General Stuff

[charlie]

I have stayed completely out of this forum and have not even read any posts until today. However, I do have a feel for what is happening with them.

Panel 2 has accepted the requirement to put AFCI protection for all 15 and 20 ampere circuits in a panelboard whenever it is replaced in an existing home. It has not accepted the requirement to put AFCI protection for all 15 and 20 ampere circuits in a new home.

The vote in the panel was indecisive in that there were a couple of abstentions and 2/3 of the members were not achieved. At this point, it is unknown whether or not the above requirements will get the 2/3 vote required to make it into the Code. If it does, it will require a higher level of protection for a one month old home if a panelboard is replaced or added than it will for a new home.

In addition, the AFCI protection will have to go to the "combination type" in 2008. The combination type will detect series faults. For what it is worth, the parallel faults have to exceed 75 amperes for the present AFCI breakers to "see" the fault and open the circuit.

 

[ryan_618]

Re: General Stuff

Hi Charlie. Thanks for filling us in.

When you say "the code will require...in the 2008 edition", I have to wonder how common that is. Is it typical to write code two cycles in advance? Why is this type of proposal even entertained?

 

[charlie]

Re: General Stuff

The technology is there for the combination type but no product is listed yet and there is nothing in the field. I think the panel wants everyone to get used to the idea before it catches people by surprise.

 

[electricmanscott]

Re: General Stuff

Massachusetts initially required afci's for panel replacements, service upgrades, and new homes on the bedroom circuits. This was quickly changed to require only for new homes or new circuit installations. Seems that there were issues with multiwire ciruits, knob and tube wiring, wiring with no egc, and the like that made the afci's malfunction when installed in old homes. I do not know the exact details but I would suspect these issues are still with us.

 

[tonyi]

Re: General Stuff

If a panel replacement type language is what's going to trigger mandatory full house AFCI retrofit, people will circumvent it by keeping old boxes alive as a backfed sub off a new main. Then no existing panel has technically been "replaced". Reconfigured perhaps, but not replaced.

The idea of attempting to address the aging wiring situation in this country is admirable, but I don't think that particular criteria will be an effective means of accomplishing it.

 

[tonyi]

Re: General Stuff

Scott, AFCI's work OK on 2-wire systems. 2-poles for multi-wire are readily available now from several manufacturers (a couple of years ago it was just CH who had them).

Most likely the issues people were complaining about were caused by old fixtures with no seperate gnd wire and a bond strap from the housing to neutral - with a metal box and a metallic cabling/pipe system, that condition is effectively a hardwired bootleg ground that's guaranteed to trip an AFCI or GFCI. I'd bet a number of the trips were genuine problems nobody bothered to investigate too

 

[roger]

Re: General Stuff

How many people on this forum (out of all 13,483 members as of 12/17/03 at 8:15 pm EDT) have trouble shot tripping AFCI's that are actually found and defined as parallel arcs?

I know you have Tony. Haven't you?

Roger

 

[tonyi]

Re: General Stuff

A better question might be how many people are doing a lot of old work and retrofitting AFCI's where there's a much greater chance to encounter the old deteriorated systems that would be prone to producing such arcs.

I doubt anyone slapping up new stuff would be in a situation to troubleshoot a tripping AFCI unless it was a lousy stuff job on a box somewhere that caused a bootleg.

The bulk of trouble shooting on stuff going up today will come 10-20 years from now when all the cheap Chinese fixtures start falling apart

 

[big john]

Re: General Stuff

The only times I've had occasion to install AFCI breakers has been when working on K&T or cloth NM.

Incidentially, the very first breaker I installed, tripped. I was full well expecting that based on the condition of the wiring, I figured there was just enough cumulative leakage current to dump the GFP function. Eventually traced it back to a brand new keyless fixture that had been installed. All that old wiring and it was this new porcelain base fixture that was causing the problem.

I never did figure out why: There was no EG in the circuit on which a parallel path could've been accidentially created. No signs of arcing. Threw on a diffrent keyless and everything came up golden.

I've been convinced, even if largely by this board, that there aren't any "false trips" out there anymore. If it trips, I trouble-shoot.

-John

[ December 18, 2003, 03:48 AM: Message edited by: big john ]

 

[pierre]

Re: General Stuff

Tonyi
How about a sheetrock screw through the wire in new work? Not much we can do to prevent that when the sheetrocker uses long screws, except to trouble shoot it later

Pierre

 

[tonyi]

Re: General Stuff

It depends on how much you want to spend on defensive measures versus the hassle of finding/fixing something like that when it happens

 

[don_resqcapt19]

Re: General Stuff

Tony,
The AFCIs currently on the market do not work effectively on 2 wire systems with a "series fault" or "glowing connection". The AFCI does not "directly" detect any type of fault other than a parallel arcing fault. Faults of other types are only "detected" when the heat has damaged enough insulation for the fault to progress into a ground fault or a parallel arcing fault. Progression of an series type fault into a ground fault or parallel arcing fault is not very likely on a knob and tube wiring system. It is somewhat more likely on a two wire NM system, but not nearly as likely as it would be on a wiring system that included an EGC. The AFCI is the least effective where it is need most, in older installations. The fire loss statistics cited by the AFCI proponents to get the AFCI requirement into the code showed that 85% of the dwelling unit fires of electrical origin occurred in homes over 20 years old.
Don

 

[al hildenbrand]

Re: General Stuff

I got called back to a gut an reset of a bedroom that I'd supplied with two circuits: one for general lighting and one for the designated computer locations. This is a metal flex job that the occupant requested for EMF control concerns.

I capped off neutrals and hots as I made up the openings, plugged in the brand new SqD AFCIs and heated up the circuits. All fine. Temporary light worked and no receptacles to worry about.

I was called back ten days later cause the computer circuit AFCI wouldn't hold. . .I figure a conductor end got loose, but I couldn't find it. I then pulled the branch circuit conductors from the breaker and it still tripped! This breaker hadn't even seen load yet, and is under the protection of layered whole house surge protection.

Just my bad luck, or suspect electronics?. . .I do know that AFCI manufacturers have been less than forthcoming about the intricacies of these microprocessors and related circuitry and programming. . .I hate having to use advertising jingoism to understand black boxes.

 

[tonyi]

Re: General Stuff

Al, I've had two Cutler BR's go bad so far. Wouldn't hold similar to what you describe. The mechanical part of the trip mechanism felt springy/funny. Both had been hot maybe 3 months when they died. No failures on the THQL's and QO's I've put in so far. The latest BR's seem to be a different physical design than the ones from a year ago.

 

[tonyi]

Re: General Stuff

Originally posted by don_resqcapt19:
Tony,
The AFCIs currently on the market do not work effectively on 2 wire systems with a "series fault" or "glowing connection".

And all the vendors have stated as much up front for some years now in their technical literature. What's your point?

Q: Do you think a 2-wire system is safer with one or without one?

They do something on 2-wire systems, and that something is better IMO, than an ordinary breaker or fuse (especially on old non-bond wire BX where the armor is good enough to let its GF trip before the armor torches up like a filament). These older places also all have metal boxes that are substantially better at containing the dreaded glowing connection/series arc than plastic boxes are.

My point is there is no electrical reason why an AFCI would cause a 2-wire system to MALFUNCTION in operation anymore than putting a GFCI on one would cause it to malfunction.

 

[al hildenbrand]

Re: General Stuff

They do something on 2-wire systems, and that something is better IMO

This seems like an article of religious faith. . .something unproveable, but believed in.

 

[don_resqcapt19]

Re: General Stuff

Tony,

And all the vendors have stated as much up front for some years now in their technical literature.

Not really. If they had made that information readily available, the AFCI requirement would not have made it into the '99 code. This information was glossed over and buried to make sure that the money making AFCI device got into the code. Right now, most people still believe that the AFCI "arc signature detector" is what opens the circuit when a high resistance heat producing fault occurs. In a new code compliant installation, there would be very few types of fire causing faults that would not be detected by a standard GFCI device. It is almost impossible, in a code compliant wiring system, to create a parallel arcing fault that is not also a ground fault. In my opinion, the vast majority of the faults cleared by an AFCI device, are actually cleared by the built in GFP circuit and not the fancy "arc signature detection circuit".
I never said that the AFCIs caused any malfunctions, just that they are not very effective. Based on the fire loss figures and other information that was published in the ROPs and ROCs, it appears that if every new dwelling unit built in 2002 in the US had AFCI protection for the bedroom circuit, that 14 fires would be prevented in the first year after the dwellings were built. And that is assuming that all of the fires that have been reported as electrical in origin were really electrical, something that I very much doubt.
Don

 

[tonyi]

Re: General Stuff

Don, how would a pinched/frayed 2-wire cord kick a GFCI? There's a vast amount of flaky worn cord/plug devices that only have two wire out in the field, and more being sold every day.

If as you claim all this technical data was "hidden", then why are we hearing about AFCI usage being considered for expansion in the future rather than eliminated? Surely by now all this secret stuff has been well circulated and investigated. Even Mike who was originally a rather vocal critic toned down his rhetoric considerably and has become somewhat of a believer in their value.

FWIW, on the 2-wire systems I won't use an ordinary AFCI. I put in the combos or use an AFCI followed by a GFCI. This better protects against shocks from a "hot box/cover" scenario if it should ever develop.

 

[don_resqcapt19]

Re: General Stuff

Tony,

Don, how would a pinched/frayed 2-wire cord kick a GFCI? There's a vast amount of flaky worn cord/plug devices that only have two wire out in the field, and more being sold every day.

Unless it is a parallel arcing fault, the current AFCIs will not detect the fault. The currently required AFCIs are not intended to provide, and will not provide, more than very limited protection beyond the outlet. When the new combo (providing both branch circuit and protection beyond the outlet) AFCIs are on the market, then these devices may have some merit.
We are hearing about AFCI usage being expanded for the same reason the AFCI was originally placed into the code. The manufacturers want to make more money.
If the code was a federal law a full cost/benefit study would have been required. I doubt if a real cost study would support the use of AFCIs. Do the prevent some fires and save some lives? Sure they do. Just like replacing our cars with tanks would prevent a lot of traffic deaths, but that would not be cost effective.
Don.

 

[tonyi]

Re: General Stuff

Unless it is a parallel arcing fault, the current AFCIs will not detect the fault.

So what? Would you rather have half a loaf or none? Seatbelts weren't perfect, then along came airbags, and at some point in the future undoubtedly something else will come along too.

There's a reason cavemen weren't driving Camaros. Technology evolves and you deploy what you got. If you wait for perfect, you'll be waiting forever.

The currently required AFCIs are not intended to provide, and will not provide, more than very limited protection beyond the outlet.

Again, something the vendors have been stating for some time now. Still, this seems pretty useful for all those 2-wire lamp cords lurking under rugs and lead cords duct taped to walls and floors.


We are hearing about AFCI usage being expanded for the same reason the AFCI was originally placed into the code. The manufacturers want to make more money.

The vast right wing conspiracy theory eh? If this is really true, why does an AFCI cost less than an ordinary GFCI of GFP breaker in many cases?

Just like replacing our cars with tanks would prevent a lot of traffic deaths, but that would not be cost effective.[/QUOTE]Well, most people can't afford a $6M tank, but a few $30 AFCIs doesn't seem beyond the means of someone spending a hundred grand or more on a place that will be standing for decades to come. If an extra hundred bucks on the pricetag of a place is going to break the bank, then they've got no business buying that house in the first place. Amortize the cost over the lifetime of a house and any cost arguments start to look pretty silly.

Would you pay a few dollars more each year to have a car that's safer? I sure would. You must be one really cheap dude to value people's property and safety so cheaply.

[ December 18, 2003, 11:07 PM: Message edited by: tonyi ]