OCPD Tolerance

[Cudal81]

I got 40 IQ7 plus inverter. max. continuous current = 1.21A
40 x 1.21 x 1.25 = 60.5A
so I need 100A safety switch with 70A fuses for my disconnect.
can I still use 60A safety switch with 60A fuses? does NEC tolerate this? if yes, by how much? Thank you!

 

[bwat]

Sadly, I don't think you can round down here, if that's what you're asking. There have been discussions in the past about where and when 220.5(B) applies, but even that wouldn't help you because it only would have allowed you to round down if it were smaller than 0.5.

220.5(B) Fractions of an Ampere. Calculations shall be permitted
to be rounded to the nearest whole ampere, with decimal fractions
smaller than 0.5 dropped.

 

[wwhitney]

Well, if we go back to the data sheet, it shows a maximum continuous apparent power of 290 VA, so dividing by 240V gives 1.20833 repeating, which the spec sheet rounds to 1.21. But if you multiply by 40 without rounding (e.g. do 40*290/240) and then multiply by 1.25, you get 60.4166 repeating amps, which you can round down to 60 amps.

Cheers, Wayne

 

[bwat]

Well, if we go back to the data sheet, it shows a maximum continuous apparent power of 290 VA, so dividing by 240V gives 1.20833 repeating, which the spec sheet rounds to 1.21. But if you multiply by 40 without rounding (e.g. do 40*290/240) and then multiply by 1.25, you get 60.4166 repeating amps, which you can round down to 60 amps.

Cheers, Wayne

Ha!! Great catch.

 

[ggunn]

Well, if we go back to the data sheet, it shows a maximum continuous apparent power of 290 VA, so dividing by 240V gives 1.20833 repeating, which the spec sheet rounds to 1.21. But if you multiply by 40 without rounding (e.g. do 40*290/240) and then multiply by 1.25, you get 60.4166 repeating amps, which you can round down to 60 amps.

Cheers, Wayne

Yes. Always do your rounding as the last step on the final result. It may or may not give you the answer you want but it will be the most correct.

 

[pv_n00b]

You run the risk of an AHJ seeing the 1.21A and requiring that be used. Most likely you will hear about this until the final inspection after the system is installed and the AHJ won't sign it off. That would be no fun. The kind of AHJ who hassles you over 0.5A will hassle you over using the 1.208A to squeak by.

 

[GoldDigger]

In any case, it can be argued that the code requires you to use the direct number from the spec sheet, not an indirect number calculated from other spec sheet values.

Sent from my Pixel 4a using Tapatalk

 

[wwhitney]

In any case, it can be argued that the code requires you to use the direct number from the spec sheet, not an indirect number calculated from other spec sheet values.

Sure, but since they list values for both 240V and 208V, it's clear they are just taking 290VA and dividing by the nominal voltage. And 290 VA is only two significant figures anyway, so they should just round to 1.2A. : - ). Perhaps Enphase could clarify if push came to shove.

But one things confuses me: Enphase requires AC side OCPD at a max of 20A, so you'd typically divide up the microinverters into groups of 13, each on a 20A OCPD. I guess the OP might have 4 strings of 10?

Interestingly, the rounding rules tell you that a string of 10 can be on a 15A OCPD, even using the datasheet value of 1.21A: 12.1 * 1.25 = 15.125, round down to 15A. Which would seem to give the nonsensical result that the combiner for (4) such strings would need an ampacity exceeding 60A.

Cheers, Wayne

 

[jaggedben]

My bottom line comment on this one: a reasonable AHJ would let it slide, but we all know that not all AHJs are reasonable.

Wayne, that's funny with the 15A breakers. I've never seen a rounding error go that direction before.

 

[pv_n00b]

The NEC rounding rules leave a lot to be desired. The following rounding is all the same on a percentage basis (4%):
10.4 to 10
104 to 100
1040 to 1000

But the NEC only allows
10.4 to 10 4%
100.4 to 100 0.4%
1000.4 to 1000 0.04%

 

[bwat]

The NEC rounding rules leave a lot to be desired. The following rounding is all the same on a percentage basis (4%):
10.4 to 10
104 to 100
1040 to 1000

But the NEC only allows
10.4 to 10 4%
100.4 to 100 0.4%
1000.4 to 1000 0.04%

Fascinating. I actually never thought about the consequences of the NEC rounding rules related to percentages.

 

[ggunn]

FWIW, if a wire size is on the hairy edge and one size up isn't, I'll use the larger size to avoid a potential argument and/or inspection failure. It's cheap insurance. That said, if my design is correct and it's already installed, I'll fight.