Derating on a roof

Status
Not open for further replies.

Fitzdrew516

Senior Member
Location
Cincinnati, OH
How do you guys handle conductors that serve RTU's and other rooftop equipment in regards to derating for sunlight. It seems to me that the NEC should allow for an exception if you are using less than a certain length of conduit/wire that will be on the roof. I thought that was what 310.15(b)(3)(a)(2) was for, but then I realized that the "(a)" references only derating for more than 3 current carrying conductors and not rooftop derating (that section is 310.15(b)(3)(c)).
I'm aware of the exception of this rule by using XHHW-2, however this exception is new to the 2014 NEC and not all jobs I work on utilize this code yet. I could always run the conduit above 36", but that seems a little ridiculous to me. I could make a note on the drawings to run all HVAC roof power inside of the curb, but notes often get overlooked. Any thoughts on this?

Thanks,
-Drew
 

Carultch

Senior Member
Location
Massachusetts
How do you guys handle conductors that serve RTU's and other rooftop equipment in regards to derating for sunlight. It seems to me that the NEC should allow for an exception if you are using less than a certain length of conduit/wire that will be on the roof. I thought that was what 310.15(b)(3)(a)(2) was for, but then I realized that the "(a)" references only derating for more than 3 current carrying conductors and not rooftop derating (that section is 310.15(b)(3)(c)).
I'm aware of the exception of this rule by using XHHW-2, however this exception is new to the 2014 NEC and not all jobs I work on utilize this code yet. I could always run the conduit above 36", but that seems a little ridiculous to me. I could make a note on the drawings to run all HVAC roof power inside of the curb, but notes often get overlooked. Any thoughts on this?

Thanks,
-Drew

If you can use the 10 ft / 10% exception, you can ignore the reduced ampacity in the smaller portion that is less than 10 ft and less than 10% of the remaining length of the circuit (whichever is less). This can be for either or both temperature corrections and ampacity adjustments. Quite common for rooftop stubs for HVAC equipment.

So if you have a 3 ft portion above the roof, and a 33 ft total length of the circuit, with 30 ft inside the building, don't worry about the reduced ampacity in stub up on the roof.

Another thing to realize is to read 110.14(C) completely. You have to use (usually) the 75C column for the termination ampacity, and if you use 90C rated wire, you get to use the 90C ampacity for your derate factors.

So consider 40A of circuit. You need #8 for the terminations, and you get to use the 55A for temperature corrections. 0.82 is a typical rooftop correction factor (33C ambient, 17C adder). 55A * 0.82 = 45.1A. So even without taking the 10%/10ft length exception, derate factors do not govern design in this example.
 
Last edited:

Fitzdrew516

Senior Member
Location
Cincinnati, OH
If you can use the 10 ft / 10% exception, you can ignore the reduced ampacity in the smaller portion that is less than 10 ft and less than 10% of the remaining length of the circuit (whichever is less). This can be for either or both temperature corrections and ampacity adjustments. Quite common for rooftop stubs for HVAC equipment.

So if you have a 3 ft portion above the roof, and a 33 ft total length of the circuit, with 30 ft inside the building, don't worry about the reduced ampacity in stub up on the roof.

Awesome. This is exactly what I was looking for. Do you have a code section to reference for this?
 

luckylerado

Senior Member
Is the "Circuit length" considered to be the total length of conductor from source through the load and back? So a single phase load 10 ft. away from the panelboard has a circuit length of 20 ft.?
 

Fitzdrew516

Senior Member
Location
Cincinnati, OH
Is the "Circuit length" considered to be the total length of conductor from source through the load and back? So a single phase load 10 ft. away from the panelboard has a circuit length of 20 ft.?

Not sure. That's a good question. I would assume it's there and back based on the language they used in the code.
 

Carultch

Senior Member
Location
Massachusetts
Is the "Circuit length" considered to be the total length of conductor from source through the load and back? So a single phase load 10 ft. away from the panelboard has a circuit length of 20 ft.?

I would say it is the 1-way length that matters. The conduit length, plus any remaining average length per phase/polarity inside other enclosures/wiring methods.

They have to make it a rule that applies consistently for single phase/DC as it does for 3-phase AC. And round trip length in AC is a more interesting calculation.
 

Fitzdrew516

Senior Member
Location
Cincinnati, OH
Does that exception cover the OPs instaltion?

I don't think it does but don't have my book with me

I believe it does. It comes before the derating sections and says this:

"Where more than one ampacity applies for a given circuit length, the lowest value shall be used.

Exception: Where two different ampacities apply to adjacent portions of a circuit, the higher ampacity shall be permitted to be used beyond the point of transition, a distance equal to 3.0 m (10 ft) or 10 percent of the circuit length figured at the higher ampacity, whichever is less."
 

pete m.

Senior Member
Location
Ohio
....however this exception is new to the 2014 NEC and not all jobs I work on utilize this code yet.

Any permit application (for non-residential) dated on or after January 1, 2015 utilizes the 2014 NEC in Ohio. You may be working on jobs that the permit application was dated prior to the adoption date of the 2014 NEC.

Pete
 

Fitzdrew516

Senior Member
Location
Cincinnati, OH
Any permit application (for non-residential) dated on or after January 1, 2015 utilizes the 2014 NEC in Ohio. You may be working on jobs that the permit application was dated prior to the adoption date of the 2014 NEC.

Pete

Thanks for the heads up, Pete! My company actually works on jobs all over the U.S. (and sometimes outside of it). This current job is in TN which is still stuck in the 2008 code.
 

Carultch

Senior Member
Location
Massachusetts
Any permit application (for non-residential) dated on or after January 1, 2015 utilizes the 2014 NEC in Ohio. You may be working on jobs that the permit application was dated prior to the adoption date of the 2014 NEC.

Pete

I would think that you would ordinarily be allowed to utilize future NEC rules, and still make a safe installation. Unless the AJH in question has a specific concern with any given future rule. I can't speak for your AHJ, but if it were up to me, I would allow utilization of most future NEC rules.
 

pete m.

Senior Member
Location
Ohio
I guess the Building Official could make that determination but I'm not sure of the legality. I would think if a future (as of yet not legally adopted) code were permitted to be used you would have to use only that edition. I don't think it would be a good idea to allow "cherry picking" from different editions.

Pete
 
Last edited by a moderator:

ggunn

PE (Electrical), NABCEP certified
Location
Austin, TX, USA
Occupation
Consulting Electrical Engineer - Photovoltaic Systems
I believe it does. It comes before the derating sections and says this:

"Where more than one ampacity applies for a given circuit length, the lowest value shall be used.

Exception: Where two different ampacities apply to adjacent portions of a circuit, the higher ampacity shall be permitted to be used beyond the point of transition, a distance equal to 3.0 m (10 ft) or 10 percent of the circuit length figured at the higher ampacity, whichever is less."
Something that sometimes confuses people reading that statement is that it is referring to the ampacity of a given wire, not the ampacity needed for a wire. If you are working the problem from the other direction, and continuous use says you need one ampacity and conditions of use tells you you need a different ampacity, it is the higher value that you must use.
 

Fitzdrew516

Senior Member
Location
Cincinnati, OH
I would think that you would ordinarily be allowed to utilize future NEC rules, and still make a safe installation. Unless the AJH in question has a specific concern with any given future rule. I can't speak for your AHJ, but if it were up to me, I would allow utilization of most future NEC rules.

I agree with you, but I have run into some pretty idiotic plans reviewers/inspectors before that don't interpret codes with logic. I had one plans reviewer burn me one time on a job where I read him the commentary out of the code book and how it is meant to be interpreted and he just said "well that's not how we interpret it." So really it always can fall back on the AHJ just requiring you to do whatever he/she wants. That's the short version of the story. I'll explain the detailed story below because, hey, why not?

So this was a tenant fit out design for a retail store, so imagine a strip center. The real "service" came from a utility transformer, underground to the building's main electric room. The service (208/120v, 3ph, 4w) terminated into a switchgear with a main disconnect, and then each switch on the switchgear supplied a seperate tenant in the strip plaza. If I'm remembering correctly each breaker in in the switchgear was 200A.

So what our client did was took over two of the tenant spaces, demo'd walls and turned it into one big space. They then utilized both 200A FEEDS (not services) to the tenant space. This is pretty standard on the retail side of things. I used to do jobs like this all the time. I'm pretty sure the total load on Feed 1 was about 160A and Feed 2 was about 170A.

So the job goes out for permit, goes through a couple rounds of building department comments (none of which mentioned this design being an issue). Finally months later we get a call from the contractor that said the inspector (also the plans reviewer) came out to do a walk through and told us that we couldn't do what we were trying to do. He said you can only have one service to the building. We called the inspector multiple times and plead our case. We used the code to explain that the "Service" was what came in from the utility and not what he was describing. As I said above at one point I literally read the commentary to him, and he still didn't agree.
So the inspector essentially told them that they had to demo the one 200A feed to the space so that it only had "One service". The issue with this being - The remaining 200A feed did not have enough capacity to handle our client's electrical demands. So the only thing that they could possibly do was to run a new (single) feed to the tenant space and demo both existing 200A feeds that were in their space (All underground btw). Obviously this is a HUGE expense on a project of this size. We tried to tell our client to get their lawyers involved because we thought they had a real case considering the installation was code compliant. In the end the client didn't want to waste the time it would take to fight it so they took the cost in the teeth, and needless to say, they were not happy about it.

So long story short - I learned my lesson that the AHJ will always have more authority than the code.

-Drew
 

Carultch

Senior Member
Location
Massachusetts
I guess the Building Official could make that determination but I'm not sure of the legality. I would think if a future (as of yet not legally adopted) code were permitted to be used you would have to use only that edition. I don't think it would be a good idea to allow "cherry picking" from different editions.

Pete

I agree with you about the not cherry picking different code editions part. By "cherry picking" different editions, one could potentially put together an inconsistent pair of rules. Rule A might be more lenient in 2014, while Rule B is more strict in 2014. If one would want to use NEC2014, they should have to follow both Rule A and Rule B. Especially if the reason why Rule A is more lenient in 2014, is precisely because Rule B is more strict. Less of a concern for completely unrelated rules.

So if one would want to follow a future edition of the code, it is probably best to follow it completely. All or nothing.

An example of two rules, which could be related such that a strict rule B allows for a lenient rule A:
Rule A: the XHHW-2/no rooftop temperature adder
Rule B: AFCI required on the circuit that is utilizing XHHW-2 in the example

If the above rules are in question, I agree that it is wrong to cherry pick them.

An example of two completely unrelated rules:
Rule A: the XHHW-2/no rooftop temperature adder
Rule B: the requirement for listed panic hardware on the exit door to the electrical room, 800A and above.

If the above two rules are in question, I see no problem with cherry picking them, given that Rule B is not a requirement in one's current applicable code.
 
Last edited:

don_resqcapt19

Moderator
Staff member
Location
Illinois
Occupation
retired electrician
...I had one plans reviewer burn me one time on a job where I read him the commentary out of the code book and how it is meant to be interpreted and he just said "well that's not how we interpret it." So really it always can fall back on the AHJ just requiring you to do whatever he/she wants. ...

-Drew
The commentary in the handbook is not an official interpretation of the code, it is just the authors opinion and there is no reason that the AHJ needs to accept it.

In your example, if there were meters for each of the 200 amp feeders, the installation would not be permitted around here by either the utility or the AHJ.
 

Fitzdrew516

Senior Member
Location
Cincinnati, OH
The commentary in the handbook is not an official interpretation of the code, it is just the authors opinion and there is no reason that the AHJ needs to accept it.

In your example, if there were meters for each of the 200 amp feeders, the installation would not be permitted around here by either the utility or the AHJ.

I know the commentary isn't actual code, but it's what makes sense. How else is one to interpret the code when designing? I feel like this was a pretty cut and dry situation. The "service" is the main feed to the building, which has it's own disconnect. In retail applications like this (Malls, strip centers, Lifestyle centers, etc.) this is how it is always done. Everything downstream of the main service is landlord owned/managed.
 

don_resqcapt19

Moderator
Staff member
Location
Illinois
Occupation
retired electrician
...I feel like this was a pretty cut and dry situation. The "service" is the main feed to the building, which has it's own disconnect. In retail applications like this (Malls, strip centers, Lifestyle centers, etc.) this is how it is always done. ....
The code does not require a single main disconnect for those applications. Sure it is often done that way and often the AHJ requires it to be done that way, but the code permits each set of service entrance conductors to have their own set of six service disconnects.
 
Status
Not open for further replies.
Top