The practical problem with AFCI (Arc Fault Circuit Interrupter) devices
is that they are supposed to sense RF.
Yes, there is a RF detector in there.
AFCI devices are available in both AFCI only, and combo AFCI/GFCI.
Now, GFCI is is a great thing, and has been around long enough that they
generally know what they are doing. Makers know how to filter them
properly, and both codes (related to equipment standards) and the
equipment makers themselves have adapted.
A GFCI will happing trigger for both arcing and non-arcing current flow,
but only so long as the flow not along the planned path.
An in-path arc will not trigger a GFCI since the current flow is still
symmetric.
Al/Cu connections without buffer materials...
That overclamped AL connection (where the Al wire flows under pressure
and leaves a loose connection)
A wire-nut connection that was not twisted where an oversize wire-nut
was used.
Loose clamp screws
Corroded connections...
Failing (internally arcing) transformers
All of those and more provide the possibility of arcing/heating along
the planned current path, and a GFCI will do nothing about them unless
current is leaking to somthing outside the path.
The AFCI device is literally a RF detector, set to trip the breaker if
it sees what it thinks might be arc caused RF on the line.
First gen/cheap/many of these are a simple RF detector.
>From what I hear, newer, future, higher end development is attempting to
not just detect RF, but build what amounts to a broadband receiver and
profile analyzer to reduce false triggers.
At the end of the day though... Every single one of these is a receiver
set to pop the breaker if they detect RF. A very different situation
that you have with a GFCI.
I am not suggesting that they should not be used, or that arc detection
is a bad thing. I have seen many bits of houshold wiring that made me
run not walk to the panel to cut power.
But, as HAMs we need to keep in mind...
1) Every single one of these is a RF detector and is supposed to be.
2) We are very early in the development cycle for these things. It will
be many years before things like RF envelope/profile analyzing (to tell
the difference between an arc and a carrier) and similar improvements
are part of the std AFCI package.
On 1/5/13 9:15 AM, Jim Brown wrote:
> On 1/5/2013 9:40 AM, rick darwicki wrote:
>> Typical government/committe expansion of what was originally a good idea
>> (maybe).
>
> Two important clarifications here. First, virtually all building codes
> in North America are based on NEC, which is written by a group of EEs
> from industry. It's a very good code. NEC is adopted by local
> governments as their Electrical building code, which gives it the force
> of law. I don't know about you, but I consider building safety codes a
> VERY VERY good thing, and a very necessary function of government. They
> make our buildings safer, so that they can withstand earthquakes, are
> less likely to burn, so that cables in them don't create noxious fumes
> when they burn, so that burning cables in vertical runs in high rises
> don't spread fires between floors in a large building (both of the last
> two things happened in the REAL "towering inferno, and causes those
> provisions to be added to NEC.
>
> I can't comment on the virtues of Arc Fault Protection, since I don't
> know anything about it, nor the hardware failures that it is protecting
> against. But I can tell you that GFCIs are a VERY good thing -- they
> protect people from electrocution in rooms where use of a faulty
> appliance could kill them -- that is, if a person touches the appliance
> and grounded plumbing.
>
> 73,
>
> Jim K9YC
> _______________________________________________
> RFI mailing list
> RFI@contesting.com
> http://lists.contesting.com/mailman/listinfo/rfi
>
_______________________________________________
RFI mailing list
RFI@contesting.com
http://lists.contesting.com/mailman/listinfo/rfi
|