220.58 (New)

Status
Not open for further replies.

George Stolz

Moderator
Staff member
Location
Windsor, CO NEC: 2017
Occupation
Service Manager
220.58 (New) New Text
Proposed Text:
220.58 Load Diversity – Dwelling Unit(s). It shall be permissible to apply a demand factor of 88% to the total calculated load, after applying demand factors allowed in other portions of this Article, where the feeder or service conductors supply all of the loads associated with an individual dwelling unit. This demand factor shall not be applied to conductors serving more than one dwelling unit.

Substantiation:

This proposal is submitted in coordination with a proposal to delete 310.15(B)(7). Currently, sizing conductors for dwelling units is an anomaly in the NEC. We perform a load calculation, select a service rating, and then select a conductor while paying no mind to the myriad of corrections that we would otherwise need to address in 310.15. Dwelling units are essentially being written a blank check, blatantly allowed to ignore ambient temperatures, adjustment factors for multiple conductors in a raceway or cable exposure to sunlight on rooftops, based on load diversity alone.

We are also left without clear direction as to how to proceed when we would like to install parallel conductors. CMP-6 has made statements in the past that using the 200A rating to size a pair of parallel conductors for a 400A service is acceptable, but the code language and Table do not support that conclusion. We are also left without clear direction as to how to cope with a neutral conductor reduced in size. If I have a 200A calculated load on a neutral conductor serving a 400A service, may I supply it with a 2/0 CU? The service rating is 400A, Table 310.15(B)(7) is telling me to either use a 400 kcmil CU or look elsewhere – but it is commonly interpreted as acceptable (and indeed, it is sensible) to size the neutral using the same diversity and Table that gave me my ungrounded conductor size.

Additionally, it is a fact that 310.15(B)(7) is commonly misinterpreted to allow 120/208V systems to make use of it. I know better, and the clever fellows serving on the CMPs knows better, but it is very common for electricians and engineers alike to see a residential table and believe it applies to all feeders and services residential. It looks like an unintentional omission to the NEC. Proposals have been made (and even accepted) to make the paragraph an easier-to-digest list format, but the 2011 still features the same obtuse paragraph that just serves to confuse.

Additionally, it is a fact that Table 310.15(B)(7) is commonly used erroneously to size feeders for panels not serving 100% of the diversified load. I have even heard of electrical contractors getting caught using the Table for sizing of feeders in clubhouses associated with apartments! If you see a service with a 90A breaker in it, it's a sure bet the EC got a surprise. If they didn't need a 100A feeder in the first place, is anyone really losing anything by applying the appropriate breaker size according to ampacity?

It has been said that “not conforming to the NEC is not an excuse for amending it” - but that statement pales when CMP-6 itself cannot maintain an interpretation from one proposal to the next within the same code cycle. In the ROP leading to the 2011, in proposal 6-86 the panel treats Table 310.15(B)(6) as a “sure nuff” ampacity table, and then proceeds to explain in the next proposal, 6-87, that the Table does not deal with ampacities. Have I sold you yet?

This table is akin to junk food – a taste here and there is fine, but it is in fact too easy to use, and at the same time too easy to misuse. I have come to the conclusion that it would be clearest to simply delete the table, put a reasonable and time-tested value to reduce our load by, and use normal rules to establish overcurrent and conductor sizes.

CMP-6 has gone on record in the 2011 proposal 6-86 saying “the conductors of a 120/240-volt, single-phase dwelling service or feeder with a calculated load of 200 amps will never carry 200 amps.” That is correct, but it is not the voltage of the circuit that creates this magic. It is the nature of the occupants of the structure that creates this diversity. We can use this new rule on all dwelling units, whether the supply is 240 or 208 volt. Previous attempts to include 120/208V in 310.15(B)(7) have met with the argument that the neutral of a 208V system carries full current when the phase conductors are fully loaded. This argument is peculiar, because load diversity and neutral current are unrelated. If we assume that the phase conductors of a 120/240V service can be protected at 112% of their ampacity or more based on faith, then both phase conductors could be possibly be overloaded 112%. Why do we lose this faith when dealing with a neutral conductor? In a 120/208 system, the neutral would be loaded to 112% as well, but no more. Dwelling units do not have a reputation for harmonic loads, so the neutral in 99.9% of the 120/208V installations can take advantage of this new rule with no practical reduction in safety. The big picture with the existing 2011 text is that people and families produce load diversity, they are the origin – not the voltage they are utilizing.

In summary, in every case except dwelling unit feeders, we select conductors based on their ampacity as it relates to the load. It makes sense to return dwelling units back to this standard. Dwelling units didn't have a special table for 80 years prior to this addition, and they will continue to get along just fine without it.

Attached is a comparison of Tables 310.15(B)(7) and 310.15(B)(16), showing the average load diversity factor currently allowed, 12%. In fact, it should be seen by looking at the analysis that large aluminum conductors were more heavily overloaded due to the sloppy fix that Table 310.15(B)(7) presented. Reducing the calculated load instead will provide for more uniform protection of these conductors.

Here's the part you're sure to love: if you delete the section, you'll never see another proposal on it again!
 

Attachments

  • 3101531016.pdf
    19.1 KB · Views: 1

Dennis Alwon

Moderator
Staff member
Location
Chapel Hill, NC
Occupation
Retired Electrical Contractor
So a #4 copper conductor rated 85 amps now can be rated 85* 112%= 95amps and since the 95 amps is not a standard size then 100 amps OCPD is allowed.

Now if someone uses SE cable thru a wall that is caulked or has insulation then we must use 60C column . Now we have 70amps *112%= 78 amps. Now that conductor is only good for 80 amps.

I like it but I am curious what can of worms this may open. The one thing I can see is that NM which previously was not allowed can now be used as se cable-- I think that the 60 capacity may change that but if se is in insulation that would be the same rating. NM is cheaper-- do you think that the wiring methods should be addressed here or not?
 

jumper

Senior Member
I am all for deleting 310.15(B)(7)-it seems to cause more trouble it is worth.

If you want to use a simple formula, I am fine with that, but not the "load diversity".

Demand factor is fine. AFAIK load diversity is not defined in the NEC. The term only appears 3 times in chapters 1-9.

The first 2 refer you to Annex B and I am not an EE, so I cannot go there.

B.310.15(B)(1) Formula Application Information. This
annex provides application information for ampacities calculated
under engineering supervision.

The third is in T520.44. Something about 50%, which I would ignore since I do not know how to figure out the percentage. I would just count the number of CCCs and apply the adjustment factor if I was using that table.
 

George Stolz

Moderator
Staff member
Location
Windsor, CO NEC: 2017
Occupation
Service Manager
Now if someone uses SE cable thru a wall that is caulked or has insulation then we must use 60C column. Now we have 70amps *112%= 78 amps. Now that conductor is only good for 80 amps.
No, you determine your load, and then pick a conductor. This factor is not used in the sizing of conductors.

So, let's say you have a "2011" calculated load of 91A. You apply the "load diversity demand factor" to the load:
91 x .88 = 80A.

You would be right to say that all houses would have a #1 AL to them now, instead of a #2 - but that is 230.42(B)'s fault. So help me, I still don't understand that rule.
230.42(B): Your ampacity of conductors can't be less than the service disconnecting means - no exceptions.
230.79(B): The service disconnecting means (i.e. the switch, not the OCPD) has to be rated 100A - no exceptions.
230.90(A) exception 5: Well, your OCPD can be less than the ampacity of your wire per 310.15(B)(7).
Gee, thanks, but you already mandated a 100A ampacity, so... :slaphead:

I like it but I am curious what can of worms this may open. The one thing I can see is that NM which previously was not allowed can now be used as se cable-- I think that the 60 capacity may change that but if se is in insulation that would be the same rating. NM is cheaper-- do you think that the wiring methods should be addressed here or not?
I see what you mean, but I don't understand why overloading an SE cable gives them a warm fuzzy feeling but overloading NM is a bad idea.

I am all for deleting 310.15(B)(7)-it seems to cause more trouble it is worth.
In case you didn't notice, the section touched a nerve for me. :D

If you want to use a simple formula, I am fine with that, but not the "load diversity".

Demand factor is fine. AFAIK load diversity is not defined in the NEC. The term only appears 3 times in chapters 1-9.
I don't understand your concern. The term appears six times between the covers, so it's apparently an accepted term. It's purpose and application is fairly evident when read in context in the proposal, isn't it?
 

jumper

Senior Member
I don't understand your concern. The term appears six times between the covers, so it's apparently an accepted term. It's purpose and application is fairly evident when read in context in the proposal, isn't it?

Nope, not to me.

3 of the appearances are in Annex B. Are you allowed in that section? I am not and I still do not have a definition or a formula that I can use.

Demand and adjustment factors, I understand and can calculate. "Load diversity" is some weird voodoo that magically appears in residential services, but seems to be forbidden for any other application.
 

George Stolz

Moderator
Staff member
Location
Windsor, CO NEC: 2017
Occupation
Service Manager
Derek, the entire reason we have a Table 310.15(B)(7) right now is because of load diversity. Isn't it better to call a spade a spade than to leave the reasoning in doubt?
 

jumper

Senior Member
Derek, the entire reason we have a Table 310.15(B)(7) right now is because of load diversity. Isn't it better to call a spade a spade than to leave the reasoning in doubt?

Really, I thought it was because of historical data provided by POCOs. They might calculate this "load diversity" concept, but how do I. What is my definition and formula?

I like demand and/or adjustment factors. I can do dat.:)
 

George Stolz

Moderator
Staff member
Location
Windsor, CO NEC: 2017
Occupation
Service Manager
Show me the definition of current and I'll change the title.

My exploration of 230.79 led me back to a proposal I made for the 2008 code that didn't pass, where I was trying to figure out why 230.79 was in the NEC at all. The panel's response led me to believe that they didn't know, either.

I am a big fan of rules explaining their existence, if it's not self evident. IMO, the title stays. :)
 

jumper

Senior Member
Show me the definition of current and I'll change the title.

My exploration of 230.79 led me back to a proposal I made for the 2008 code that didn't pass, where I was trying to figure out why 230.79 was in the NEC at all. The panel's response led me to believe that they didn't know, either.

I am a big fan of rules explaining their existence, if it's not self evident. IMO, the title stays. :)

You got me on this one. NEC and Merriam-Webster are lacking.:):):)

I screwed up replying to your post, I hope I corrected it back right. :ashamed1:
 
Status
Not open for further replies.
Top