2-to-1 combining of source circuits, multiple stages

[Carultch]

Given 4 source circuits, with a module maximum fuse size of 15A, with an Isc at STC of 2A, is it permissible to gang up all four circuits with two stages of branch connectors, and no in-line fuses?

2A*1.25*1.25 * 4 = 12.5A

Circuit A = Circuit 1 and Circuit 2 in parallel
Circuit B = Circuit 3 and Circuit 4 in parallel
Combine Circuit A and Circuit B.

I know it is generally the case that you can do this with two source circuits. But what about the rare situation where the Isc is so low, that a combination of four source circuits is below 15A?

The reason I ask, is I have BIPV modules with an unconventional mix of voltage and current. So I need numerous source circuits to connect in a code-compliant manner that is much more than combiners and inverters at this scale are built to contain.

 

[ggunn]

Given 4 source circuits, with a module maximum fuse size of 15A, with an Isc at STC of 2A, is it permissible to gang up all four circuits with two stages of branch connectors, and no in-line fuses?

2A*1.25*1.25 * 4 = 12.5A

Circuit A = Circuit 1 and Circuit 2 in parallel
Circuit B = Circuit 3 and Circuit 4 in parallel
Combine Circuit A and Circuit B.

I know it is generally the case that you can do this with two source circuits. But what about the rare situation where the Isc is so low, that a combination of four source circuits is below 15A?

The reason I ask, is I have BIPV modules with an unconventional mix of voltage and current. So I need numerous source circuits to connect in a code-compliant manner that is much more than combiners and inverters at this scale are built to contain.

There is no difference between doing what you describe and combining all four strings at once. Combine more than two and you need fuses.

 

[310 BLAZE IT]

Just be careful you never want to bring different voltages to the same mppt, you'll have drastic losses. I think this right

Sent from my SM-N920V using Tapatalk

 

[ggunn]

There is no difference between doing what you describe and combining all four strings at once. Combine more than two and you need fuses.

Self correction here. While it is indeed true that what you describe is electrically identical to ganging them all together at once, 690.9(A) Exception (b) appears to allow the combination of PV source circuits without fuses as long as the total of the short circuit currents from all sources do not exceed the ampacity of the conductors or the maximum fuse size allowed by the module manufacturer. In your example, if Isc is 2A and the maximum fuse size is 15A (which seems weird to me, but if you say so), the maximum short circuit current from four strings is (4)(2A)(1.25) = 10A (the second 1.25 multiplier you used is not applicable), so it looks like you'd be OK.

 

[Carultch]

Self correction here. While it is indeed true that what you describe is electrically identical to ganging them all together at once, 690.9(A) Exception (b) appears to allow the combination of PV source circuits without fuses as long as the total of the short circuit currents from all sources do not exceed the ampacity of the conductors or the maximum fuse size allowed by the module manufacturer. In your example, if Isc is 2A and the maximum fuse size is 15A (which seems weird to me, but if you say so), the maximum short circuit current from four strings is (4)(2A)(1.25) = 10A (the second 1.25 multiplier you used is not applicable), so it looks like you'd be OK.

Yeah, that's what I was asking. I wanted to know if there was something special about doing this with exclusively two circuits, or if it was just about total Isc (with either 1.25 or 1.56 as factors where applicable) in general, not exceeding the circuit ampacity.

 

[jaggedben]

The two string rule of thumb is a consequence of the code language, except that it wouldn't apply if you oversize the conductors per the exception. There's an argument for changing the rule to be actually two strings, but I thinknwhat you're proposing is okay under the current rules.

 

[ggunn]

The two string rule of thumb is a consequence of the code language, except that it wouldn't apply if you oversize the conductors per the exception. There's an argument for changing the rule to be actually two strings, but I thinknwhat you're proposing is okay under the current rules.

I don't think it's so much an oversizing of the conductors (don't we all use #10?) but the very large disparity between the module Isc and the maximum allowable series fuse size. That's a new one on me, but I have very little experience with thin films modules.

 

[Carultch]

(the second 1.25 multiplier you used is not applicable)

I get confused about when I'm supposed to only use one of the 1.25 multipliers and when to use both.

Here's reasons where I know I need to use both factors:
OCPD sizing in general
110.14(C) terminal sizing

Here's reasons where I know I only need to use one 1.25 factor:
90C wire ampacity requirements with conditions of use
OCPD sizing where listed for continuous duty
Sizing of unfused disconnect ampacity, where the disconnect is specifically listed for continuous duty.
Sizing of the EGC in applications governed by NEC 2014 or later, when there is no applicable OCPD.
Adjusted maximum short circuit current on labels

Here's some examples of where I have my doubts:
Sizing of unfused disconnect ampacity, where not specifically listed for continuous duty
Maximum strings in parallel for inverter inputs
Maximum strings in parallel for charge controller inputs
Maximum strings in parallel for combiners and AFCI combiner contactors
Any time I use a product that isn't manufactured for the US market
Anything involving a product that isn't specifically manufactured for PV applications, and specifies a maximum current. Trying to ask whether or not these factors apply is agony.

 

[ggunn]

I get confused about when I'm supposed to only use one of the 1.25 multipliers and when to use both.

It's actually pretty simple. OCPD and wire size are calculated from Imax whether it is AC or DC.

Imax for an inverter is the published maximum current output of the inverter.
Imax for DC PV source and output circuits is 1.25 times the applicable Isc, whether it is for a single string or multiple strings.

Once you get to Imax, treat them all the same.

 

[Carultch]

It's actually pretty simple. OCPD and wire size are calculated from Imax whether it is AC or DC.

Imax for an inverter is the published maximum current output of the inverter.
Imax for DC PV source and output circuits is 1.25 times the applicable Isc, whether it is for a single string or multiple strings.

Once you get to Imax, treat them all the same.

So when an inverter datasheet specifies maximum input current, is there an industry standard for how this is supposed to relate to a module datasheet? When I ask manufacturers for clarification, they struggle to understand my question.

 

[ggunn]

So when an inverter datasheet specifies maximum input current, is there an industry standard for how this is supposed to relate to a module datasheet? When I ask manufacturers for clarification, they struggle to understand my question.

No offense intended, but maybe it is just that you do not understand their answer. When an inverter specifies a maximum input current, that's the most it will take from the array irrespective of how much the array is capable of producing. If conditions are such that the array can produce more than the inverter can take, the inverter will simply clip off what it cannot use. As I said recently in another thread, some clipping when conditions are ideal may be OK if the clipping losses are more than made up for by more production on the shoulders of the power curve.

I guess that the answer to your question is that there is no hard and fast relationship between module Isc and inverter maximum current. It is the same question we went through earlier about the optimum amount to overload an inverter with modules; it's situational.

 

[SolarPro]

So when an inverter datasheet specifies maximum input current, is there an industry standard for how this is supposed to relate to a module datasheet? When I ask manufacturers for clarification, they struggle to understand my question.

Inverter manufacturers are all over the map with this. If you see a "max Isc" or "max short-circuit current" value, this is a hard-stop design, much like maximum voltage. (Of course, some manufacturers of 600 or 1,000 volt nominal inverters publish an even higher higher maximum.) In any case, the max Isc value is the limit that the vendor used used for NRTL testing and it is intended to limit fault currents within the inverter. If you exceed this limit, you are not using the device in accordance with the listed instructions.

But what if the manufacturer does not specify a maximum array Isc in their literature? What does max inverter input current actually refer to in this case? In most cases, it is an operational limit as ggunn suggests. Considering that most inverters these days have multiple mppt inputs, each one of which is typically NRTL tested at a max Isc value, you're highly unlikely to bump up against a hard-stop design limit in practice.

(Just don't try to build the 1 MW system plugged into a 5 kW inverter. Yes, the inverter will power limit at 5 kW. But if there is a fault at the inverter input—that would be bad; it would vaporize the inverter.)

 

[jaggedben]

Inverter manufacturers are all over the map with this. If you see a "max Isc" or "max short-circuit current" value, this is a hard-stop design, much like maximum voltage. (Of course, some manufacturers of 600 or 1,000 volt nominal inverters publish an even higher higher maximum.) In any case, the max Isc value is the limit that the vendor used used for NRTL testing and it is intended to limit fault currents within the inverter. If you exceed this limit, you are not using the device in accordance with the listed instructions.
...

But just to be clear, if the inverter states a max Isc, then it's the nominal STC Isc on the module datasheet to which one compares, right? None of the code required multipliers apply, or any other multipliers based on the environment? I think that's what Carultch is asking.

In my experience one is as likely to see a nominal power limit as a max Isc rating or no limit stated. But regardless, I have always used the modules' STC numbers to compare.

 

[ggunn]

But just to be clear, if the inverter states a max Isc, then it's the nominal STC Isc on the module datasheet to which one compares, right? None of the code required multipliers apply, or any other multipliers based on the environment?

Yes, I am pretty sure that's correct. Isc is a specific number.

 

[pv_n00b]

Since most thin film modules are used in large installations without AHJ review they do first level combining though harnesses that connect multiple strings together without fusing. Anything else is too expensive when dealing with high voltage 2A Isc modules. I would recommend you look into a wiring harness for your system, they make them with and without fuses. It’s less expensive than a bunch of DC combiners with 2A fuses.

It was more common for modules to have high reverse current ratings back maybe 10 years ago, I’ve seen data sheets showing values up to 25A on some old modules. At that time is was normal to put more than 2 strings in parallel without fuses. These days modules seem to be rated just high enough to use a correctly sized fuse. Probably something in the UL listing test changed that made it worth keeping the size as small as possible. The NRTL does not tell the module manufacturer what the current rating is, the module manufacturer tells the NRTL what current to use in the test and the module just has to pass.

 

[pv_n00b]

But just to be clear, if the inverter states a max Isc, then it's the nominal STC Isc on the module datasheet to which one compares, right? None of the code required multipliers apply, or any other multipliers based on the environment? I think that's what Carultch is asking.

In my experience one is as likely to see a nominal power limit as a max Isc rating or no limit stated. But regardless, I have always used the modules' STC numbers to compare.

Inverter manufacturers are not doing a great job communicating this information. Almost all list a maximum DC input current which is the most current the inverter will accept and convert to AC, but not a limit on the most current the array can produce. This is an operational design issue and not a code issue. It might be a warranty issue if someone put a very small inverter on a large array.

I’ve seen DC/AC ratio limits where the inverter can’t be loading more than 1.X times the AC rating. They generally do not state what array DC rating they are using but the industry standard is STC with no corrections. If this is in the installation manual it now falls under the code requirement that the instructions be followed. If not then it’s a warranty limit. Sometimes a manufacture will raise this ratio for a specific installation in consultation with the engineer’s designing the system.

Then there is the maximum DC short circuit that can be available to the inverter. This is a hard limit that has to do with the listing and must be followed. It includes Isc and the 1.25 multiplier for over insolation since that’s what the NEC considers the maximum current from the array. I can’t think of a reason it should require the 1.25 continuous current multiplier since that’s an operational adjustment, not a fault adjustment.

 

[jaggedben]

...
Then there is the maximum DC short circuit that can be available to the inverter. This is a hard limit that has to do with the listing and must be followed. It includes Isc and the 1.25 multiplier for over insolation since that’s what the NEC considers the maximum current from the array. I can’t think of a reason it should require the 1.25 continuous current multiplier since that’s an operational adjustment, not a fault adjustment.

See this what we're asking about... How is anyone supposed to know that you would have to take 1.25 of the module nameplate Isc if the manual doesn't spell that out?