General Stuff

Status
Not open for further replies.

iwire

Moderator
Staff member
Location
Massachusetts
Re: General Stuff

Tony it is interesting that you are the only one on the forums that never budges in their opinion.

Have you ever admitted there are two sides to anything?
 

pierre

Senior Member
Re: General Stuff

Tonyi

You were debating the issue quite well until your last statement, that was uncalled for. You have no idea what anyone on this forum may be like, personal statements should be left out of the mix here.

Pierre
 

don_resqcapt19

Moderator
Staff member
Location
Illinois
Occupation
retired electrician
Re: General Stuff

Tony,
Still, this seems pretty useful for all those 2-wire lamp cords lurking under rugs and lead cords duct taped to walls and floors.
How is it useful? The device is not effective under those conditions.
Amortize the cost over the lifetime of a house and any cost arguments start to look pretty silly.
Cost arguments are not based on any individual installation. They are based on the total cost of the rule and compared to the total benefit of the rule. If you assume an average cost of $75 to comply with 110.12 in 100% of the new dwelling units built each year, the yearly installation cost would be about $120 million per year. Using the fire statistics from the ROPs and ROC, over a 10 year span the AFCIs could be expected to prevent about 750 dwelling unit fires. The total AFCI installation cost over these 10 years would be about $1.2 billion. That is a cost of over $1.6 million per fire prevented.
Don
 

roger

Moderator
Staff member
Location
Fl
Occupation
Retired Electrician
Re: General Stuff

Tony, I wish I knew thousands of people like you, I could make a fortune selling snake oil and promises.

I agree with Pierre, (not that Don needs us to fend for him) you seem to be so disapointed in having your belief in this item crushed, that you are throwing personal shots.

Roger
 

don_resqcapt19

Moderator
Staff member
Location
Illinois
Occupation
retired electrician
Re: General Stuff

Guys,
To be fair to Tony, I don't think that he is being any more stuborn in this issue than I am. We just happen to be on the opposite sides of the issue.
Don
 

tonyi

Senior Member
Re: General Stuff

So how is it that two wire cords can't develop parallel arcs detectable by the device?

An AFCI can't tell the difference between fixed wiring and a cord.

Don, people buy houses, not statistics. My brother (who's not a dumb guy - a PhD mathemetician) asked me last year why people wouldn't do the best job possible when wiring a place when lives and property are at stake. After I stopped laughing I clued him in to the reality. He was shocked because he'd willingly pay a more for a safer job versus a minimal job.

The problem is that developers and people doing the work have taken this decision away from the person who has a vested interest in the quality of the job because they're going to have to live with it 7/24 for many years to come. The buyers and homeowners aren't even being told that there are quality and safety options that are available. They understand things like carpet/tile and appliance selection, but wiring is something the average guy has to trust someone else on - and this is often a misplaced trust! The price of a house doesn't mean much either, a $2M place can wind up being roped and backstabbed with $.39 devices just like a small $70K bungalow might.

With cars, people have this choice - if they can afford it and value their safety they buy something a bit more expensive that does well in crash tests. What developer ever asks a buyer if they want to spend a grand more for a safer wiring job with better quality components?

Heck, people are spending $10-15K more for SUV's that they perceive to be "safer" and that's an investment that's going to be recycled into a tuna can long before their house is flattened by a dozer. The boomer generation buying houses today is willing to pay for quality if you let them know it exists. Heck, my brother paid $2500 for some supposed child proofing expert to child proof his house. If he knew safer wiring was possible, he'd have gladly paid that much more and then some.
 

bennie

Esteemed Member
Re: General Stuff

My daughter also hired a safety expert to inspect and install all known child safety devices in her home.

I thought the expert did a fine job. However my beautiful little one year old granddaughter immediately defeated and bypassed most of the devices.

My complaint with the AFCI's is they did not enter the market based on their own merit. They are mandated by law. There is pending law suites regarding the methods used to sell the questionable devices.

The credibility of the devices is in question, along with the NEC.
 

don_resqcapt19

Moderator
Staff member
Location
Illinois
Occupation
retired electrician
Re: General Stuff

Tony,
So how is it that two wire cords can't develop parallel arcs detectable by the device?
The impedance of the circuit is too high to permit the required 75 amp fault current to flow. The branch circuit and feeder type of AFCI (the type currently on the market) must see a 75 amp arcing fault for at least 8 half cycles before it will open the circuit.
Don
 

charlie

Senior Member
Location
Indianapolis
Re: General Stuff

Originally posted by Don:
We just happen to be on the opposite sides of the issue.
Don, you are not alone. Right now, the new changes will add the following:
210.12(C) "Lighting and Appliance Branch Circuits in Dwelling Units. When a panelboard that contains the overcurrent protection devices for branch circuits is replaced, a listed arc-fault circuit interrupter shall protect each branch circuit that existed prior to the replacement that serves 120-volt, single-phase, 15- and 20-ampere outlets for lighting and for appliances."

To my way of thinking, the Code would not only be retroactive but it would require something more for a home that is one week old than for a new home.

We still have hope, the written ballets are not in yet (The comment with the panel action is required to have a 2/3 majority in order to pass.), it has not passed the TCC, it has not passed the Electrical Section meeting, it has not passed the floor of the NFPA Annual meeting in Salt Lake City, it has not passed the Standards Council on appeals, and it has not passed the NFPA Board of directors on appeals. Have I ever mentioned that the Code process can be complicated? I also left out some steps to keep it simple. :D
 

tonyi

Senior Member
Re: General Stuff

Originally posted by don_resqcapt19:
The impedance of the circuit is too high to permit the required 75 amp fault current to flow.
You gotta be kidding! Plug in a lamp then cut the cord with a pair of Kliens. I bet you a buck you produce a magnetic trip (unless its a FPE, then maybe you just welded your kliens to the cord and started your house on fire :D ).

There's a lot more than 70 amps available on a cord, the only thing limiting it is any intrinsic resistance in the recept, cord, branch, feeders and what the poco can actually deliver.

If a cord is parallel arcing, the served load is
completely out of the picture as a limiting factor. Series arcs are load limited, but not parallel ones.
 

don_resqcapt19

Moderator
Staff member
Location
Illinois
Occupation
retired electrician
Re: General Stuff

Tony,
You gotta be kidding! Plug in a lamp then cut the cord with a pair of Kliens.
That is not an arcing fault. Any standard OCPD will open under that condition. If a parallel arcing fault would draw that amount on current there would be no need for AFCIs as the standard breaker would do the job, but the arc itself limits the current. We are not talking about short circuits (bolted faults), we are talking abut arcing faults. Even the AFCI manufacturers recognize this issue. There was a proposal from a SquareD guy for the 2002 code to require both AFCI breakers and AFCI receptacles because the AFCI branch circuit and feeder devices do not provide adequate protection on the load side of the branch circuit wiring. UL has standards for at least 3 types of AFCI devices. They are "branch circuit and feeder", "outlet" and "combination" AFCI devices. Only the second two types are tested to provide protection on the load side of an outlet, unfortunately, these types have not yet reached the market.
Don
 

Ed MacLaren

Senior Member
Re: General Stuff

Only the second two types are tested to provide protection on the load side of an outlet,
I don't see what the difference would be (electrically) from the line side to the load side of a receptacle. The receptacle is just another connection, equivalent to a wirenut, unless - - - - they must be referring to the possible absence of the EGC in a two-wire cord?

Ed
 

tonyi

Senior Member
Re: General Stuff

Originally posted by don_resqcapt19:
That is not an arcing fault. Any standard OCPD will open under that condition.
Obviously. Its graphic demonstration of the available amps that you're claiming don't exist at the cord.

Will you now agree that a cord can supply at least ~70A when damaged in the manner I described? If it can magnetic trip an ordinary breaker, then I certainly hope your answer is going to be yes, otherwise "you got some splaining to do Lucy" :D

If so, then good. Now position said 2-wire cord under a throw rug in a high traffic area for a period of years where the floor and traffic act on it like a slow motion sander and abrade the insulations away to nothing.

Is this or is this not a situation where a parallel arc can now occur? I say the answer is that it is. The Klien experiment demonstrated that sufficient amp potential exists, and there is now a bare hot and neutral in close proximity. It seems about as close to a lab constructed parallel arc situation that you can get.

Obviously you feel differently and I'm keenly interested in the electrical explanation as to why you feel this way.
 

roger

Moderator
Staff member
Location
Fl
Occupation
Retired Electrician
Re: General Stuff

Tony, how far away from the source is your "klien" calamity taking place?

BTW, is this being done by plumbers? :D

Roger
 

don_resqcapt19

Moderator
Staff member
Location
Illinois
Occupation
retired electrician
Re: General Stuff

Tony,
The whole basis for requiring AFCIs is that the parallel arcing fault does not draw enough current to open a standard OCPD. I'm not the one saying that the AFCI won't trip for a parallel arc beyond the outlet. The AFCI manufacturers and UL are saying this. If the manufacturers and UL were sure that the branch circuit and feeder type of AFCI would trip on the type of fault that you have described, why would there be a need for an "outlet" or "combination" type of AFCI? I think that with the frayed cord, there is so little contact that the current does not reach the 75 amp level, but the arcing is producing enough heat to cause a fire. With the branch circuit wiring, even when the parallel arcing current is less than the 75 amps, the heat of the arc soon causes a ground fault which opens the device.
Don
 

don_resqcapt19

Moderator
Staff member
Location
Illinois
Occupation
retired electrician
Re: General Stuff

Ed,
The arcing current testing levels for the two types of AFCIs (that are not yet on the market) that provide protection beyond the outlet are at the 5 and 10 amp level and not the 75 and 100 amp levels that are used in testing the branch circuit and feeder type of AFCI.
Don
 

tonyi

Senior Member
Re: General Stuff

It would seem to me that direct contact produces an ordinary dead short (with one brief arc as the wires approach and contact is made). An air gap and ionization is what makes an arc an arc.

What UL and the manufacturers are doing is plainly legal CYA. I wouldn't want to be responsible for all the weird stuff and abuse people put cords to either. In the real non-litigous world, there is plainly very little difference between a frayed cord outside a box and decayed cloth/rubber insulations inside on a 2-wire system.

An AFCI doesn't have any sort of cord detection logic I know of to prevent it from tripping and I can't envision how a programmer would write code to tell the difference if they wanted to. If the amps/signature/cycles meet its trip criteria, it'll trip. There's no black magic here.

The need for the other planned devices is obvious - arcs of less than 70A have proven capable of producing ignition.

What we have now is the 1st generation and there's nothing wrong with that. Its the nature of technological evolution.

Were you equally critical of the early generation of breakers that only tripped thermally and not magnetic? Were they somehow intrinsically defective and not worth installing?

With a more powerful micro and more sophisticated algorithms, better AFCI's will appear that can sort out things like motor arcing from genuine series arcs. This is more of a pattern recognition problem though than the more simplistic and more restrictive criteria in the current device hardware.

To make the current generation USABLE in the field and not cause major problems with existing cord/plug stuff its detection criteria HAD to be downgraded from what was theoretically possible with the current hardware implementation.
Obviously there is a very good reason why this ~70A figure was chosen as the implementation threshold. I'm sure thousands of different normally functioning devices and fixtures were tested to see what they were drawing under normal operation.

Undoubtedly some common ones were peaking in the range of 50A, so the AFCI threshold had to be set somewhat above whatever the high water mark the test samples produced. The installed base must be supported to an exceptionally large degree. Its like backwards compatibility on an operating system upgrade. If your new upgrade doesn't run the vast majority of your old applications, its going back for a refund. More compute power, sophistication and discrimination logic in later generations could easily have this threshold below an amp. But that's not the device I can buy and install TODAY.

As I've said previously - if you insist on waiting for what's on the horizon, you'll be waiting forver and never deploy anything because there's ALWAYS a better next generation in the works.

At some point a manufacturer has to freeze a design and ship it, and at some point a buyer needs to decide that some particular rev provides enough value that its worth deploying.
 
Status
Not open for further replies.
Top