AFCI – Cutler-Hammer Responds

Dear Mike:

I last contacted you on AFCIs by letter dated April 8th, with Attachments.

In recent weeks I note that your website has contained much material on AFCIs. This has included a selected Comment from the National Electrical Code, and minutes from a recent IAEI Chapter Meeting. It has also contained Square D material in support of the technology.

With respect to the Code, you published the negative Comment of Mr. Larry Brown of the National Association of Home Builders. Mr. Brown’s proposal to the 2002 NEC was defeated 11 – 1. In fairness to your readers, it would have been useful to note that there were many Proposals and Comments, both pro and con, during the 1999 Code cycle (ROP Proposals 2-128 through 2-130, ROC Comments 2-55 through 2-85) and during the 2002 Code cycle (ROP Proposals 2-101 through 2-120, ROC Comments 2-67 through 2-82). Cutler-Hammer considers that your readers would have been better served if you had directed them to the total package of Proposals and Comments rather than exposing them to an individual negative Comment. In particular, NEC Code Making Panel 2 studied all of the Proposals and Comments in depth during two Code cycles. These thorough deliberations resulted in the dwelling-unit bedroom-outlet requirement that appears in 210-12.

With respect to the minutes of the George Washington IAEI Chapter Business Meeting, the two-page article in the July/August 2002 IAEI News essentially summarizes the extensive notes made by the Chapter Secretary, David Shapiro. A more complete account of the meeting can be found on David Shapiro’s website Your readers might be interested in that more complete report of the March 19, 2002 meeting.

With respect to Square D, we were pleased to see that you had finally presented your readers with some solid technical information about AFCIs. As previously mentioned, Cutler-Hammer had sent you similar information by personal letter in April 2002 and we had also offered to host a visit to our Technology & Quality Center with a view to demonstrating and explaining the technology in detail. That visit has not taken place but the offer remains.

At this time, Cutler-Hammer chooses to reiterate the protection afforded by currently available UL listed AFCIs located at the load center (Branch / Feeder AFCIs.) For the total circuit, including cords beyond the outlet, they protect against all unwanted arcs to ground and against all unwanted arcing in parallel with the load. These Branch / Feeder AFCIs are tested to decrease the risk of fire associated with various types of arcing in accordance with the UL 1699 Safety Standard, First Edition, February 26, 1999 A Standards Technical Panel reviewed this standard in 2001 and UL 1699 is now an American National Standard (ANSI) with revisions current through July 2002. The Branch / Feeder AFCI tests include unwanted arcing in series with the electrical load in typical non-metallic sheathed cable (NM-B plus ground) installed wiring. In practice, this is accomplished by the detection of leakage currents to ground prior to cable combustion. The tests also include the detection of arcing to ground. In particular, the tests include the detection and interruption of parallel arcs in both NM-B and two conductor STP cords. It is these high-energy parallel arcs that are most likely to create a fire hazard. Branch/Feeder AFCIs do not detect series arcs in two conductor extension cords. These arcs are typically low current and short duration, and are far less hazardous than parallel arcs due to the subsequent low energy content. Distinguishing such arcs from normally occurring switching arcs would be extremely difficult. Low current arcs of long duration are highly unlikely in 120V circuits.

An important side-benefit of the ground fault detection (typically 30 to 100 milliamps) included in commercially available Branch / Feeder AFCIs is the significant protection against the high-resistance series faults that occur, for example, at loose connections or at wire-breaks that subsequently reconnect. The resulting heating (glowing connections) causes insulation degradation, with resulting parallel arcing or leakage current to ground that trips the AFCI. A UL Special Services Investigation for Cutler-Hammer entitled “Branch / Feeder AFCIs Incorporating Equipment Ground Fault Protection” (May 31, 2001) indicates the value of commercially available AFCIs in protecting against these high-resistance series-fault overheating situations.

AFCI technology has been made possible, both technically and economically, by the miniaturization of electronics. The development of the technology has been motivated by the responsible safety concerns of circuit breaker manufacturers and of organizations such as CEMA (EIA), NEMA, NFPA, NASFM, CPSC, UL, and CSA. The problem is real; with more than 60,000 electrical fires per year in dwelling units, with hundreds of deaths and with tens of thousands of injuries. The technology is real, and presents a real opportunity to reduce these appalling statistics. And it is websites such as yours that have a real opportunity to contribute to consumer education with a subsequent real improvement in fire safety. We trust that your future personal website-comments will reflect that opportunity.

Yours sincerely,


Manager, Applications and Standards


Mike Holt’s Comment: Mr. Kimblin thank you for your comments. By the way, I was reviewing a paper (Home Electrical Protection Beyond the Code- Arc-Fault Circuit Interrupters) by Dr. Joseph C. Engel, Manager, Electronics Research and Development, Cutler-Hammer, Technology and Quality Center

In reviewing this letter there were some statements that were made that frankly surprised me.

Page 2.“C-H research engineers identified eleven hazards [including the study] in residences that can cause either high-resistance series faults (which are not hazards in themselves, but can become one)”

Maybe I’ve been in the field too long but I’m not sure I would agree with this statement. My experience is that loose connections (which are not detected by an AFCI device) can create a fire and this condition is hazardous.

Page 6. “‘Dual-Listed’ breakers provide the highest level of protection against all of the eleven faults, regardless of the type of fault (series, arcing, or ground) or its location (in-wall or in-room wiring). For example, the breaker functions as follows when a glowing contact develops at a receptacle terminal in an outlet box:”

Okay the way I see it, the paper is saying that loose connections, which Dr. Engel calls glowing connections are not a hazard but they becomes hazardous the moment a ground fault or short circuit develops. Is this paper implying if I used a 20A fuse or standard circuit breaker, the protection device will not clear the fault before a fire but an AFCI will?

Are you telling me that there is no fire hazard with loose connections and that it’s only a problem when the short circuit or ground fault occurs? For all of these years, I thought the loose connections caused the fire and the short circuit or ground fault caused the protection device to operate, but not before the fire.

Appendix II: In reviewing your test, I wonder what would have occurred if you had preformed the same test using standard circuit breakers and fuses. I also noticed that all of the tests were done with the wires exposed and not contained in an enclosure in a wall, in other words not the real world.

I’m sorry Mr. Kimball, but I just feel the manufactures of AFCI devices are selling us something that does not solve a problem. All I see is a set of tests designed to give predicted results, and I would have had the same results with a standard circuit breaker or fuse.

I would be willing to visit your facility to see a demonstration where a non-AFCI protected circuit would not protect against a fire in each of the 11 conditions that CH identifies as a hazard. Then show me how the AFCI protection device would prevent the fire. I ask that this demonstration be open to all those that are interested in getting the facts.

If you can demonstrate where a fire can be created with a non-AFCI protected circuit and how an AFCI protected circuit would prevent that from occurring, then I have to tell you I will be your greatest supporter.

I’m tired of the technical mambo jumbo. I just want to see how this works in the real world.

Mike Holt

Copyright © 2002 Mike Holt Enterprises,Inc.
1-888-NEC-CODE (1-888-632-2633)