AFCI (yes again)

[don_resqcapt19]

...
Is there any historical data out there yet that shows less electrical related fires in homes with AFCI's? That will be about the only thing that will convince me they are worth their extra cost. Do insurance companies offer lower insurance rates if you have AFCI's installed? That would indicate they see them working as well.

It will be many years before you can prove or disprove that. If you run the numbers for fires that the AFCI would prevent and use the fact that 85% of the dwelling unit fires that are said to be of electrical origin occur in dwelling units at least 20 years old, you will find that the AFCI, even assuming that they actually would prevent 100% of the fires on the protected circuits, that you are preventing a very very small number of fires.

I ran the numbers for the first year of full compliance with the 2014 NEC as far as AFCIs. Based on an expected construction of just over 1,000,000 new dwelling units, you could expect the AFCIs to prevent 55 fires. Note that if you assume the additional costs for the AFCIs to be $600 per new dwelling unit, you have spent 600 million dollars to prevent 55 fires.

Of course as time goes by more fires are prevented each year, but even at the end of a 20 year time frame, with the construction of 21 million dwelling units, the total number of dwelling unit fires that you could expect the AFCI to have prevented, assuming 100% effectiveness of the AFCI, would 13,900 fires. Even if I keep the AFCI cost flat at $600 per dwelling unit, the cost to prevent each of those 13,900 fires would be about $925,000. The average dwelling unit fire loss for an unsprinklered dwelling unit is just short of $50,000.

 

[mbrooke]

One thing this thread will not bring up is how do we know AFCIs provide protection to begin with? First off I have heard of electricians doing bench top tests and nothing trips them. They can hold long standing series or parallel arcs, some will trip others will not. It seems that AFCIs are hit or miss, more like make all arc signatures un-concerning so nuisance tripping goes down.



As I have said before, to accurately detect arcs you need the computing power of at least an I phone if not a computer. AFCIs do not use mathematical morphology algorithms let alone high sampling rates, and that's just beginner's terminology. I have seen real arc fault logic used in utility relaying intended for reclosers and Medium voltage feeder protection. Once you see the anatomical/theory part behind it without a doubt wishful thinking in AFCI breakers becomes blatantly obvious.

To expand on the above: The irony behind arc fault logic (sensitive earth fault detection in MV & HV) actually came from not being able to use low zero sequence current/neutral CT settings for line and feeder protection because the grounding system is used to carry neutral current from primary wye grounded distribution transformers. The result from this is often down conductors going undetected creating an extreme hazard as well as systems reliability concerns. The solution came about from a long known fact that most downed conductors will produce an arcing signature which differentiates from others. The end result was relays that could sense several amperes of arcing signature on a feeder regardless of zero sequence (GFCI) values and trip when needed.

In so far this technology does not nuisance trip and can even give you a darn good co computed location of the fault.



While I am comparing a domestic service to a medium or high voltage utility system the concept is still there. It takes real logic to discriminate between a motor and an arc, in addition to the fact that most dangers/wring errors/ect are removed from 30ma GFCI protection, not arc signature analysis.


Second part Id like to add is surge tolerance in AFCIs. I have heard from others that AFCIs will fail over 126 volts. Lightning or even surges can do the same. How do we know that 20 years from now half of these AFCIs will still be functioning?


The circuit boards are just protected with a mediocre MOV, and I highly doubt these breakers are designed beyond marginal.


To someone who doesn't understand electrical theory or has not seen what most of use might call exotic technology (its not) reach the conclusion AFCIs are some type of miracle. They are not, they are a deceptive smoke and mirror scam. AFCIs are a fraud. I can think of bigger concerns, but with profit dwindling in the residential sector why not take a skewed statistic, inflame it, pass some covert money and reach financial gain?


I don't believe the 30ma GFI logic in itself is a farce, if anything it has caught sloppy electricians. AFCI logic on there other hand is, which is way to primitive to be code required. And I know CMP has little concerns over 30ma logic in that some manufacturers are even taking it out rendering what good AFCIs might bring giving something totally worthless to the end user.

 

[mbrooke]

Of course as time goes by more fires are prevented each year, but even at the end of a 20 year time frame, with the construction of 21 million dwelling units, the total number of dwelling unit fires that you could expect the AFCI to have prevented, assuming 100% effectiveness of the AFCI, would 13,900 fires. Even if I keep the AFCI cost flat at $600 per dwelling unit, the cost to prevent each of those 13,900 fires would be about $925,000. The average dwelling unit fire loss for an unsprinklered dwelling unit is just short of $50,000.

I respect that But that is assuming they will work. (Which I know you brought up)

But What if they aren't? What if they fail? And lets not forget AFCIs stress wiring to the max. How so? To offset cost contractors are now wiring to code minimum 3va per foot if not less where inspectors don't catch it. How is that moving forward in safety? In Canada homes have double the if not triple the 120 volt circuits. Outlets are limited to 12 (If I remember correctly)for a 15 amp circuit. In France its some think like 8 outlets per 20amper circuit which translates to something like 4800 watts per room (see link) If you ever see a French electrical panel in an American size home: :jawdrop: Yet AFCIs are pushing wiring back in time. :rant:

 

[MasterTheNEC]

To someone who doesn't understand electrical theory or has not seen what most of use might call exotic technology (its not) reach the conclusion AFCIs are some type of miracle. They are not, they are a deceptive smoke and mirror scam. AFCIs are a fraud. I can think of bigger concerns, but with profit dwindling in the residential sector why not take a skewed statistic, inflame it, pass some covert money and reach financial gain?

You have GOT to be kidding me....You are SO technically advanced above all developers and producers of AFCI's that you can make that statement?......All you have provided is your OPINION and then make outlandish statements to that fact. Everyone always say...."I know a guy"...."who did some tests"....but have YOU ever done the so called TESTS?

 

[kwired]

You have GOT to be kidding me....You are SO technically advanced above all developers and producers of AFCI's that you can make that statement?......All you have provided is your OPINION and then make outlandish statements to that fact. Everyone always say...."I know a guy"...."who did some tests"....but have YOU ever done the so called TESTS?

Sorry I have to ask, have you done the tests?

Do you trust they are being 100% honest with you when they have a financial reason to push their product.

I tend to trust third parties that intend to remain neutral on opinions and get paid the same no matter how their results come up.

Is the technology there to detect some things that can be a hazard - probably. Do they detect things that are considered "normal" and fail to discriminate the condition which ultimately causes undesired tripping? Seems as though they do. Do consumers get tired of this "nuisance" tripping and have the device replaced at some point with one that doesn't provide any "advanced" protection abilities? Yes. How does that prevent the problems these devices are supposed to protect us from, and who got extra money in their pocket from all of this? Who convinced the code panels to make this a rule to put these into code? The idea is great, just don't put a requirement in code before the product is finished is all I am asking for. They still are not finished, and the consumers are a part of the R&D lab and the consumers are paying for R&D, which I guess they ultimately will anyway but in this case they are also the lab rats, plus they had to pay for the devices that are being tested in their own homes:roll:

 

[Dennis Alwon]

My bet is the electricians who did the test do not know that an afci needs to see about 4 or 5 amps before it will trip under af conditions. Unless the electrician has experience with this type of testing I would not count on it. I have seen afci trip when they were suppose to trip. As Don has stated many times we often don't know if the gfp caused the trip or the afci. GE does not have GFP in their afci so if they trip it is a good guess it would be under afci conditions.

I have heard rumor that the NEC may require gfp to be put back in the afci. Not sure about that

 

[kwired]

I have heard rumor that the NEC may require gfp to be put back in the afci. Not sure about that

I honestly believe it should be listing standards that should make such a requirement, not the NEC, but I guess NEC does cover design of several other items as well. Any sub article titled "construction" kind of does this but IMO this should be up to listing requirements and any such "construction" articles should only apply to items that are not/not required to be listed.

 

[Dennis Alwon]

I honestly believe it should be listing standards that should make such a requirement, not the NEC, but I guess NEC does cover design of several other items as well. Any sub article titled "construction" kind of does this but IMO this should be up to listing requirements and any such "construction" articles should only apply to items that are not/not required to be listed.

It does not much matter who does it to me. The NEC could put pressure on the manufacturer to do it-- they do that all the time. Let's face it the NEC is not really what they claim- it is becoming more of a design manual as time goes on.

 

[MasterTheNEC]

Sorry I have to ask, have you done the tests?

Do you trust they are being 100% honest with you when they have a financial reason to push their product.

I tend to trust third parties that intend to remain neutral on opinions and get paid the same no matter how their results come up.

Is the technology there to detect some things that can be a hazard - probably. Do they detect things that are considered "normal" and fail to discriminate the condition which ultimately causes undesired tripping? Seems as though they do. Do consumers get tired of this "nuisance" tripping and have the device replaced at some point with one that doesn't provide any "advanced" protection abilities? Yes. How does that prevent the problems these devices are supposed to protect us from, and who got extra money in their pocket from all of this? Who convinced the code panels to make this a rule to put these into code? The idea is great, just don't put a requirement in code before the product is finished is all I am asking for. They still are not finished, and the consumers are a part of the R&D lab and the consumers are paying for R&D, which I guess they ultimately will anyway but in this case they are also the lab rats, plus they had to pay for the devices that are being tested in their own homes:roll:

Actually...I have (Dennis can privately elaborate how I might have been privy to such tests)...I have been in the lab in controlled environments, witnesses the tests and in fact know more about their limitations than most probably do. Let's just say I have been blessed to be close enough to those who do these types of tests on a regular basis in a lab designed specifically for such test. Now, I will stop at that because to me it is not if the AFCI's are 100% perfect....no but then again when something trips i never would call it a "nuisance" trip, if they trip their is a reason and regardless of that reason it is investigated to determine the reason but just labeled as a "nuisance".

I label my SON as a "nuisance"...but have no plans on replacing him or turn him in as defective:angel:

 

[MasterTheNEC]

My bet is the electricians who did the test do not know that an afci needs to see about 4 or 5 amps before it will trip under af conditions. Unless the electrician has experience with this type of testing I would not count on it. I have seen afci trip when they were suppose to trip. As Don has stated many times we often don't know if the gfp caused the trip or the afci. GE does not have GFP in their afci so if they trip it is a good guess it would be under afci conditions.

I have heard rumor that the NEC may require gfp to be put back in the afci. Not sure about that

I am sure the NEC can't do that...but UL 1699 can.

As for GE, they use a specific algorithm to replace the GFP function and pass the UL 1699 test criteria. UL 1699 only cares about passing the test....not how they design the AFCI devices.

 

[kwired]

but then again when something trips i never would call it a "nuisance" trip, if they trip their is a reason and regardless of that reason it is investigated to determine the reason but just labeled as a "nuisance".

I understand that, and for other items like motor overloads or GFCI's I generally can find a solution to the problem. AFCI's just don't work out that way, and if I can't find a solution I can be stubborn and refuse to remove/replace the AFCI and tell the HO that his appliance is very likely defective - and maybe it is but I am not necessarily able to prove it, what stops him from thinking I am just full of it (and expensive for no more then I did for his needs) and then Mr. handiman or somebody else comes in removes the AFCI and because there is no more inconvenience trips, in the HO eyes that guy is now a genius

 

[Dennis Alwon]

I am sure the NEC can't do that...but UL 1699 can.

As for GE, they use a specific algorithm to replace the GFP function and pass the UL 1699 test criteria. UL 1699 only cares about passing the test....not how they design the AFCI devices.

Didn't the NEC require the afci to be both parallel and series??? How is this different

 

[don_resqcapt19]

...
As for GE, they use a specific algorithm to replace the GFP function and pass the UL 1699 test criteria. UL 1699 only cares about passing the test....not how they design the AFCI devices.

Eaton has also removed the GFP from one of their lines of AFCIs.


As you said before the UL standard does not and never has required the GFP function, however, in my opinion, that is the most effective part of the device and it needs to be required by the standard.

The manufacturer's have been very resistive at both the NEC and UL standards making processes to this requirement.

 

[GoldDigger]

All the NEC needs to do is require Class B or better GFI for all circuits required to be AFCI protected. That way it does not have to part of the AFCI breaker. But market forces will lead to the same result.

 

[MasterTheNEC]

Didn't the NEC require the afci to be both parallel and series??? How is this different

No, the NEC demanded a "combination" device that by proxy gives you the parallel and series protection. I would assume that if UL 1699 is lobbied as such to UL 1699 and once it went to the CSDS process then it would make it into the standard. At this point all we do know is that as long as they pass the testing criteria set out in UL 1699 they are good to go.

Why should be manufacturer go above the minimum standard?....many electricians go their entire life just meeting the minimum standards of the NEC.

 

[MasterTheNEC]

All the NEC needs to do is require Class B or better GFI for all circuits required to be AFCI protected. That way it does not have to part of the AFCI breaker. But market forces will lead to the same result.

That would be a "Combination PLUS" device....lol......

 

[mbrooke]

You have GOT to be kidding me....You are SO technically advanced above all developers and producers of AFCI's that you can make that statement?......All you have provided is your OPINION and then make outlandish statements to that fact. Everyone always say...."I know a guy"...."who did some tests"....but have YOU ever done the so called TESTS?

Nope I am serious. The fact you are saying my statements are "outlandish" obviously means you have no clue what I am talking about. I have seen and worked with AFCI technology far more advanced than the ones in residential breakers. I don't have to give an opinion because I personally have worked on projects, seen it applied, and spoken to engineers who work with this type of AFCI logic. I take it you never have, or even seen the equations?

Its ok if you haven't heard of any of this (heck tell me anything about nuclear engineering or medicine and it will go right over my head), but to say something that is real is just outlandish is out of place.

Granted, yes, the bench top tests I mentioned are not scientific in this argument holding much less weight. But still, it gives me something to ponder over. The rest is black and white fact that can be Googled supporting my opinion that AFCI are to premature to be in code.

 

[MasterTheNEC]

Nope I am serious. The fact you are saying my statements are "outlandish" obviously means you have no clue what I am talking about. I have seen and worked with AFCI technology far more advanced than the ones in residential breakers. I don't have to give an opinion because I personally have worked on projects, seen it applied, and spoken to engineers who work with this type of AFCI logic. I take it you never have, or even seen the equations?

Its ok if you haven't heard of any of this (heck tell me anything about nuclear engineering or medicine and it will go right over my head), but to say something that is real is just outlandish is out of place.

Granted, yes, the bench top tests I mentioned are not scientific in this argument holding much less weight. But still, it gives me something to ponder over. The rest is black and white fact that can be Googled supporting my opinion that AFCI are to premature to be in code.

They are claims of a single or select group.....Defined as Outlandish in terms of an individual making a global statement for a much bigger group of people than a few electricians who have sour grapes over the technology. That kinda of Outlandish...

Also it appears you missed my post....I have talked to plenty of them......even been in the testing labs used to produce the very tests that UL 1699 is confirmed on. Yeah......I have seen my own tests and I think some would like a device to do something it was not designed to actually do...100%

Just Sayin

 

[MasterTheNEC]

They are claims of a single or select group.....Defined as Outlandish in terms of an individual making a global statement for a much bigger group of people than a few electricians who have sour grapes over the technology. That kinda of Outlandish...

Also it appears you missed my post....I have talked to plenty of them......even been in the testing labs used to produce the very tests that UL 1699 is confirmed on. Yeah......I have seen my own tests and I think some would like a device to do something it was not designed to actually do...100%

Just Sayin

HOWEVER...being that everyone has opinions and I prefer to NOT get into a chest beating...My lab is better than your lab situation..lol.....Lets just agree to disagree and I eagerly await your PI's to remove AFCI protection from the NEC. Good Luck with that:slaphead:

 

[mbrooke]

They are claims of a single or select group.....Defined as Outlandish in terms of an individual making a global statement for a much bigger group of people than a few electricians who have sour grapes over the technology. That kinda of Outlandish...

Perhaps this select group has no financial or political reasons to defend AFCIs. And, in addition to that we ask questions rather than blindly follow. I see nothing outlandish about making an educated statement.

As I said other AFCI technology in service has the ability to differentiate between normal and abnormal arcs on a mathematical level. Home AFCIs do not, and they cant with the limited logic. That is a hard fact that can be demonstrated.

Also it appears you missed my post....I have talked to plenty of them......even been in the testing labs used to produce the very tests that UL 1699 is confirmed on. Yeah......I have seen my own tests and I think some would like a device to do something it was not designed to actually do...100%

Just Sayin

And I have seen some stuff as well in that regard Both us know that the tests are nothing more than having a breaker trip on a defined condition. Nothing more nothing less. There are no tests that says said breaker being tested can differentiate between a new plasma TV and what they are supposed to trip on. "I think some would like a device to do something it was not designed to actually do...100%" like? Perhaps be designed to detect arcs reliably? Whats the point of having them if they trip for none arcs?


And if you don't believe me here is another real world example that I come across daily of the nuisance AFCI are becoming. Just posted today:


http://www.diychatroom.com/f18/new-...ging-computer-watching-tv-224065/#post1575993