AFCI Headache

[ramsy]

I do not see how you could use a TDR in place of an AFCI since they have to be monitoring "Real Time". (not all faults are continuous, some are intermittent).

TDR's have pinpointed faults, while monitoring live cables carrying 60 Hz, for several years [6].. Its called Spread Spectrum Time Domain Reflectometry for Wire Fault Location.

IEEE published a nice SSTDR analysis, and designed an Application-Specific Integrated Circuit (ASIC), for live Arc fault mapping at 400 Hz. IEEE SENSORS JOURNAL, VOL. 5, NO. 6, DECEMBER 2005

The code rule tells us that we have to use the combination type

The code panels could mandate future improvements: with the ASIC economy of scale demonstration of SSTDR's --noise & data tolerant-- AFCI, GFCI, Overload, & Bolted fault monitoring in one place, potentially reducing all circuit breakers to inexpensive relays.

 

[ELA]

Ramsy,
Thanks for the interesting read.
Here is another link from an article at a later date that seems to indicate it is not so promising a technique.

http://www.ece.utah.edu/~cfurse/Publications/fray%20paper.pdf

here is the conclusion from that article:

"V. CONCLUSION
This paper provides simulations and measurements (TDR,
FDR, and SSTDR) for a variety of fray conditions. The most
significant observation of this paper is that frays on wires have a
reflectometry signature that is smaller than ordinary impedance
changes on the wire and are, therefore, going to remain invisible
to reflectometry methods. Even with further developments in the
methods that might be able to make them sensitive enough to locate
frays in idealized test environments, the normal impedance
variation in the environment of the cable is greater than the
impedance change due to the fault itself."

It is interesting to see that they have made attempts to make this work but from what I have read it seems elusive.

At any rate it would be expensive. My point about current AFCI breakers was that they are too primitive and could conceivably be made more reliable with more complexity and cost added.

Another quote from this article:

"The implications of continual monitoring of live wires include
some sizeable advantages as well as several challenges.
Methods to miniaturize SSTDR systems to enable their implementation
within aircraft circuit breakers, power control systems,
avionics boxes, connectors, etc. are currently under development.
Conversion of existing circuit boards to integrated
circuits will significantly reduce their size, cost, and power requirements.
Even so, the costs in size, weight, dollars, and complexity
of monitoring every wire in a legacy aircraft are probably
prohibitively large."

 

[jdoggg]

arc fault

arc fault

99% of the time when I have had a problem with the arc faults I ended up taking apart the whole circuit and usually finding a bare ground real close to the neutral when the receptical was pushed into the wall.

 

[ActionDave]

This is true, and was the source of almost all my Arc-Fault trouble shooting victories, however, I have seen with my own eyes an AFCI breaker trip only because of the equipment that was plugged into it.

 

[ramsy]

It is interesting to see that they have made attempts to make this work but from what I have read it seems elusive.
Use of Reflectometry for Fray Location
http://www.ece.utah.edu/~cfurse/Publ...ay%20paper.pdf

Amazing. Pinpointing a fray within troubled KAPTON & Poly-X aircraft wiring is nicer than pass/fail insulation tests, but replacing Megger testing with live SSTDR?

The article concludes SSTDR for hard faults is more reliable than for frays, within the miles of higher-noise/impedance, aircraft-wiring bundles, so we shouldn't throw our Megger testers away any time soon.

At any rate it would be expensive. My point about current AFCI breakers was that they are too primitive and could conceivably be made more reliable with more complexity and cost added.

No arguments here. Thanks for the article.

 

[GUNNING]

Back to basics...

Back to basics...

Back to basics. Wire temperature was not to rise beyond 5 Degree Celsius. That was the standard for the ampacity tables which we use to get our breaker sizing. Fuses were great because they opened when the amperage heated up the metal and opened the circuit.
Then breakers came along. Same principle, bi metallic strip, spring, and renewable.
Now we have a kludge that is designed by committee. A kludge is a computer with a box of wires hanging off the side of it to make it work better. Seems to me The horse that looks like a camel really isn't any better than a fuse that looks like a breaker.
KISS, Keep it simple stupid. If the AFCI's are not performing as stated lets relieve the end user of the additional $55 cost per breaker and put in that $3 breaker. Solid state is great but cost prohibitive. I think its called the tail wagging the dog or costs out weighing the benefits of this application.
We have had a lot of fires recently. Most of them have been from the mortgages getting too close to the fire insurance and causing spontaneous combustion. I haven't seen where a lot of fires are caused by electricity except by lightning and negligence. Besides the concept of AFCI's being high tech, so they must be better, why not go back to the tried and true ol Circuit Breaker? I'm sure the Manufacturers have recouped their initial costs. If not too bad. This product just doesn't work as advertised.
It' like putting out a fire by wetting the wood first and saying "see no fire!"
Where are the statistics of less electrical fires to prove these things are better.

 

[don_resqcapt19]

Back to basics. Wire temperature was not to rise beyond 5 Degree Celsius. That was the standard for the ampacity tables which we use to get our breaker sizing. ...

Where does the 5?C rise come from? It is my understanding that if you load a conductor to its rated ampaciy under the conditions stated in Table 310.16 that the conductor temperature under those conditions will approach the rated insulation temperature. That would be a 30?C rise for a conductor rated at 60?C.

 

[ramsy]

Now we have a kludge that is designed by committee. ..a computer with a box of wires hanging off the side..

A kludge that thinks loads are faults, or ignores all faults when simple power strips with RFI-surge filters masks the signal.

Seems to me The horse that looks like a camel really isn't any better than a fuse that looks like a breaker.

We have had a lot of fires recently. Most of them have been from the mortgages getting too close to the fire insurance and causing spontaneous combustion.

With any surge filter they can jab the cables, burn down the building, then burn the contractor's license, bonding, & future insurability.

UL listed Surge-Protected outlet w/ RFI-signal mask

 

[ELA]

With any surge filter they can jab the cables, burn down the building, then burn the contractor's license, bonding, & future insurability.

UL listed Surge-Protected outlet w/ RFI-signal mask

Ramsy,
Just "any Surge filter" will not do. It must incorporate an EMI filter that can filter out frequencies below 100 khz. That is usually a pretty good quality filter.

Where did the "w/RFI-signal mask" text come from in your link?

Do you have first hand experience using the Series 140 ? Can you point me to a data sheet?

If this device incorporates an EMI filter sufficient to mask the RFI components required by an AFCI in order to detect an arc it could be handy to those experiencing issues with nuisance trips.

 

[Mgraw]

http://www.leviton.com/OA_HTML/ibeCCtpItmDspRte.jsp?item=134520&section=10355&minisite=10027

 

[ELA]

Thanks for the data Mgraw.
-30dB @ 500Mhz is not a particularly good EMI filter unless the offender is very high frequency.

Anyone had experience using it? If so what situation did it cure?

 

[ramsy]

Where did the "w/RFI-signal mask" text come from in your link?

See link to published brief, this thread, post #47.
re: In 1997 Square D published a testing brief of AFCI's that used RFI filters to mask the signal.

RFI specs. appear on product listings rated in (db).
30-80db is common for surge protectors with RFI filters, most likely used in the Square D study above, and the RFI-surge duplex available at any home center, shown in the previous post.

My apologies for missing if that Square D report also used EMI filters for masking the AFCI.

 

[ELA]

I use EMI and RFI filter somewhat interchangeably, I was not trying to make any point about a difference there.

What I was interested in was the complete specification required for a device to be considered an effective "RFI-signal mask" for AFCIs.

The series 140 is not a very effective filter for lower frequencies (below 100Khz). It is not enough to simply list a range of dB for an EMI/FRI filter.
The specification must include the frequency range as well.

I suspect that any EMI filter used in the Square D test that effectively masked AFCI operation had a good attenuation figure below 100Khz.