at the arc signature unless the current exceeds 75 amps for a couple of cycles. I guess the thinking is that even though a 75 amp arc can start a fire, it takes some time and the AFCI should open the circuit before then. I think this 75 amp limit is one of the reasons the arc fault "testers" don't always work...there many not be 75 amps of current available at the receptacle where the "tester" is connected. The word "tester" is in quotes because there is no listed device for testing AFCIs. The combination type AFCI that is required starting 1/1/08 will look at the arc signature when the fault current exceeds 5 amps.
Don

 

[wptski]

Let's see if I can remember the correct values on the Ideal 61-165 tester which has a ACFI test fuction. If anyone considers their method correct is debatable!

It sends out current pulses of 91-141A for no longer than 8.3ms which is a half period or hertz being that at 60 hertz, 16.6ms/hertz.

A 1hp single phase capacitor start/capacitor run motor that I just checked has a inrush current of 74.3A for .5P.

EDIT:
Here's a interesting link on AFCI's and testing: http://www.mikeholt.com/mojonewsarc...I-Testers-versus-AFCI-Indicators~20050728.php

 

[Greg Avakian]

OK, I have another question: Can an AFCI device on the cord of an air conditioner detect a fault or arcing condition on the circuit it's plugged into?

=====
The reason I ask is that when an air conditioner is plugged into a new receptacle that was recently installed, the AFCI trips on the AC unit's cord.

I plugged it into the 'old' receptacle (on an extension cord!) and it worked just fine.

So ...I plugged my tester into both receptacles to be sure they were both wired properly. It turns out the 'old' receptacle wasn't grounded.
so I thought that there must be a problem between the hot and ground in the AC unit and that this would trigger a trip in the AFCI protection when the AC unit was plugged into a grounded receptacle.
To quickly check my theory, I used a 3-prong adapter in the grounded receptacle -making sure that the metal terminal for the screw didn't touch the receptacle (IOW, I created a non-grounded condition on the 'new' circuit without taking it apart). It still tripped the AFCI protected cord. That seems to indicate that the problem is between the hot and the neutral after all -but before the AFCI protected cord???
Now I'm thinking that there's a loose connection and the voltage is fluctuating -but when I tested with my multimeter, it seemed to be getting a steady 112 volts across the hot and neutral.

This is a mystery to me. What am I missing here?

 

[mdshunk]

OK, I have another question: Can an AFCI device on the cord of an air conditioner detect a fault or arcing condition on the circuit it's plugged into?

That's not an AFCI. It's an LCDI.