Does table 310.15 B6 allow #2 Aluminum SER cable to be protected by a 100 amp breaker to feed a subpanel? (common practice around here)
By "subpanel", I mean a panel that is fed from the panel that contains the main disconnect.

I can't figure out if this table only applies to feeders between the main disconnect(s) and panelboard(s), or all feeders in a dwelling.

Feeder: "All circuit conductors between the service equipment....and the final branch circuit overcurrent device." (NFPA 70, 2002 Article 100 definition)



Peter

Peter I will say yes to your question, but wait for other replies as this one confuses me too and I do not use this table at my work.

I base my answer on this part of 310.15(B)(6)


Notice the (s) on feeder and panelboard.

Bob

In my opinion, you are correct (as is Bob).

I, like Bob, think the "main power feeder" must originate at the service equipment and end at a lighting and appliance panel. If you had a 200 amp panel (outside service equipment), it would be legal to feed an indoor panel with 4/0 AL, but it would be a violation to leave the interior panel and feed another panel with 2 AWG (unless the terminals were rated 75 degrees and you use the round-up rule). Once you have a feeder that originates outside of the service equipment, you are in the land of 310.16, not 310.15(B)(6).

Anything; be it a device, appliance, down stream panel, etc... fed from the lighting and appliance branch-circuit panelboard(s), will fall under "other" articles, and can not use 310.15(B)(6).

Here we go.

Roger

I say just remove Table 310.15(B)(6) and have us all use 310.16.

Bob

I vote Bob for president

You can not round up 90 amp rated 2 AWG AL to 100 amps as 90 amps is a standard size OCPD.

Now I feel bad for my last post.:O

Bob

Bob, the truth hurts. I'm sure Ryan will still vote for you (maybe ), I know if you get 310.15(B)(6) removed (or my preference, have it reworded) I will.

Roger



[This message has been edited by Roger (edited 04-08-2004).]

I didn't even look at 240.6, just assuming that 90 is an off the wall number and wouldn't be considered standard. One of these days I'm gonna stop shooting from the lip and start looking in the code before I post :0

This question comes up allot on these forums! The answer is NO; you cannot use that table for subpanels. The NEC does a TERRIBLE job of defining this situation. Buy "Electrical Wiring Residential (14th edition)" from this site's bookstore if you want some clarification.

The "(S)" and the “feeders” in the code pertain to certain specific situations. I'll give two examples. If you had a 320A meter socket feeding a large house, you could use that residential table to size the wires feeding from the meter to both a 200A and a 100A panel inside the home (two panels, hence the “(s)”). Also, imagine your service entrance conductors first terminate in a fused disconnect at the closest point inside the house. Then a single “feeder” left that disconnect and went some distance away to a panelboard (so that the wire ran inside the home was fused). In this case, both the S.E.C. and the feeder could be sized using 310.15(B)(6).

Some people will dispute my claim that the aforementioned table cannot be used for subpanels. However, nobody has ever been able to site a source that backs up his or her side of the issue. The last time this topic was raised, I practically begged the "opposition" to provide the name of an outside resource that substantiates their view. Nobody could. They were all just making their own interpretation of that VERY ambiguous part of the code (one of many such parts).

I would very much like to use table 310.15 (B)(6) for residential subpanels. Therefore, I’d like to offer $50 to anyone that can give me proof that this can be done. Provide me with a RELIABLE source that contradicts the one I already have. If the source you give me is extremely detailed, I will make it $100. No tricks or oddball cases. My offer applies to the average residential subpanel fed from the average residential main panel.

Let’s see it!

Peter, seriously, you must forget the article 100 definition of "FEEDER" and in this case, use the definition in 310.15(B)(6).

Roger

Triple no offense but what makes your interpretation more correct than others?

200 amp service, the service entrance conductors 2/0 cu are sized by 310.15(B)(6) now I add a 200 amp sub panel.

I will use 2/0 cu right?



[This message has been edited by iwire (edited 04-08-2004).]

will that be Paypal?

Iwire, I provided a source. I am not just giving my opinion. Please provide your source (if yours differs from mine). I am definitely not trying to say that I am better at interpreting the NEC than anyone else. I leave the difficult interpretations up to well-known, published experts. I realize that such people can also be wrong but their opinions hold far more weight, than those of the average electrician posting on a forum.

I want to reiterate that I do not consider myself to be an expert in any realm. However, I do rely on the "expert" view of those published few (relatively) to provide some degree of analysis since the codebook itself is far too vague.

Please people; do not dance around the issue yet again. If subpanels can be fed using 310.15(B)(6) then surely there is a book written by SOMEBODY, SOMEWHERE that says so. Give me the name of this book.

Triple I do not know the answer to this question.

My only point is that whatever book you are reading or whatever book I might find to support the other side is not the NEC so it is not an official interpretation, it is just the authors [options] edit opinions.

Their opinions must be based on the same confusing wording that you and I read.

The only true interpretation as to come from the NFPA or in some States the local AHJ.



[This message has been edited by iwire (edited 04-08-2004).]

Tripple, you and I agree with the jest of this thread, but at the risk of offending some "published experts", I don't agree with your statement
I don't agree!

Now, I do agree with Iwire here, .

With all this agreeing and disagreeing the published people aren't very convincing. .

Roger

Iwire, if you reread my posts slowly you will see that I also agree that anything from an outside source is only the author's opinion. Also, I provide a specific source to back me up, not just "whatever book". There are over 6 billion people in this World so surely there must be at least one, well-known, published writer that supports the use of that table for subpanels if indeed it can be done. If no "expert" is willing to write it down, then how could it possibly be true?

I have to believe that the code writers know EXACTLY what that table pertains to. Thus, there has to be an exact and precise answer to this question. We are not talking about a subject as wide open as "what is the meaning of life". The code is not meant to be a book of riddles even if is can sometimes seem that way. The code is written by humans (not passed down by God on stone tablets) so its meaning is known PRECISELY by those that write it. If it was not then my life and work are based on a myth!


[This message has been edited by triple (edited 04-08-2004).]

All I am saying triple is you gotta go to the source if you want the real answer.


Of course the CMP knows what they mean, but you can not get that info from any outside source, it must come from the correct agency.

Do you know that if someone here at the forum was a member of a code making panel that they would have to put a disclaimer up that says that their post is their opinion and not that of the NFPA, this would even be true if the person literally wrote the words in question.

Of course another way to handle this would be to use a screen name instead of their real name.

Roger, the forum members here could possibly run the gamete from having never run a wire in their life to being an electrician studying the code for 60 years. The average person could be someone that can only squeak by on his or her state Journeyman test. Can you honestly say that this person's interpretation of a part of the code will hold as much or more weight than a published writer selling books on the same site you moderate? If a moderator is publicly stating that the books sold on this site have that little merit then I definitely want my money back!

I'm just jerking you around. I realize, as a moderator, you probably feel compelled to stand behind fellow members. I'm not trying to put anybody down, as I don't feel I am any more knowledgeable than the "average" forum user. Unlike the average person though, I don't rely on my own guesses and accept them as fact.


[This message has been edited by triple (edited 04-10-2004).]

Thanks for all the answers, guys. My head is spinning a little less now.


Ok, it's forgotten. But it's still confusing. That's why I looked up the definition of a feeder in the first place. I guess this article modifies the NEC definiton of a feeder, right?

I agree with you Roger that this article should be rewritten. IMO, it should be rewritten to allow Table 310.15 B6 to pertain to feeders from the main panelboard as well.

Confession: The subpanel I added at my home a few years ago is #2 AL SER protected by a 100 amp breaker. No, I'm not going to change it to a 90, in case anyone is wondering.

Peter

[This message has been edited by CTwireman (edited 04-08-2004).]

Triple, yes I know your just "jerking me around" and I'll give you credit to be able read and translate my post as I meant it.

Look at where I said that you seemed to have (in a blanket) put all your faith in.

As to having the privilege of being a moderator here, I was never asked to patronize anyone.

Roger




[This message has been edited by Roger (edited 04-08-2004).]

(Alas I do not have my copy of the NEC handy, so I am working from memory...I promise I'll check it later)

This issue comes up because the 'load calculations' produce a single number (KVA), which we presume is a real value, rather than a nominal value which is used to size conductors and OCPD. A 12ga conductor is _not_ a 20A conductor; it is a conductor with a particular time/temperature curve, with particular characteristics in extreme overload, etc. In general use, it is limited to operation on a 20A circuit breaker, again with its own time/temperature curve, with various restrictions on its use.

For things like welders or electric motors, the various articles describe how you can use one value for the conductor ampacity, and a _different_ number for the OCPD rating, this based upon the time characteristics of the load, the internal overload protection, etc.

The same applies to 'dwelling' loads. To supply sufficient power for a 'dwelling' you need conductors of a given ampacity and OCPD of a given rating...but rather than have load calculations which produce these two separate numbers, we have a load calculation which produces a single number, and then table 310.15(B)(6) which gives different conductor sizes to use for the single number produced by the load calculation.

If I do a load calculation on a residence, and it comes out to 82A, then it is certainly _not_ the case that this residence will be drawing 82A continuously. The 82A number simply tells us what OCPD is required and what size wire is required. This is rather different from a single device which draws 82A continuously, and both the conductor and OCPD sizes will be different for these two applications.

My interpretation of the matter: 310.15(B)(6) applies any feeder which supplies the dominant load of a single dwelling, and is intended to better match the load characteristics of a dwelling. It does not apply to feeders which do not have the same overall characteristics of a dwelling, including feeders within dwellings, eg. for air conditioning or a hot tub.

If there is a subpanel in this dwelling that does _not_ carry a 'dwelling type' load, then 310.15(B)(6) would not, and could not, apply. For example, a subpanel to a workshop is rather different from a subpanel which includes the kitchen, the air-conditioning, the heating, the lighting and the various house branch circuits. So, for example, if you have a main disconnect at the meter, and then feeders to two subpanels, one a workshop in the garage, the other the 'main circuit breaker panel' for the house (meaning the panel that has the branch circuit breakers for all of the residence circuits), then 310.15(B)(6) would apply to the feeder to the residence side of things but not to the feeder to the garage.

Then, in addition to the loads which 310.15(B)(6) is designed to properly size the conductors for, there is language that _sometimes_ extends this sizing to other feeders in residential occupancies. In particular, there is the wording that says that no feeder need be larger than the service entrance conductors. I take this to mean that you can size the service entrance conductors using 310.15(B)(6), then continue to use _this_ size for any subpanel feeders.

For example, a 200A subpanel in a home with 200A service. The feeder to the subpanel need not be larger than the service entrance conductors, even if normally you would have to use 310.16 and use larger conductors. This makes sense in that the current to the subpanel must be less than or equal to the main service current. However this note does not extent 310.15(B)(6) to other feeders. For a 100A subpanel to a home with 200A service, 310.15(B)(6) could not be applied to the subpanel feeder.

-Jon

[This message has been edited by winnie (edited 04-09-2004).]

It is very clear here that the NEC book does a very poor job of explaining its position. I understand that outside sources cannot be enforced as if from the NEC. However, in cases such as this, what else is there? How many of you have bought books on grounding or motors and relied on the info they provided?

I don't have a very large library of electrical books. I'm sure many ECN members have much larger ones. Since nobody has come forward (in this thread or in previous ones) with a source of opposition, then I can only assume such a source does not exist.

Not a single published person agrees that the residential table can be used for subpanels?????????????????????????????

How can that be?

Use your common sense and an answer is readily apparent.

I say just the opposite, use your common sense and the answer is more elusive.

If you have a combo meter / service disconnect outside and a panel inside this panel inside is a 'subpanel' (separate grounded and grounding conductors) you may use the table for the conductors running between the service disconnect and this 'subpanel'.

Your right the wording is tough.

Bob

Iwire, you took my words out of context. When I said, "use your common sense" I was only referring to the fact that not a single person has written a book supporting your side of the issue.

The NEC does not (to my knowledge) define what a subpanel is. The situation you describe would not be a subpanel in the eyes of table 310.15 (B)(6). Since the conductors you are referring to would carry the entire supply to each residence then the afformentioned table could be used.

Iwire, since you apparently don't believe in the use of sources outside the NEC then how can the word subpanel even be in your vocabulary? Do you have a selective memory;...only information that supports your personal position is absorbed?

I hope that nobody takes offense to my comments. I feel that "the truth is out there". So if your view is the truth then where is it? Do you want me to believe that the truth is a secret or perhaps a conspiracy? If it is not meant to be hidden, then where is it?

I believe that Roger and Iwire both have more written resources available to them then I do. So please take 20 minutes, refer to them, and get back to this thread with your newfound information.

Do I have to beg yet again


[This message has been edited by triple (edited 04-10-2004).]

Triple, that was good on the subpanel thing.

The application you describe (the 320 meter socket with two panels and two feeders) is without a doubt one way to apply this table.

Now as the only thing that matters is what is written in the code book (as you just nailed me with ) not what other sources say, lets really look at the wording here, totally forget what we may think the intent is, just what is written down.


As it is written in this section there is nothing to prevent me form using this for 10 feeders and panel boards one after the other instead of side by side like your 320 example.

In both cases these panels will not be suppling the main power to the building, in your 320 example each panel would be suppling about half the power.

You, me, and the authors of the books can only go by what its written in that article beyond that it is time for an official interpretation.

What I said about common sense was not a swipe at you at all, what I mean is electrically it makes no common sense that tis can apply to the main panel and not an additional panel.

Both panels sizes and loads are calculated by the same methods.

Like I said before take the table right out and have all of us use 310.16.
Bob

Our local inspector has a book of proposed changes to the NEC. I don't know what it is called but I'm sure one could find it at NFPA. But, we thumbed through it to try to pick up on some of the intentions of the CMP. There was a proposal to make an exception to the recpt. requirement on a 2' section of wall if that section of wall was buried behind the swing of a door. Then it shows why they rejected the proposal with an explaination.

The reason I bring this up, if anyone has this book, there may have already been a proposal to eliminate that section with a rejection and an explaination directly from the CMP. If so, (and hopefully it is clear enough) that may be enough to satisfy all parties.

Great point Jps1006. The way I first learned that 310.15 (B)(6) could not be used for residential garages was by reading the proposed changes. Someone suggested allowing that table’s use for such a purpose and the panel provided reasons why it could not. Is there a place where a person can easily look up all previous proposed changes and subsequent rejections on a given code issue (I know the most recent ones are always available but that may not be enough)? That may be the best way to resolve this issue once and for all.

Please people, I know I am not the only person with a book that deals with this subject. Quickly review those books and let us know your findings. Am I the only one that reads anymore? Does everybody else just slide through life assuming they already know it all?

Triple, do you really believe everything you read or hear?

Look at the NEC handbook, it is only opinions.

If you insist on being so insecure that you need a published document to substantiate your direction, than so be it.

If Tesla would have settled for the published words, we wouldn't be talking about this now.

Roger

[This message has been edited by Roger (edited 04-10-2004).]

Roger, how many times do I have to acknowledge my understanding that writings outside the actual NEC are not code? Between this thread and previous ones I must have said it 10 times! Therefore, when you write, “do you really believe everything you read or hear” you already know my standing. Obviously you are making that statement because its audacity makes me sound ignorant.

Tell me Roger, did you learn everything you needed to know about your electrical career by simply reading the codebook? In other words, did you one day just pick up a copy of the NEC, read it cover to cover, and then call yourself an electrician? Of course you didn’t. Like most (all?) electricians you probably began your career by working under those already experienced in the trade. You learned not only the procedures involved but also the code from (gasp) the electrician you worked under. Shame on you for not ignoring those outside sources!

This is the way I learned. I have since found that some of what I was taught was incorrect but most was not. I know that if all my electrical related info came straight from the NEC, I would have misinterpreted and misapplied about 50% of what I read! I salute you if you’ve been able to independently implement the NEC from day one on the job.

For argument sake, let’s say the 100 top NEC savvy individuals in the US (subjective, I know) were asked to weigh in on this issue. If all 100 agreed that subpanels did not fall under the residential table then would you be willing to accept this ruling from outside the NEC? Or would you feel more comfortable with your own interpretations? I’m not making a comment on your standing on the table issue (especially since I believe you stated you agreed with me) but rather on your apparent disinterest in outside sources.

Don’t get me wrong. I’m not saying that the residential wiring book I received my information from has to be correct. I don’t know the author from Adam. For all I know, he could be a kook with an agenda involving the spreading of misinformation for his own amusement. For every expert that leans this way, there could be 10 others that lean the other. What I was hoping to accomplish here was to get others involved in doing a tiny amount of research. Apparently I am asking too much.

Dude, Chill out a little bit.

Forget about the outside sources and read the article out of the NEC itself and tell me (or us) what part supports this view you are getting so upset about.

No one here thinks your ignorant, you obviously care a great deal about your trade.

But please forget about my opinion, Rogers opinion and your books opinion and read carefully word for word what the NEC article says.

I know a lot of people share the view you do, I just do not see anything in the code book to support it.

The more I read it it looks like both your 320 example and mine would be a violation because neither of those applications supplies all the main power.

But... we have those darn (s) to deal with.

Bob,

BTW Roger, are you comparing your decision-making methods to those of Tesla's? You also "need a published document to substantiate your direction". That document is the NEC! You are not dreaming up your standing all on your own.

On another point, I didn't go in search of a document that would "substantiate" my "direction". Actually, I had always used the residential table to wire subpanels until I read on a forum that this was not code compliant. Only then did I search out an outside source (after repeated attempts to decipher the NEC myself). I had to change my "direction" after doing this research. Like I said before, I very much want to use table 310.15(B)(6) for subpanels so it's not like I set out to debunk my previous view.

Simple question, hopefully a simple answer. (I doubt it though.) What is the danger of using this table for "subpanels" in a residence? As I stated in my intro, this is commonly done around here. I know that doesn't make it right, but common sense should prevail.

Peter

The danger that I see in using the table in question for anything less than "serve as the main power feeder to a dwelling unit" would be the possibility that the lack of diversity in the load thus served could cause chronic overloading of the conductors. The example I would give is the service call I did in a dwelling with a feeder supplied panel that supplied the air conditioning loads for a home having three separate AC systems. All the loads on that panel were 240 volt. During a particularly brutal hot spell the feeder faulted out. Since the feeder had been run in flexible metallic conduit the failure was confined to the feeder itself. Rather than just repair the feeder by replacing the conductors I checked the size against the ampacity tables and found they were sized in accordance with Table 310.15(B)(6) using number four copper RHW conductors for the feeder that was protected by a 100 ampere breaker. When first called out on the sunday afternoon I had used a 2/0 Al service entry cable as the temporary feeder to that panel. It was the only wire that I had on hand that seemed safe enough for a one day fix. Once I was able to research the problem and found that the original feeder was overloaded I purchased number three THHN - THWN as the replacement conductors. Since my suspicion had been aroused I megged the feeders for the kitchen panel and the panel to the detached garage. The feeder to the all electric kitchen failed at the 500 volt test level. It was also sized at number four copper for a 100 ampere breaker. This was also replaced with number three THHN. The feeder to the garage workshop was number 2 Al UF. Since it had megged clear at 500 volts I simply replaced the breaker with a 70 ampere breaker which was within the ampacity of that wire type under table 310.16. I think that diversity is the critical factor here. If the circuit in question supplies "the main power" to a dwelling unit then it is unlikely that so many of the loads will be used at once that the dwelling feeder sized circuit will be overloaded. On the other hand if you feed only part of the load with the feeder diversity is not assured or even perhaps likely depending on the type of load that the feeder supplies.
--
Tom Horne

The danger applying 310.15(B)(6) to subpanels:

IMHO this all comes down to load diversity and the current versus time statistics of a given load.

A residence has a rather different load characteristic than an industrial plant, and the whole residence has a rather different load characteristic than the electric heaters inside that residence.

Depending upon the load characteristics, the OCPD protecting a set of conductors may or may not be matched to the continuous ampacity of those conductors. An example would be the use of oversized OCPD to protect an electric motor, where the thermal overload protection of the motor itself is permitted to provide the overload protection for the conductors, and the OCPD is used more for short circuit protection.

IMHO 310.15(B)(6) assumes the time characteristics of residence loads to mean that the conductors will be properly protected where the OCPD rating is greater than the continuous current rating of the conductors as determined using 310.16.

Thus IMHO if the load has a time characteristic that is anything like that of a normal residence, then there will be essentially no danger from using 310.15(B)(6) for a subpanel...it may not be 'code' but it will be okay.

It is only if the load is very different from an average residence, very high (in proportion to the feeder rating) and continuous that I believe that you will see problems.

In the worst case, the conductors will end up at 90C or above, and the insulation will degrade rapidly.

-Jon

Jon & Tom understand I have no inside knowledge on this subject and do not have the option of using this table at my job.

You both bring up some interesting points to think about, the problem I have is there is nothing in that article that talks about diversity.

Another problem I have is if you calculate the loads for both panels in the same way it will not make a difference.

As an example a subpanel sized to feed only two outside air conditioner compressor condenser units.

They both have a minimum circuit ampacity of 50 amps, we install a 100 amp subpanel using Table 310.15(B)(6) which requires 4 AWG CU with 75 C or 90 C conductors.

This panel at most see a load of 80 amps, as the tag on the HVAC units already includes the 125% increase required by 440.

310.16 lists 4 AWG CU @ 75 C as 85 amps, so how would this be a problem?

triple,

I understand what you are saying about outside sources. I think they are important too. Afterall every code ever written was once the opinion of an outside source.

With all do respect, guys, your point is pretty clear. But triple's point is a separate, but valid point.

Sorry to stick my nose in if it doesn't belong, but I understand triple's frustration even if it is seen as him needing to chill out.

Now Bob,

I don't think you could call a panel with 2 50A air conditioner circuits and nothing else on it a 'lighting and appliance panelboard' (just yanking the chain a bit)

I agree that the code in 310.15(B)(6) doesn't mention load diversity or any sort of load time characteristics; I guess that I am reading this into the code based upon the similarity to things like the explicit reduced ampacity of conductors for welders and the like, related to the duty cycle of the welder...remember though that I am a theory guy rather than someone with experience (again *grin*, Tom was writing his post while I was writing mine, and my jaw dropped when he described as a service call that he did, something that I'd theorized was possible...)

It is not as though the electrons going into a residence are somehow different from those going into an office building, such that the actual continuous amapcity of the conductors changes (again, tongue firmly in cheek). The difference in the situation has to be that conductors into residences are not expected to be used with the same time distribution as conductors in other occupancies. Again, this is just me reading into code.

In the example that you gave, I think that _in general_ there would be a problem putting a _continuous_ 80A load on 85A conductors with 100A OCPD. The general requirements are that both the OCPD and the conductors be increased in size to meet the 125% of continuous load requirements. For motors the requirements are even stricter; you don't have the '100% rated OCPD' exception. The only time that it is okay to have conductor ampacity > load amps but < OCPD rating 'in general' is when you are using the round-up rule for the OCPD.

Where to go with this: I still believe that applying 310.15(B)(6) to anything that doesn't look like a residence _as a whole_ is at the very least a 'bad idea', and depending upon interpretation violates code. As Bob noted, however, the '(s)' in the actual code text leave a big hole open for interpretation, as does the concept of 'main power feeder'. Clearly the air conditioning panelboard that Tom described is fed by a feeder in a residence, and if there were a couple of general purpose lighting circuits on it then it would be a lighting and appliance panelboard.

Here is perhaps a different interpretation of 'main power feeder' to multiple panels...not saying that it's right, just presenting another possibility. If 310.15(B)(6) is used to size multiple feeders to multiple panels in a residence, then _all_ of the feeders must be sized using 310.15(B)(6) for the ampacity of the _entire_ service.

This clearly does not match 'common sense', unless the use of the '(s)' was intended to address multiple panels and feeders in series rather than in parallel.

-Jon

My opinion of the (s's)is as follows.

This article specifically allows reduction to the first lighting appliance
branch-circuit panel or panels immediatly down line of the service equipment.

Using a lateral in this example

400A service
2-200 amp lighting appliance branch-circuit panels.

Service lateral --400KCM copper per article 310.15(b)(6) to a 400A OCPD.

We have not made it to our lighting appliance branch-circuit panels (note plural) yet.

For sake of coversation we install 2-200 amp enclosed breakers adjacent to
this 400 amp OCPD and feed these from parallel lugs.

We use 2/0 copper per 310.15(b)(6).

We still have not made it to our lighting appliance branch-circuit panels
yet.

We feed 2-200 amp lighting appliance branch-circuit panels with 2/0 copper
per 310.15(b)(6)and make all terminations.

We have finally made it to our lighting appliance branch-circuit panels.

Now 310.15(b)(6) has seen it's end and if we feed any sub panel, say a 100 amp, from one of these aforementioned lighting appliance branch-circuit
panels, this 100 amp sub panel can not use 310.15(b)(6)

Just my unplublished opinion.

Roger



[This message has been edited by Roger (edited 04-12-2004).]

I've been watching this thread with interest, as this seems to cause more stress than just about any issue. I believe Tom Horne and Jon/Winnie have the technically correct answer, but the code words are not clear.

Roger - would your answer be any different if the two 200A breakers were changed to 4 100A breakers on #4 copper (all from parallel lugs and assume each 100A feeder terminates in a 100A MLO lighting and appliance panelboard)?

In dwelings, there is some implicit diversity as the load increases (you don't see feeders under 100A in table 310.15(B)(6)). There is also some implicit diversity in a lighting and appliance panelboard. However, the 10% threshold to change from a power panel to a lighting and appliance panel seems way too small to me. In the largest panels, you only need four or maybe five 30A or less circuits with neutrals to make a lighting and appliance panelboard. You only need 1 in a 10 circuit panel.

If I was NEC god, you could leave the words to 310.15(B)(6) as is if either lighting and appliance panelboards were redefined to have more <=30A breakers (these are the most diverse loads) or you mandated that the target panelboard had some minimum number of circuits. I'd be inclined to decree you could use that table if there were 4 or more single pole breakers rated 20A or less -or- there are at least 6 breakers of any size. I would also consider tossing in a restriction that the largest breaker in the feeder target panel could be no larger than 50% of the feeder rating.

But I'm not NEC god...

Welcome to the forum Mark.
If tap rules were followed, any number of lighting appliance branch-circuit panels
can be fed using 310.15(B)(6) immediatly downstream of the service equipment.

If we look at a 30 unit condo, (any number can be used) from the Metering Equipment, and disconnects at this location, the individual feeders to the lighting appliance branch-circuit panel in each unit could use 310.15(B)(6).

Roger

So if multiple 100A panels are OK, what would be wrong with two 100A feeders off of a 200A panel each feeding half of a dwelling? How about 3 100A panels from a 320A disconnect each supplying 1 floor of a large house? This seems to be the issue -- at what point is a feeder supplying a diverse enough portion of a dwelling to qualify for 310.15(B)(6)?

This seems difficult to codify easily as it would be possible for the dumb electrician to put a bunch of electric heat, water heater, and one 15A outside branch circuit all in one panel fed with #4 copper. Conversely, if all of the general purpose branch circuits and all of the required ones that have no calculated load (outside, basement, bathroom, etc) were put on a 100A panel with #4 copper, you'd porbably never have more than half of that being used for any significant time.

Mark, nothing would be wrong with this if the 200 amp panel were a "Power Panel", if it were a "lighting appliance branch-circuit panel", 310.15(B)(6) has already been used.


This would be fine in that the feeders would fit the discription of "Main Power Feeders" per 310.15(B)(6).

Take note that my opinion of this issue is simply the way I interpret the poor wording of this article, which I think many of us agree is the case.

Roger

Guys I have to point out that the plural in "feeder(s) between the main disconnect and the lighting and appliance branch-circuit panelboard(s)" is meant to apply to multiple dwellings that are supplied with single phase 120/240 service. This is a very common situation in small apartment buildings. A good example is a two story and basement building code construction type III (masonry bearing walls with wood floors and roof framing; generally called ordinary), with two units on each upper floor, one unit in the basement and a basement laundry room. With a calculated load of 370 amps. The service entry conductors to the meter bank could be 400 MCM copper. The service entry conductors from each meter to the five apartment units could be number four copper. But the service entry conductors to the 100 ampere house panel that serves the laundry room, heating plant pumps, and so forth would have to be number three copper.

What is obvious to me is that the language of section 310.15(b)(6) needs to reflect a one dwelling unit one feeder approach. I've called the NFPA; since I'm a member; and I've gotten the forms for a request for formal interpretation. Short of a change in the code language that is the only authoritative way to resolve this. In my conversation with the engineer in the electrical section he assures me that the intent is one dwelling one set of conductors. Before you tell me that it is only his opinion please keep in mind that he is staff to the code making panel that is responsible for that section. He even pointed out that parallel conductors for these services cannot be sized under Table 310.15(B)(6).

One correct way to install the entrance conductors for a four hundred ampere service using two, two hundred ampere, panels is to equip one of the panels with double barreled lugs, run 600 MCM Al or 400 MCM Cu to those lugs from the meter can, then run 3/0 Cu or 250 MCM Al to the second panel under exception two to 230.40. But before going to the expense of that size of conductor remember that the size of the service entrance conductors serving more than one disconnect is governed by the calculated load rather than by the sum of the ratings of the two to six disconnecting means. If your calculation is for a 250, 300, or 350 ampere service then there is no need to use four hundred ampere conductors.
--
Tom Horne


[This message has been edited by tdhorne (edited 04-14-2004).]

[This message has been edited by tdhorne (edited 04-14-2004).]

Hello Tom, now we have another controversy to the wording.

Let us know if you have any satisfaction with the formal interpretation.

As of right now and the wording provided, I don't agree with this person, (yes, I know he is a member of the CMP for this section) although the intent may be otherwise, it isn't written to reflect it.

This will take more than one question, and would almost take rewriting 310.15(B)(6) in a series of questions for any interpretation/s (note the plural) to put this to rest.

I have heard (hearsay only) that formal interpretations must be worded as YES or NO questions, this will cause confusion in itself.

BTW, I don't think parallel conductors are questioned since they are not mentioned in the table, and therefor could not be considered anyways.

Roger



[This message has been edited by Roger (edited 04-14-2004).]

Tom, you may be right, maybe you do have inside info on this, maybe your contact is 100% correct.

Here is the problem, all electricians are supposed to be able to use and apply the code without 'inside info'.

Right now there is no language in the code article to support (or dispute) your stated reasons for the (s).

As they do not tell us what the (s) is for, we can use it as we like.

And I agree, this table does not apply to parallel conductors.

Bob

I have been staying out of this until now but will give my opinion even though it’s not worth much.

I have debated this issue with Don a couple of times over the last couple of years and have always believed it to be exactly how Tom stated. In previous additions of the NEC the only feeder conductors that could be sized according to table 310-15(B)(6) had to carry the entire load of the service. I believe it was the 96 addition when the change was made to allow feeders that carry the main power to the dwelling to use the reduced sizes. Before the 96 addition of the NEC apartment feeders could not use this table and had to be sized according to table 310.16. In my area we always have over current protection in residential meter sockets. Often there are two 2-pole service disconnect breakers, 1 of the breakers will feed a panel in the house and the other will feed an exterior hot tub or swimming pool. In this case table 310.15(B)(6) can be used for the feeder supplying the panel in the house because it carries the main power for the dwelling since the pool or hot tub are not part of the dwelling.

Curt

The problem that I have with this section is that there is nothing in the definition of "main power feeder" that requires this feeder to supply the main power to the dwelling unit. There is nothing about the diversity of the load. The wording only requires one thing: that the feeder be between the main disconnect and the lighting and appliance branch circuit panels. It doesn't even say that this feeder must be directly between the main disconnect and the lighting and appliance branch circuit panels. If you just apply the words as written, then all feeders within the dwelling unit are "main power feeders". This is very poor code writing. Yes, the intent was that these feeders are required to provide the main power to the dwelling unit, but it is not easy to enforce intent. The inspector can only enforce what is written in the code book and the words as written permit all feeders that feed lighting and appliance branch circuit panels to the sized per Table 310.15(B)(6).

Don

I have submitted the request for formal interpretation. I'm told it will take a couple of months for that to be processed.
--
Tom H

[This message has been edited by tdhorne (edited 05-04-2004).]

Tom, I for one anxiously await the answer/s

Roger

Tom, is there any update to your request for the formal interpretation?

Roger

WOW! The file tag was flaming for a reason.

Just for information porposes...
Not trying to get into this...
The Handbook commentary...
Not that it has anything to do with it... Just a refferance...
http://home.mindspring.com/~e57/310156.htm

The following is a paste from the ROP and the ROC for the 2005 NEC to Artcle 310.15 (B) (6)

I have left the names in the paste to give credit to those who attempt to improve the readability of the Code. This is a public document so I do not believe it is an issue.

I have checked the TCR's and ROP's and ROC's back to 1992 and their is no specific language attempting to clarify the issue (untill this cycle). I could go further back but it's too late to back to work. I don not believe that this information will solve the issue but it may give insight to the complexity of writing clear concise enforseable code

ROP
Final Action: Accept in Principle in Part
Submitter: Paul Dobrowsky Holley, NY
Recommendation:
Revise as follows:
(6) 120/240-Volt, 3-Wire, Single-Phase Dwelling Services and Feeders. For individual units of one family, two-family and
Multifamily dwelling units, conductors, as listed in Table 310.15(B)(6), shall be permitted as 120/240-volt, 3-wire, single-phase
service-entrance conductors, service lateral conductors, and feeder conductors that serve as the main power feeder to a each dwelling unit
and are installed in raceway or cable with or without an equipment grounding conductor. For application of this section, the main power
feeder shall be the feeder(s) between the main disconnect and the lighting and appliance branch-circuit panelboards(s). The feeder
conductor to a dwelling each unit shall not be required to be larger than their service-entrance conductors. The grounded conductor
shall be permitted to be smaller than the ungrounded conductors, provided the requirements of 215.2, 220.22, and 230.42 are met.
Substantiation:
The present definitions in Article 100 for One, Two, and Multifamily dwelling units, literally limit this application to a One Family
Dwelling in a separate building.
Panel Meeting Action: Accept in Principle in Part
Revise the recommended text to read as follows:
(6) 120/240-Volt, 3-Wire, Single-Phase Dwelling Services and Feeders. For individual dwelling units of one family, two-family and
multifamily dwellings, conductors, as listed in Table 310.15(B)(6), shall be permitted as 120/240-volt, 3-wire, single-phase
service-entrance conductors, service lateral conductors, and feeder conductors that serve as the main power feeder to each dwelling unit
and are installed in raceway or cable with or without an equipment grounding conductor. For application of this section, the main power
feeder shall be the feeder(s) between the main disconnect and the lighting and appliance branch-circuit panelboards(s). The feeder
conductors to a dwelling unit shall not be required to be larger than their service-entrance conductors. The grounded conductor shall be
permitted to be smaller than the ungrounded conductors, provided the requirements of 215.2, 220.22, and 230.42 are met.
Panel Statement:
The panel does not agree with changing "a dwelling" to "each" since a dwelling unit is defined in Article 100.
Number Eligible to Vote: 11
Affirmative: 11 Ballot Results:

ROC
Submitter: Frederic P. Hartwell, Hartwell Electrical Services, Inc.
Comment on Proposal No: 6-41
Recommendation: Accept the panel action in principle. Clarify the permissible
application of the multiple feeder allowances as one of the following four
options:
1) “… the main power feeder shall include the feeder(s) serving only loads
associated with a single dwelling unit and running to but not originating in the
lighting and appliance branch-circuit panelboard(s) serving the dwelling unit.”
OR
2) “… the main power feeder shall include the feeder(s) serving only loads
associated with a single dwelling unit and running to the lighting and appliance
branch-circuit panelboard(s) serving the dwelling unit.” OR
3) “… the main power feeder shall include the feeder(s) serving only dwelling
loads and running between the main disconnect and the lighting and appliance
branch-circuit panelboard(s) serving the dwelling unit.” OR
4) “… the main power feeder shall include the feeder(s) serving only dwelling
loads and running to but not originating in the lighting and appliance branchcircuit
panelboard(s) serving a particular dwelling unit.”
Report on Comments — May 2004 Copyright, NFPA NFPA 70
Substantiation: By clarifying that this note applies to dwelling units within
multifamily housing, which is well advised, the proposal raises important questions
as to exactly which panelboard feeders are within the scope of this allowance.
Options 1 and 2 exclude feeders that are comprised of dwelling loads, but
that serve multiple dwelling units. Options 3 and 4 allow such a feeder. Options
1 and 2 as a group and options 3 and 4 as a group sort out whether this allowance
applies to subpanel feeders within a dwelling unit. Dwelling unit subpanel
loads do not present the same diversity as dwelling unit panels serving the
entire dwelling unit, and thereby undercut one of the traditional supporting
assumptions underlying these allowances. However, all of these interpretations
are possible given the ambiguous “(s)” endings on the word “feeder” and
“panelboard.” CMP 6 needs to clarify exactly which feeders qualify for this
allowance.
Panel Meeting Action: Hold
Panel Statement: The panel is holding the comment for further study, because
it presents new material that has not had public review.
Number Eligible to Vote: 11
Ballot Results: Affirmative: 11

It would appear that those who write the code also find the wordin ambiguous

And as a result the CMP has put action on hold

It will be interesting to see the text of this section in 2005.

I think it will appear unchanged.

Tom, have you heard anything on your request?

Roger

Not yet.
--
Tom