Equipment and AC/DC advice - 225-foot run to panel, ground mount

[bosco]

Hi,
I've found one forum posting on a ground mount system that was quite a distance from the grid-tie in, but it doesn't exactly answer my questions. Any advice is greatly appreciated.

I'm doing a DIY 15KW ground mount system ~225 feet from my grid-tie in. 42 panels, 4 strings, two inverters (each with 2 MPPT) Total system specs shown at the bottom of this post. And panel (360W) and inverter (7000-MTLP) specs attached.

Most ground mount systems I see in my geography have the inverter(s) at the array, but based on my internet research it seems as if the advice for "long runs" is to remain in DC to keep wiring size smaller and put the inverter(s) near the grid-tie in. Based on an online wire size calculator - if I keep the 225-foot run in DC, I'm at 6awg. If I put the inverters at the array, I need to jump up to a wopping 1/0 wire for the 225 run.

My specific questions:
1. Do I have a bust in my math for the AC/DC wire size difference noted above? Ma
2. Assuming you concur that my 225-foot run should be DC, I will have 8 wires (4 pos, 4 neg) plus ground. What equipment/boxes/subpanels/combiners, etc do I need at the array(s) and at the house before I tie into my main breaker box? Just looking for wiring and equipment strategy advice.

Thank you and looking forward to hearing ideas/advice.
Steve

2 X Growatt 7000MTLP-US

42 X AS-6M-360W (360W)

(1) String of 11 (MPPT #1) (1) String of 10 (MPPT #2) (INV #1) and (1) String of 11 (MPPT #1) (1) String of 10 (MPPT #2) (INV #2)

15,120W

13,475W

58A (Continuous Current) 72.50A (Max 125% Current)

Load-side tie-in to existing MSP, with main breaker derated to 150A

24,404kwhr

[Mike90250]

I don't download unknown files, but here's my thoughts based on best practices not code in your area.

1) do those inverters have dual MPPT inputs or merely 2 inputs into 1 MPPT ?

2) what DC voltage/Amps are in the DC run ?

3) unless local code requires it, I would not ground either DC line.

4) if you calculated #6 copper , I'd suggest # 4 aluminum, if the inverter terminals are CU/AL rated. (alum wire needs 2 sizes larger than copper, cheaper and easier to manage)

5) go a bit oversize on the required DC conduit, easier to pull.

6) Keep the a array strings within the inverter MPPT spec, not just the "operating" spec

7) Grounding.. The local inspector rules here. I'd make the panel frames and racking, it's own lightning ground, using the UFER principal if you are using concrete, otherwise, a ground rod at each corner of the array bonded with #4 copper. Do not embed copper into concrete.
Before you fill the trench, make sure the inspector does not need a bonding wire from the array to the main building ground rod.
a) the wire prevents voltage gradients in case of lightning strike,
b) the wire invites the lightning next to your main panel ( WOW ! )

[bcroe]

Those numbers sound a lot like mine. With 400VDC my losses over that distance are way less
than at similar distance and 240VAC. You can compensate with really big wire and hardware to
match, but it will be costly. If your strings are identical electrically and in regard to shading, and
the inverters allow it, you might save by running only 2 larger gauge pairs, with 2 strings in
parallel.

I would also ask if snow is any concern, there is a lot that can be done about snow on a ground
mount. Bruce Roe

[bosco]

Hi Mike and Bruce - thank you both.

I'll try to answer Mike's questions first (in the same order).

1. The inverter specs show two, independent MPPTs.
2. The 360W panels are (NOCT) -- Vmp of 34.8V and Imp of 7.65A. With 42 panels and the inverter specs, my max string can be 11 panels. So was thinking I'd do two arrays of 21 panels - each with a string of 11 and string of 10. 11 panels would be 383 volts DC (11x34.8) and the amps would be 7.65 since running in series, correct? But I think amps are 1.25*(7.65) - so 9.6, correct (125% rule)? If I'm supposed to be using the STC specs, Vmp is 38.1V and Imp is 9.45A.
3. I thought NEC required grounding at the array (grounding bar) and also grounding back to the my main house panel.
4. Interesting - I hadn't looked at aluminum.
5. Ok - thank you
6. ok - thank you
7. I thought I had to run one ground wire and attach to each panel...then to the ground rod (in soil) and also run this back to the house panel.

Follow up questions:
1) would I run my 4 strings (4 pos + 4 neg wires) into a combiner box at the array - then run the aluminum 4awg (one pos and one neg?) 225 feet to another combiner box at the inverter - and split the two 4awg wires into 4 pos and 4 neg to connect to the two inverters (each have 2 independent MPPT inputs)?
2) would you put a DC disconnect on each string at the array(s)? My AHJ has been allowing the AC disconnect to be the new breaker in the main house panel.

Bruce -
As mentioned above, I'll have 4 strings - two will have 11 panels and 2 will have 10 panels. Both arrays will be in an area with full sun - no shade. I'm totally open to advice on how to connect everything at the array and then how to connect everything at the house/inverters.
Regarding snow - we get dusted usually 1x per year but the AHJ does not require snow loading for my racking design.

thank you both!!

Steve

[bcroe]

Once you have combined pairs, you will not be separating them again. To get it down to 2
pairs, you could parallel the strings with 10 panels, the other pair to the strings with 11 panels,
10 and 11 connected in parallel would cost you efficiency.

Only one snow dusting, not to worry. With 20 heavy snow falls a year here, I have a sticky on
how to keep panels clear with little effort. It used to take me 30 hours a year to clear it. But
do have reasonable ground clearance. Arrays can be sloped, best not to move a lot of dirt
to level them.

Note your 383 VDC is loaded down, you must allow for open circuit which might be 480 VDC.
Then allow for an increase at your lowest temp, my July 360 rises to 420 in Feb.

With a long run, likely wire heavy enough for good efficiency will also be able to tolerate
1.25 multiplier which would be a fault condition. Check it out. Bruce Roe

[bosco]

Hi Bruce,
thank you for the input. My panel specs show Voc=46.7V Isc=9.89A at STC and Voc=43.1V Isc=8.01A at NOCT.

Can you "dumb it down" even further for me on the connections/boxes and transitions.

How do you go from 8 wires (4 pos and 4 neg) from the four strings down to the next step? With a box like this?

https://www.wholesalesolar.com/8500295/midnite-solar/disconnects/midnite-solar-transition-box-1000-v-gray-aluminum-3r-rainproof-enclosure

Am I going from the 8 12awg string wires to 4 (large) wires (2 pos and 2 neg) for the long run to the two inverters?

Hopefully my questions make sense. My connections/transitions and # of wires as I transition is where I'm clueless.

thank you
Steve

[Salts]

I didn't see anyone mention a rapid shutdown system. It was my understanding that if the inverter was attached to the home, it had to have a rapid shutdown installed. Rapid Shutdown is not cheap last time I looked.

Not sure if this matters or not.

[bcroe]

11 panels in series at standard conditions, open circuit, can develop 514 V. Then figure the increase
at your lowest temp, it must still be below the inverter max input V.

Check that your inverters can accept full power into a single MPPT input. If so you may simply tie
2 strings in parallel. More than 2 requires fusing for each. Plenty of MC4 Y or splitter connectors
are available. If the feed pair is too large for an MC4 connector some kind of combiner/junction
box is needed. My OUTBACK PV12 COMBINER box would handle all circuits in separate halves.
Its twice your required capacity, also supplies fuses. Bringing all those circuits over that distance
without the access a box provides, might turn out to be difficult to deal with.

Once combined run the 4 wires of proper gauge for efficiency and current capacity, either in
conduit (like mine) or direct burial wire. Those wires have no outdoor exposure here. Bruce Roe

[Ampster]

I didn't see anyone mention a rapid shutdown system. It was my understanding that if the inverter was attached to the home, it had to have a rapid shutdown installed. Rapid Shutdown is not cheap last time I looked.

Not sure if this matters or not.

Good question. In California it depends on where the panels are located. I have some panels on my roof and another system on a patio cover. My local building department says I dont need RSD on the patio cover since no Firemen will ever be on that patio cover during a fire.

[bosco]

Thanks guys

I'm not sure how to define "full power" (STC versus NOCT) to determine if I can use a single MPPT or if I need both. If I need both, my DC wires for the long run would double if I'm understanding correctly.

The specs for the inverter max show power input per MPPT as 7000/4800W.
Two strings (21 panels) are 360Wx21 = 7560W (STC) and 266Wx21=5586W (NOCT). STC is not realistic, correct?

The rapid shutdown will be the new breaker in my house panel per my AHJ. I thought I would need a separate, stand-alone switch, but I guess not.

thank you again
Steve