New Solar Power System - Bus Rating Question

[Jackbenimble7]

Hello all,

I have a question regarding bus sizing when installing new solar power systems. I do a lot of commercial electrical design, and I've found myself questioning whether I'm sizing AC panels correctly. Typically, we will install multiple inverters that feed back to a single AC panel. For example, we may install (3) 20KW inverters. These will all have their own separate runs back to an AC panel. I will size this new AC panel's main breaker based off the wattage of the inverters. This gives me a main breaker size of 100A (at 480V 3ph.). I would then run conduit from that 100A breaker back to a disconnect at the existing Main Switchboard, and then supply side interconnect the existing bus.

I have been sizing the new AC panel (with the 100A main circuit breaker with three inverter generators) bus at 200A. This is how I've been taught to do it, the theory being that the bus needs to be sized for backfeed from the utility. The idea is that if the utility is producing a load through that 100A main breaker and the inverters are also generating power, then hypothetically you could have more than 100A on that bus. Does this make sense? Or am I needlessly increasing the bus size of new AC panels?

This isn't really an issue at low wattage, but for bigger systems, I may have an 800 main circuit breaker with a 1600A bus. There's no way the utility will every provide 800A of power to a panel that's only connected to solar inverters, right? And the inverter output would be capped at 800A due to their individual breakers. Ultimately, I believe these AC panels should be rated off the main circuit breaker size (100A MCB = 100A bus, 800A MCB = 800A bus). I'm basically doubling the size of the bus to compensate for load flow from two different directions. Anyone have any opinions?

Let me know if you need clarification, this can be a little confusing to write out. Thank you.

 

[GoldDigger]

The double bus current from both sources is only possible if you connect that much load to the same panel bus. If there is no load, the current will only be flowing into the bus from one direction at any given time.
The only possible problem would be a highly unlikely fault on the bus which somehow draws between 100 and 200 amps and keeps at that level for minutes.

Sent from my XT1585 using Tapatalk

 

[Carultch]

Hello all,

I have a question regarding bus sizing when installing new solar power systems. I do a lot of commercial electrical design, and I've found myself questioning whether I'm sizing AC panels correctly. Typically, we will install multiple inverters that feed back to a single AC panel. For example, we may install (3) 20KW inverters. These will all have their own separate runs back to an AC panel. I will size this new AC panel's main breaker based off the wattage of the inverters. This gives me a main breaker size of 100A (at 480V 3ph.). I would then run conduit from that 100A breaker back to a disconnect at the existing Main Switchboard, and then supply side interconnect the existing bus.

I have been sizing the new AC panel (with the 100A main circuit breaker with three inverter generators) bus at 200A. This is how I've been taught to do it, the theory being that the bus needs to be sized for backfeed from the utility. The idea is that if the utility is producing a load through that 100A main breaker and the inverters are also generating power, then hypothetically you could have more than 100A on that bus. Does this make sense? Or am I needlessly increasing the bus size of new AC panels?

This isn't really an issue at low wattage, but for bigger systems, I may have an 800 main circuit breaker with a 1600A bus. There's no way the utility will every provide 800A of power to a panel that's only connected to solar inverters, right? And the inverter output would be capped at 800A due to their individual breakers. Ultimately, I believe these AC panels should be rated off the main circuit breaker size (100A MCB = 100A bus, 800A MCB = 800A bus). I'm basically doubling the size of the bus to compensate for load flow from two different directions. Anyone have any opinions?

Let me know if you need clarification, this can be a little confusing to write out. Thank you.

Be careful. Sometimes the inverters have a higher current rating than you would ordinarily calculate from either P=I*V or P=I*V*sqrt(3). One reason is that the inverters have a higher nominal current to allow full nameplate power when grid voltage is on the low end of the margin. Another reason is power factor management. In any case, the datasheet value overrides P=I*V or P=I*V*sqrt(3).

Per NEC2014, the maximum you should ever have to size a dedicated PV system panelboard for, is "the sum of the breakers, excluding the main supply, shall not exceed the ampacity of the busbar". So just pretend all branch circuits are all inverters, add up the trip ratings, and size the busbar per that sum. Exclude the main supply.

Even this is overkill, in my opinion, as it requires you to add up all those rounding errors, that will be caught by the main breaker before there really is a problem.

 

[Jackbenimble7]

The double bus current from both sources is only possible if you connect that much load to the same panel bus. If there is no load, the current will only be flowing into the bus from one direction at any given time.
The only possible problem would be a highly unlikely fault on the bus which somehow draws between 100 and 200 amps and keeps at that level for minutes.

Sent from my XT1585 using Tapatalk

So it sounds like you're thinking along the same lines as I am, that the bus only needs to be sized to the MCB. I tend to agree, though I do still worry a bit about backfeed from the utility. What do you think?

 

[GoldDigger]

So it sounds like you're thinking along the same lines as I am, that the bus only needs to be sized to the MCB. I tend to agree, though I do still worry a bit about backfeed from the utility. What do you think?

The exact applicable guidelines will depend on your code cycle. And more importantly whether your inspector feels that the 120% rule still applies to an AC combiner panel with no load breakers.
It varies!

Sent from my XT1585 using Tapatalk

 

[Jackbenimble7]

You're telling me something varies in the world of electrical contracting!?? No way, I don't believe it!!

Thanks for the feedback, if anyone else has experience with this, please let me know.

The exact applicable guidelines will depend on your code cycle. And more importantly whether your inspector feels that the 120% rule still applies to an AC combiner panel with no load breakers.
It varies!

Sent from my XT1585 using Tapatalk

 

[electrofelon]

So it sounds like you're thinking along the same lines as I am, that the bus only needs to be sized to the MCB. I tend to agree, though I do still worry a bit about backfeed from the utility. What do you think?

Well are you interested in what the code says, or opinions on what the code should be? The 2014 is a big improvement as it gives us some more flexibility. IMO, Panelboards and services of 150 amps or more in residential applications should be able to be backfed to the rating of the service. I also would like to see the 120% rule raised a bit. I think it is overly conservative.

 

[Jackbenimble7]

Well are you interested in what the code says, or opinions on what the code should be? The 2014 is a big improvement as it gives us some more flexibility. IMO, Panelboards and services of 150 amps or more in residential applications should be able to be backfed to the rating of the service. I also would like to see the 120% rule raised a bit. I think it is overly conservative.

To be honest with you, I'm most interested in creating the most cost effective design. In my first post, I stated that this is generally for 3 phase commercial applications, not sure if that changes your opinion on this topic.

I'm still of the notion that doubling the bus rating is far too conservative, though I need to do further research.

 

[jaggedben]

To be honest with you, I'm most interested in creating the most cost effective design. In my first post, I stated that this is generally for 3 phase commercial applications, not sure if that changes your opinion on this topic.

I'm still of the notion that doubling the bus rating is far too conservative, though I need to do further research.

At the very least you could add the sum of the utility feed and 125% of inverters and divide that by 1.2. (120%). Doesn't make much difference in your example but it could in others.

The 2011 code can be read to require the oversize bus, and the 2014 code pretty much eliminated that interpretation.

 

[ggunn]

...I do still worry a bit about backfeed from the utility. What do you think?

You have OCPD at the point of interconnection that limits the fault current available from the utility on the feeder between the service and the AC combiner. All the OCPD on the AC side is to protect the conductors from the utility; the inverters are incapable of endangering the conductors if they are sized correctly.