AFCI and GFCI Kitchens

[mbrooke]

A curious, if not inscrutable, allusion. . . :?

Money and power blinds. So can ignorance. Fear of losing reputation or embarrassment seals the deal.

The arguments for increased safety (the Holy Grail of better Code) aside, the workings of individual solutions to the AFCI Grand Concept remain behind firewalls to protect "proprietary" knowledge.

Id argue that part of it is liability of self incrimination.

No field usable general use third party tech has been created to test any AFCI device, or created to troubleshoot suspect wiring and apparatus for arc-fault signal signatures. I believe this absence of "test-ability" to be the heart of the transparency needed for the Pro and Con AFCI camps to begin reconciliation.

I think part of the problem is we have yet to define what a dangerous arc waveform ought to look like. "Arcing" is far, far from simple. Both normal current ripples and those said to come from dangerous arcing each have their own unique finger print. Catching them alone takes enormous computing power, let alone the research to separate each current signature into "safe" vs "dangerous" and then to compare the two weeding out differences which discriminate one from the other. This of course assume dangerous arcing is the norm in residential wring which Id argue its not. Paschen's law seems to agree.

A really good, messy, public product liability lawsuit would do wonders to move things forward off this impasse, a lawsuit that would enter into public record just how a particular AFCI discriminates between benign and malevolent arc signatures.

The manufacturers have stone walled us for fifteen years. This has to end.

Its the only way to end it outside of local amendments. What I myself would like to know is who, or how, was it determined that 30,000 home fires are the result of arcing. I have yet to see a single shred of evidence, let alone official theory to back this up.

As more and more Electrical Contractors are awakened to how they are trapped into bank-rolling the equipment and AFCI manufacturers troubleshooting, the more heightened the contradiction becomes.

And the general public will be joining in eventually, as they face the bills that come in for the troubleshooting, and endure the lost time, money and convenience for product service, return and/or replacement.

At this point contractors and the general public are nothing more then uncontested guinea pigs in an experiment. Research and development starts, not ends, after AFCIs are installed.

And here, I return to the non-GFP AFCI. The very firewall that hides the "test-ability" of AFCI is the mechanism that declares the non-GFP AFCI to be fully 100% as safe as an AFCI with GFP. This, in my estimation, is the petard that the manufacturers will be hoisted upon. . .

A none GFP AFCI is simply an empty fire extinguisher. GFP was forced into AFCIs only because at the time researchers could not develop or find any reliable arc analysis technology to pass UL1699 testing. Many still rely on it to this day. Which yet again validates the 30/50ma GFP in AFCIs is the only worthwhile function.

When the rest of the world had similar concerns, they choose GFP, not AFCI and at this point its working well for them as do GFCIs for us. When a GFCI trips, its actually an electrical fault. I still cant for the life of me figure out why GFP was not mandated instead of AFCIs if the concern was so great.

 

[mbrooke]

I rarely get call backs on GE afci but recently we did a huge home with lots of led's and I had to change a few out to a different generation unit.

Which confirms what I have been saying: the nuisance tripping is coming from the signature analysis component in AFCIs, not the GFP component.

 

[mbrooke]

The cry of alarm is worthless, also. The juggernaut of expanded AFCI enforceable ordinance that has occurred over the last fifteen years has demonstrated repeatedly that rational and dispassionate, or emotional and passionate pleas for change are value-less.

The installing electrician, and the local electrical inspector, have been FORCED to accept the AFCI "TEST" button as the ONLY proof that statute required protection IS, IN FACT, in place.

Period.

I agree, and unfortunate reality. :happysad:

AFCI manufacturers and Utilization Equipment manufacturers FORCE their dissatisfied customers into a communications triangle that will frustrate, exhaust and defeat most customers before any support is actually given.

This can't be sustained, if the devices don't, in fact, provide real safety improvements. . . actionable events, out here where we work and live will occur, and the subsequent litigation will be directed at the manufacturer's.

The only people to blame imo are the NFPA. They had all the power to say no, ask questions, and wait until a working trouble free product was developed if the concern of arcing was real.

The non-GFP combination AFCI is DEFINED by UL and the manufacturer as providing 100% of the UL safety when I install it and press the TEST button. The more of these non-GFP combination AFCIs are installed, the quicker we will get to the actionable event(s) that, after litigation, will result in redress.

Either the non-GFP combination AFCI is actually working, OR, the fuse is burning towards the hoisting of the manufacturer on its own petard.

Either way is likely a win for the ECs and the customers. (Except for those involved in the actionable event.)

Even if all nuisance tripping is resolved, their is still the issue two issues:

1. Will the device catch dangerous arcing if #2 is yes.

2. Is dangerous arcing even a major concern at 120 volts?

My answer is no. This video actually proves much of what I have been saying. First, what we see at 3:15 is not arcing but rather sparking. Notice how the sparking is not self sustaining. It only takes place only under mechanical intervention and Paschen's predicts that. At 120 volts AC self sustained arcing can not take place unless you are physically making and breaking the connection. There is of course an exception where NRTLs use 15,000 volts to "carbonize" institution to the point 120 volts will conduct on its own, but thats another story. Second, as badly damaged as the NM cable is, I do not see any significant charring, major discoloration or heat damage of the cable or wood stud. Good chance the circuit stopped working at that point and they were called. Yes the cable has some discoloration right around the staple, but chances are its normal rust from the staple itself. Third, the use of arcing to describe this is a misnomer showing that electricians are being heavily misinformed about arcing in an effort to create a perception arcing is a major crisis in home wiring. Fourth, with the bare EGC present in NM cable, GFP would have tripped on the damaged section.

Comments or opinions welcome.


https://www.youtube.com/watch?v=Dt14lhLgBzY

 

[mbrooke]

What does that even mean? (With respect, specifically, to the non-GFP combination AFCI.)

I think, and will let RJ answer, that the only viable function inside AFCIs is the 30/50ma GFP for those that have it.

 

[al hildenbrand]

A non GFP AFCI is simply an empty fire extinguisher.

While this may be true, I think the declaration is a distraction. "Empty fire extinguisher" is a meme. Just like the 2002 National Association of State Fire Marshall's rabidly pro AFCI media campaign's slogan that "AFCIs detect fires" was a meme.

Educating my clients about the Code requirements for putting AFCI in their Premises Wiring (System) routinely includes the description of how I will charge for the troubleshooting. I also explain that the products they use are potential sources of "arc signals" that the signal processor in the AFCI is reacting to.

Most people just want to get on with their lives, but I do what I can to keep placing the information in ever wider circles.

Until evidentiary rules breach today's proprietary information firewalls (or a manufacturer actually becomes forthcoming), we can't determine if there is ANY meme that applies to AFCIs.

 

[mbrooke]

While this may be true, I think the declaration is a distraction. "Empty fire extinguisher" is a meme. Just like the 2002 National Association of State Fire Marshall's rabidly pro AFCI media campaign's slogan that "AFCIs detect fires" was a meme.

Educating my clients about the Code requirements for putting AFCI in their Premises Wiring (System) routinely includes the description of how I will charge for the troubleshooting. I also explain that the products they use are potential sources of "arc signals" that the signal processor in the AFCI is reacting to.

Most people just want to get on with their lives, but I do what I can to keep placing the information in ever wider circles.

Until evidentiary rules breach today's proprietary information firewalls (or a manufacturer actually becomes forthcoming), we can't determine if there is ANY meme that applies to AFCIs.

I think you I get where your going with this. We cant call AFCIs anything because we has no proof or evidence to support what they will or will not do? Not disagreeing, I agree with most if not everything you have said thus far.

 

[romex jockey]

I think, and will let RJ answer, that the only viable function inside AFCIs is the 30/50ma GFP for those that have it.

Thank you for waiting for me MBrooke, i'm way over my head with the holiday deadlines here

Basically a Toroidal Coil is a CT

This is the heart of afci, gfci, current leakage or rcd technology......over 1/2 century old in ocpd's globally

The afci incorporates two , one whole, one with maybe 20% of the ring gone.

But it, like our gfci's can only detect what a toroidal coil can detect ,it has zero oscilloscopic ,thermodynamic or nonsinusoidal abilities regardless of it's electronica , and quite frankly would be in a world of trouble with our power quality here if they did.


It detects parallel events only, which is why we literally have it's past r&d men , who admitted their failure publicly , are lobbying cmp-2 to clarify it's functionality


Earl Roberts can explain it better than i in his book>

If there's any true substantial change , it's the lower mag trip ratings..... (be then magnetic or electronic)

For the cynics (guilty as charged) this means parallel faults appearing like they might have been a series incident

Which is why the manufacturers, as well as nema members who frequent this forum are rather hush hush

~RJ~

 

[kwired]

I rarely get call backs on GE afci but recently we did a huge home with lots of led's and I had to change a few out to a different generation unit.

At your own/customers own expense I assume?

 

[al hildenbrand]

We cant call AFCIs anything because we has no proof or evidence to support what they will or will not do?

A lot of good people have assembled a lot of good evidence that various Make and Models of AFCI are not performing to the expectations created by those who are Pro-AFCI. To date, this evidence has not been able to alter the Definition that if the TEST button works on an individual AFCI device that 100% of AFCI "protection" is present.

This has us in our current situation.

This is like a raceway on the outside of a structure, a structure in the Death Valley, being defined in Article 100 as a Wet Location. The Definition is real, Reality is not real.

Unlike raceway in Death Valley, AFCIs are more likely to be involved in property loss, injury and/or death which elevates the likelihood of litigation. . . and, if I may say, the dispelling of this cloud of illusion.

 

[mbrooke]

A lot of good people have assembled a lot of good evidence that various Make and Models of AFCI are not performing to the expectations created by those who are Pro-AFCI. To date, this evidence has not been able to alter the Definition that if the TEST button works on an individual AFCI device that 100% of AFCI "protection" is present.

This has us in our current situation.

This is like a raceway on the outside of a structure, a structure in the Death Valley, being defined in Article 100 as a Wet Location. The Definition is real, Reality is not real.

Unlike raceway in Death Valley, AFCIs are more likely to be involved in property loss, injury and/or death which elevates the likelihood of litigation. . . and, if I may say, the dispelling of this cloud of illusion.

I agree, and Id say that IMHO it has to do with the fact we have yet to define what a dangerous current signature looks like in the real world.

 

[al hildenbrand]

But it, like our gfci's can only detect what a toroidal coil can detect ,it has zero oscilloscopic ,thermodynamic or nonsinusoidal abilities regardless of it's electronica , and quite frankly would be in a world of trouble with our power quality here if they did.

This is where you lose me.

A CT's output is designed to match the electrical characteristics of the electronics that are to receive the output. . . an output that contains the same wave form as the waveform of the load current passing through the primary of the CT and is proportional in magnitude to the amount of current in the primary. The waveform is the information. . .

Combine the current waveform information with the voltage waveform information and manipulate the data therein to get results. That's the heart of this Arc Fault Detection and Discrimination Methods paper from 2007 published by and downloaded from Siemens. The paper was delivered at the IEEE Holm Conference.

You see, to me, AFCI is (or is supposed to be) a signal processor that acts upon the information contained within the signals passing by the processor. The paper by Mr. Restrepo offers a snapshot of the concepts that Siemens allowed to be released in 2007.

 

[al hildenbrand]

Id say that IMHO it has to do with the fact we have yet to define what a dangerous current signature looks like in the real world.

Yes. Somehow, an infinite number of malevolent arcs must be differentiated from an infinite number of benign arcs
in a "standard".

 

[mbrooke]

Yes. Somehow, an infinite number of malevolent arcs must be differentiated from an infinite number of benign arcs
in a "standard".

Which requires so much data and mind numbing research that carrying it out on 120/240 residential services is impractical. The library or dangerous vs safe arcing is still a work in progress. This is one why reason why 15 years latter manufactures still have trouble making device that are free of nuisance trips.

 

[mbrooke]

This is where you lose me.

A CT's output is designed to match the electrical characteristics of the electronics that are to receive the output. . . an output that contains the same wave form as the waveform of the load current passing through the primary of the CT and is proportional in magnitude to the amount of current in the primary. The waveform is the information. . .

Combine the current waveform information with the voltage waveform information and manipulate the data therein to get results. That's the heart of this Arc Fault Detection and Discrimination Methods paper from 2007 published by and downloaded from Siemens. The paper was delivered at the IEEE Holm Conference.

You see, to me, AFCI is (or is supposed to be) a signal processor that acts upon the information contained within the signals passing by the processor. The paper by Mr. Restrepo offers a snapshot of the concepts that Siemens allowed to be released in 2007.

From the above link:

Beginning with the 1999 National
Electrical Code (NEC), there have been requirements for the
use of AFCI’s, capable of detecting arc faults limited by an
available current of 75A or higher.

You do know where that number came from, right?

 

[al hildenbrand]

Which requires so much data and mind numbing research that carrying it out on 120/240 residential services is impractical. The library or dangerous vs safe arcing is still a work in progress. This is one why reason why 15 years latter manufactures still have trouble making device that are free of nuisance trips.

Yes. In the first iteration the data and research is mind numbing. But, look to the model of understanding electrical engineering theory. One goes up through math through differential and integral calculus, applies it to electrical circuits and then use the Laplace Transform to side step the grunt work of the calculus,

Or, in asynchronous three phase systems applying positive, negative and zero sequence vector analysis to reduce infinitely variable sine waves into three sets of "rotating vectors".

This is what, I trust, is being sought after. . . the method to condense, extract and manipulate the information in the signals present that the AFCI device can process. The AFCI device is listening to everything, so part of the difficulty is the sources that are benign that mimic malevolent. . .

Given the present legal definition forcing us to accept AFCI protection being 100 % present in each breaker successfully "TEST"ed, we, the con-AFCI, are held sequestered, YET, time passes and the fuse continues to burn. . . leading to the exposure of what the firewall protects.

 

[mbrooke]

Yes. In the first iteration the data and research is mind numbing. But, look to the model of understanding electrical engineering theory. One goes up through math through differential and integral calculus, applies it to electrical circuits and then use the Laplace Transform to side step the grunt work of the calculus,

Or, in asynchronous three phase systems applying positive, negative and zero sequence vector analysis to reduce infinitely variable sine waves into three sets of "rotating vectors".

This is what, I trust, is being sought after. . . the method to condense, extract and manipulate the information in the signals present that the AFCI device can process. The AFCI device is listening to everything, so part of the difficulty is the sources that are benign that mimic malevolent. . .

Given the present legal definition forcing us to accept AFCI protection being 100 % present in each breaker successfully "TEST"ed, we, the con-AFCI, are held sequestered, YET, time passes and the fuse continues to burn. . . leading to the exposure of what the firewall protects.

At least you know what you are quantifying with those vectors along with a legit reason why.

Speaking of testing, I wonder what these guys know or do >>>

http://www.afcimaster.com/En/ProductView.asp?ID=6

 

[mbrooke]

For the cynics (guilty as charged) this means parallel faults appearing like they might have been a series incident

Which is why the manufacturers, as well as nema members who frequent this forum are rather hush hush

~RJ~

What absolutely sealed the deal for me was when a NEMA member stated on his DIY radio podcast that glwoing connections just burn themselves "spring" open. That is absolutely untrue, and it shows just how much electrical theory is missing from their understanding... in fact it would explain why they push AFCIs altogether. They just don't know any better.

 

[kwired]

What absolutely sealed the deal for me was when a NEMA member stated on his DIY radio podcast that glwoing connections just burn themselves "spring" open. That is absolutely untrue, and it shows just how much electrical theory is missing from their understanding... in fact it would explain why they push AFCIs altogether. They just don't know any better.

From what I have seen, glowing connections always burn themselves open, eventually, but sometimes ignite something else nearby before that gets a chance to happen.

 

[mbrooke]

From what I have seen, glowing connections always burn themselves open, eventually, but sometimes ignite something else nearby before that gets a chance to happen.

Correct, they sometimes can and do lead to fire. Sadly what they wont tell you is that the majority of glowing connections do not produce an arc signature when glowing red. Arcing is often the end stage of a glowing connection, not the beginning. And even more important is that glowing connections can take place below 5 amps which is where the series arc detection threshold begins to end on AFCIs.




Glowing connection = series arc fault

Short circuit = parallel arc fault

Nothing more then common everyday occurrences relabeled as "arcing".

 

[al hildenbrand]

Speaking of testing, I wonder what these guys know or do >>>

http://www.afcimaster.com/En/ProductView.asp?ID=6

Given that they are a Chinese venture, and the devices shown are specified at 220 - 240 Volt, 50 Hz, 16 - 32 Amp, I would submit that a language and regulatory hurdle must be surmounted to extract meaningful content Stateside.

I do note, with interest, that they offer an "AFCI Net Controller" that, to me, intimates a form of linking of AFCI protective devices to other devices through something akin to the Internet of Things (IoT). Each AFCI device may include WiFi capability to send and receive information. The "AFCI Net Controller" can interact will up to 99 different AFCI protective devices.

Pretty cool. Love to see that take hold here.