Separate conduit - NEC interpretation

[ggunn]

I am working on a PV system that will have 30 SMA inverters and 1062 modules. Does each inverter have to have its own unshared conduit from its PV array and to the AC combiner?

Please hold your questions about why it's being done this way and not with a central inverter. If I could, I would.

 

[iwire]

I am working on a PV system that will have 30 SMA inverters and 1062 modules. Does each inverter have to have its own unshared conduit from its PV array and to the AC combiner?

No, it does not have to, but if you mix them the derating will kill you.

 

[ggunn]

No, it does not have to, but if you mix them the derating will kill you.

How so? Even at the maximum derating (41 conductors and above), which is .35, it only takes me to #10, and I was going to have to use #10 for voltage drop, anyway.

Isc= 7.72A. Top ASHRAE temperature is 100 degrees F. For four inches off roof add 40 degrees F. Derate to 140 degrees F is 0.71.

(7.72A)(1.25)/(0.71)(0.35) = 38.8A.

#10 is good to 40A, unless I have to connect to 75 degree terminals, which only bumps me to #8 worst case.

Am I missing something?

But can you point me to a code article which allows this? 690.4(B) allows mixing conductors "directly related to a specific system" but this is in essence 30 separate systems.

 

[infinity]

Isc= 7.72A. Top ASHRAE temperature is 100 degrees F. For four inches off roof add 40 degrees F. Derate to 140 degrees F is 0.71.

(7.72A)(1.25)/(0.71)(0.35) = 38.8A.

Can you explain this calculation, is it PV specific?

 

[ggunn]

Can you explain this calculation, is it PV specific?

Part of it is. The maximum current from a PV string is the module short circuit current (Isc) multiplied by 1.25 to account for insolation (sunlight) that may exceed the 1000W/m2 standard test conditions. The rest is just temperature and conduit derating. The max ambient expected temperature (100 degrees F) is adjusted up 40 degrees F for conduit 4" off the roof per 310.15(B)(2)(c) and the conduit fill derating is maxxed at 0.35 per 310.15(B)(2)(a) for >40 conductors.

I think I'm doing this correctly, but feel free to point out anything that is incorrect. It is more important to me to be correct than right, if you get my drift.

 

[tkb]

Do you really have terminations that are rated 90?c?
Not very common.

 

[Smart $]

...adjusted up 40 degrees F for conduit 4" off the roof per 310.15(B)(2)(c)...

... feel free to point out anything that is incorrect. ...

3 1/2" to 12"" above the roof is a 30?F adder.

 

[Smart $]

...

#10 is good to 40A, unless I have to connect to 75 degree terminals, which only bumps me to #8 worst case.

Am I missing something?

....

Do you really have terminations that are rated 90?c?
Not very common.

Doesn't matter.

Derating is permitted from the #10 40A 90?C rating, and his ampacity only has to be 7.72 ? 125%, which is well under the maximum 75?C termination ampacity for #10 at 35A.

 

[infinity]

Doesn't matter.

Derating is permitted from the #10 40A 90?C rating, and his ampacity only has to be 7.72 ? 125%, which is well under the maximum 75?C termination ampacity for #10 at 35A.

According to his calculation the #10 just makes it. 40 amps*.71*.35=9.94 amps, 7.72*125%=9.65

 

[iwire]

Wouldn't it be Isc +125% +125%?

690.8(A)(1) and 690.8(B)(1)

Beyond that I am sorry I said you could mix the systems, that was a mistake.

 

[Smart $]

Wouldn't it be Isc +125% +125%?

690.8(A)(1) and 690.8(B)(1)

Correct.

Beyond that I am sorry I said you could mix the systems, that was a mistake.

690.4(B) says not with feeders or branch circuits, and the PV system conductors are not feeders or branch circuits.

 

[iwire]

First off we should all mention what year NEC we are using as Article 690 is full of changes.

My references will be the 2008 NEC.

690.4(B) says not with feeders or branch circuits, and the PV system conductors are not feeders or branch circuits.

I agree but looking at my 2008 a 'PV System' is pretty clearly defined and other 'systems' cannot be run together.


IMO it can go either way as the wording is ambiguous.

 

[Smart $]

First off we should all mention what year NEC we are using as Article 690 is full of changes.

My references will be the 2008 NEC.




I agree but looking at my 2008 a 'PV System' is pretty clearly defined and other 'systems' cannot be run together.


IMO it can go either way as the wording is ambiguous.

Same here, 2008 NEC.

Yes PV system is well defined:

Solar Photovoltaic System. The total components and subsystems
that, in combination, convert solar energy into
electric energy suitable for connection to a utilization load.

...then where in 690 are PV system conductors called feeders or branch circuits???

(B) Conductors of Different Systems. Photovoltaic
source circuits and photovoltaic output circuits shall not be
contained in the same raceway, cable tray, cable, outlet box,
junction box, or similar fitting as feeders or branch circuits
of other systems, unless the conductors of the different systems
are separated by a partition or are connected together.

You would only have feeders or branch circuits on the load side of the AC combiner (assuming this is how the PV system will be connected to the service-supplied system).

 

[iwire]

You would only have feeders or branch circuits on the load side of the AC combiner (assuming this is how the PV system will be connected to the service-supplied system).

Conductors of Different Systems

Would the conductors of a second PV system be conductors of a different system?

IMO yes, even though they are not feeders or branch circuits.

I really don't know which way to read it other than saying I wish it was clearer.

 

[Smart $]

Would the conductors of a second PV system be conductors of a different system?

First it would have to qualify as a second PV system. Even then the conductors would not be feeders or branch circuit conductors.

IMO yes, even though they are not feeders or branch circuits.

I really don't know which way to read it other than saying I wish it was clearer.

Clearer would be better

 

[iwire]

My 2011 is at the office but I see this in the ROPs

(B) Conductors of Different Systems. Photovoltaic source circuits and
PV output circuits shall not be contained in the same raceway,
cable tray, cable, outlet box, junction box, or similar fitting as conductors,
feeders or branch circuits of other non-PV systems, unless the conductors of
the different systems are separated by a partition.

I can't tell you if that is in the code as written.

 

[Smart $]

My 2011 is at the office but I see this in the ROPs





I can't tell you if that is in the code as written.

Here's the 2011 version:

(B) Identification and Grouping. Photovoltaic source circuits
and PV output circuits shall not be contained in the
same raceway, cable tray, cable, outlet box, junction box, or
similar fitting as conductors, feeders, or branch circuits of
other non-PV systems, unless the conductors of the different
systems are separated by a partition. Photovoltaic system
conductors shall be identified and grouped as required
by 690.4(B)(1) through (4). The means of identification
shall be permitted by separate color coding, marking tape,
tagging, or other approved means.

 

[ggunn]

3 1/2" to 12"" above the roof is a 30?F adder.

You are correct, sir. I misread the table.

 

[ggunn]

Wouldn't it be Isc +125% +125%?

690.8(A)(1) and 690.8(B)(1)

That is for overcurrent protection. According to published articles by John Wiles, who is the guru when it comes to installing PV systems to code, the way to size PV source circuit conductors is either to include the second 1.25 factor for continuous use or calculate and include deratings for conditions of use. Actually, he says to calculate it both ways and use the result which is the more conservative of the two. In my case, the conditions of use factors combine to more than 125%.

I have emailed him my question about combining source circuits for different inverters in a common conduit, since the code looks ambiguous to me.

Film at eleven.

 

[iwire]

That is for overcurrent protection. According to published articles by John Wiles, who is the guru when it comes to installing PV systems to code,

I know who John is, he has even posted here before.

But what you describe is not what the 2008 NEC says, it is 125% twice for a total of just over 156%.

Take a look at 690.8 I don't see anything limiting it to just the OCPD.