Using a Recombiner as the NEC rated protection on a fused Inverter

[Kevin Whitten]

I am currently trying to size the following inverter to meet NEC Code for a project:

500kW Inverter
- 1750A max dc current
- 14 inputs with 125A fuses

285W Panel
- 8.53A Isc

The fuses cannot be changed in this inverter. The typical current per input that I have would be roughly 110A, however when I size the inputs for oversized current according to the NEC, I obtain roughly 173A (8.53A x 13 strings/combinder box on said input x 1.25 x 1.25). With this number, I cannot use this inverter because of the fuses, yet it is required equipment so I must utilize it.

So, my question is this: If I put a recombiner on the inverter?s DC input with 14 inputs/outputs with 14x200A breakers, can I still utilize the 500kW inverter even though its? internal fuses do not meet NEC requirements?

 

[GoldDigger]

If the fuses are part of listed equipment and you provide external protection for the wires that satisfies the NEC, I do not see a problem. The fuses in the inverter would be secondary/optional OCPD.
But the wire sizes connected to each input would have to meet NEC ampacity rules based on the PV panel array output rating.

Tapatalk!

 

[jaggedben]

I am currently trying to size the following inverter to meet NEC Code for a project:

500kW Inverter
- 1750A max dc current
- 14 inputs with 125A fuses

285W Panel
- 8.53A Isc

The fuses cannot be changed in this inverter. The typical current per input that I have would be roughly 110A...

With the disclaimer that I don't design systems this large...

I think you have a problem right there, regardless of the NEC. A good 'edge of cloud effect' could blow your inverter fuses.

 

[GoldDigger]

With the disclaimer that I don't design systems this large...

I think you have a problem right there, regardless of the NEC. A good 'edge of cloud effect' could blow your inverter fuses.

On the other hand, you will see that his 110A is based on the Isc of the panels rather than the Imp, so there will be some additional cushion.
And the inverter's control circuitry may well keep the draw below 125A regardless of what the panel array could deliver.

The inverter manufacturer is not bound by the double-dip markup from the worst case (Isc) value the way the installer of field wiring is. (Short circuit is possible so use Isc. Higher insolation is possible so multiply by 1.25, and the short circuit could be continuous as far as the panels are concerned, so multiply by 1.25 again.)
My bet is that the inverter would clip before the point where the fuses would blow. Or if there is a fault in the inverter you would just as soon have them blow, although they probably will not.
Worth investigating further.

 

[jaggedben]

On the other hand, you will see that his 110A is based on the Isc of the panels rather than the Imp, so there will be some additional cushion.
And the inverter's control circuitry may well keep the draw below 125A regardless of what the panel array could deliver.

The inverter manufacturer is not bound by the double-dip markup from the worst case (Isc) value the way the installer of field wiring is. (Short circuit is possible so use Isc. Higher insolation is possible so multiply by 1.25, and the short circuit could be continuous as far as the panels are concerned, so multiply by 1.25 again.)
My bet is that the inverter would clip before the point where the fuses would blow. Or if there is a fault in the inverter you would just as soon have them blow, although they probably will not.
Worth investigating further.

Good points, but...

The rated Max DC that the OP stated is exactly the combined rating of the 14 fuses. And what if one string exceeds 125A while a few others do not, and the combined current is still 1750A or less?

I'm struggling to understand how an inverter manufacturer designs something with 14 125A inputs, knowing that by the NEC this would limit it to less than it's rated power with typical solar modules (that being when utility scale designers always want excess DC power).

@Kevin

I agree with Goldigger that the answer to your question is 'yes', but does the inverter literature address recommended max current per string?

 

[GoldDigger]

I am speculating that the "inverter" is really a box containing multiple parallel GTIs, or at the very least has an independent MPPT module for each fused input. That would allow input clipping on each input in addition to allowing for unmatched or differently shaded panel arrays.
What would not make sense to me would be to simply parallel the DC inputs to a common bus after the fuses.


Tapatalk!