I am guessing they were just trying to avoid some nuisance tripping by making the GF protection higher. I do belive most of the real protection is coming from the GFP protection, not the snake oil in arc detection. Ground fault protection is a very simple technology and it responds to the first fault, not having to sort out locked rotors and inrush.
I dug out some notes and reread a couple sources. The sources are:
Series arcs should be able to be detected at a 5A arc level. (Sounds like a major challenge.)
Branch/Feeder AFCI Required for bedrooms in the 2002 NEC. Install as a circuit breaker (or at the panel) and detects arcs on branch circuits and feeders. They also detect arcs on cord sets plugged into a receptacle. The arc detection is at the 75A level so it detects parallel arcs. But there is also a test for "carbonized path" ?series? arc on Romex. Since Romex has a ground wire, I believe this is found by the ground fault detection, which IIRC is required to be 50mA but is commonly 30 mA. If the arc was H-N, it would fairly rapidly also become a ground fault. (UL did limited tests of "glowing connection" series arcs on grounding type receptacles, and about half eventually tripped an AFCI from the ground fault current that developed.)
Outlet circuit AFCI Installs as a receptacle and may have wire through (like a GFCI receptacle). Arc detection is at the 5A level, so they will detect series and parallel arcs. Can detect upstream series arcs.
Combination AFCI The 2005 NEC requires all AFCIs installed after 1-1-08 to be Combination type. Installs as a circuit breaker and combines the protection of Branch/Feeder and Outlet Circuit. Will detect series arcs.
are:
Thank you for the excellent references. (I guess I am going to renew my IAEI membership again.) The IAEI article was excellent. There are a lot of rumours and misinformation about AFCI's and it is going to take some convincing arguments to convince people the 2008 NEC is correct.
Ground fault protection at "moderate" (as compared to 15/20/30 amp currents could likely be accomplished with 100% electro-mechanical stuff (just like the CBs) if you don't load on the requirement to detect G-N "shorts."
If the folks writing the NEC really want to install "smart" CB boxes perhaps a "ground up" solution can be devised.
I believe that about 20 years ago some folks were talking about "smart" wiring whereby an appliance would not receive any juice until the "smart wiring" determines that it's "safe" to so do. Does anyone know what happened to that?
Back to the AFCIs: the more I hear about these things the more convinced I am that there is a "hidden agenda." Not of the actual "fact" make sense.
I don't remember if it was obvious, but the IAEI article has 3 parts.
There are 3 posters in this thread (including me) that think the 2008 NEC is wrong.
PolyTech Forum website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.