ECN Forum Forums Code Related Discussion NEC & other Code issues Arc Fault Breakers

Brenden,
Thanks for addressing my some of my questions. I'm not really trying to be a pain in the ***, but have real concerns that these devices do what you and the other manufacturers say they will. I don't doubt that they will clear the arc faults that you claim, I just don't think that those faults are the cause of very many residential fires.

I don't believe that homeowners test their GFCIs and don't think that they will test the AFCIs. Again I ask how many of you out there reading these posts regularly test your GFCIs. I know that the ones in my house are not tested monthly as required by the instructions.

I'm not saying that there are better statistics, just that there is a "when in doubt, its electrical" mentality in the fire investigation process. Very few fires are investigated by people with any real knowledge of electrical systems. Even where the cause of the fire has been correctly identified as electrical, there is no detailed information available as to the exact type of electrical fault that caused the fire. Some of the data submitted by Mr. Clarey of Cutler-Hammer in comment 2-68 in the 98 ROC says that 36% of the home fires were started by the fixed wiring system. This is not broken down into the exact types of fires. His comment goes on to say that AFCI breakers could be expected to prevent 40% of these fixed wiring system fires. Now we are down to the AFCI breaker being able to prevent less than 15% of the home fires. Data from the US Fire Administration's National Fire Incident Reporting System (NFIRS) which is used by 140,000 fire departments in 40 states shows that the home fire loss caused by electrical fires is only 13.75 of the total loss and the loss caused by the fixed wiring systems is only 5.5% of the loss. The fixed wiring system loss included all of the fixed wiring within the walls....junction boxes, lighting outlets, receptacle outlets, distribution panels and the wiring between these connection points. There is nothing in the data that tells us how many are parallel faults and how many are series faults.

I don't believe that very many of the fires are originating in the conductors between outlet points. Can you please describe the types of faults that will create parallel arcing between outlet points? I can't think of very many things that could make that happen. I can't think of any type of parallel fault that could occur between outlets in a code compliant wiring system that would not be cleared by a GFCI. Can you list some parallel faults that occur between outlet points that won't be cleared by a GFCI?


The engineers who did the study in Japan that is cited on the Fire-Fighter site had no trouble making these "glowing connections" occur. Are you saying that a glowing connection is not an arcing condition? Is a glowing connection a series arc? Are you saying that a fire cannot be started by the heat produced by a faulty connection? I'm sure that all of us who have done any service work have seen the melted parts that have been caused by a poor connection. Are you down playing the hazard of series arc at points of connection because the AFCI breaker won't detect them?

Just because you have lots of good people and are a well respected company doesn't automatically mean that you are right. Edison made that statement that AC current was only good for powering electric chairs. He made this statement based of both what he believed to be technically correct and because his company only produced DC equipment. I didn't say that I would support the Fire-Fighter product, just that it has led to more questions. As a matter of fact, I have even more reservations about that product then I do about AFCIs.

If AFCI breakers can do it all, why did one manufacturer of AFCI breakers submit a proposal (proposal 2-111, 2001 ROP) that would have required both AFCI breakers and AFCI receptacles? The fact that this was done also makes me question the AFCI requirement.
Don(resqcapt19))

[This message has been edited by resqcapt19 (edited 12-09-2001).]

I’ve been monitoring this post for the past week. Lots of good information thanks to all who post.

rkukl;
I've been reading with great interest too

motor-T;

In review, the 96' NFPA 72 2-3.24, will not allow a residential smoke(s) off a GFI.

Perhaps smokes were not considered in the 11th hour revisal of 210.12 to stipulate 'outlets' instead of receptacles.

I've also been following this with interest, but keeping fairly quiet as I've not had any experience of these devices.

Brendan,
Are you aware of any AFCI being exported to or manufactured in Europe? There's no sign of them in any British manufacturer's range as yet.

Re: Test buttons,
I've found that the majority of people here have no idea that their RCD/ELCB (GFI) even has a test button. I always try to explain, but I doubt most of them go on to test even once a year.

Personally, I'm obsessed with testing GFCI's before I use them...

It's just something I've always done... Since 1976 when my Mom bought the double wide with the funny outlets...

Before I became an electrician, I didn't even realize I was doing a good thing! I was just curious about how it worked...(poke-pop!-heh heh heh... repeat)

I can't speak for anyone else and I am probably the exception to the rule...



[This message has been edited by sparky66wv (edited 12-13-2001).]

I have read these posts with a great deal of intrest. Since use of the AFCI breakers- made by your company Brendan- cost me over 45 manhours in callbacks, replacing the 5 ea AFCI breakers in the house twice,(total 15 AFCI breakers),a new hair dryer for mrs homeowner, 5 single pole 20 amp breakers, and a lost customer who I have known for 15 years. Untill I see proof that these devices work and will prevent fires by at least 15 years experience, I will NOT ever warrent them in my company.
The supply house and local rep refused to replace the breakers or refund the money I paid for this junk claiming lightning caused the problem, even though 1 set or breakers failed without any storm in the area.
To say I am angry is correct.
I understand what you and many others are trying to do. You are trying to improve safety in electrical installations. But I must agree that the average home owner will NOT test every month- They do not test GFCI outlets in plain sight. What makes you think that they will open the panel cover to test a breaker then have to reset the alarm clock by the bed? What about apartment dwellers- most of whom could care about maintence less?
It is my belief that the primary reason the AFCI breakers are in the code is so the manufacturer can recoup there R&D costs and make a profit. If this product is so great sell it on its own merits. Do not force it down our throuts legally when the 02 code is adopted.

The stat's collected and presented as validation exemplify the old addage.

House fires do not recieve the forensic's of airline crashes. There are few, if any extensive investigations. Most do not even warrant an investigation at all, so an educated quess by the firechief, or deputy goes unquestioned.
Even if said fire WAS electrical in nature, the factors involved are rarely pursued, i.e: was there a bona-fide electricain? was there a permit? was there an inspection done? Was there any DIY'er involvement? The standards vary so much from locale to locale that a generic compiling does this no justice.
The insurance industry simply pays off, and the matter is dropped & Our trade is held to blame
Yet manufacturer's will use this info , and grab the public by the lapels by advertising firefighters haulin' little girls out of burning buildings in our trade mags!

sparky
former FF, .......bark worse than bite

Brendan,

Hope your week has gone well. I'm sure that you have discovered that the audience here is critically astute and looking for precise information about your product. So much of what has been available out in the trade journals, in my opinion, has the flavor of hand waving of a math proof where one or two steps are glossed over by the professor.

Perhaps you can explain a missing step or two. As most of us are dealing with AFCI installations, we want them to be durable and dependable for what they do, so that John Q. Public may have a quality experience of life and electric power and we can move on to the next installation with out looking back.

In addition to the questions in the thread above, please tell us about the AFCI microprocessor, how is it protected from the harsh environment it services? Are there any concerns about needing to change the microprocessor's read only memory of criteria for interruption? And, if so, would this be done to the installed product, or is it, once installed, unsupported?

Respectfully,

Al

I apologize for my lapse in replying. I have been traveling and attending to my more than full-time job!

I will try to take a few moments to answer some questions. For some of the more technical questions, I am going to our design engineers, and at this time of year many folks are taking needed breaks.

For ElectricAL:
The AFCI continually monitors the voltage. We monitor the voltage trace waveform across the resistive element in the breaker. We monitor the voltage drop and convert to current. We are looking for the spikes in this waveform that are indicative of arcing faults. When we see that there is evidence of an arcing fault of 50A RMS or greater, we analyze the current waveform for the signature of an arcing fault. If we see this, the breaker is "armed". We then look for 8 arcing half cycles over the next 1/2 second, (these do not have to be consecutive half-cycles, and they can be positive or negative. If we see 8 half cycles of arcing, we trip the breaker. If the electronics does not see 8 arcing half cycles in that time, the breaker is no longer armed, and goes through the process again. All of this is done with a custom analog integrated circuit. Additionally, we monitor arcs to ground for the parallel arcs in NM-B cable at a 30-mA level. There are MOV's on the board to protect from surges. Hope I've answered your questions.

For Motor-T and Sparky:
Smoke detectors would be covered by the changes made to 210-12 for the 2002 code that changed the word "receptacles" to "outlets".

For pauluk:
We are beginning discussions to export the technology to Europe in the IEC breakers. There is a problem in Europe that the appliances are more flammable over there due to the fact that Euro manufacturers do not use as many fire-resistant chemicals in their plastics due to concern over the environment. Someone earlier stated that they do AFCI in Europe. That is a misstatement. European standards use earth leakage, which is not AFCI.

For nesparky:
First, I apologize for your difficulties. I would like for you to email me directly at BrendanAFoley@eaton.com with your location, and the distributor and sales office that you were dealing with. I need to let them know that they are not following our "no-quibble" warranty process. In the same email, I'd like you to send your name and address so I can have the Cutler-Hammer sales engineer get in contact with you.

You did not provide a lot of details about your problems, whether the five breakers were all replacements on the same circuit or not. If you email me directly, I can go through this in more detail. I cannot be sure until we talk further, but I would suspect one of three things was causing the circuit breakers to trip:
1) There was ground current in the circuit. This could be because there was a grounded neutral somewhere in the circuit.
2) There was a shared neutral in the circuit.
3) There was arcing in the hair dryer.

The vast majority of the "defective" GFI breakers that we receive back have no problem whatsoever. We test the devices, and they test within design parameters…that is, they are functioning properly.

To all, I will try to do a better job of replying in a more timely fashion. I know that you have concerns, and I will try to address them.

Happy Holidays,
Brendan Foley

Brendan,

Thanks for the info. I guess we'll start seeing AFCIs over here within the next few years then, although I expect they'll be given a different name, as with GFCI/ELCB/RCCB/RCD.

A few days ago I think someone (Sparky?) mentioned the possibility of using a single AFCI as a sub-feed to several regular breakers to minimize costs.

We already have whole-house GFI set-ups here (and elsewhere in Europe), and I have to say that I don't particularly like the approach. It can be a real pain when an earth fault on one ciruit knocks out power to the whole building.

What sort of price range are we talking for a single-pole 15 or 20A AFCI compared to a similar GFCI?