Wire size calculation check

[difranco999]

I'm looking at placing solar in an open field that's about 300' from our house. The panels will be on a ground mount, and there will be an insulated shed to house the inverter and batteries. Currently looking at using a EG4 1200xp (spec list for quick reference).

In the diagram below you can see a rough idea of what I'm envisioning.

My primary question right now is if my AC wire size calculations are correct. Also interested in general feedback as to whether this design makes sense or if there's a better way.

I used this wire size calculator... Plugged in 300' @ 50 amps w/ 3% voltage drop, it reported back 3awg.

For the ground wire, I used this calculator which reported 10awag.

So in summary, I would need 5 separate 500' runs of 3awg wire:

Wire 1: L1 from house to inverter
Wire 2: L2 from house to inverter
Wire 3: Neutral from house to inverter
Wire 4: L1 from inverter to house
Wire 5: L2 from inverter to house

And 1 500' run of 10awg wire for ground

Is this accurate?

 

[sunshine_eggo]

Why in god's name would you locate the inverter and batteries 300' from the house?

Why not locate them at the house and then simply run high voltage DC the 300'?

 

[difranco999]

Why in god's name would you locate the inverter and batteries 300' from the house?

That was the initial plan. Then I was watching this video.

One of the commenters wrote: "60V and a 250' run hurts my soul." There were a few other comments along this line. (I'm not seeing them currently so it's possible the video author has removed them).

This lead me to researching about AC vs. DC wire when it comes to long runs and I read the following:

Wire size/cost
• DC wiring over long distances requires thicker, more expensive wires to prevent voltage drop.
• AC wiring can use smaller, more affordable wires, making it more cost-effective for long runs.

Safety:
• Long DC runs carry safety risks, such as higher fire hazards and the potential for arcing.
• AC wiring is generally safer and easier to manage over long distances.

But I guess in my proposed setup, I'd have to have more AC runs to/from the inverter than I would if I did the run in DC, so the cost savings is a moot point?

Apologies for the ignorance - doing hours of research a day, but obviously still a lot to learn.

 

[spiker611]

3 AWG wire is ~$2.58/ft at home depot near me. So you're looking at 1500 feet at $3870 total for just wire, which is bonkers.

I think your concern about DC wire is misplaced.

> DC wiring over long distances requires thicker, more expensive wires to prevent voltage drop.

Not really, you can run your PV (DC) wires at a much higher voltage. With higher voltage you run less current, requiring a thinner wire.

> AC wiring can use smaller, more affordable wires, making it more cost-effective for long runs.

There's not much difference between DC and AC except for what voltage you run. AC you're limited to 120/240V, DC can be up to 480V for your inverter.

> Long DC runs carry safety risks, such as higher fire hazards and the potential for arcing.

Yes there are safety concerns but they can be mitigated with proper installation and equipment.

I'd recommend you look into running the PV (DC) wires from your array to the inverter placed next to your load panel. You haven't shared the array configuration so we can't help you there.

 

[zanydroid]

Few major issues:
- Wrong voltage used for DC voltage drop calculations. 60V is less than two panels in serial. Only hipsters use 60V high current design with lots of panels in parallel, the mainstream is many panels in series with maybe two series strings in parallel. In fact I’m pretty sure you need north of 100V for that MPPT
- 12000XP is only IP20. Has to be indoors. Good luck building a shed out at the array
- DC voltage drop is less of a big deal than AC for operating equipment and if you ever convert to grid tie. The inverter MPPT will step up the DC voltage unless you go below the minimums. You need enough margin for voltage drop and for temperature voltage drop of solar panels

YES, DC needs to be respected. You will need an IMO disconnect at the array and at your inverter so you can easily isolate the system. I prefer DC in hard conduit. When it enters home it has to be in metallic conduit (FMC or EMT as minimum depending on protection level needed in location, which is higher than for AC under electrical code). PV DC is allowed to be direct buried but that scares the crap out of me.

If anything AC needs bigger wire than DC, if at frequency where AC skin effect matters (irrelevant for solar)

 

[LakeHouse]

Wire size/cost
• DC wiring over long distances requires thicker, more expensive wires to prevent voltage drop.
• AC wiring can use smaller, more affordable wires, making it more cost-effective for long runs.

This looks like it was written by someone for whom 'DC' means 12V RV or car batteries. Yeah, it's expensive to move power that way.
But that's not what you're trying to do. You want to move power from your array to your inverter, and you aren't going to do that at 12V, you're going to do it at hundreds of volts.
As others have said, locate the inverter close to the loads. Having the panels 300' away isn't a big deal. Size wire for Isc of your setup, and feel free to completely ignore voltage drop for the PV run. It just doesn't matter for PV; if your array puts out 300V or 270V, your SCC is fine with either, right? So if your array is putting out 300V, and 30V is lost as voltage drop in 600' of wire, your inverter sees 270V... And doesn't care at all.
Granted, the power your inverter can pull from the array will be less, but only when it's creating a lot of power since voltage drop depends only on current. From a cost perspective, it's almost certainly cheaper to add more panels than to upsize the PV wire to limit voltage drop.

 

[sunshine_eggo]

AC/DC doesn't matter.

Voltage, distance, wire gage, and current are what determines losses. To minimize losses:

Maximize voltage.
Minimize current.
Maximize gauge.
Minimize distance.

Your MPPT can go up to 450V (adjusting for temperature) and will operate at close to 300V. That's higher than AC, and you only need ONE path rather than two, i.e., panels to AiO vs. grid to AiO and back to house.

288V, 18A, 300', 8awg:

Voltage Drop Calculator

This free voltage drop calculator estimates the voltage drop of an electrical circuit based on the wire size, distance, and anticipated load current.

www.calculator.net

2.36% voltage drop. Luscious.

 

[sunshine_eggo]

@difranco999

Please purge everything about that video from your mind. it's simply a commercial for the Aker F3800 and has no value to you.

 

[zanydroid]

@difranco999

Please purge everything about that video from your mind. it's simply a commercial for the Aker F3800 and has no value to you.

A shame, since Benjamin has some pretty good videos and is IIRC pretty careful about code

 

[PVGeezer]

Don't ignore voltage drop. But don't be afraid of it either. For sure you can just size the wire for an ampacity of 1.56xIsc which is code minimum and be done.

But if you really want to know the ideal wire size what you want to do is minimize the cost of the wire + the value of the energy lost due to voltage drop in the wire.

The cost of the wire is easy. Cost up your wire runs for the code minimum wire. Then do it again for the next wire size up.

Next, calculate the percent voltage drop for each of the two wire sizes. Use 2/3 of Imp for this calc. Why 2/3 of Imp? Because the weighted average operating current of a typical PV array, taking into account solar incidence angle and cloudiness, is around there, give or take. Multiply the percent voltage drop for each wire size by your total budget for the system. That's the lifetime value of the electricity lost in the wire.

Add that to the cost of the wire for each size. You're looking for the lowest total cost of the two combined. If it's the larger wire size, repeat for the next size up. Stop when the sum of the costs starts to go up. Use the wire size that gives you the lowest cost.

In my case the code minimum was 12awg but cost was minimized with 10awg, so that's what I used.

If you look at the math here you can see that the result is going to be the same regardless of wire length, because both terms in the equation are linear with wire length.

 

[sunshine_eggo]

A shame, since Benjamin has some pretty good videos and is IIRC pretty careful about code

Never heard of him, and I don't care. One should not use a sponsored video for advice... especially not the comments section... well...unless the comment tells you not to use sponsored product advice.

 

[zanydroid]

Another thing to consider with wire is future proofing (taking into account 1.56 Isc multiplier). Esp if pulling spare wire for expansion.

#12 is quite rough with the wider cells nowadays. It would likely not be up to code with the higher output ones.

#10 can handle any panel in 1P, but 2P starts getting rough. And if you go above 3 circuits in a conduit the extra derate makes 2P definitely non viable

#8 works well for 2P but in many cases with derate it won’t work for 3P so it’s not future proof either

For both the #10 and #12, you can potentially get a tiny bit more headroom by using code edge cases, IF that headroom is not needed for derate

I think with the max size IMO you might be able to get #4 aluminum in (didn’t see if it’s rated for aluminum)

 

[zanydroid]

Never heard of him, and I don't care. One should not use a sponsored video for advice... especially not the comments section... well...unless the comment tells you not to use sponsored product advice.

He hasn’t shilled enough to overcome the variety of good builds I’ve seen on his old videos. Unlike Risinger, who shills a lot and talks lot like a manager (pretty sure he has a hustle as a builder/GC for fancy houses so it makes sense)

 

[sunshine_eggo]

He hasn’t shilled enough to overcome the variety of good builds I’ve seen on his old videos. Unlike Risinger, who shills a lot and talks lot like a manager (pretty sure he has a hustle as a builder/GC for fancy houses so it makes sense)

You're making me proud of my YouTube habits. I don't know who Risinger is either.

 

[wheelman55]

OP. Are you going to put your wires in pvc conduit? Or direct burial?

If in conduit, consider using 2 inch pipe with long radius 90 degree bends and bury it deep. We buried ours 4 feet. our total run is just under 300 feet. 150 feet DC (high voltage/low amp) and 135 feet AC. We used 10 awg for the dc, and 2 awg pure copper for the AC. Use a voltage calculator. Do your own research and don’t rely purely on what you hear or see on the internet.

Good advice further up.

Check out the recent grounding thread that Will started sometime this past week.

Good luck with your project!

 

[zanydroid]

Instead of #2 copper I would much rather just use aluminum, unless the PVC is size constrained. And even then, my POCO requires aluminum and upsizing PVC to compensate (and their rules govern infrastructure that’s supposed to stand for decades).

There are also special code provisions governing mixing AC and solar DC in the same conduit. They are restrictive (meaning can burn you when you want to reconfigure) well as rather confusing to understand. I would suggest having two conduits if you have both DC and AC.

You can trust what you see on the internet if it’s from one of the handful of solar installers, educators that run $$$ seminars, or NEC.

 

[robbob2112]

Instead of #2 copper I would much rather just use aluminum, unless the PVC is size constrained. And even then, my POCO requires aluminum and upsizing PVC to compensate (and their rules govern infrastructure that’s supposed to stand for decades).

There are also special code provisions governing mixing AC and solar DC in the same conduit. They are restrictive (meaning can burn you when you want to reconfigure) well as rather confusing to understand. I would suggest having two conduits if you have both DC and AC.

You can trust what you see on the internet if it’s from one of the handful of solar installers, educators that run $$$ seminars, or NEC.

Huh? I can see them wanting aluminum at the disconnect, maybe if they are worried about galvanic corrosion.

Aluminum generally requires upsizing at least 2 sizes.

 

[zanydroid]

Huh? I can see them wanting aluminum at the disconnect, maybe if they are worried about galvanic corrosion.

Aluminum generally requires upsizing at least 2 sizes.

Save money. But not as much as going with DC.

If you’re going all the way up to #2 copper. And budgeting for it/using equipment that can accept that size. You can just use 2 size up aluminum, as you said. Usually the terminals at that size can handle it

For DC it’s a harder sell. I think you can maybe justify going to #4 aluminum if you really wanted to future proof to 2P of monster panels in 2030. It would still be able to handle, with 4 circuits in a conduit derate, 2P of

75*.7/1.56/2 =16.8A ISC

 

[robbob2112]

Maybe I have more money than sense... I would do copper myself.

 

[Hedges]

I've always used copper, but might use some huge aluminum for a future battery setup.

For OP, if you do run PV DC the long distance as we suggest, don't worry much about limiting to 3% drop.
You'll only see maximum power under ideal conditions, and if you lose 10% or 15% on occasion that could be OK if you save enough on wire.

But do plan for future expansion as suggested.

Consider a PV array with SE and SW facing PV strings connected in parallel; more hours, lower peak current.

And consider inverters that work up to 600V or even 1000V. (less likely to find AIO at 1kV; some GT PV like Fronius are.)

 

[difranco999]

Thanks everyone for the input and discussion. Very helpful! Will post again soon with updated design.