Eaton AFCI two thumbs DOWN!!!

[ActionDave]

Hi,
This is my first post. I'm very active in product safety engineering and work for a major appliance industry involved in several safety committees (UL - CSA - AHAM, etc).

Welcome. I hope you stick around. It sounds like you have a good perspective to add to the conversations.

Obviously, I support AFCI's because they will improve over time.

It's the present state of AFCI's that is problem. At this stage of development I don't think they should be mandated by law.

Are any of you old enough to remember all the problems when automotive first came out with electronic ignition? Man, I think every mechanic (me included at the time) wanted to tar and feather all engineers who were ever involved with the early development. Nowadays, you hardly ever hear of automotive ignition failures...

That is not an example of a technology being mandated by law. Sticking with the automotive analogy, air bags might be more fitting.

 

[ActionDave]

...In regards to AFCI's nuisance tripping, I agree with the one fella's post of to use the most recent AFCI. ...

Who do you suggest pays for that?

 

[mbrooke]

Hi,
This is my first post. I'm very active in product safety engineering and work for a major appliance industry involved in several safety committees (UL - CSA - AHAM, etc). In regards to AFCI's nuisance tripping, I agree with the one fella's post of to use the most recent AFCI. It is a delicate balance of making sure the AFCI performs it's function when it's supposed to and yet not trip when there is no safety issue. From an engineers prospective there are 4 outcomes:
-Device does not trip when there is no safety issue - Good
-Device does trip when there truly is a safety issue - Good
-Device does not trip when there is a safety issue - Very Bad and known as a Beta Error
-Device trips when there is a safety issue (nuisance tripping) -Bad and known as an Alpha Error.

I've briefly looked at some of the algorithms with AFCI suppliers (looking at the "current signature" and making a determination if it's a dangerous arc). These algorithms are very complex, but they are constantly striving to improve them. If they don't improve, they will be out of business. Thus, I'm interested in your comments.

Obviously, I support AFCI's because they will improve over time. Are any of you old enough to remember all the problems when automotive first came out with electronic ignition? Man, I think every mechanic (me included at the time) wanted to tar and feather all engineers who were ever involved with the early development. Nowadays, you hardly ever hear of automotive ignition failures...

You need a lot more computing power and mathematical sophistication to accurately detect arcs than what is being applied to AFCI breakers. Its like trying to get a computer to work with only a few transistors instead of a normal microprocessor. Its just not possible.


Second, even if AFCIs aren't perfect I don't see anyone going out of business. AFCI are mandated by code so profit will still come in regardless.



My apologies, I am aware that you are new and Welcome to the forum but I think many are frustrated by the nuisance tripping as I am myself.

 

[don_resqcapt19]

...
I'm a firm believer of the produce but I am still waiting for the insurance industry and fire protective services to gather history and documentation to support the importance of the AFCI. Until that happens there will be complainers as well as naysayers.

When you run the numbers on how few fires the AFCI will prevent, based on the fact that the fire cause and origin data shows 85% of the dwelling unit fires that are said to be of electrical origin, were in dwelling units at least 20 years old, you will see that it will be many, many years before there will be any real data.
Using the available data, over the first 20 years of compliance with the 2014 AFCI rule, you would only expect to prevent ~14,000 fires. Also contrary to what many believe based on the pro-AFCI people, the percentage of dwelling unit fires that are said to be of electrical origin is less than 15% of the total dwelling unit fires.

Given the small percentage of dwelling unit electrical fires, we would be far ahead in preventing residential fire deaths and injuries, and in reducing fire loss costs if we would require residential fire sprinkler systems. Our city had added an amendment to our electrical code saying that if you install a NFPA compliant sprinkler system in a dwelling unit, you don't need to install AFCIs. We feel that the overall safety is improved by using sprinklers over using AFCIs. The elimination of the AFCIs provides about 25% of the money needed to install the sprinkler system. While there may be a couple more fires because the AFCI was not there to prevent them, the losses from all of the fires will still be less, and most all fire deaths would be eliminated, many injuries would be eliminated.

 

[templdl]

c

c

When you run the numbers on how few fires the AFCI will prevent, based on the fact that the fire cause and origin data shows 85% of the dwelling unit fires that are said to be of electrical origin, you will see that it will be many, many years before there will be any real data.
Using the available data, over the first 20 years of compliance with the 2014 AFCI rule, you would only expect to prevent ~14,000 fires. Also contrary to what many believe based on the pro-AFCI people, the percentage of dwelling unit fires that are said to be of electrical origin is less than 15% of the total dwelling unit fires.

Given the small percentage of dwelling unit electrical fires, we would be far ahead in preventing residential fire deaths and injuries, and in reducing fire loss costs if we would require residential fire sprinkler systems. Our city had added an amendment to our electrical code saying that if you install a NFPA compliant sprinkler system in a dwelling unit, you don't need to install AFCIs. We feel that the overall safety is improved by using sprinklers over using AFCIs. The elimination of the AFCIs provides about 25% of the money needed to install the sprinkler system. While there may be a couple more fires because the AFCI was not there to prevent them, the losses from all of the fires will still be less, and most all fire deaths would be eliminated, many injuries would be eliminated.

And this is what is so frustrating. I would like to think that the AFCI may be more affective when installed in older dwelling than with new ones. But, that would only by as my theory in using AFCI technology in preventing electrical fires if it can be documented that they would be effective. But we know that it is a big gray area at best.
I was a beta test site for Eaton AFCIs for which I have one trip because the electrician crammed #12 wires into a 4x4 box that an EGC folded back around the left side of the duplex outlet and came into contact with the neutral screw. This didn't seem dangerous up front but did provide a point where neutral current would be turned loose on the EGCs.
Sprinklers would be the best way to address this but we also now it's a 'pipe dream' which would be more practically used in new construction but would be costly. I personally would like sprinklers but living in Wisconsin freezing water lines is the ceilings and wall would create an issue. Water damage from an accidental activation or leaks would not be welcome with the insurance companies.
It may be damned if you do, damned if you don't issue.

 

[don_resqcapt19]

And this is what is so frustrating. I would like to think that the AFCI may be more affective when installed in older dwelling than with new ones. But, that would only by as my theory in using AFCI technology in preventing electrical fires if it can be documented that they would be effective. But we know that it is a big gray area at best.

They might be effective in older buildings, but the code is not retroactive.

I was a beta test site for Eaton AFCIs for which I have one trip because the electrician crammed #12 wires into a 4x4 box that an EGC folded back around the left side of the duplex outlet and came into contact with the neutral screw. This didn't seem dangerous up front but did provide a point where neutral current would be turned loose on the EGCs.

It remains my opinion that the GFP part of the AFCI is the most effective part, but it is not required by the standard and at least two manufacturers have removed the GFP protection from their AFCIs.

Sprinklers would be the best way to address this but we also now it's a 'pipe dream' which would be more practically used in new construction but would be costly. I personally would like sprinklers but living in Wisconsin freezing water lines is the ceilings and wall would create an issue. Water damage from an accidental activation or leaks would not be welcome with the insurance companies.
It may be damned if you do, damned if you don't issue.

Sprinklers are now required by two model building codes (ICC and NFPA) for new dwelling units, but the sprinkler requirement is being deleted by most areas when they adopt one of the model building codes. With the plastic pipe now used for dwelling unit sprinkler systems, damage from freezing is much less likely, but of course the system is not effective if the water can't flow. As far as insurance issues with accidental water damage, if that was a real issue, why do all of the insurance companies offer a reduction in the fire insurance premium if you install a sprinkler system?

 

[mbrooke]

I think if the money spent on AFCIs and research were invested on developing something to stop glowing connections a real win-win would exist.


Further, I agree that firesprinklers are a better option in all fires.



Off topic, but do resi fire sprinkler codes require a siamese?

 

[don_resqcapt19]

...Off topic, but do resi fire sprinkler codes require a siamese?

Typically just a connection to the domestic water supply with a backflow preventer.

 

[kwired]

When you run the numbers on how few fires the AFCI will prevent, based on the fact that the fire cause and origin data shows 85% of the dwelling unit fires that are said to be of electrical origin, were in dwelling units at least 20 years old, you will see that it will be many, many years before there will be any real data.

Which brings up the question of whether or not the AFCI will still work after 20+ years (assuming it works now).

FWIW I have come across many GFCI breakers that are 25+ years old that do not trip when you press the test button, GE seems to be most common one I've seen that won't trip out of the older ones.

 

[GoldDigger]

Which brings up the question of whether or not the AFCI will still work after 20+ years (assuming it works now).

FWIW I have come across many GFCI breakers that are 25+ years old that do not trip when you press the test button, GE seems to be most common one I've seen that won't trip out of the older ones.

Which brings up the second question of whether the TEST button (mechanical contact) or the logic (semiconductors plus a relay) has failed.

 

[kwired]

Which brings up the second question of whether the TEST button (mechanical contact) or the logic (semiconductors plus a relay) has failed.

Yes, last one I recall coming across maybe a couple months ago, I put a test load hot to ground - no trip - no GFCI protection. It was a GE probably installed mid 1970's. Was supplying bathroom receptacles - I was there for a bath receptacle issue - bad backstab - It got replaced with a GFCI receptacle and the GFCI breaker was left installed.

 

[mbrooke]

Which brings up the question of whether or not the AFCI will still work after 20+ years (assuming it works now).

FWIW I have come across many GFCI breakers that are 25+ years old that do not trip when you press the test button, GE seems to be most common one I've seen that won't trip out of the older ones.

I think its a gamble. High voltage surges and lightning will damage them. Ive heard people claim that reps have told them anything over 126 volts will fry them.

Consider the fact European RCDs are usually none electronic...

 

[kwired]

I think its a gamble. High voltage surges and lightning will damage them. Ive heard people claim that reps have told them anything over 126 volts will fry them.

Consider the fact European RCDs are usually none electronic...

I would sure hope they are designed to withstand maybe at least 135 volts for at least a few hours. Around here we can easily see 130 to 135 at times on nominal 120 volts when a POCO regulator is malfunctioning.

 

[templdl]

I think its a gamble. High voltage surges and lightning will damage them. Ive heard people claim that reps have told them anything over 126 volts will fry them.

Consider the fact European RCDs are usually none electronic...

Being a sales and applications engineer for 2 leading manufacturers I wouldn't give you a plug nickel for some of our reps. It is unbelievable some of the hot air they spewed, construction salesmen in particular. It was very common for them to be technically challenged to say the least. I found it sad but somewhat entertaining as basic peddlers who are good a BS and their know it all attitude. They are interested in promotions while gaining little experienve along the way.
As such one should be somewhat cautious by taking what they say as you would as a grain of sand unless you can varify what they say.

 

[Aleman]

I have yet to even see a AFCI breaker. I consider that a good thing. Having browsed this forum for awhile now I have formed the opinion that they are a work in progress.

 

[mbrooke]

Being a sales and applications engineer for 2 leading manufacturers I wouldn't give you a plug nickel for some of our reps. It is unbelievable some of the hot air they spewed, construction salesmen in particular. It was very common for them to be technically challenged to say the least. I found it sad but somewhat entertaining as basic peddlers who are good a BS and their know it all attitude. They are interested in promotions while gaining little experienve along the way.
As such one should be somewhat cautious by taking what they say as you would as a grain of sand unless you can varify what they say.

Ok, that is nice to know then. I hope they can take more than I was lead to believe.

Does anyone know?

I have yet to even see a AFCI breaker. I consider that a good thing. Having browsed this forum for awhile now I have formed the opinion that they are a work in progress.

Yes they truly are :happyno:

 

[ActionDave]

I think its a gamble. High voltage surges and lightning will damage them. Ive heard people claim that reps have told them anything over 126 volts will fry them.

Consider the fact European RCDs are usually none electronic...

I know 240V will. I worked on a place that suffered an open neutral on the main feed. Fried every AFCI in the panel, even the ones that had no load on the circuit at the time of the incedent.

 

[kwired]

I know 240V will. I worked on a place that suffered an open neutral on the main feed. Fried every AFCI in the panel, even the ones that had no load on the circuit at the time of the incedent.

I'm guessing they still closed the circuit when switched on but the AFCI components no longer worked?

 

[gjneumann]

Wow, these are all awesome comments!

Wow, these are all awesome comments!

Thanks Guys!!! This is the exact stuff the NEC and UL committees needs to know before implementing new mandated requirements (which I'm on some of them and new to some of them). As for "action daves" comment of "who do you suggest pays for that". golly I have no good answer except for the following:

As for damage to homes/products from Arcing, I can tell you it's REAL! We (the appliance industry) see it a lot. Lots of Property damage. A sustained Arc Tracking Event (one that sounds like an Arc Welder) is very energetic and often produces over 1000 watts of power for several seconds. Electrical Enclosures have a hard time containing these violent events. Particularly with the increased use of plastics. There is only so much you can do with plastics regarding flame resistance. The 3rd party safety agencies (IE UL, IEC, CSA etc) know this as well. FYI, I have no problem replicating Arcing Events the burn thru those blue PVC receptacle boxes (in my laboratory).

As far as Glowing connections (as you folks call them), yes they too are a big problem. Particularly with high current applications. At this time, there seems to be no good practical way of detecting a glowing connection. We can only try to prevent them and contain them.

I look at Electrical Fire safety consisting of Prevention (quality), Detection (GFI, AFCI), and Containment (enclosures, sprinklers). We need them all... There is no one solution (even though our government thinks so) that will cover all safety failure modes.

Again, thanks for the comments!

 

[kwired]

Thanks Guys!!! This is the exact stuff the NEC and UL committees needs to know before implementing new mandated requirements (which I'm on some of them and new to some of them). As for "action daves" comment of "who do you suggest pays for that". golly I have no good answer except for the following:

As for damage to homes/products from Arcing, I can tell you it's REAL! We (the appliance industry) see it a lot. Lots of Property damage. A sustained Arc Tracking Event (one that sounds like an Arc Welder) is very energetic and often produces over 1000 watts of power for several seconds. Electrical Enclosures have a hard time containing these violent events. Particularly with the increased use of plastics. There is only so much you can do with plastics regarding flame resistance. The 3rd party safety agencies (IE UL, IEC, CSA etc) know this as well. FYI, I have no problem replicating Arcing Events the burn thru those blue PVC receptacle boxes (in my laboratory).

As far as Glowing connections (as you folks call them), yes they too are a big problem. Particularly with high current applications. At this time, there seems to be no good practical way of detecting a glowing connection. We can only try to prevent them and contain them.

I look at Electrical Fire safety consisting of Prevention (quality), Detection (GFI, AFCI), and Containment (enclosures, sprinklers). We need them all... There is no one solution (even though our government thinks so) that will cover all safety failure modes.

Again, thanks for the comments!

Where does government come in other then when AHJ's adopt the NEC as the law? NEC is what requires AFCI's not so much any government agency - in a direct manner anyway.