Wind Generation Panel Sizing

[Shoe]

I have a situation where I'm sizing a panel/feed to connect (10) 1KW wind generators grid-tied to a large commercial facility. The output of each of the (10) inverters is 277V. My question has to do with Article 705.12:

How do I size the feeder and panel bus to feed the (10) inverters? Each wind generator would have a 277V, 20A circuit going to it from a 277/480V, 3-phase panel.

Do I have to comply with the 120% rule, in which case, I'd have to multiple 20amps*10*120%? = 250 panel and feeder? That seems ridiculous, given the generating capacity is less than 15 amps at 480v, 3-phase.

Please let me know where (or if) I'm going wrong here.

 

[jaggedben]

How do I size the feeder and panel bus to feed the (10) inverters? Each wind generator would have a 277V, 20A circuit going to it from a 277/480V, 3-phase panel.

Do I have to comply with the 120% rule, in which case, I'd have to multiple 20amps*10*120%? = 250 panel and feeder? That seems ridiculous, given the generating capacity is less than 15 amps at 480v, 3-phase.

First of all, you are reading the 120% rule backwards. The sum of the breakers supplying the panel can be up to 120% of the panel rating, so the calculation would be the (sum of supply breakers)/1.2, not (sum of supply breakers)*1.2.

Secondly, you are presumably going to put 4 inverters on one bus of the 3phase panel and 3 inverters each on the other two. So your calculation would involve only 4 inverters, not 10.

There has been some disagreement on this forum about whether you can ignore the breaker supplying the panel if the panel is only used as an "AC combiner" for the inverters. If your AHJ let's you ignore it, then your bus calculation should be 20A*4/1.2 or 66A minimum. If he doesn't, then your calculation would be (80A+panel supply breaker)/1.2.

As far as sizing the feeder for an unbalanced system like this, I have to admit that the math is a bit beyond me (I'm not an EE), but look at this article:
http://solarprofessional.com/article/?file=SP3_1_pg66_Hamon
It's written for solar but is completely applicable to wind as well.
I think you're going to be looking at a 25A or 30A breaker to supply the panel, so a 125A panel ought to do, but I have not done the calculation. Your calculation of less than 15A at 480V strikes me as incorrect, as if it were a balanced system with 9 inverters then the max amps on each line would be 18.6.

Finally, I'm not sure why you have decided to use 20A breakers for 1kW inverters. I see only 3.6A per inverter at 277V. If you can downsize these breakers you can use a smaller panel.

 

[Shoe]

Thank you for the excellent response. I've reviewed all points over the weekend and agree with your analysis.

I plan to use 15 amp breakers, which will reduce my overall feeder requirements. The panel is a standard 125 A panel, so no worries about bus capacity.

When I read the code originally (much too quickly I might add), I panicked when I started adding up the 20 amp breakers without regard to phasing or correctly interpreting the 120% rule

 

[gadfly56]

I don't know if this will affect anything, but you can count on never getting more than 30-35% of the nameplate capacity of the wind generators in actual output.

 

[jaggedben]

I don't know if this will affect anything, but you can count on never getting more than 30-35% of the nameplate capacity of the wind generators in actual output.

As far as the NEC breaker sizing for the inverters is concerned, it doesn't matter. You have to use the nameplate output. He is oversizing his breakers (and presumably his wires) anyway.

Secondly, I don't think what you're saying isn't strictly true. 30-35% is often quoted as the average 'capacity factor' of wind generation, but what this means is that sometimes generation is well over 35% of nameplate output and sometimes it is well under. At some point there will probably be good wind and the generator will deliver well above 35% percent, if not 100% of nameplate output. Actual performance depends on the site, of course.

If he thinks his turbines are very rarely going to produce full power, he could undersize the feeder and breaker to the combiner panel, but then the breaker might trip when he is getting his best output, and is that worth the money saved?

 

[ggunn]

If he thinks his turbines are very rarely going to produce full power, he could undersize the feeder and breaker to the combiner panel, but then the breaker might trip when he is getting his best output, and is that worth the money saved?

Doesn't he have to size the conductors and OCPD based on the nameplate rating on the inverter (or whatever it is that converts the power in from the turbine to the correct AC voltage and phase) just as we have to do with a solar inverter irrespective of how many modules are wired to it?

 

[jaggedben]

Doesn't he have to size the conductors and OCPD based on the nameplate rating on the inverter (or whatever it is that converts the power in from the turbine to the correct AC voltage and phase) just as we have to do with a solar inverter irrespective of how many modules are wired to it?

For each inverter output circuit, yes. But for the feeder to his 'AC combiner'? I'm not sure the code has a requirement about that.

 

[ggunn]

For each inverter output circuit, yes. But for the feeder to his 'AC combiner'? I'm not sure the code has a requirement about that.

It only makes sense to me that if you must size the conductors and OCP from the inverter to the combiner based on 125% of the maximum rated current of the inverter, then the conductors from the combiner to the interconnection point need to be sized for the sum of those currents. That's certainly the way that I design them, anyway.

Caveat: I only do PV; if it's different for wind, I wouldn't know.

 

[jaggedben]

It only makes sense to me that if you must size the conductors and OCP from the inverter to the combiner based on 125% of the maximum rated current of the inverter, then the conductors from the combiner to the interconnection point need to be sized for the sum of those currents. That's certainly the way that I design them, anyway.

Caveat: I only do PV; if it's different for wind, I wouldn't know.

705 applies to both wind and PV. If the code requires what you're saying, it's not in 705.

For solar, it seems reasonable that you're going to get close to the maximum output many times a year on sunny days, so it makes sense to design systems as you are doing regardless of whether code requires it.

For wind, I'm not so sure. But I don't do wind either.

 

[ggunn]

705 applies to both wind and PV. If the code requires what you're saying, it's not in 705.

It's in 690.8. Looking at 694.12, it appears that the same is true for wind.

 

[jaggedben]

All right, you're correct. What clinched it for me was not 690.8, but rather reviewing the definition(s) of an inverter output circuit in 690 and 694. It includes the wiring all the way back to the service where that is present.