AC disconnect on roof - San Diego plans require it

[d_1d]

Hi, All. I've gotten comments back on my plans (3kw micro inverter system)

The reviewer requires an A/C disconnect on the roof. I haven't seen this yet in my casual studying of the industry. What do people recommend? Where is it mounted? - to rails? To roof?

Any input? Thanks a ton!!

The other major comments are that I need to provide UL1703 certificates for the system. After reading a bit more, it seems to mean submit the letters (available at IronRidge and Canadian Solar) showing the cert. Note, that it is important to match up the module "Type", since the "system" means the rack and module were tested together. Luckily, for IronRidge, it is compatible with types 1,2,3, and the CS6P modules are Type 1.

[emartin00]

I can't say I've ever heard of needing an AC disconnect on the roof. DC disconnect, yes, but not AC.

As far as the certifications go, UL1703 is a listing for modules only. Before 2703 came out, the module manufacturers had to list acceptable grounding methods and fire ratings in their install manuals.
The 2703 listing information that IronRidge should trump anything from 1703, although you may need the module data sheet to show it's ratings.

[SunEagle]

The AC disconnect is needed because the micro inverters convert the panel DC voltage to AC voltage as opposed to a string inverter which gets the DC from the panels.

The possible reason may have to do with being able to manually disconnect the power generated by the panels as close to the array as possible instead of down at the main AC panel.

Although the NEC 2014 requires an automated way of reducing the voltage generated by the panels within 10 feet of the array. SD requirements may include an additional manual disconnect as well.

I would think the AC disconnect would need to be mounted on the array framework or close to it and be rated for wet weather (minimum NEMA 3R). You might ask the AHJ what they want instead of just guessing.

[paul65k]

The AC disconnect is needed because the micro inverters convert the panel DC voltage to AC voltage as opposed to a string inverter which gets the DC from the panels.

The possible reason may have to do with being able to manually disconnect the power generated by the panels as close to the array as possible instead of down at the main AC panel.

Although the NEC 2014 requires an automated way of reducing the voltage generated by the panels within 10 feet of the array. SD requirements may include an additional manual disconnect as well.

I would think the AC disconnect would need to be mounted on the array framework or close to it and be rated for wet weather (minimum NEMA 3R). You might ask the AHJ what they want instead of just guessing.

Seems like it would be more useful at ground level properly labeled so a utility or emergency worker would have access.........wondering about the real world the utility of having it on the roof???

[SunEagle]

Seems like it would be more useful at ground level properly labeled so a utility or emergency worker would have access.........wondering about the real world the utility of having it on the roof???

It may be nothing more then a way to de-energize all the cables from a pv system as close to the array as possible. That is what the Article 690.12 concerning Rapid Shutdown is supposed to do.

It basically limits the area of "energized" wire that is running between the array to the main panel. Having a disconnect at the ground level is probably also required.

[d_1d]

I emailed the plan reviewer for clarification.
There is not a separate AC disconnect required at ground level except for the backfed breaker. Makes sense to reduce the failure points and simplify the design. As soon as this breaker, or the main breaker is flipped, the micros all stop and bring 0-volts at the rooftop AC due to auto-de-energizing. Furthermore, to service a particular panel/inverter, the micro can be disconnected (a disconnect, if I read Enphase documents correctly) via the Engage cable). So, I don't see why the city wants the roof disconnect. However, as the saying goes, the inspector/reviewer is always right

[SunEagle]

I emailed the plan reviewer for clarification.
There is not a separate AC disconnect required at ground level except for the backfed breaker. Makes sense to reduce the failure points and simplify the design. As soon as this breaker, or the main breaker is flipped, the micros all stop and bring 0-volts at the rooftop AC due to auto-de-energizing. Furthermore, to service a particular panel/inverter, the micro can be disconnected (a disconnect, if I read Enphase documents correctly) via the Engage cable). So, I don't see why the city wants the roof disconnect. However, as the saying goes, the inspector/reviewer is always right

It might come down to the inspector not trusting the automatic shutdown of the micros yet so they request a manual switch to "make sure" the wires are d-energized.

Back in the 80's I ran into something similar with Variable Frequency Drives. They were designed to stop working and d-energize the motor leads but due to being unfamiliar with a VFD we were required to install electrical contactors which would open the circuit and isolate the motor leads from the VFD. That was my first "belt and suspenders" design. IMO while it added to "safety" it was a PIA and an extra cost for our company. But you have stated, you can't argue with the inspector.

[d_1d]

Nema 3R enclosure orientation?

Thanks, everyone.
Ok, I got an answer from the AHJ.
They indeed want a "pull out" type disconnect on the roof. The ones I can find are NEMA 3R, but I'm still struggling to figure out how to mount them. Since they are rated 3R in a vertical orientation, if I have them perpendicular to the roof surface, wouldn't that be a violation? I've looked for NEMA 4 enclosures, but they are all large combiners and way overkill. Since the roof is 12:4, I suspect the slope would be enough to keep out the water, but it's still not officially "right". I really don't want to mount it upright, as that's a huge eyesore.

[SunEagle]

Thanks, everyone.
Ok, I got an answer from the AHJ.
They indeed want a "pull out" type disconnect on the roof. The ones I can find are NEMA 3R, but I'm still struggling to figure out how to mount them. Since they are rated 3R in a vertical orientation, if I have them perpendicular to the roof surface, wouldn't that be a violation? I've looked for NEMA 4 enclosures, but they are all large combiners and way overkill. Since the roof is 12:4, I suspect the slope would be enough to keep out the water, but it's still not officially "right". I really don't want to mount it upright, as that's a huge eyesore.

Too bad you don't have a structure like a chimney or wall to mount the disconnect to. Keeping it verticle is part of the requirement for NEMA 3R classification.

Depending on the amperage and voltage rating there are small "pull out" boxes used by the HVAC industry that is nothing more then a twin blade with a handle that can be pulled out or pushed back in to complete the circuit. Check to see if that type would be acceptable to the AHJ.

[d_1d]

The plan reviewer specifically said "pull type". It may be OK. Still sketchy with the box laying down, but I don't know what else to do. The HVAC boxes are perfectly sized, and I've seen them recommended elsewhere on the inter-webs for this task. I already emailed my expected supplier to get their feedback.

[inetdog]

It may be nothing more then a way to de-energize all the cables from a pv system as close to the array as possible. That is what the Article 690.12 concerning Rapid Shutdown is supposed to do.

It basically limits the area of "energized" wire that is running between the array to the main panel. Having a disconnect at the ground level is probably also required.

The Rapid Shutdown requirement applies only to the PV conductors, which in NEC terms are the DC module output and combined conductors all the way up to the inverter input. Inverter output circuits (AC) are subject to different rules, found in Article 705.

Article 690 confuses some people with its reference to PV source circuits and PV output circuits. A close reading of text shows that PV output circuits refer to the output of a fused or unfused combiners or optimizers, not to the AC inverter output circuits.

However, because microinverters are in fact grid tie inverters disconnecting the AC side from the grid anywhere in the system wiring will make the AC output of the micro's stop.
So an AC disconnect down at the service or main panel results in a shutdown of the micros' AC output while the DC side never gets beyond the boundary of the array.

SolarEdge optimizers react to loss of the connection to the string inverter by reducing the DC output of each module to 1V or less, satisfying the Rapid Shutdown requirement as long as the optimizers have capacitor bleed down resistors in place.

[inetdog]

The plan reviewer specifically said "pull type". It may be OK. Still sketchy with the box laying down, but I don't know what else to do. The HVAC boxes are perfectly sized, and I've seen them recommended elsewhere on the inter-webs for this task. I already emailed my expected supplier to get their feedback.

For What It's Worth, at least there are pull-out (visible contact break) disconnects with load breaking capability are readily available for AC. Trying to put in a DC disconnect would be much more demanding.

And it is not all that hard to fabricate a sheet metal bracket that will hold the 3R enclosure vertical rather than either parallel to or perpendicular to the roof.

[d_1d]

Here's another bit of info I found on the subject that specifically talks about San Diego.

http://forums.mikeholt.com/showthread.php?t=156797

[paul65k]

For What It's Worth, at least there are pull-out (visible contact break) disconnects with load breaking capability are readily available for AC. Trying to put in a DC disconnect would be much more demanding.

And it is not all that hard to fabricate a sheet metal bracket that will hold the 3R enclosure vertical rather than either parallel to or perpendicular to the roof.

Stupid question but..........since this discussion is specifically about micro inverters wouldn't the disconnect NEED to be AC and not DC?? As soon as that PV powergoes through the micro inverter it comes out AC.......Now I'm really confused

[d_1d]

Stupid question but..........since this discussion is specifically about micro inverters wouldn't the disconnect NEED to be AC and not DC?? As soon as that PV powergoes through the micro inverter it comes out AC.......Now I'm really confused

haha.. I think inetdog was just saying "I know that's annoying you have to put in an unnecessary AC disconnect, but at least it's not DC, which would be harder".