Brain teaser...Help with DC wire sizing @ 300 ft conduit run.

Status
Not open for further replies.
Hi All,

I'm trying to verify what my outsourced system designer has sent for us to submit.

Project characteristics:

58.5Kw - (180) Hyundai 325W ground mounted

4 inverters - (3) 12.5kWW Fronius Primo 240V single phase and (1) 15.0Kw

12.5kW - (4) Strings of 11 Paralleled to 1 output for reduced wiring - These paralleled circuits (3) will travel in 300 ft of PVC, what size conductor?
15.0Kw - (4) Strings of 12 Paralleled to 1 output for reduced wiring - This paralleled circuit (1) will travel in 300 ft of PVC, what size conductor?

Inverters are 300 ft away from the Array and DC combiner. Combiner Rack Output = (4pos) + (4neg) + Ground. Same Conduit for DC PV to Inverter Run (300ft.)


How to calculate and verify?

Here's what my designer has so far:

elect table.JPG wire segments.JPG

Can you find any discrepancies? Any help/advice is humbly welcomed!
 

jaggedben

Senior Member
Location
Northern California
Occupation
Solar and Energy Storage Installer
Well, we'd have to look up the module specs for starters, which you didn't provide.

I haven't done any math. At first glance nothing seems off to me, except that I noticed that the inverter model for Array #4 is not listed correctly in bottom part of the electrical table. (It's clearly supposed to say Primo 15.0-1, as it does further up.) Otherwise I'm just looking at the numbers and thinking they look normal, if the modules are 72 cell. That 597V for max system voltage on the one inverter is cutting it awfully close though, if the max is 600V. Choice of design temperature might be worth asking about, as well as whether the inverters have overvoltage protection. Curious why it's not a 1000V system but maybe there's a good reason.

Since you asked about wire sizing for a 300ft run I gather you are concerned about voltage drop? The numbers you provide don't address voltage drop and its not readily apparent whether your designer accounted for it. However unless Vmp is near the bottom of the inverters MPPT window, there's not necessarily any justification for increasing wire size to reduce voltage drop. A paper in Solar Pro magazine argued that in most cases you won't get your money back.

http://solarprofessional.com/articl...-selection-in-pv-designs?v=disable_pagination
 

Carultch

Senior Member
Location
Massachusetts
That 597V for max system voltage on the one inverter is cutting it awfully close though, if the max is 600V.

Borderline situations like this come up all the time in my office on this issue, where it would be preferable to do a certain string size that technically does exceed the 600V limit, but only by a few volts. And it would be a situation where a string size of 14 modules would be as elegant as it can be, while a string size of 13 or 12 would make it significantly more complicated.

Everyone knows that all equipment that matters is built to withstand significantly more than 600V, even though the rating on paper is only 600V. It's also more of a "Murphy proofing" rule than anything else. Because as soon as a significant amount of sunlight is present, the modules will heat up and the voltage will move away from this point. And if this voltage does cause current to flow, the point on the IV curve will move away from the open circuit condition to a condition of lower voltage anyway.

Choice of design temperature might be worth asking about, as well as whether the inverters have overvoltage protection.

Interesting how you say "choice of design temperature". Because it isn't like an engineer can simply choose for nature to not get any colder than a certain temperature. The most you can do is choose which nearby data site is most applicable to a location that is in between two sites. Such as choosing to model Concord MA as Boston instead of Worcester. I usually take a distance average for the in-between places.

I have wondered how the ASHRAE design temperatures are what matters. Because I know from experience that it has been both hotter and colder than both of those temperatures, in just about every place I've lived.
 

ggunn

PE (Electrical), NABCEP certified
Location
Austin, TX, USA
Occupation
Electrical Engineer - Photovoltaic Systems
Borderline situations like this come up all the time in my office on this issue, where it would be preferable to do a certain string size that technically does exceed the 600V limit, but only by a few volts. And it would be a situation where a string size of 14 modules would be as elegant as it can be, while a string size of 13 or 12 would make it significantly more complicated.

Everyone knows that all equipment that matters is built to withstand significantly more than 600V, even though the rating on paper is only 600V.
It's a gamble; some inverters have a "black box" register in their electronics that records the highest DC voltage it has seen and if it's over the published absolute high voltage limit you will have issues with the warranty if you should ever need to use it.

Personally, I never push that limit right up to the line with ASHRAE numbers; I know that it can get colder than the the ASHRAE Tmin.
 

jaggedben

Senior Member
Location
Northern California
Occupation
Solar and Energy Storage Installer
Borderline situations like this come up all the time in my office on this issue, where it would be preferable to do a certain string size that technically does exceed the 600V limit, but only by a few volts. And it would be a situation where a string size of 14 modules would be as elegant as it can be, while a string size of 13 or 12 would make it significantly more complicated.

Everyone knows that all equipment that matters is built to withstand significantly more than 600V, even though the rating on paper is only 600V. It's also more of a "Murphy proofing" rule than anything else. Because as soon as a significant amount of sunlight is present, the modules will heat up and the voltage will move away from this point. And if this voltage does cause current to flow, the point on the IV curve will move away from the open circuit condition to a condition of lower voltage anyway.

To me it's a bit less concern about the practical functioning of the system, and a bit more about whether I'm voiding the inverter warranty (primary concern) or the module UL listing (could be an AHJ issue). Warranty issues could really depend on the manufacturer and their fine print.

Interesting how you say "choice of design temperature". Because it isn't like an engineer can simply choose for nature to not get any colder than a certain temperature. The most you can do is choose which nearby data site is most applicable to a location that is in between two sites. Such as choosing to model Concord MA as Boston instead of Worcester. I usually take a distance average for the in-between places.

I have wondered how the ASHRAE design temperatures are what matters. Because I know from experience that it has been both hotter and colder than both of those temperatures, in just about every place I've lived.

Yeah, using actual extreme record temperatures is more conservative than ASHRAE or another average, and yet we're often told that using extreme records is not necessary. I guess it depends on how careful you want to be, and yeah, that does seem to be a choice. The design temps in the OPs attachments seem to be ASHRAE or something similar, as I know of nowhere in California that the record temps are 91F and above freezing.

Of course, nowadays one should account for the average high and record high temps to be more likely to be exceeded than the the low temps.
 

Carultch

Senior Member
Location
Massachusetts
Hi All,

I'm trying to verify what my outsourced system designer has sent for us to submit.

Project characteristics:

58.5Kw - (180) Hyundai 325W ground mounted

4 inverters - (3) 12.5kWW Fronius Primo 240V single phase and (1) 15.0Kw

12.5kW - (4) Strings of 11 Paralleled to 1 output for reduced wiring - These paralleled circuits (3) will travel in 300 ft of PVC, what size conductor?
15.0Kw - (4) Strings of 12 Paralleled to 1 output for reduced wiring - This paralleled circuit (1) will travel in 300 ft of PVC, what size conductor?

Inverters are 300 ft away from the Array and DC combiner. Combiner Rack Output = (4pos) + (4neg) + Ground. Same Conduit for DC PV to Inverter Run (300ft.)

Worst case scenario for voltage drop, is the strings of 11 in series.

Since all DC feeders have 4 strings in parallel, all DC feeders will have the same "minimum local size" as I like to call it. From what I've typically seen, #4 wire is realistic for this. I've looked at your numbers, and will let you know that you could install #6 if your terminations on both sides were listed and labeled otherwise for 75C, which is usually the case, but you do have to check. If either one isn't, you'd need #4.

Your DC voltage drop calculation is as follows:
%Vd = 2*L*r*Imp/1000/Vmp * 100%

where:
The 2-factor applies because current travels round trip on a DC or single phase circuit. If it is three phase AC, this would be sqrt(3) instead.
L is the 1-way length
r is the ohms/1000 ft that you look up for this particular wire size
Imp is the operating current at the optimal power point on the IV curve, multiplied by string qty in parallel
Vmp is the operating voltage at the optimal power point on the IV curve, multiplied by module qty in series


2*(300 ft) * (0.321 ohm/kft) * (34.4A) / ((416V) * (1000 ft/kft)) * 100% = 1.59% voltage drop, which is reasonable.

Remember that there is also voltage drop on the string wiring before you get to the combiner box. All voltage drop adds up, and must be within the limit you need to follow.

Had you used #6, you would have 2.44% on the DC feeder. Following the 2%/2%/3% rule, I wouldn't accept this.
 
Status
Not open for further replies.
Top