Are you saying you are backfeeding circuits with a male-to-male drop cord????? Just to test them or for temp power???

This doesn't sound good....

Yes, pety_racer, using a male-to-male drop cord to make the receps live just to test them. When we plugged into any circuit with an AFCI, it sparked - just like there was a direct short. I am wondering if this is just the nature of the AFCI breaker, that it cannot be backfed in this manner.

why not use a 240v drop cord to feed the PANEL, then test each circuit as it would normally run? that is much easier and MUCH safer!

Actually, that is exactly what we will do if we don't get the grading (it is raining here for the first time we weeks) and the electrical hook-up done soon. In the mean time, I was (and still am) wondering why the AFCI-protected circuit behaved as it did.

I have NO idea why this happened as I have never heard of this, nor have I ever tried doing what you are doing, or ever heard of anyone doing this.
It is obviously a product of the AFCIs being back fed. I hope you did not fry the AFCI breakers.

Sorry, but I have too much common sense than to even attempt to make a male-to-male cord of ANY sort.
I know this will start a flame war but I don't care.

"Sorry, but I have too much common sense than to even attempt to make a male-to-male cord of ANY sort. "

I would also NEVER make up a male-male cord carrying AC household voltage. It's ridiculously dangerous.

No flame war here. I value your experience and opinion far too much. Making the special cord for this (and only this) purpose does not seem that dangerous to me - if we are very careful with it.

I am confused about one thing. If you "have NO idea why this happened" how can you say that is it "obviously a product of the AFCIs being back fed" ?

Testing at this stage does not require the presence of a breaker. I have not tested any circuit with an AFCI in this manner, never needed to do so, never plan to do so. So from my personal experience, I will never have an answer to your question.

"Making the special cord for this (and only this) purpose does not seem that dangerous to me - if we are very careful with it."

Tom, my worry wouldn't be with you using a male-male cord, it would be with a kid or uninformed adult finding it and plugging it in. It is impossible to have all of your tools and materials under 100% constant supervision at any jobsite or storage location. No flame war here either, just wanting to let other forum readers know that what you are suggesting and doing is not a safe practice.

I had some new fluorescents in bedroom closets that would not work. I wanted to test for power to the fixture and for proper equipment ground. My favorite tester is two 15 watt/120 volt lamps in series. With the AFCI breaker, placing the test leads on the "hot" and the ground resulted in an instant trip. That did indicate good ground. I do not know if the instant trip resulted from AFCI sensing or GFCI sensing. The fluorescent problem turned out to be a non-grounded ballast due to heavy paint coat on the fixture. It probably was the powder-coated process.

Oh, I do agree with that. We don't even have the breakers on at the temp power pole unless we are there.

tom, i think the point is that someone could plug in the wrong cord EVEN when you are there and thus energize the other end, resulting in someone getting hurt. for example, you have your cord run out to the temp pole, and a framer send his guy out there to plug up the compressor. guy sees 2 cords laying there, and not knowing which one, he plugs in both. you pick up the other end of your "dead" cord, and get your d*%* knocked in the dirt.

it is much safer to wire a regular cord to backfeed the main panel thru a couple of breakers and test that way. plus you won't have nuisance trips of GFI and AFI recepts and breakers. when you are done testing, unplug the cord AND the breakers and put a note at each end that only you can plug either up.

install a temporary twist lock male plug next to your panel. put a twist lock female on the cord, and then you can easily remove the cord when not testing. this would also give you the capability in the future of plugging in a temporary genny during a power outage. of course, if you leave it permanent you should have it properly configured so that the male plug can never have commercial power on it, interlocking breakers for manual transfer are made for this very reason.

"I am confused about one thing. If you "have NO idea why this happened" how can you say that is it "obviously a product of the AFCIs being back fed" ?"

Because you said youself you've done this before and it is only happening with AFCI breakers. This is straight from the original post.
Why with only the AFCIs? I don't know.

I will agree and stand with Bus Driver's post of 9:57 this morning.

Thanks for the help. I was not trying to be a smart aleck. It just seems to me that a nail or drywall screw through the wire somewhere between the breaker and the first receptacle could cause the same behavior. I suspect the AFCI breaker behaves this way because there is some sort of connection inside the breaker - be it a resistor or whatever circuitry the AFCI uses. With a regular breaker, no possibility of a connection exists because only the hot is connected to the breaker.

I plan to just check it out with a meter.

I never said this was the best way to check the circuits. It's just what I was doing at the time.

Thanks!

"My favorite tester is two 15 watt/120 volt lamps in series. With the AFCI breaker, placing the test leads on the "hot" and the ground resulted in an instant trip. That did indicate good ground."

You are using a ground current of only about 0.06 A.
The ground should reliably carry the full load current of the circuit and MORE to ensure tripping of the breaker if a short circuit occurs.

Test equipment is available that will apply full circuit current to the ground and measure the voltage drop, and some will even place overloads on the ground and measure the trip time.
This is usually only used in industrial settings to verify solid grounding.

My point was that this appears to be a low-cost, non-destructive, means of testing an AFCI in an operating circuit. Discovered accidentally. But that result was quite incidental and secondary to the process at that time of troubleshooting the non-functioning fluorescents. My assumption is that a ground sufficient to supply current for my tester is sufficient for the operating function it serves for a fluorescent light.

I don't think backfeeding an AFCI is a particularly good idea

If you are backfeeding recepticals with a male to male cord (which i agree with everyone else = bad bad bad) The breaker should have been turned OFF otherwise you are backfeeding the whole panel throught the AFCI breaker, in which case it will probably trip because there will be a load difference on the neutral since your panel is likely bonded to ground, yet you are introducing a potential difference on the neutral line, essentially creating a big ground loop.. no GFCI (of which an AFCI provides also GFCI protect, but not of Class A) will function if the neutral and ground wires come into contact downstream from the GFCI..

a dead short between hot and neutral or hot and ground within the circuit should be easily detectable with a simple ohm meter while the circuit is powered down.

I think you will need to replace your AFCI breakers now as you may have somehow shorted out the internal circuity in them

If you insist on continuing your testing method, remove the AFCI breaker from the circuit completely (disconnect its the hot, cap it off), also disconnect the neutral wire on this circuit as you are supplying a neutral with your illegal cord and try again. If you isolate the AFCI breaker out of the circuit and it still sparks you know its a wiring problem.

As others have suggested, hooking temp power to your panel is a much better idea. You could do it with a cord as well if you wish, and just test each side of the panel one at a time, you shouldn't actually need "real" power to test your circuits though, an ohm meter should tell you everything you need to know in terms of shorts.

Re: "Question: Is this happening because of the backfeed on the AFCI breaker? OR should we be looking for a wiring problem? "

Most likely because you are backfeeding the AFCI breaker. However, as suggested above feed the line side at the breaker panel to do your testing so that you can verify whether or not there is a wiring problem.

It is also likely that you destroyed the AFCI breaker by feeding the load side.

Attached is a block diagram for a AFCI breaker.

http://www.pfeiffereng.com/The%20Arc%20Fault%20Circuit%20Interrupter.pdf

If the breaker was open and you back fed the output side it is very likely you have destroyed the electronics.
The power supply is on the line side of the breaker.
You applied voltage to the electronics in the breaker without the power supply be on.
This is often results in damage to electronics unless very careful design steps have been taken to limit the currents that flow under the circumstances into the electronics.

Thanks. Can either of you (mike13, brickeyee) tell me how to test whether I destroyed the breaker? I checked the circuit with a meter and the wiring is fine. Across the breaker, I did get a reading. It was the same for the AFCI on the circuit that I was messing with as it was with the other AFCI breakers.

simple, feed the panel with a temporary feed as described above and check the breaker. if it works, then you did not destroy it. if it does not, then you did.

From September 2007 JLC

In the News
Expanded AFCI Requirements Spark Controversy
In June, the National Fire Protection Association which publishes the National Electric Code  voted to greatly expand requirements in the 2008 code for residential use of arc-fault current interrupters (AFCIs). These devices prevent fires caused by faulty wiring; unlike conventional circuit breakers (which trip on gross faults) or ground-fault circuit interrupters (which cut off power to a circuit if they detect an imbalance between the hot and neutral conductors), AFCIs trip in response to unintentional arcs in household wiring. Already, the more stringent rules are provoking debate within the industry  even though they won't take effect until January.

Back story. Requirements for AFCIs aren't new: The devices have been cropping up in the NEC for several years, ever since an amendment to the 1999 code mandated their use in all bedroom receptacles. The 2002 code extended that requirement to all bedroom outlets, including light fixtures, receptacles, and smoke alarms. And the 2005 code reduced current levels and required for the first time that the devices detect both series and parallel arcs. (Series arcs occur when the current jumps a gap, as when a wire is cracked; parallel arcs form when damaged insulation allows the current to jump between conductors, or from a conductor to ground.)

Now the pending 2008 code has gone even further, specifying combination-type AFCIs  the type capable of detecting both kinds of arc  in all 120-volt 15- and 20-amp circuits that supply "dwelling-unit family rooms, dining rooms, living rooms, parlors, libraries, dens, bedrooms, sunrooms, recreation rooms, closets, hallways, or similar rooms or areas."

Advocates tout fire prevention. Gerard Winstanley of the National Electrical Manufacturers Association (NEMA) contends that the new AFCI requirement will prevent numerous fatal fires. "I expect them to have a similar effect to smoke alarms on reducing fire deaths," he says. "That data will be collected 10 or 20 years down the line." He refers to figures on NEMA's consumer-information Web site (www.afcisafety.org) that ascribe 67,800 fires, 458 deaths, and $868 million in property losses to home electrical problems each year. (It should be noted, however, that estimates of losses from fires of electrical origin vary widely according to source; the Underwriters Laboratory Web site, for example, puts the figures at 32,000 fires, 220 deaths, and $674 million in damages, while a 2006 NFPA report cites 19,100 fires, 140 deaths, and $349 million in damages.)

According to Winstanley, the difference in cost between a conventional circuit breaker and a combination-type AFCI is typically about $15, meaning it should cost $200 or less to install a full complement of AFCI breakers in a typical new home. "That's a fairly small outlay," he says. "Once it's explained, consumers will accept them."

Opponents predict glitches. Other industry observers, however, are skeptical; they question not only NEMA's numbers but also its motivation, noting that the electrical manufacturers themselves are in the business of selling the mandated products. Independent NEC expert Mike Holt, for example, contends that AFCI technology is "not even close" to being ready for general use. Few electrical contractors, he says, have experience with the new combination AFCIs, making it impossible to know whether normal arc-producing events like distant flashes of lightning or the operation of motorized devices will lead to widespread nuisance tripping. "I'm totally convinced there will be a huge problem," he says.

The pending 2008 NEC will require AFCI circuit breakers  like the ones shown here from four manufacturers  to protect 15- and 20-amp circuits in most rooms of the house. Opponents claim the devices are not ready for general use, and that nuisance tripping will be a problem.

Moreover, Holt scoffs at NEMA's cost figures. "Look," he says, "if they're smart, electrical contractors are going to figure out their real costs and add some money for service calls to the job." He figures that the AFCI provision will cause most subs to up their bids by $500 to $1,000 per home. "Electrical contractors will be okay," he says, "but they don't have to sell that increase to the buyer. I still can't believe home builders didn't fight this more than they did." Â Jon Vara