Got the power turned on to my first home under the new code requiring AFCI protection for bedrooms. Out of curiosity, I pushed the ground fault test button on my GFI plug tester while plugged into a bedroom plug and the breaker tripped! I thought these things were not suppose to trip with a ground fault test? Anyone else had this problem? Also, one of our inspectors here is under the impression that AFCI circuits cannot leave the room they are intended to protect; that is to say that you cannot wire a bedroom bath combination of AFCI. He is saying that every bedroom must have it's own AFCI breaker and it cannot feed the master bath in that bedroom or an adjacent bedroom. Am I missing something in the new code? How would one go about interconnecting smoke detectors if the circuit is not allowed to leave the bedroom? Any ideas???

I believe that AFCI's have GFI protection built into them as well.

There is no limit on what the AFCI circuit serves. The code only requires that any branch circuit feeding bedroom outlets have AFCI protection. As far as the GFCI tester tripping the AFCI, I don't know. The ground fault function that is built into (and that really provides much of the protection) the AFCI is not a Class A GFCI. It is a GFP that trips at about 30 mA of ground fault current. The GFCI tester should only have a ground fault current of less then 10 mA. GFCIs are required to trip at 4 to 6 mA.
Don(resqcapt19)

Last fall I bought the first AFCI breakers my supply house sold. I installed one in my shop feeding a single Hubbell CR-20 duplex receptacle. I plugged everything I could find into this outlet to try to trip the breaker with no success. My Ideal GFI plug-in tester was the only thing that tripped it other than a direct line to neutral short or a line to EG short. I posted my results on several of the forums at the time but got little response. I installed 3 of these in a new custom home this spring. The inspector made sure to see if they were in the panel but said he had no way of testing them. They haven't spontaneously tripped yet and I don't believe they will. I just hope they function as a standard 20A over-current device should if the need arises.

The AHJ in my area are requiring the AFCI to feed bedrooms and are requiring our smokes (of course) to be on one of the Bdrm circuits. I adimately agree with Don that the code only requires Bdrms to be AFCI protected and does not limit the use of these new products to "protect only the Bdrms" as long as the load does not exceed the Max. allowed for the device. I am however, curious about the GFI tester tripping these devices. Perhaps (hopefully) a faulty tester.
Andrew
Also, shouldn't the smokes be on their own circuit? I know that it is not cost effective, however my understanding of these new devices leads me to believe that "what if" little suzie Q were supposed to be sleeping in her bed but instead she was playing with her toys and just happened to short one of the recepts in her room tripping the breaker. The lights are already out so she doesn't give it a second thought. BTW the batteries in the smokes went bad last summer and the smokes started chirping so jo-blo took them out without replacing them. So right about now it's 12:00 AM and everyone is asleep in the house including Grandma who fell asleep and forgot about the food she was cooking on the stove!! If the smokes are on little suzies bdrm circuit what will happen?

[This message has been edited by N.O.Minded (edited 05-07-2002).]

I have recently put in 2 AFCIs and they both tripped with my Ideal GFCI tester too. I didn't think about anything being wrong though. The ones I had were Murray brand.

Bill

NO Minded, wow that is some theory. All that was left out was the second shooter on the grassy knoll. Seriously though I beleive the thinking is that if the smokes are on their own circuit and the circuit trips for whatever reason, or Grandma is making cookies for little Suzie and sets of the smoke detectors, she then turns of the circuit because of the noise, forgets to turn it back on. It would be more likely to be turned back on if something else was turned off like lighting or receptacles which would be more obvious than the smoke detectors alone.

Some of the city's I work in have required the smoke detectors to be on their own circuit and I have always been against it. I agree with Electricmanscott that if they are tied on with lights or receptacles it is not very likely that this circuit will be turned off for very long. If they are on their own circuit the only way to know that the circuit is off would be to look for the little LED light in the detectors. How many home owners look to see if the light is on in their smoke detector? As to the comment about the batteries being removed, if the batteries are low OR removed the detector will beep until a good battery is installed.

All commercially available AFCI's also have equipment level GFI protection inherent in the design. This would explain why the GFI tester trips the breaker. I am not familiar with the specs of these testers, but our sensitivity level, is 30 mA.

The code does not limit AFCI protection to the bedroom, that is merely the minimum standard. I agree with Don that there is no limit to which circuits can have AFCI protection. I have them throughout my home on every 15 and 20 amp 125v circuit.

Tying in the smokes to a lighting circuit makes sense as a trip indication. However if the only thing that is on the circuit is the smokes, you would probably be less likely to have an event that would cause a trip as long as all of the connections are good, and the insulation integrity of the circuit is good.

The AFCI's will trip on overloads and short circuits just like standard circuit breakers. However, there is added protection for arcing faults and ground faults.

The proper way to test the breaker is to push the test button. I cannot speak to the other manufacturers' designs, but our AFCI test button puts a simulated arc current wave form into the custom integrated circuit, thus tripping the breaker. If upon pushing the test button, the breaker does not trip, then the breaker should be replaced.

Brendan Foley
Eaton | Cutler-Hammer

Hmm, I can see both sides of the argument for smoke detectors on their own circuit or combined with lights.

I suppose the ultimate system to get the best of both worlds would be an individual circuit for the detectors fed from a C/B with an integral alarm contact which sounds a separate buzzer if it trips, a little like the alarm fuses used in telephone exchanges.
That's probably getting carried away for residential though.

Brendan,
How did Cutler-Hammer come up with 30mA as the trip threshold for integral ground-fault protection in your AFCIs? Is is just coincidence that 30mA happens to be the standard ground-fault threshold for GFI in Europe now, or were your breakers designed with IEC specifications in mind as well as UL/ANSI?

30 mA is also the standard level for equipment GF protection on this side of the pond mate. The GFI protection enhances the breaker by sensing current leakages to ground in NM-B as well as ground faults caused by the heat generated by glowing contacts.

I like the idea of the alarm contact. We'll have to talk about your royalties.

Brendan,
I would expect that these testers provide a ground fault load that only slightly exceeds the maximum 6 mA trip point for a Class A GFCI device. I know that the internal resistor in a GFCI receptacle creates a 8 mA fault to test the device. Would there be any other reason that these testers would trip the AFCI?
Don(resqcapt19)

I have always put smoke detectors on a separate circuit. I have also put in audible alarm on the system, in case the breaker trips. Never had a problem, either way. The reason I put the audible trip alarm on, is at the request of the home owner. I think and I feel comfortable with putting them on a dedicated circuit. Think about, the battery backup, the smoke detectors will chirp and the home owner will call (hopefully an Electrician) or find out why they are chirping/making noise. Granted, they are chirping and the batteries still work, until changed or power restored.

Brendan,
wouldn't this need to be reworded to
'your glowing contacts will continue to glow , until it goes to ground'?

RockyJo, are you implying that the smoke detectors will chirp if they loose their primary building power? If so that is not the case with any of the detectors I have used. They will chirp if the battery is missing or low but they only way to know that the primary power source is off is to look for the indicator light inside the detector.

Sparky, you could add that the glowing contact would continue until it caused a ground fault, but I do not think that my reply implied otherwise. But to be fair, you would also need to add that with a standard breaker or fuse that the glowing contact would continue until the home caught fire.

Don, the ony thing I see causing the trip is the ground current.

Brendan;
to be 'fair' it should not be refered to as an arc-fault, as it will not universely do as the name implies.
Clarity in marketing as a contractor is condusive to good buisness and customer relations.
Note the LCDI here, a far better description.

440.65 Leakage Current Detection and Interruption (LCDI) and Arc Fault Circuit Interrupter (AFCI).
Single-phase cord-and-plug-connected room air conditioners shall be provided with factory-installed LCDI or AFCI protection. The LCDI or AFCI protection shall be an integral part of the attachment plug or be located in the power supply cord within 300 mm (12 in.) of the attachment plug.


[This message has been edited by sparky (edited 05-10-2002).]

From the Cutler-Hammer Training Page

NFPA Proposal/Comment On-line Submission Confirmation
Submitter: Steve Mancuso
Chester Electric inc
1181 Roach Road
Chester, VT 05143
USA
E-mail: electric@vermontel.net
Representing:

--------------------------------------------------------------------------------

ROP/ROC Option: Print

--------------------------------------------------------------------------------

Prop. or Comm.: Proposal
Document Number: 70
Date Submitted: 5/14/2002
Section/Paragraph: 210.12(A)
Change Recommends: Revised Text
Original Material: No
Reference as:

--------------------------------------------------------------------------------

Recommendation: Clarify the definition as to the specific 'arcing' that the device will address, re; parrallel, series, or both.

(A) Definition. 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.

Substantiation: AFCI's will not trip on a 'series' arc, only a 'parralel' arc. The public, installers , ect, may be misled due to an inacurate definition.

What about a smoke detector that has a built in non replaceable rechargeable battery that will power the smoke detectors for about 30 days and recharge after a power failure and have the home owner replace the unit when the battery alarm goes off? at least this will take care the problem of home owners removing the batteries.

[This message has been edited by hurk27 (edited 05-16-2002).]