Combiner Box Locations & Quantity

[mikerbutler]

I've encountered an interpretation of 690.14(C) that has created quite a challenge. I've seen this particular code discussed in other posts here on this forum but I'm still not convinced the AHJ's interpretation is correct.

We've installed some ground-mounted 1 MW arrays in the past, under 2008 rules. (The basics: 4 PVP 260 kW inverters centrally located, DC & AC disconnect in the inverters, 8 combiner boxes feeding each inverter, & each inverter has a dedicated fused AC disconnect tied into a transformer on the inverter pad.) Our 32 combiner boxes were scattered throughout the 5 acre field and they did not include DC disconnects.

We are planning another 1 MW system. This time, there will be two AE 500TX inverters with 12 combiners each. Again, the inverters will be located on a pad, with a transformer, DC & AC disconnects on the inverters, and a fused AC disconnect for each inverter. The 24 combiner boxes will include a DC disconnect. The inverter pad will have a canopy (like a picnic shelter).

The AHJ has interpreted that our inverter pad qualifies as an "other structure" and thus 690.14(C) applies. The AHJ believes that all combiner boxes must be located at the inverters, and there can be no more than 6. I'm not sure how to proceed. If the AHJ is correct, I'm not sure how anyone can install a PV system utilizing string inverters.

I'm open to suggestions.

 

[ggunn]

I've encountered an interpretation of 690.14(C) that has created quite a challenge. I've seen this particular code discussed in other posts here on this forum but I'm still not convinced the AHJ's interpretation is correct.

We've installed some ground-mounted 1 MW arrays in the past, under 2008 rules. (The basics: 4 PVP 260 kW inverters centrally located, DC & AC disconnect in the inverters, 8 combiner boxes feeding each inverter, & each inverter has a dedicated fused AC disconnect tied into a transformer on the inverter pad.) Our 32 combiner boxes were scattered throughout the 5 acre field and they did not include DC disconnects.

We are planning another 1 MW system. This time, there will be two AE 500TX inverters with 12 combiners each. Again, the inverters will be located on a pad, with a transformer, DC & AC disconnects on the inverters, and a fused AC disconnect for each inverter. The 24 combiner boxes will include a DC disconnect. The inverter pad will have a canopy (like a picnic shelter).

The AHJ has interpreted that our inverter pad qualifies as an "other structure" and thus 690.14(C) applies. The AHJ believes that all combiner boxes must be located at the inverters, and there can be no more than 6. I'm not sure how to proceed. If the AHJ is correct, I'm not sure how anyone can install a PV system utilizing string inverters.

I'm open to suggestions.

You may have to do two levels of combining with combiners at the array and recombiners with six handles max at the inverter.

 

[SolarPro]

Interesting interpretation. The six-handle rule certainly applies to services. While you could have service disconnecting means requirements on the ac side of a PV system—after all, you are allowed to dedicate a service to a PV system—it seems to defy logic to apply those requirements on the dc side of a PV system. Look at the definition of service in Article 100.

Under 2011 Code requirements you would need to provide additional disconnecting means at the inverters for fuse servicing purposes. See 690.16 in the 2011 NEC. The intent is to enable a service technician to quickly and conveniently de-energize the dc bus in the event that a fuse needs to be replaced.

A lot of BOS providers are selling circuit breaker recombiners as a fuse and disconnect replacement for this purpose. If you review the specs on those products you will notice that they commonly include more than 6 circuit breakers, suggesting once again that the six-handle rule does not apply to the aggregation of dc circuits.

Here is an example:

http://www.bentek.com/cutsheets/BentekSolar-Bundled-Circuit-Breaker-Recombiners-IDC.pdf

If you are on the 2011 Code, then the AHJ is correct that you do need some additional disconnecting means at the inverter pad—just not for the reason given.

If you are not subject to the 2008 Code, then you may be able to make a compelling case that no additional disconnects are required. Just be aware that the system is more difficult and dangerous to service without fuse servicing disconnects. We've been encouraging designers to build to these new requirements even if they are not yet required by Code.

Here's a link to an article by Bill Brooks about the new fuse servicing requirements:

http://solarprofessional.com/articl...use-servicing-requirements-and-combiner-boxes

 

[jaggedben]

I'll be blunt; the AHJ is bonkers.

The single DC disconnect in the inverter disconnects the "photovoltaic system conductors" from "all [other] conductors" in the structure, and thus satisfies 690.14(C). End of discussion. As you stated, if the AHJ is correct, no one can install... well, no one can install rooftop solar, period. So the AHJ is clearly wrong about that.

Second, the single DC disconnect in the inverter is "the photovoltaic system disconnecting means", and it's only one handle. All other handles on the DC side are irrelevant if that one handle disconnects all PV conductors from the inverter.

If it came down to it, I would fight tooth and nail to change his mind on those points. (Granted, for the reasons SolarPro pointed out, it might not come down to it.)

 

[James@CHA]

I've encountered an interpretation of 690.14(C) that has created quite a challenge. I've seen this particular code discussed in other posts here on this forum but I'm still not convinced the AHJ's interpretation is correct.

We've installed some ground-mounted 1 MW arrays in the past, under 2008 rules. (The basics: 4 PVP 260 kW inverters centrally located, DC & AC disconnect in the inverters, 8 combiner boxes feeding each inverter, & each inverter has a dedicated fused AC disconnect tied into a transformer on the inverter pad.) Our 32 combiner boxes were scattered throughout the 5 acre field and they did not include DC disconnects.

We are planning another 1 MW system. This time, there will be two AE 500TX inverters with 12 combiners each. Again, the inverters will be located on a pad, with a transformer, DC & AC disconnects on the inverters, and a fused AC disconnect for each inverter. The 24 combiner boxes will include a DC disconnect. The inverter pad will have a canopy (like a picnic shelter).

The AHJ has interpreted that our inverter pad qualifies as an "other structure" and thus 690.14(C) applies. The AHJ believes that all combiner boxes must be located at the inverters, and there can be no more than 6. I'm not sure how to proceed. If the AHJ is correct, I'm not sure how anyone can install a PV system utilizing string inverters.

I'm open to suggestions.

Are you using input fuses at the inverter, or just the upstream fused DC disconnect and landing the conductors on the DC bus?

We have recently been working on system using the AE250kW unit. Having the separate DC disconnect is nice for maintenance of inverter fuses, because even after you've turned off the internal inverter DC disconnect and pulled the fuses, you still have DC voltage in the line side. But i do not think the six disconnect rule would apply here.