Solar PV Panel Feeder

[dereckbc]

Ok I have a unique situation I think I know the answer to but want a sanity check.

I am bidding out a 15 Kw grid tied system, but what makes it unique is the panels are 500 feet from the building mounted on a pole with 2-axis tracking. The system will consist of 7 strings. In the RFQ I quoted the feeder from the panels to the building be a 250 MCM to keep the power losses under 4%.

OK I get one bid back from a contractor that quotes running 15 #6 AWG or one pair of 6 AWG from each string plus 1 ground. I quized him about why and he replies the labor and cable will be less expensive. OK I don't have a problem saving money, but I am not sure running 7 feeders will comply with 690. I cannot find anything that prohibits it, nor do I find anything that allows it either. I just have never seen done this way.

What do you think.

 

[ty]

For starters, just to clarify, where are the inverters in your design, and where are they in his?

 

[mcclary's electrical]

Ok I have a unique situation I think I know the answer to but want a sanity check.

I am bidding out a 15 Kw grid tied system, but what makes it unique is the panels are 500 feet from the building mounted on a pole with 2-axis tracking. The system will consist of 7 strings. In the RFQ I quoted the feeder from the panels to the building be a 250 MCM to keep the power losses under 4%.

OK I get one bid back from a contractor that quotes running 15 #6 AWG or one pair of 6 AWG from each string plus 1 ground. I quized him about why and he replies the labor and cable will be less expensive. OK I don't have a problem saving money, but I am not sure running 7 feeders will comply with 690. I cannot find anything that prohibits it, nor do I find anything that allows it either. I just have never seen done this way.

What do you think.

How are the 15 #6 ran? Are they tied together on each end? If so, that's a violation to be smaller than 1/0. Other than that,I can't picture what's happening.

 

[dereckbc]

For starters, just to clarify, where are the inverters in your design, and where are they in his?

The inverters are on an outside wall near the service panel some 500 feet from the panels.

 

[dereckbc]

How are the 15 #6 ran? Are they tied together on each end? If so, that's a violation to be smaller than 1/0. Other than that,I can't picture what's happening.

Guess I should have been clear. It is 7 individual circuits. So nothing is parallel. I see where you are going if it was parallel conductors I would have known better right off the bat.

What is bugging me is usually you see a String Junction Box installed close to the panels with each string having its own circuit breaker, then one fused feeder with a disconnect down to the inverters. What he is doing is bringing all seven strings to the inverter, and joining them all together. Each string will have its own breaker at the pole where the panels are mounted. I just have never seen it done this way before and it caught my attention.

I guess what is bugging me the most is there is not a single switch that can be thrown at the pole where the panels are located to disconnect all the panels. Instead 7 separate breakers at the pole. However there will be a single switch to disconnect at the inverter after all the strings are combined there.

Actually I like the idea of being able to isolate each string for testing purposes, but just not sure it is compliant.

 

[mcclary's electrical]

Guess I should have been clear. It is 7 individual circuits. So nothing is parallel. I see where you are going if it was parallel conductors I would have known better right off the bat.

What is bugging me is usually you see a String Junction Box installed close to the panels with each string having its own circuit breaker, then one fused feeder with a disconnect down to the inverters. What he is doing is bringing all seven strings to the inverter, and joining them all together. Each string will have its own breaker at the pole where the panels are mounted. I just have never seen it done this way before and it caught my attention.

I guess what is bugging me the most is there is not a single switch that can be thrown at the pole where the panels are located to disconnect all the panels. Instead 7 separate breakers at the pole. However there will be a single switch to disconnect at the inverter after all the strings are combined there.

Actually I like the idea of being able to isolate each string for testing purposes, but just not sure it is compliant.

It sounds compliant

 

[ty]

So on his proposed #6's per string, this will be DC for 500ft? That's a pretty long run with voltage drop.
The voltage drop on the DC wires is going to drop the efficiency down greatly. (code compliant or not)

Can't you install the inverters closer to the array and run AC?

 

[dereckbc]

Can't you install the inverters closer to the array and run AC?

Well yes but that would incur even higher losses. The DC voltage will be at around 450 VDC in summer, and around 490 in winter. If it were AC then it would have to be 240. The DC requires about half the current. Each string will be at around 4.7 amps each over 1000 feet of #6 AWG loop which works out to 22 watts on each feeder or 154 watts total between the 7 feeders. That works out to about 1% loss. Not to shabby IMO.

I like the guys bid he is the lowest. Guess it does not really matter to me if it is code compliant or not. If not he will have to deal with it on his nickle. He has done work for me before, not solar, but he does good work.

 

[BillK-AZ]

Solar PV Panel Feeder

You need to check the number of current carrying conductors per conduit per NEC Table 310.15(B)(2)(a). The #6 seems to be based on all wires in one conduit. Three smaller conduits will reduce the fill adjustment factor from 50% to 80%.

Ultimately, it comes down to $/watt difference between two wire sizes/costs/losses vs. the cost of the installed PV modules in $/watt.

Voltage drop on the DC side is not as critical as on the AC side, the inverters have a wide input range.

 

[elohr46]

Guess I should have been clear. It is 7 individual circuits. So nothing is parallel. I see where you are going if it was parallel conductors I would have known better right off the bat.

What is bugging me is usually you see a String Junction Box installed close to the panels with each string having its own circuit breaker, then one fused feeder with a disconnect down to the inverters. What he is doing is bringing all seven strings to the inverter, and joining them all together. Each string will have its own breaker at the pole where the panels are mounted. I just have never seen it done this way before and it caught my attention.

I guess what is bugging me the most is there is not a single switch that can be thrown at the pole where the panels are located to disconnect all the panels. Instead 7 separate breakers at the pole. However there will be a single switch to disconnect at the inverter after all the strings are combined there.

Actually I like the idea of being able to isolate each string for testing purposes, but just not sure it is compliant.

More than two strings you are required to fuse each string which normally go into the combiner box and like you said one feeder from that point to the Inverter. I don't see a violation, however, how would you terminate all of those conductors at the inverter?

 

[drive1968]

<<<I guess what is bugging me the most is there is not a single switch that can be thrown at the pole where the panels are located to disconnect all the panels.>>>

Some jurisdictions might require a shut-off switch where the panels are located, particularly since they are so far away from the inverter. You might want to ask your inspector about this. That could be done though without too much problem with a switch for each of the seven branches right at the tracking mount.

I can't see anything wrong from a code perspective though about running the 15 smaller wires.