pv source circuit wire sizing

[Zee]

Well i always thought I understood PV wire sizing, but as y'all know it gets a bit complex.
I am combining two PV strings on roof. Each string is 10.5 A Isc each = 21 A Isc combined.
There are existing #10 THHN conductors in conduit from roof to inverter. I would like to use those for labor saving.
Can I?

Overall, I would do the calcs like so:
De-rate wire ampacity down and multiply current up, then compare. Current must be less than equal to ampacity.

Specifically,
A. AMPACITY
Start with 40A ampacity to derate #10 THHN wire.
2 de-rates to consider:
-- Only 2 CCC in conduit, so no derate for # of wires in raceway.
-- Ambient TEMP: Average high Temps are just about 70-80F here, so no derate. (I recall it is average high temps that are to be used. Am i right on this? or record high Temp must be applied? Big difference!)

===> 40A x 1.0 x 1.0 = 40 A allowable ampacity (although ultimately 30A max allowed by NEC in end no matter what)

Now....can I further de-rate the 40A ampacity by 0.8 PV use factor - rather than multiply the Isc and continuous use factor by 1.25?
That gives me 32A allowable. This of course gets knocked to 30A absolute max allowable.

B. CURRENT
I multiply 21 A Isc by 1.25 continuous use factor = 26.25 A.
This is under the above 30A, so I am golden.

Otherwise if I must multiply by 1.56 (1.25 continuous and 1.25 PV use factors), then 1.25 x 1.25 x 21 A pushes me over 30A by a smidge. (1.56 x 21 A = 32.81 A) Ok.... by 2.81 smidges.

My recent reading says the 1.56 had been implied by the NEC, but not quite that simple..... and has been since clarified.
In other words, there is something about applying the 1.25 factors separately one at a time, but I never quite got that.
Any insight appreciated gents!

 

[ggunn]

You are overthinking this. For fewer than four #10 wires with 90 degree insulation in conduit:

If you derate for ambient temp of 100 degrees F (common around here) your 90 degree conditions of use ampacity is (0.91)(40A) = 36.4A

Your 75 degree ampacity derated for continuous use is (0.8)(35A) = 28.0A (and yes, you can either multiply the ampacity by 0.8 or the current by 1.25; the result is the same).

36.4A and 28.0A are both greater than (21.0A)(1.25) = 26.25A, so you are good to go.

I don't follow everything you are saying, but it appears that you may be derating twice.

 

[wwhitney]

(although ultimately 30A max allowed by NEC in end no matter what)

I think this is where you are mistaken. The above 30A limit applies to OCPD sizing for #10 Cu for most applications (but not motors). However, per 690.9(A), if the conductor has sufficient ampacity no OCPD is required for PV Source Circuits.

Then 690.8(A)(1)(1) tells you that the PV Source Circuit current is 1.25 * 2 * 10.5 = 26.25A.

And then 690.8(B) tells you the ampacity required: before adjustment and correction, at least 1.25 * 26.25 = 32.8A. [As far as I can see, this requirement should be deleted, as a 1.25 factor for continuous current is only relevant when non-100% rated OCPD is involved.] After adjustment and correction, at least 26.25A.

So if you are #10 Cu with 90C insulation, 32.8A < 40A, fine. And as long as your adjusted and corrected value, starting at 40A base ampacity, exceeds 26.25A, fine.

Note that this is a reason not to multiply ampacities by 0.8 for continuous circuits, as the 125% factor only applies to one of the two ampacity checks required.

Cheers, Wayne

 

[ggunn]

I Note that this is a reason not to multiply ampacities by 0.8 for continuous circuits, as the 125% factor only applies to one of the two ampacity checks required.

Six vs half a dozen, and some people get confused as to which 125% you are talking about with DC conductors; the 125% due to "excess insolation" applies to them both. I derate 90 degree ampacity for conditions of use and 75 degree ampacity for continuous use, and both have to exceed maximum current. That seems pretty straightforward to me, but YMMV.

 

[wwhitney]

I derate 90 degree ampacity for conditions of use and 75 degree ampacity for continuous use, and both have to exceed maximum current.

So what is the termination temperature rating for an MC4 connector (if that's what is typically used at the PV panel)? And how about typically at the inverter?

To my understanding, there's no need to apply a 125% continuous use factor at the terminations. So if the "maximum current" is 125% of Isc, then with 90C conductors and 75C terminations, the necessary requirements are:

1) 90C ampacity, no adjustment or correction factors, must not be less than 125% of maximum current [the rule I see no reason for.]
2) 90C ampacity, with adjustment and correction factors, must not be less than the maximum current.
3) 75C ampacity (terminations), with no adjustment or correction factors, must not be less than the maximum current.

Cheers, Wayne

 

[ggunn]

So what is the termination temperature rating for an MC4 connector (if that's what is typically used at the PV panel)? And how about typically at the inverter?

To my understanding, there's no need to apply a 125% continuous use factor at the terminations. So if the "maximum current" is 125% of Isc, then with 90C conductors and 75C terminations, the necessary requirements are:

1) 90C ampacity, no adjustment or correction factors, must not be less than 125% of maximum current [the rule I see no reason for.]
2) 90C ampacity, with adjustment and correction factors, must not be less than the maximum current.
3) 75C ampacity (terminations), with no adjustment or correction factors, must not be less than the maximum current.

Cheers, Wayne

We just assume all connectors are 75 degrees, but what you speak of is a non issue for us; we use minimum #10 everywhere on the roof and we don't do any 2:1 combining.

 

[Carultch]

-- Ambient TEMP: Average high Temps are just about 70-80F here, so no derate. (I recall it is average high temps that are to be used. Am i right on this? or record high Temp must be applied? Big difference!)

There is an informational note to use ASHRAE as a data source for this. It isn't the record high, it is a statistical measure of a typical high, such that only 2% of the time will exceed it. You can get the data for high and low temperatures from solarabcs.com.

There also is a square root formula in the same section of the NEC, that occasionally has an advantage in its results when compared to using the table for looking up temperature correction factors. You can either use the table, or the square root formula, whichever is convenient for you.

Now....can I further de-rate the 40A ampacity by 0.8 PV use factor - rather than multiply the Isc and continuous use factor by 1.25?

The 0.8 is just an algebraic manipulation of the 1.25 multiplication that the NEC specifies. There are multiple 1.25 factors that may apply, and they can compound, depending on which calculation you are doing.

Derate factors are generally less than 1, so they generally reduce the ampacity of a conductor. Derates apply to wires, but not to terminations, which is the primary reason why there is value in 90C wire despite 75C terminations. It gives you headroom for your derate calculations. I know the NEC no longer uses the term derate, but it is the easiest way to explain it.

There is a 4-step calculation to determine the required ampacity of a conductor for "wild PV". I.e. PV circuits directly from the panels, or combined strings of panels, with no power electronics in between.
1. Calculate 1.56*Isc. Size your OCPD and terminations (generally 75C) from this value. An informational note tells you 1.25*1.25 becomes 1.56.

The reason for the two 1.25 factors:
The enhancement factor for more than 1kW/m^2 of sunlight. This applies for "wild PV", but not for outputs of current-limited devices.
The continuous load factor, this applies to sizing terminations and OCPDs, but does not apply to step 2 below.

2. Calculate 1.25*Isc/total derate. Look in the 90C column for a size that meets or exceeds this value.

3. Calculate the derated wire ampacity. 90C column value * total derate. This must "round up" to the OCPD you are using, if you are using one. If not, this step doesn't apply. See 240.4(B) for specifics of what I mean by "round up".

4. Confirm that the termination ampacity (usually 75C) also "rounds up" to the OCPD you are using.

It is the norm of ratings that terminations are 75C and wires are 90C, so adjust the above accordingly as applies to you. It is rare for today's wire types to be rated less than 90C. You may see 90C terminations, but they are a lot less common, and you have a burden of proof to confirm 90C terminations on both sides. 110.14(C) spells out the default termination ratings. The 60C part of the rule is more academic than practical, because most equipment is listed for the 75C rating. You still have a burden of proof, but it is rare to not meet it, for the 75C rating at 100A and less.

Remember, if part of equipment, the entire product needs to be listed for 90C terminations to take credit for it, and not just lugs marked AL9CU. To answer the question about MC4 connectors, as stand-alone connectors, they are 90C rated, but if built-in to a product, you have to follow the instructions of that manufacturer to know whether to use the 90C or 75C rating.

 

[wwhitney]

The continuous load factor, this applies to sizing terminations and OCPDs, but does not apply to step 2 below.

Do you have a reference for the idea that the continuous load factor applies to terminations? I don't see that in 110.14. Principally relevant for PV source conductors that have no OCPD.

Also, do you see a physics justification in applying a continuous load factor for the minimum ampacity of PV source conductors without OCPD? I don't. I only see the continuous load factor as coming from OCPD limitations. Ampacity is already a continuous rating.

To answer the question about MC4 connectors, as stand-alone connectors, they are 90C rated, but if built-in to a product, you have to follow the instructions of that manufacturer to know whether to use the 90C or 75C rating.

So do PV panels typically come with +/- pigtails with MC4 connectors on them, and if so what temperature rating applies to those pigtails?

Cheers, Wayne

 

[macmikeman]

MACTIP OF THE DAY.... Run #8. This tip is not about safety, it's about customer satisfaction. They are looking for the cheapest monthly bill you can provide them.

 

[ggunn]

Once again we have proved that we can make a simple concept complicated. If we say we want to rearrange the deck furniture on the Titanic someone will say that we need a permit for that, and someone else will introduce calculations for the leverage necessary to pick up a deck chair, and someone else will say that we are not allowing for the wind resistance to moving a deck chair...

Engineers, gotta love 'em. Of course, I am one.

 

[jaggedben]

MACTIP OF THE DAY.... Run #8. This tip is not about safety, it's about customer satisfaction. They are looking for the cheapest monthly bill you can provide them.

Probably not actually worth the extra cost in wire though.
A few years back there was an article in Solar Pro that did the math and basically proved that bigger DC wire size doesn't pay for itself. It was focused on utility scale, but still.

 

[ggunn]

MACTIP OF THE DAY.... Run #8. This tip is not about safety, it's about customer satisfaction. They are looking for the cheapest monthly bill you can provide them.

Is that because of voltage drop in the DC home runs? I don't think that in most residential cases the difference in output would justify the increase in cost. In PV source circuits the Vd losses in #10 wire are usually pretty low already.

 

[wwhitney]

And then 690.8(B) tells you the ampacity required: before adjustment and correction, at least 1.25 * 26.25 = 32.8A. [As far as I can see, this requirement should be deleted, as a 1.25 factor for continuous current is only relevant when non-100% rated OCPD is involved.]

I overlooked the exception to 690.8(B)(1), which has been in the NEC since at least the 2017 NEC. It reads:

Exception: Circuits containing an assembly, together with its overcurrent device(s), that is listed for continuous operation at 100 percent of its rating shall be permitted to be used at 100 percent of its rating.

I assume the DC inputs on a string inverter are rated for continuous operation at 100% of their rating? If so, this exception eliminates the need for the 125% continuous factor on ampacity for PV Source Circuits without OCPD.

I am of course reading "together with its overcurrent devices(s)" as meaning "overcurrent device(s) if any," rather than indicating that there must be an OCPD. Because that is the logical requirement.

Cheers, Wayne

 

[Zee]

Well, quite a discussion was stirred up. Thank you. It always amazes me how seemingly simple .......and yet not........... wire sizing is.

 

[Fred B]

Once again we have proved that we can make a simple concept complicated. If we say we want to rearrange the deck furniture on the Titanic someone will say that we need a permit for that, and someone else will introduce calculations for the leverage necessary to pick up a deck chair, and someone else will say that we are not allowing for the wind resistance to moving a deck chair...

Engineers, gotta love 'em. Of course, I am one.

Don't forget the OSHA restrictions on max lifting for one person. Might need to people to move the furniture. Or labor laws that limit how long they can work without time off. Lol

 

[ggunn]

Well, quite a discussion was stirred up. Thank you. It always amazes me how seemingly simple .......and yet not........... wire sizing is.

It is simple, IMO. Try programming a PowerWall sometime.

 

[Carultch]

Do you have a reference for the idea that the continuous load factor applies to terminations? I don't see that in 110.14. Principally relevant for PV source conductors that have no OCPD.

Also, do you see a physics justification in applying a continuous load factor for the minimum ampacity of PV source conductors without OCPD? I don't. I only see the continuous load factor as coming from OCPD limitations. Ampacity is already a continuous rating.


So do PV panels typically come with +/- pigtails with MC4 connectors on them, and if so what temperature rating applies to those pigtails?

Cheers, Wayne

Read 690.8 and 690.9, that are the PV-specific sections. Part 1 and part 2 of my 4-step process, come directly from these sections.

As for the factory-installed MC4 connectors on PV panels, they usually are rated for 90C, but rarely (if ever) does this govern a design. Obviously, they have to have an ampacity rating for the product they are attached to, and unless you shine 2 suns of irradiance on them, you aren't going to overload those connectors from the ampacity.

Where you terminate your field-installed home runs on the other side of the wire, is more likely going to be what governs you to use the 75C rating.

 

[Carultch]

I overlooked the exception to 690.8(B)(1), which has been in the NEC since at least the 2017 NEC. It reads:

Exception: Circuits containing an assembly, together with its overcurrent device(s), that is listed for continuous operation at 100 percent of its rating shall be permitted to be used at 100 percent of its rating.

I assume the DC inputs on a string inverter are rated for continuous operation at 100% of their rating? If so, this exception eliminates the need for the 125% continuous factor on ampacity for PV Source Circuits without OCPD.

I am of course reading "together with its overcurrent devices(s)" as meaning "overcurrent device(s) if any," rather than indicating that there must be an OCPD. Because that is the logical requirement.

Cheers, Wayne

One place you would see this, is if you are using a SolarBOS recombiner for a central inverter, that has continuous duty rated breakers in it, and is listed as an assembly to allow you to take credit for this. I've only seen this where the OCPD's are breakers, and I haven't seen it where the OCPD's are fuses.

There is a burden of proof to see this in documentation, to allow you to forgo the 125% continuous load factor. If it isn't specified, you treat every fuse or breaker as a standard OCPD, and size the OCPD to 1.56*Isc for "wild PV", and also apply the continuous load factor to all other parts of the calculation where it applies. Likewise, if it isn't specified that terminations in equipment where OCPD doesn't apply aren't specifically rated for continuous duty, then you also apply the continuous load factor

 

[wwhitney]

Read 690.8 and 690.9, that are the PV-specific sections. Part 1 and part 2 of my 4-step process, come directly from these sections.

Sure, but you made a broader statement that the 125% continuous load factor applies to termination ampacity checks, and I don't believe that's the case.

One place you would see this, is if you are using a SolarBOS recombiner for a central inverter, that has continuous duty rated breakers in it, and is listed as an assembly to allow you to take credit for this.

Thanks for the real-world example with OCPD. But what about the more usual case on string inverters where there is no OCPD, the strings connect directly to the inverter. Wouldn't the exception apply there, as the inverter is rated for continuous current at 100% of its rating (I assume)?

Cheers, Wayne

 

[Carultch]

Sure, but you made a broader statement that the 125% continuous load factor applies to termination ampacity checks, and I don't believe that's the case.

690.8(B) applies to the ampacity of the conductor system in general, which includes the wire, the termination on device 1, and the termination on device 2. The first part is the compounded pair of 1.25 factors, which both the wire and termination ampacities would therefore need to meet or exceed. The second part only uses the 1.25 enhancement factor, and the derate calculations that apply to the wire insulation rating, but not to the termination rating.

Thanks for the real-world example with OCPD. But what about the more usual case on string inverters where there is no OCPD, the strings connect directly to the inverter. Wouldn't the exception apply there, as the inverter is rated for continuous current at 100% of its rating (I assume)?

Unless you have product instructions to the contrary, I would not count on the conductor sizing algorithm that takes credit for the inverter being meant for continuous operation, when you directly connect to it on a circuit without OCPD. The fact that an inverter is a continuous load by default, governs the 125% factor on its AC side for tying into a standard rated AC OCPD, but does not automatically exempt you from accounting for this factor on the DC side.

The exception does not specify "or lack thereof where not required". It only refers to 100% continuous duty rated equipment with overcurrent devices.