Enphase microinverter branch max circuit size?

[tortuga]

Greetings all I have only ever installed a dozen or so microinverter systems, and always have used 20 amp branch circuits and 12/2 wire (perhaps also 10/2 for voltage drop).
I have a job where another electrician roughed in a nice beefy 8/2 romex for the 'future solar' and now that the future is here and the solar is being installed I put it on a 30A breaker, just because I had it in the van, then I got to thinking I have never used a circuit larger than 20 Amps so I looked over the enphase documentation figuring I would see something saying I need a 20 Amp circuit max but instead it says:
5 "Limits may vary. Refer to local requirements to define the number of microinverters per branch in your area."

In my area its pretty much just the NEC, Is there anything in the NEC that limits the microinverters branch circuit to 20 amps?
For this particular system its tiny only 3.7 kVA, but I was thinking of larger systems in the future.
Thanks in advance and happy friday

 

[wwhitney]

figuring I would see something saying I need a 20 Amp circuit max

It's there, just at the end of the Input section for some reason: "AC side protection requires max 20A per branch circuit"

Cheers, Wayne

 

[tortuga]

It's there, just at the end of the Input section for some reason: "AC side protection requires max 20A per branch circuit"

Cheers, Wayne

Thanks Wyane

 

[jaggedben]

The 8/2 goes to the roof?
One option is to use the Enphase 'Aggregator' to split a larger roof feed into smaller circuits on the roof.
If the 8/2 goes to the combiner location then you can run 20A circuits to the roof.

 

[tortuga]

The 8/2 goes to the roof?

Its just coiled up in the attic

One option is to use the Enphase 'Aggregator' to split a larger roof feed into smaller circuits on the roof.
If the 8/2 goes to the combiner location then you can run 20A circuits to the roof.

Yeah good point that #8 gives them options for future expansion.
Right now its just a minimal 10 micros.

 

[jaggedben]

It's also good for lower voltage drop. Half a percent of the electricity those panels will generate over 25 years may add up to more than the additional cost of wire. I've never seen 12 awg brought to the roof.

 

[tortuga]

I've never seen 12 awg brought to the roof.

What do you typically see? #10?

 

[jaggedben]

Most typically 10awg, yes, sometimes 8, especially for longer runs.

 

[tortuga]

Most typically 10awg, yes, sometimes 8, especially for longer runs.

Good to know thanks.
Yeah now that voltage drop is 'in the code' allot of my 'rules of thumb' are out the window.

 

[jaggedben]

There's still no code requiring a minimum voltage drop, as far as I know. But the other reason for 10awg was derating conductors for rooftops. Many AHJs required a .58 temp derating factor so 12 awg wasn't good enough, especially for DC conductors subject to the 1.56 multiplier of Isc. So a lot of it is 'tradition' in the PV industry, and reduced voltage drop also a benefit.

 

[tortuga]

There's still no code requiring a minimum voltage drop, as far as I know.

Voltage drop is enforceable all state energy codes that I know of
For the 3 west coast states its:

California title 24 energy code: 130.5 (c) Voltage Drop. The maximum combined voltage drop on both installed feeder conductors and branch circuit
conductors to the farthest connected load or outlet shall not exceed 5 percent.

For Oregon its
Oregon Energy code:

90.1-2019
Section 8 – 8.4.1
Voltage Drop

Two types of conductors
• Feeder conductors
– Connect service equipment to the branch circuit breaker panels
• Branch circuit conductors
– Run from the final circuit breaker to the outlet or load
Feeder conductors and branch circuits combined to be sized
for a maximum of 5% voltage drop total

State of Oregon: Energy in Oregon - Energy Code & the Built Environment

www.oregon.gov

2018 Washington State Energy Code:

C405.11 Voltage drop in feeders and branch circuits. The total voltage drop across the combination of
feeders and branch circuits shall not exceed five percent.

https://www.sbcc.wa.gov/state-codes-regulations-guidelines/state-building-code/energy-code

But the other reason for 10awg was derating conductors for rooftops. Many AHJs required a .58 temp derating factor so 12 awg wasn't good enough, especially for DC conductors subject to the 1.56 multiplier of Isc. So a lot of it is 'tradition' in the PV industry, and reduced voltage drop also a benefit.

Interesting makes sense.

 

[jaggedben]

Voltage drop is enforceable all state energy codes that I know of
For the 3 west coast states its:

Yeah, but interestingly a microinverter is neither an outlet or a load. So they missed including them.

 

[Solar Guy]

Depending on the rating of the panel your microinverters are installed under, the micros will generate maybe 1.8A. So, 10 of them in parallel and you have ~18A, and the rule is to size the circuit for 125%, so you will need #10AWG minimum and a 25A 2pole breaker minimum. The #8AWG is a nice gift, and so is the "free" 30A breaker you had. The breaker protects the branch wiring, not the micros.
You should also check the actual voltage being delivered by the utility. The inverters will raise their voltage to a little above the utility so as to push their power in preference over the utility, and doing that plus fighting a big voltage drop over small feeders could overtax them.

 

[wwhitney]

Depending on the rating of the panel your microinverters are installed under, the micros will generate maybe 1.8A. So, 10 of them in parallel and you have ~18A, and the rule is to size the circuit for 125%, so you will need #10AWG minimum and a 25A 2pole breaker minimum.

That's not the way it works. The microinverters get counted at their maximum output current regardless of connected panel. And for Enphase products I've seen the branch circuit (feeder?) is limited to 20A, so the maximum number of microinverters on one circuit is limited by that.

I.e. the product in the OP says its output current is 1.45A at 240V, and 16A / 1.45A = 11.03, which is why the spec sheet specifies a maximum of 11 microinverters on a 20A circuit.

Cheers ,Wayne

 

[Solar Guy]

Wayne, you're probably right given that you know the model # and have the spec sheet for this job, and I don't. We also limit our residential strings to 10 or 11 micros, because we hook them up with #12AWG modular connectors. If we have more strings than that, we add a combining panel and THEN go to a larger feeder. I'll plead a bit of weekend ignorance as I work mostly on the commercial side with ~440watt panels, which a micro can't harvest all of. Or would die trying.

 

[jsherwin]

Odd question - can you combine microinverter circuits in parallel (power block)? or does code require over current protection on each circuit?

 

[wwhitney]

Odd question - can you combine microinverter circuits in parallel (power block)? or does code require over current protection on each circuit?

A string of microinverters is always connected in parallel. You can have as many on a string as the spec sheet allows--typically, as in the OP, the spec sheet listed a maximum 20A OCPD and that determines the maximum count.

Cheers, Wayne

 

[jsherwin]

A string of microinverters is always connected in parallel. You can have as many on a string as the spec sheet allows--typically, as in the OP, the spec sheet listed a maximum 20A OCPD and that determines the maximum count.

Cheers, Wayne

Hi Wayne - My apologies. I wrote that poorly.
I meant that ask if the code allows you to combine strings of modules with microinverters in parallel without overcurrent protection on the individual branches. The home run would have a single ocpd.
Thanks.

 

[wwhitney]

I meant that ask if the code allows you to combine strings of modules with microinverters in parallel without overcurrent protection on the individual branches. The home run would have a single ocpd.

I think what you are asking is whether the physical wiring needs to be a daisy chain downstream of the OCPD, or whether it could be a star topology, say, or otherwise branched.

The answer is that electrically those are all parallel connections, and the physical arrangement doesn't matter. The only requirement is that downstream of the 20A OCPD you have no more than the allowable number of microinverters.

Cheers, Wayne

 

[tortuga]

I think what you are asking is whether the physical wiring needs to be a daisy chain downstream of the OCPD, or whether it could be a star topology, say, or otherwise branched.

The answer is that electrically those are all parallel connections, and the physical arrangement doesn't matter. The only requirement is that downstream of the 20A OCPD you have no more than the allowable number of microinverters.

Cheers, Wayne

Agreed, here is their sample diagram, the limiting factor is the required 20A OCPD in the breaker panel: