125% Output current is 0.14 over a standard breaker size

[ggunn]

Sorry for the duplication, but this from the other thread is more pertinent to this one:

I saw a system recently that used (5) PV inverters that each had 48.25A max output current. (1.25)(48.25A) = 60.3A, and 60A inverter OCPDs were used per 220.5(B). For the combined inverter output, however, (5)(1.25)(48.25A) = 301.6A, so no rounding; a 350A OCPD was used.

FWIW, in the above PV system a note that 220.5(B) was used to size the inverter OCPDs was clearly marked on the electrical drawing, and although the AHJ in this instance was a stickler for NEC compliance, the use of 220.5(B) was not questioned. I don't see why the fact that the inverter is a source would make any difference; the inverter OCPD is there to protect the conductors from fault current from the service, the same as for a load; it's not needed to protect them from the inverter.

 

[wwhitney]

the use of 220.5(B) was not questioned. I don't see why the fact that the inverter is a source would make any difference

Procedurally (as opposed to physics-wise), because 220.5(B) only comes into play where the NEC wording uses the word "load". The requirements in Article 690 under consideration do not use the word "load". So Article 690 is silent on the question of rounding.

Cheers, Wayne

 

[ggunn]

Procedurally (as opposed to physics-wise), because 220.5(B) only comes into play where the NEC wording uses the word "load". The requirements in Article 690 under consideration do not use the word "load". So Article 690 is silent on the question of rounding.

Cheers, Wayne

Be that as it may, there is no difference between an OCPD protecting the conductors to a load and one protecting the conductors to an inverter. They both protect the conductors from current coming from the service during a fault; the direction of the current in normal operating conditions is irrelevant.

 

[jaggedben]

Be that as it may, there is no difference between an OCPD protecting the conductors to a load and one protecting the conductors to an inverter. They both protect the conductors from current coming from the service during a fault; the direction of the current in normal operating conditions is irrelevant.

If someone says "I'm not interested in the physics, just what the code says" then you will say "But physics!".

And if someone says "Let's leave aside the code for a minute and talk about the physics" you will say "But code!".

 

[ggunn]

If someone says "I'm not interested in the physics, just what the code says" then you will say "But physics!".

And if someone says "Let's leave aside the code for a minute and talk about the physics" you will say "But code!".

Um, OK; is there something wrong with that?

If the code were to prohibit rounding down 125% of the maximum inverter current to the nearest whole number if the difference is less than half an amp, I would of course comply, but it doesn't.

 

[pv_n00b]

There have been a number of proposals to make that rule apply through out the code, but they have been rejected and that rounding rule only applies to for the purposes of load calculations and not to OCPDs and conductor sizing.
The rejection statements say the rounding rule only applies within Article 220.

I've been one of those putting in proposals to add rounding to 690 and 705 and they always get shot down. The CMP does not want rounding applied to sources, only load calculations.

 

[pv_n00b]

If someone says "I'm not interested in the physics, just what the code says" then you will say "But physics!".

And if someone says "Let's leave aside the code for a minute and talk about the physics" you will say "But code!".

Physics is phun, but the code feeds the bulldog. Really the bottom line is that the code is conservative and the arguments to look at the physics and look at the math are ways to come to a less conservative but still accurate answer. An answer that always will allow cost savings or why bother. But they are not technically code compliant. So maybe you get one past the goalie and maybe not if you try to out physics the code.

 

[ggunn]

Physics is phun, but the code feeds the bulldog. Really the bottom line is that the code is conservative and the arguments to look at the physics and look at the math are ways to come to a less conservative but still accurate answer. An answer that always will allow cost savings or why bother. But they are not technically code compliant. So maybe you get one past the goalie and maybe not if you try to out physics the code.

It's not "physics". The protection of conductors by an OCPD is precisely the same for loads and grid tied PV inverter output circuitss. OCP on inverter output circuits has nothing whatsoever to do with the direction of current flow during normal operating conditions. Under fault conditions there is no difference between them - under fault conditions the circuit IS a load; that is why the OCPD is at the opposite end of the conductors from the inverter. I am not "getting away" with anything when I invoke 220.5(B) in sizing an inverter OCPD, and it could make a big difference when qualifying a combiner bus with 705.12(B)(3). I am steadfastly against taking advantages of "loopholes" in the code which can result in hazardous or unsafe installations, but this is most assuredly not one of them.

 

[wwhitney]

The protection of conductors by an OCPD is precisely the same for loads and grid tied PV inverter output circuitss. OCP on inverter output circuits has nothing whatsoever to do with the direction of current flow during normal operating conditions.

Sure, but the sizing of the OCPD is not about fault conditions, it's about normal operating conditions. It is logically consistent for the NEC to say: "in a load calc, the inputs are somewhat uncertain, and there is diversity, so rounding down is quite unlikely to cause nuisance tripping and is allowed. While in inverter output circuits, there is less uncertainty, less margin for error, and so rounding down is not allowed."

Now, I'm not actually saying that the NEC is saying that, just arguing that your claim of equivalence is not accurate. Seems like the NEC has a position on rounding with respect to requirements using the word "load" (220.5(B)) and is silent on all other rounding questions (anything involving 690 or 705).

Cheers, Wayne

 

[Zee]

If the 125% continuous Amps come to 20.9A, can I use a 20A breaker?

 

[ggunn]

Sure, but the sizing of the OCPD is not about fault conditions, it's about normal operating conditions. It is logically consistent for the NEC to say: "in a load calc, the inputs are somewhat uncertain, and there is diversity, so rounding down is quite unlikely to cause nuisance tripping and is allowed. While in inverter output circuits, there is less uncertainty, less margin for error, and so rounding down is not allowed."

Now, I'm not actually saying that the NEC is saying that, just arguing that your claim of equivalence is not accurate. Seems like the NEC has a position on rounding with respect to requirements using the word "load" (220.5(B)) and is silent on all other rounding questions (anything involving 690 or 705).

Cheers, Wayne

As you say, the NEC does not say that. Inverter output circuits are current limited, and there is a 125% margin on the selection of OCPD, so whether it is sized to 125% of the inverter output or only 124.4%, a nuisance trip is impossible. It is a distinction without a difference. Under fault conditions, which is the only time that the OCPD has any effect on the circuit, load circuits and grid tied PV inverter output circuits are equivalent.

 

[ggunn]

If the 125% continuous Amps come to 20.9A, can I use a 20A breaker?

No. Read 220.5(B).

 

[wwhitney]

As you say, the NEC does not say that. Inverter output circuits are current limited, and there is a 125% margin on the selection of OCPD, so whether it is sized to 125% of the inverter output or only 124.4%, a nuisance trip is impossible.

For continuous currents, the NEC has judged that a 125% factor will reduce the risk of nuisance trips with non-100% rated breakers sufficiently. If you only provide 124.4%, you have not met the (arbitrary) threshold, and you have to draw the line somewhere. Textually in the NEC, 220.5(B) has no bearing on articles that don't use the word "load", like 690 or 705.

Either way, if the ambient temperature of the breaker exceeds 40C, which is entirely plausible in an outdoor panel in direct sunlight in a warm climate during the summer (where your inverter DC/AC ratios are higher enough that you are still maxing out the inverters), you may very well get nuisance trips.

Cheers, Wayne

 

[ggunn]

For continuous currents, the NEC has judged that a 125% factor will reduce the risk of nuisance trips with non-100% rated breakers sufficiently. If you only provide 124.4%, you have not met the (arbitrary) threshold, and you have to draw the line somewhere. Textually in the NEC, 220.5(B) has no bearing on articles that don't use the word "load", like 690 or 705.

Either way, if the ambient temperature of the breaker exceeds 40C, which is entirely plausible in an outdoor panel in direct sunlight in a warm climate during the summer (where your inverter DC/AC ratios are higher enough that you are still maxing out the inverters), you may very well get nuisance trips.

Cheers, Wayne

I thought about that as well, and if 220.5(B) were to show an exception for continuous circuits I might be persuaded, but it doesn't.

 

[wwhitney]

I thought about that as well, and if 220.5(B) were to show an exception for continuous circuits I might be persuaded, but it doesn't.

Not following what you mean by the above.

BTW, I'm not saying it's wrong to necessarily round for Article 690 or 705 the way Article 220 says to round for loads. I'm saying that 220.5(B) is not definitive as it is not directly applicable. Rounding may be reasonable and supportable, and you can argue by analogy to 220.5(B). But it's a bit of a grey area.

Cheers, Wayne

 

[ggunn]

I don't have any more to add; I've said my piece and you know where I stand.