Enphase IQ series +7 inverter

[wwhitney]

Instead, each phase current Ip also has a vector component Ipq that's at 90 degrees from Ipi, and with a magnitude sin(30) x Ip = 0.5 Ip. This current flows between the two phases at a delta connection. And although it does not appear in the line output, this current Ipq goes through the buss when the delta is constructed by connecting backfed breakers to the bus.

I agree with your point about the bus stabs, although I would be surprised if that were a significant factor in the overall bus heating. But I'm going to disagree with the above.

Say the A bus is an idealized line, and it has just 3 connections, in order: (1) a 2 pole breaker on AB phases, (2) a 2 pole breaker on AC phases, and (3) the 3 pole main breaker. With 3 single phase inverters connected in a delta arrangement, say with an output of 60A, the current on the segment 1-2 will be 60A, and the current on segment 2-3 will be 60 sqrt(3), not 120. The current Ipq you are talking about does not flow through the bus.

Remember, these vectors are just short hand for waveforms, and if you add the actual waveforms, you will see that the current does not exceed 100 sqrt(3) on the bus. Of course, I don't know anything about the 3D effects of an actual bus, but again I would be surprised if that variation from the idealized model caused significant heating.

Cheers, Wayne

 

[ggunn]

I agree with your point about the bus stabs, although I would be surprised if that were a significant factor in the overall bus heating. But I'm going to disagree with the above.

Say the A bus is an idealized line, and it has just 3 connections, in order: (1) a 2 pole breaker on AB phases, (2) a 2 pole breaker on AC phases, and (3) the 3 pole main breaker. With 3 single phase inverters connected in a delta arrangement, say with an output of 60A, the current on the segment 1-2 will be 60A, and the current on segment 2-3 will be 60 sqrt(3), not 120. The current Ipq you are talking about does not flow through the bus.

Remember, these vectors are just short hand for waveforms, and if you add the actual waveforms, you will see that the current does not exceed 100 sqrt(3) on the bus. Of course, I don't know anything about the 3D effects of an actual bus, but again I would be surprised if that variation from the idealized model caused significant heating.

Cheers, Wayne

Really, all this is beside the point, which is compliance with NEC2014 705.12(D)(2)(3)(c). Current is not mentioned, only breaker ratings. Whether the code is written as it should be is a completely different discussion.

 

[wwhitney]

Really, all this is beside the point, which is compliance with NEC2014 705.12(D)(2)(3)(c). Current is not mentioned, only breaker ratings. Whether the code is written as it should be is a completely different discussion.

Well, the question was whether "adding" the breaker ratings under that section should be done as scalars or vectors.

However, upon further reflection on 705.12(D)(2)(3)(c), I agree that if an AHJ insists on adding them as scalars, I can't really fault that. The purpose of the section, as I see it, is to provide a simple rule that someone can use to verify the bus is protected by just looking at the panel without opening it up. A 2-pole breaker could be used as an MWBC and put its full rated current on each busbar in a way that doesn't partially cancel. Knowing the phase angle of the current put on the bus in order to do vector addition requires knowing more of the circuit details than just the number and quantity of breakers in the panel, and I think that in the spirit of keeping 705.12(D)(2)(3)(c) simple, the rule doesn't want to require that extra data to check compliance.

Of course, if the panel only has 2-pole breakers to which only L-L single phase inverters are connected, I could still hope an enlightened AHJ would agree to allow the addition vectorially.

Cheers, Wayne

 

[pv_n00b]

If you are looking at doing an AC combiner with 705.12(D)(2)(3)(c) then all the discussion about vector math is moot. The code just adds up the CB ratings, and really is assuming that 3ph services are getting 3ph inverters and split phase services are getting single phase inverters. At this point the CMP is probably not even thinking about making a distinction for single phase inverters on 3ph panelboards. It's just not that common anymore and would require too much wording in the code. Which is one way of saying don't bother doing it unless you like oversized AC combiners.

 

[wwhitney]

BTW, what about using (1) 100A 3-pole breaker instead of (3) 60A 2-pole breakers? I'm thinking that a short piece of 100A conductor on each breaker could go to a polaris connector picking up two 60A L-L legs not in phase. Then the outdoor tap rule of unlimited length would be used for the 60A conductors. You might need to put a 60A main breaker in the combiner boxes, I'd have to check the tap rules.

Cheers, Wayne

 

[ggunn]

If you are looking at doing an AC combiner with 705.12(D)(2)(3)(c) then all the discussion about vector math is moot. The code just adds up the CB ratings, and really is assuming that 3ph services are getting 3ph inverters and split phase services are getting single phase inverters. At this point the CMP is probably not even thinking about making a distinction for single phase inverters on 3ph panelboards. It's just not that common anymore and would require too much wording in the code. Which is one way of saying don't bother doing it unless you like oversized AC combiners.

I'm not sure what you are driving at, but the last time I dealt with it the AHJ accepted totaling the breaker ratings landing on each bus of the panel for compliance with the article in question.

 

[MWh_Pro]

Figured I'd take a swag at answering some of the other parts of the original question:

2. Inverter is on roof so is aggregator which has fuse. If I provide single disconnecting means at aggregator per NEC 2014 Article 690.17 then would I still need to comply per NEC 2014 Article 690.15(A)(B)(C)?
--> Yes, but if you install an AC switch there you shouldn't have a problem. 690.15(A) "a single disconnecting means in accordance with 690.17 shall be permitted for the combined ac output of one or more inverters or ac modules in an interactive system." --> See 690.17 --> 690.17(A) = N/A because your branch circuit(s) are inverter output circuits, not "PV conductors." 690.17(B)-(E): make sure it simultaneously opens the poles (B), is externally operable (C), doesn't disconnect the grounded conductor (C) [N/A because Enphase IQ7+ doesn't have a neutral/grounded conductor in output circuit], and (E) rated for the max current/voltage/kA of the system. Back to 690.15(A). (1) N/A, we're talking about AC circuit. (2) your AC disconnect on the roof should meet this requirement. should be within sight of the microinverters. You need the switch on the roof, because the inverters don't have an AC disconnect built into them. (3) You need a PV system disconnect per 690.13(A) in a readily accessible location, elsewhere, NOT on the roof. --> You need another AC switch. (4) A plaque shall be installed in accordance w/ 705.10.

3. Do I need to comply per NEC 2014 Article 690.16?
--> Yes. If your "aggregator" is a fusible switch or some type of disconnect-combiner with finger-safe fuseholders then the requirements for fuse serviceability should be met. Otherwise, you need a way to isolate each side of the fuse, which is best done with a load-break switch (on one side of the fuse), and then you could have a pullout, or "non-load break disconnect" as simple as a lockable connector that can be opened to isolate the other side of the fuse(s) prior to working on it. Alternately, you could use an aggregator that uses 240V/20A circuit breakers and avoid the fuse serviceability requirement.

4. Solar sub panel is on roof remote from main service disconnecting means which is inside building electric room. Do I need to comply per NEC 2014 Article 690.13(A)(E)? Their is exception for 690.13(A) but not for 690.13(E) is confusing.
--> Yes, you need a "PV system disconnect." This is the main switch or breaker that isolates the PV system from the rest of the building. It must also be in a readily accessible location per 690.13(A). From the look of it, you don't have any disconnect between your solar subpanel and the taps from the line side of your main service disconnect. So it seems a service-rated AC disconnect that is also your "PV system disconnect" should go in front of your solar subpanel, and should probably have overcurrent protection, depending on the length of the tap. Since your solar subpanel is on the roof, I think your best bet is to install this separate AC switch (690.13(D)) in the electrical room. Finally, I don't think you can use the exception to 690.13(A), which cites 690.31(F) (think it's a typo in my book), which references small conductors and doesn't make any sense. Anyway, you'll need that PV system disconnect, since it will also be your rapid shutdown "initiating device"...

5. Any one know ENphase IQ +7 inverter is on roof how does it comply per rapid shutdown Nec 2014 Article 690.12? How does one initate rapid shutdown?
--> The Enphase IQ7+ (and others) are UL listed to comply with Rapid Shutdown requirements. Their marketing claims NEC 2014 and 2017 compliance, and you can verify that your microinverter (IQ7PLUS...) is in their UL file QIJW.E341165. By removing AC power via your PV system disconnect the inverters should shut off and conform to 690.12. The PV system disconnect should be labeled per 690.56(B): "PHOTOVOLTAIC SYSTEM EQUIPPED WITH RAPID SHUTDOWN." BUT! Please verify with the manufacturer on how to do this correctly. As with all things free--take my advice for direction, but please verify with professionals (inspectors, engineers, and the manufacturer) that it works for your specific situation.

Besides that, good luck! D

 

[jaggedben]

BTW, what about using (1) 100A 3-pole breaker instead of (3) 60A 2-pole breakers? I'm thinking that a short piece of 100A conductor on each breaker could go to a polaris connector picking up two 60A L-L legs not in phase. Then the outdoor tap rule of unlimited length would be used for the 60A conductors. You might need to put a 60A main breaker in the combiner boxes, I'd have to check the tap rules.

Cheers, Wayne

I like this idea. Not sure it fully checks out but worth making a note to self.

 

[ggunn]

BTW, what about using (1) 100A 3-pole breaker instead of (3) 60A 2-pole breakers? I'm thinking that a short piece of 100A conductor on each breaker could go to a polaris connector picking up two 60A L-L legs not in phase. Then the outdoor tap rule of unlimited length would be used for the 60A conductors. You might need to put a 60A main breaker in the combiner boxes, I'd have to check the tap rules.

Cheers, Wayne

But of course that means your conductors would have to be good for 100A protection all the way back to the inverters.

 

[jaggedben]

BTW, what about using (1) 100A 3-pole breaker instead of (3) 60A 2-pole breakers? I'm thinking that a short piece of 100A conductor on each breaker could go to a polaris connector picking up two 60A L-L legs not in phase. Then the outdoor tap rule of unlimited length would be used for the 60A conductors. You might need to put a 60A main breaker in the combiner boxes, I'd have to check the tap rules.

Cheers, Wayne

...--> Yes. If your "aggregator" is a fusible switch or some type of disconnect-combiner with finger-safe fuseholders ...

But of course that means your conductors would have to be good for 100A protection all the way back to the inverters.

ggunn you know this, but that's not what a tap means. You could run three sets of 60A conductors to each subpanel on the roof. They'd all be taps. Or you could run 100A to some point along the way (first subpanel) and then tap three sets of 60A conductors from there. Whatever works.

Please all of you take a look at what the aggregator actually is.
https://enphase.com/en-us/support/enphase-q-aggregator-quick-install-guide

(Note: it appears Enphase may no longer be producing this product. Perhaps because it costs as much as a subpanel, and more than just running three circuits to the roof.)

The aggregator cannot be the termination for a tap because of 240.21(B)(5)(2). (One circuit breaker or set of fuses.) However one could replace the aggregators with small single-phase subpanels with 60A main breakers. Cost of said subpanels containing 4 2-pole breakers is about the same as the aggregator price where I'm at.

 

[ggunn]

I think you are mistaken (a la Kirchoff), but I will look this over when I have more time. In the meantime it is a non-sequitur when it comes to 705.12(D)(2)(3)(c); there is no mention of current in the article, only the ratings of load and supply breakers.

I stand corrected; Kirchoff is still in effect. I rarely deal with three phase systems, and every time I do it seems I have to go back to first principles to get it all straight in my head. While it is true that a PV breaker connected to phases A and B and one connected to phases B and C both contribute to the current on phase B, they do not do so in phase with each other, hence the sqrt3 factor.

My bad.

It doesn't change the way section (c) is written and interpreted, though.