NEC Changes For #14 Ampacity

[iwire]

Numbers do not lie; especially when conductor impedance, circuit length and breaker trip curves are both known and documented.

Lie? No numbers don't lie. But incorrect assumptions and application of numbers can produce meaningless results.

On the other hand you have not backed up your theory.

It will be at least the weekend before I have time to set some stuff up.

 

[mbrooke]

Lie? No numbers don't lie. But incorrect assumptions and application of numbers can produce meaningless results.

Ok, I guess UL is assuming also. They have tables which actually show the available fault current per length of home run. Anyone doing resi knows most home runs exceed 50 feet.

It will be at least the weekend before I have time to set some stuff up.

Are you testing fault current of temps btw? (just curious)

 

[mbrooke]

The table, starting at page 53:


http://newscience.ul.com/wp-content...lity_to_Mitigate_Parallel_Arcing_Faults_2.pdf

 

[mbrooke]

Oh, just to give you an idea. mag vs thermal trip. Looks like the trip solenoid does not work on DC:




https://www.youtube.com/watch?v=S6BRo5fjtVk

 

[iwire]

Ok, I guess UL is assuming also. They have tables which actually show the available fault current per length of home run.

I think it is pretty funny how you will disparage UL info in one post and use UL to support your views in another.

Anyone doing resi knows most home runs exceed 50 feet.

Wait, make up your mind, a few post ago you where over 100'

Are you testing fault current of temps btw? (just curious)

No spoilers.

 

[mbrooke]

I think it is pretty funny how you will disparage UL info in one post and use UL to support your views in another.

I have NEVER disparaged UL. (Ok there is UL1699 but thats the manufactures telling them how to test)

Wait, make up your mind, a few post ago you where over 100'

Yes, assuming infinite buss. When 500amps is available at the panel run length is even lower. My point being that come real world several different factors make none magnetic tripping a decent probability.

No spoilers.

 

[romex jockey]

That 'real world' is full of impedance, of either original installation distances, or degradation over time.

~RJ~

 

[mivey]

Of course. However keep in mind that Id say roughly about 1/3 to 2/3 of all branch circuits in North America will not trip a breaker magnetically at the furthest point.

You might have your work cut out backing up that assumption with facts, even if it is true.

Fault currents can get pretty low (say less than 100-200 amps) on some circuits on some old, small buildings with small transformers and long runs but I would like to see how you spin that into a 1/3 to 2/3 of ALL branch circuits in North America.

 

[mivey]

I can claim a rough guesstimate.

Claiming is one thing...

Both us know in a typical 2,000 sqft dwelling the average home run length is about 125 feet from the panel to the furthest point.

If you believe that then... well I just don't know what to say that would be acceptable.

BTW, look up what a home run is.

When did you last wire a 2000 sq ft house that had an average like that? I'm guessing never. Which leads to a different point that I'll just choose to ignore. I like where you are going with the analysis but feel free to ask for help when you need it please as it taints your point.

 

[mivey]

It will be at least the weekend before I have time to set some stuff up.

FWIW, I believe I found mid/upper 100s to mid 200s on an older home we had. Seems like it was 400-600 on a newer home, maybe even 800+ at some points. Pretty anecdotal since I can't recall the exact figures. I could make some new measurements tomorrow for the fun of it if I don't forget.

 

[mivey]

Anyone doing resi knows most home runs exceed 50 feet.

Now that is getting closer to the truth but not always.

 

[romex jockey]

I've spent 3 decades running rx in rural homes , some of which are rather far from their serving Xformers.

In fact, using my Vdrop meter ,many compliant installs fail (yes i know it's just a FPN)

But that doesn't bother me as much as all the antiquated wiring i deal with, much of which megs out dismal

Plenty of evidence exists where some junction ,device, or termination suffered insult , if not repeated insult

That said, addressing the physics behind it all seems prudent.

Personally, i'd like to hear more about the magnetic trip function of an ocdp please. ....


~RJ~

 

[FionaZuppa]

That is true. Therefore it would make sense the NEC bases the tables to assume worse case scenario.

whats the worst case scenario, infinite R value with no heat capacity? how does the NEC use worst case to the derive usable #'s for real world applications?

 

[mbrooke]

You might have your work cut out backing up that assumption with facts, even if it is true.

Number of homes x the typical sizes in each home x breaker trip thresholds. Anyone with who has wired a home or has blueprints can figure out short circuit with with a basic computation.

Fault currents can get pretty low (say less than 100-200 amps) on some circuits on some old, small buildings with small transformers and long runs

So you admit it is possible

but I would like to see how you spin that into a 1/3 to 2/3 of ALL branch circuits in North America.

I will let you answer this:

1. What are circuit lengths in most homes? Min from max

2. Knowing the length, what is the total AC impedance of the fault loop?

3. What the typical fault current from a none network service or what is the impedance of the transformer and service drop?

4. What is the typical mag trip threshold of new single pole 15 and 20 single pole amp breakers?

5. What is the typical mag trip threshold in old 15 and 20 amp single pole breakers?

Using 1-5 what is the fault current of the mins? The maxes? which ones will trip the old breakers? The new ones?

 

[mbrooke]

Now that is getting closer to the truth but not always.

There are circuits less then 50 feet, and anyone wiring a home knows circuits can also exceed 50 feet.


50 feet however is not an arbitrary number, certainly not in 210.12 (A) 3 (b):

a. The branch-circuit wiring shall be continuous from
the branch-circuit overcurrent device to the outlet
branch-circuit arc-fault circuit interrupter.

b. The maximum length of the branch-circuit wiring
from the branch-circuit overcurrent device to the
first outlet shall not exceed 15.2 m (50 ft) for a
14 AWG conductor or 21.3 m (70 ft) for a 12 AWG
conductor.

This number was derived from the above UL research where it is was determined that under worse case scenario 50 feet is the maximum length at which a fault is guaranteed to trip a new breaker magnetically. Therefore over 50 feet is where the gamble begins, and with basic computations over 125 feet is where magnetic trip starts to become unlikely even with infinite fault current available (something not technically possible).

 

[mbrooke]

FWIW, I believe I found mid/upper 100s to mid 200s on an older home we had. Seems like it was 400-600 on a newer home, maybe even 800+ at some points. Pretty anecdotal since I can't recall the exact figures. I could make some new measurements tomorrow for the fun of it if I don't forget.

With those numbers anything over 50 feet is hit or miss, over 75 its more miss then hit.

But dont my word for it. Those numbers can be applied to the table made by UL in one of the reports:

 

[mbrooke]

I've spent 3 decades running rx in rural homes , some of which are rather far from their serving Xformers.

In fact, using my Vdrop meter ,many compliant installs fail (yes i know it's just a FPN)

But that doesn't bother me as much as all the antiquated wiring i deal with, much of which megs out dismal

Plenty of evidence exists where some junction ,device, or termination suffered insult , if not repeated insult

That said, addressing the physics behind it all seems prudent.

Personally, i'd like to hear more about the magnetic trip function of an ocdp please. ....


~RJ~

What would you like to know?

whats the worst case scenario, infinite R value with no heat capacity? how does the NEC use worst case to the derive usable #'s for real world applications?

To be honest I dont know how the NEC came about those numbers, but my theory is around realistic worse case conditions in the real world NM in High R value insulation carrying load 24/7.

 

[mbrooke]

:lol:

Great movie. :thumbsup:

And yes, relying only calculations and meters is pure gibberish.

If you truly mean that you just outed yourself. :happyno: You do know engineers apply number to every building you work in so it doesn't cave in? I guess the've been doing it wrong all these years.

 

[GoldDigger]

There are circuits less then 50 feet, and anyone wiring a home knows circuits can also exceed 50 feet.


50 feet however is not an arbitrary number, certainly not in 210.12 (A) 3 (b):




This number was derived from the above UL research where it is was determined that under worse case scenario 50 feet is the maximum length at which a fault is guaranteed to trip a new breaker magnetically. Therefore over 50 feet is where the gamble begins, and with basic computations over 125 feet is where magnetic trip starts to become unlikely even with infinite fault current available (something not technically possible).

Note that whatever the UL related justification for the 50' number may be, it occurs only in the section which allows the use of a receptacle or outlet type AFCI in conjunction with a breaker which is NOT a full branch circuit protection AFCI.
There is no such limitation on the run to a first outlet in a circuit which does not require AFCI or is protected by a full branch circuit AFCI breaker.
And I do not want to get into the arguments on the relative merits of low magnetic trip versus arc signature detection which is only enabled in the presence of a current that is already an overload.

I mostly want to inject a measure of skepticism into the analysis of any Code language related to AFCIs. (The whole barrel of apples have been spoiled at this point IMHO.)

 

[iwire]

If you truly mean that you just outed yourself. :happyno: You do know engineers apply number to every building you work in so it doesn't cave in? I guess the've been doing it wrong all these years.

Of course but how did they arrive at those numbers in the first place?

Real world testing.