My experience also, and IMO good reason why connections/splices are supposed to be in junction boxes or where open conductors are allowed they must have clearances to other objects, like the connections to a service drop for example.
I have never accepted fire and smoke as being a real world issue. Just part of the ongoing code fights between metal clad cables and raceways and non-metallic cables and raceways. When you step back and look at the finished and furnishings of the building as compared to the mechanical and electrical system, you soon understand that the mechanical and electrical systems are such a small part of the total fuel load, it really does not make much difference.
The stats used to force the AFCIs into the code said that 40% of the electrical fires are caused by the fixed wiring of the building.
Maybe they took 1000 or even 10,000 incidents, then found 40 that were caused by issues in fixed wiring, then picked 60 others at random and then said those 100 were what was evaluated:blink:
So, if they meant 40% of the total 9% that would mean only 3.6% of house fires were caused by premise wiring, pretty low in my book.
If you were an AFCI manufacturer you think mentioning 3.6% to code making panels is as convincing at getting them to accept requiring installation of your product as mentioning 40%, even though you didn't clarify exactly what it is 40% of?
The numbers came from the NFIRS (National Fire Incident Reporting System). These are the fire cause reports submitted by fire departments.
Depending on whose numbers you are looking at, the percentage of dwelling unit electrical fires attributed to electrical is between 9 and 15, but yes the number of those fires caused by the fixed wiring of the structure would be 40% of that.
The AFCI people never said that their device could only prevent 40% of the dwelling unit electrical fires. They strongly implied that they could prevent the vast majority of them. Note that the original proposals pretty much said that the original branch circuit and feeder device could do what they now tell us that the combination AFCI can do....the only problem with that is that the combination device did not exist until some 13 years after the original proposals.
But did they clearly state 40% of 9 to 15% or did they put more emphasis on getting that 40% in CMP's minds.
The proposals in support of AFCIs never really brought up the 40% number. It is just a number that you can find when you research the same data that they used to get the AFCIs into the code. I believe the number was brought up in some of the comments on the proposals.
Certainly not conductors/cables, other then at a termination point that got hot for whatever reason. I hear a lot of emphasis put on undersized conductors at times - reality is I don't think that in itself causes as much trouble as many think, I think most the issues it does cause is at terminations, but if those are inside an enclosure like they are supposed to be - that greatly reduces fire hazard.
