AFCI Success Story with Knob & Tube (well, mostly)

[brycenesbitt]

There's a lot of bad press on AFCI on this forum, I though I'd share a semi-success story:

In a 1930's construction home a double-pole AFCI was added the legacy multiwire shared neutral Knob & Tube circuit. The AFCI trips immediately, and the sparky replaces it with a regular breaker. But on second look two things are found:

1) Plumbers added a circuit, and managed to wire nut in both sides of an unused 24 volt doorbell wire. This 24 volt wire then sat in the basement just waiting to spark.
2) One K&T neutral is interconnected to a modern breaker neutral. And of course it's all done wrong. The bad wiring then goes and actually sparks.

Thus the AFCI lead to the discovery of two sparking wire segments. Sort of.
Once the faults are cleared the original AFCI was reinstalled, and worked fine with no callback.

----
Without disagreeing with much of the bad AFCI sentiment, I can see the merit of AFCI for knob & tube, given the way wires enter metal junction boxes. Some vendors seem to agree "Eaton recommends the use AFCI in older homes with knob and tube wiring systems. Contact Eaton's Residential Application Team for assistance".

Those complaining of AFCI false trips perhaps could submit reports to: http://afcisafety.org/report.html

 

[templdl]

Would it possible for you to post a one line or wiring diagram of the location of the arcing failure in relationship to the AFCI?
Personally, I have been a big proponent of the AFCI. However I have become discouraged that there have been little or no reports of AFCIs preventing fires which is why I'm very interested in your post.
Your description appears to describe that there is a 24vac doorbell transformer involved that's why I asking for a circuit diagram so that I can better understand your description of the event.

 

[GoldDigger]

Would it possible for you to post a one line or wiring diagram of the location of the arcing failure in relationship to the AFCI?
Personally, I have been a big proponent of the AFCI. However I have become discouraged that there have been little or no reports of AFCIs preventing fires which is why I'm very interested in your post.
Your description appears to describe that there is a 24vac doorbell transformer involved that's why I asking for a circuit diagram so that I can better understand your description of the event.

I think there was no transformer. Just abandoned doorbell wire which was open on both ends until one end got connected to 120V.

 

[ptonsparky]

The big question is "What caused it to trip?" Overcurrent, AF, GF, or a bootlegged neutral. Small but Oh so important details that templdl is asking for.

 

[GoldDigger]

The big question is "What caused it to trip?" Overcurrent, AF, GF, or a bootlegged neutral. Small but Oh so important details that templdl is asking for.

Absolutely! Particularly whether there really was an arc or just a current mismatch that tripped the GFPD function of the breaker.
It is fairly difficult for a crossed neutral problem to cause arcing, unless it is a mechanically bad connection too.

 

[templdl]

I think there was no transformer. Just abandoned doorbell wire which was open on both ends until one end got connected to 120V.

That took some brains.

 

[GoldDigger]

That took some brains.

The OP did say that they were plumbers.

 

[templdl]

The OP did say that they were plumbers.

S--t does run down hill.Knowing that you must have a 1/4"/ft grade as I can recall is what needed.

 

[PetrosA]

I think one of the main reasons insurance companies like AFCI systems is that they'll usually trip on wiring mistakes after installation. We don't have a certification system here like many other countries do so something like an AFCI breaker is an alternative to that. Once the system is up and running without bad splices, it "should" continue to be safe. Underwriter happy.

 

[templdl]

I think one of the main reasons insurance companies like AFCI systems is that they'll usually trip on wiring mistakes after installation. We don't have a certification system here like many other countries do so something like an AFCI breaker is an alternative to that. Once the system is up and running without bad splices, it "should" continue to be safe. Underwriter happy.

Strange that I was a Beta sight in 1998 for AFCIs and......one tripped.
After ringing out the 20a circuit I found an EGC that was touching the neutral screw of a duplex outlet which had been there since the house was built. After being corrected I have never had on trip since.
I attribute that to a wiring error. Some may think this error was no big feel but when a neutral comes into contact with an EGC other than the SE it makes those neutral current available of any electrical device in that structure that uses the EGC as a ground.
The EGC had been allowed to become a current carrying conductor by sharing the neutral current..

 

[kwired]

Strange that I was a Beta sight in 1998 for AFCIs and......one tripped.
After ringing out the 20a circuit I found an EGC that was touching the neutral screw of a duplex outlet which had been there since the house was built. After being corrected I have never had on trip since.
I attribute that to a wiring error. Some may think this error was no big feel but when a neutral comes into contact with an EGC other than the SE it makes those neutral current available of any electrical device in that structure that uses the EGC as a ground.
The EGC had been allowed to become a current carrying conductor by sharing the neutral current..

But GFCI technology had been in place for quite some time and would also have found the same problem. You can not necessarily claim the AFCI technology uncovered this problem.

 

[cowboyjwc]

But GFCI technology had been in place for quite some time and would also have found the same problem. You can not necessarily claim the AFCI technology uncovered this problem.

Yeah, but you would have only found it with GFCI if it was on a circuit that was protected by GFCI.

 

[kwired]

Yeah, but you would have only found it with GFCI if it was on a circuit that was protected by GFCI.

Well he wouldn't have found it without the AFCI either which probably had some extent of GFCI combined within it.

 

[cowboyjwc]

Would it possible for you to post a one line or wiring diagram of the location of the arcing failure in relationship to the AFCI?
Personally, I have been a big proponent of the AFCI. However I have become discouraged that there have been little or no reports of AFCIs preventing fires which is why I'm very interested in your post.
Your description appears to describe that there is a 24vac doorbell transformer involved that's why I asking for a circuit diagram so that I can better understand your description of the event.

How would you know if it stopped a fire? The problem may have been there for years before it started a fire. An AFCI may not stop a fire even if it works like it's supposed to. If the event was enough to trip the breaker it may have been enough to spark the fire, under the right conditions. So even though the breaker trip you could still have a fire.

You also will never know how many fires were prevented by AFCI, the same way you will never know how many people were not electrocuted because there was a GFCI. Those events are not reported as a rule. You only get the report if there's a fire or a death.

 

[kwired]

How would you know if it stopped a fire? The problem may have been there for years before it started a fire. An AFCI may not stop a fire even if it works like it's supposed to. If the event was enough to trip the breaker it may have been enough to spark the fire, under the right conditions. So even though the breaker trip you could still have a fire.

You also will never know how many fires were prevented by AFCI, the same way you will never know how many people were not electrocuted because there was a GFCI. Those events are not reported as a rule. You only get the report if there's a fire or a death.

I agree. Now if we start to notice there are fewer fires or electrocutions we may be able to relate them so some extent to AFCI or GFCI, but it is hard to tell if a specific installation was saved from such an incident at some time.

Just like the controversy over whether seat belt use saves lives, sometimes people are killed in a car crash even though they were wearing a belt. Other times people are not wearing a belt and survive a crash. Occasionally someone buys the winning lottery ticket also, even though all that bought a ticket paid same price for the ticket.

 

[templdl]

But GFCI technology had been in place for quite some time and would also have found the same problem. You can not necessarily claim the AFCI technology uncovered this problem.

I didn't claim AFCI technology did. I stated that the AFCI discovered it. It is obvious that the GFCI element sensed it. The reason that the AFCI was given the opportunity to detect my EGC to N issue was that I had been give the opportunity to be BETA sight for AFCIs. I would have never applied a GFCI and if it wasn't for the AFCI the problem would have never been detected. The requirement for AFCIs use forces the use of GF detection in areas that previously never required it such as GFCIs are used.
I am well aware that the AFCI is more likely to detect an arcing fault to ground as that appears to be more likely to occur anyway.

 

[templdl]

How would you know if it stopped a fire? The problem may have been there for years before it started a fire. An AFCI may not stop a fire even if it works like it's supposed to. If the event was enough to trip the breaker it may have been enough to spark the fire, under the right conditions. So even though the breaker trip you could still have a fire.

You also will never know how many fires were prevented by AFCI, the same way you will never know how many people were not electrocuted because there was a GFCI. Those events are not reported as a rule. You only get the report if there's a fire or a death.

I think that you got my point. GFCIs are more likely to be recognized as their use to be imperative but it is very difficult to show similar justification for the AFCI in the prevention of fires. If an AFCI were to have tripped why wouldn't the cause of trip be explored where it may be discovered that it prevented a fire.
Reporting such events may not be commonly done but the AFCI is such a unique device that if a the prevention of a fire was attributed to the AFCI it most likely would be publicized and we would know about it. With as many installations out there, there should be many more positive comments about it.

 

[kwired]

How many times have GFCI's tripped, and yet we really don't know just how many of those times actually saved a life had there not been a GFCI there? You may notice there is little or no electrocution incidents where GFCI is involved, but it will take some time for those stats to become obvious, which the GFCI has been around long enough that those stats probably are there, but AFCI has not been around long enough to see any significant results. One may be able to find more cases where AFCI did not do what it was intended to do at this point than you can find cases where you can claim it definitely saved lives or property.

 

[PetrosA]

How many times have GFCI's tripped, and yet we really don't know just how many of those times actually saved a life had there not been a GFCI there? You may notice there is little or no electrocution incidents where GFCI is involved, but it will take some time for those stats to become obvious, which the GFCI has been around long enough that those stats probably are there, but AFCI has not been around long enough to see any significant results. One may be able to find more cases where AFCI did not do what it was intended to do at this point than you can find cases where you can claim it definitely saved lives or property.

At least with GFCIs, there's a person around to say "Yeah, I got shocked but then the power shut off." At that point we can have a huge amount of certainty that the GFCI did it's job. I kind of hope we do start finding proof that AFCIs are failing, but the likelihood is low as so many of the danger bases are covered in modern homes by updates to the NEC and other building codes - extension cords, socket multipliers, old appliances, space heaters, dried up wiring, etc. are all at a minimum or non-existant in new homes and new homes evidently burn so fast and so hot that there's very little evidence left to examine after the fact anyway. Boooooo.

 

[templdl]

How many times have GFCI's tripped, and yet we really don't know just how many of those times actually saved a life had there not been a GFCI there? You may notice there is little or no electrocution incidents where GFCI is involved, but it will take some time for those stats to become obvious, which the GFCI has been around long enough that those stats probably are there, but AFCI has not been around long enough to see any significant results. One may be able to find more cases where AFCI did not do what it was intended to do at this point than you can find cases where you can claim it definitely saved lives or property.

And that's what concerns me about AFCIs. In theory as may be verified in the lab AFCIs show good results. But even though are they now required there has been little or no indication from the field that they are effective as are GFCIs.
Being involved with the release of the AFCI into the market I was excited about the technology but have been extremely disappointed that I haven't seen any real indication as to their actual contribution to the prevention of fires and saving lives as they have been on the market for more than 10 years now. When you consider the installed base now and the hundreds on thousands of dollars in expense I was looking for a few heroic stories about them except the countless complaints about nuisance tripping.