Charge controlller "80% rule" where is it?

[ggunn]

The majority of inspectors I've run across do not understand the concept of short-circuit current rating.

You see? There's no problem.

 

[jaggedben]

In some locations and orientations it is fairly routine to load inverters to 150% DCW STC to ACW. Since voltage is what it is, that overload is all Isc. In some AHJ's that ratio is limited (California, for example), but my impression was that the rule exists to keep rebates that are based on installed STC DC ratings under control.

That may well have been true of the CSI rebate program. (Their automated system probably would have spit out an error or automatically limited the rebate if the DC-AC ratio exceeded a certain limit. I recall something like this happening once or twice when I entered erroneous data). But the 2013 California CEC is just the 2011 NEC in all relevant aspects to this discussion. So, no actual code required limit.

The concept you guys are trying to force feed me with is not all that hard to comprehend. I think its more a matter of terminology, or rather misuse of terminology. Refer to 90.3, the definition of short-circuit current rating in Article 100, and 110.10, which covers short-circuit current ratings and listed equipment.

When I look at the last sentence of 110.10 I think it supports everything I've been saying. Not to mention that there's something weird about applying the concept of 'fault levels' to this situation.

 

[Smart $]

...

When I look at the last sentence of 110.10 I think it supports everything I've been saying. Not to mention that there's something weird about applying the concept of 'fault levels' to this situation.

It supports your point of view in your mind. :lol:

Short-circuit current rating of an array is no different [well... slightly different :blink:] than how available short-circuit current is perceived from other sources.

 

[ggunn]

It supports your point of view in your mind. :lol:

Short-circuit current rating of an array is no different [well... slightly different :blink:] than how available short-circuit current is perceived from other sources.

It is different. A PV module is a virtual current source over most of its IV curve; lowering the load resistance does not draw more current once you get past the knee of the curve. That's more than a slight difference from short circuit current from most other sources.

 

[jaggedben]

It supports your point of view in your mind. :lol:

Short-circuit current rating of an array is no different [well... slightly different :blink:] than how available short-circuit current is perceived from other sources.

Didn't you say you were done debating?

I agree with ggunn that it's not the same, for additional reasons. 'Clearing a fault' with a PV inverter or charge controller would typically only involve GFDI, and that would not typically involve current levels in excess of the module spec Isc. A real short-circuit fault from a PV array would happen somewhere out toward the solar panels and bypass the inverter or charge controller entirely. That would have to be cleared by some other equipment (which is not actually required by code). So it seems just a little weird to apply 110.10 to this situation at all. In any case, to repeat, the last sentence takes care of all the issues here, so... back to 110.3(B) and the equipment instructions.

 

[Smart $]

Didn't you say you were done debating?

I agree with ggunn that it's not the same, for additional reasons. 'Clearing a fault' with a PV inverter or charge controller would typically only involve GFDI, and that would not typically involve current levels in excess of the module spec Isc. A real short-circuit fault from a PV array would happen somewhere out toward the solar panels and bypass the inverter or charge controller entirely. That would have to be cleared by some other equipment (which is not actually required by code). So it seems just a little weird to apply 110.10 to this situation at all. In any case, to repeat, the last sentence takes care of all the issues here, so... back to 110.3(B) and the equipment instructions.

I'm not debating. I'm just commenting. There is a difference, you know.

And what of a fault internal to the inverter or charge controller. How does one implement GFD on input terminals and their jumpers or PCB bus before the GFDI electronics? That's kind of putting the cart ahead of the horse, don't you think?! This equipment does have a short-circuit current rating buried in the listing somewhere. All electrical equipment have an SCCR (for the sake of this statement only, I'm going to say AIC = SCCR).

 

[jaggedben]

I'm not debating. I'm just commenting. There is a difference, you know.

And what of a fault internal to the inverter or charge controller. How does one implement GFD on input terminals and their jumpers or PCB bus before the GFDI electronics? That's kind of putting the cart ahead of the horse, don't you think?!

Well, ground-fault combiners are a possibility (may already exist?). As for a fault internal to the inverter or CC, I'll acknowledge the possibility, (even though it seems way, way less likely than a fault beforehand). In fact, funny you should ask, it just so happens I may have witnessed one today, for the first time ever. But again, I just don't see how 110.10 is supposed to apply beyond following the instructions.

This equipment does have a short-circuit current rating buried in the listing somewhere. All electrical equipment have an SCCR (for the sake of this statement only, I'm going to say AIC = SCCR).

I've not disputed such a point. Still think SCCR as applied by the SMA manual refers to module spec Isc.

 

[Smart $]

...
I've not disputed such a point. Still think SCCR as applied by the SMA manual refers to module spec Isc.

And that has been my point all along (though with a nuance yet to be accepted ). The manual specifies an SCCR. That is, the available short-circuit current of the array (IOW, array Isc) cannot exceed the SCCR given in the manual. The array Isc is calculated from module spec Isc (series 1:1, parallel x:1). So if we agree on this, the only matter unresolved is whether the array Isc is to be factored 125% per 690.8(A)(1)...

 

[Smart $]

FWIW, I believe I read somewhere that a series-connected array Isc can be less than module spec Isc when factoring in the resistance encountered from each member panel. I'd have to research that in order to present it as fact, but it could have some bearing on the matter.