diy solar

diy solar

10.6KW bifacial off grid ground array powering multiple locations 500 feet away

AlaskanNoob

Solar Enthusiast
Joined
Feb 20, 2021
Messages
900
Hello all,

I had previously been planning on installing PV at our cabin that we just built last summer as detailed in this thread. But thanks to the many great inputs from people we got in that discussion, we've changed our original vision a lot.

Now we are planning on using LG440N2T-E6 bifacial panels in two different strings of 6s2p, wired into a nearby Victron 450/200 SCC, connected to a nearby Victron Multiplus II, charging a nearby 48V battery bank with a nearby Honda EU7000 generator plugged in to backup charge the battery bank as required.

The Victron equipment and battery bank will be buried in a root cellar structure with a small cabin above it, to keep the batteries from getting too cold in the winter and to keep from having to use PV to heat them.

Worst case, the PVWatts calculator shows this ground mount in the new location should produce 12KW of power on the darkest day (not accounting for any gains from the panels being bifacial on a snowy day). That will work great for our energy needs.

But this new location that sees the sun year round is 500 feet away from our cabin, and 600 feet from our guest yurt, and 600 feet away from where we hope to build another guest cabin. It's only 30 feet away from a nearby structure. We'd like to trench wires so that we can provide power to all these various locations.

Questions for the charitable:

1. Am I correct that connecting the structures to the power means trenching electrical wires from each structure to the Victron Multiplus II?

2. I am vaguely familiar with this wire size calculator, but I'm not sure how to calculate the voltage and amps that will travel through the wires to the several structures. I assume the Multplus II will be providing the power to the wires, and if so, what voltage and amperage does it supply power at?

3. I assume the wires will connect to a breaker box at each structure, but is there any other component required between the Multiplus II and the wire to the structures?

4. Does this plan pass the smell test?

Many thanks for any continued help with our evolving system design. We're hoping to order all this equipment within the next month so we can be ready to install it this spring. Got lots of trenching to do...
 

Attachments

  • Solar Ground Array v1.001.jpeg
    Solar Ground Array v1.001.jpeg
    95.8 KB · Views: 18
Last edited:
Your panels put out a Vmp of 38V, at 9.2A, so here is the calculator results with one solar string.
1641748865148.png
If you wire the two solar strings in parallel to run down one pair of wires, you are doubling the amperage, and voltage drop does not reach an acceptable level to you reach 4 gauge.
1641748998990.png

Assuming your Victron unit has a Voc limit of 450, at -40F that limit drops to 360Voc. If you can add additional panels to each string and raise the voltage up into the low 300s, you get less voltage drop.
 
Your panels put out a Vmp of 38V, at 9.2A, so here is the calculator results with one solar string.
View attachment 78988
If you wire the two solar strings in parallel to run down one pair of wires, you are doubling the amperage, and voltage drop does not reach an acceptable level to you reach 4 gauge.
View attachment 78990

Assuming your Victron unit has a Voc limit of 450, at -40F that limit drops to 360Voc. If you can add additional panels to each string and raise the voltage up into the low 300s, you get less voltage drop.
Thank you for that.

The distance from the PV to the SCC and Multiplus II and 48V battery bank is small. The distance from the SCC, Multiplus II, and PV to the various structures that needs the power is large (500-600 feet).

I'm still confused on what is supplying power to the 500 foot trenched wire. My thought was the PV arrays go to the SCC which goes to the Multiplus II and that would supply the AC power to the 500 foot or so wire to the structure. Is that correct? And if so, what volts and amps does it use to supply power to the line?
 

Attachments

  • Solar Ground Array v1.001.jpeg
    Solar Ground Array v1.001.jpeg
    95.8 KB · Views: 4
You run the highest voltages across the longest wire run, if at all possible. For you, that's the wire from the solar array to the SCC.
 
You run the highest voltages across the longest wire run, if at all possible. For you, that's the wire from the solar array to the SCC.
But I don't think the solar array to the SCC is the longest run. It will be ten feet or so. The longest wire run will be from the SCC, Multiplus II, and battery bank to the structures I'm trying to power. Those distances will be 500-600 feet.

What I'm envisioning is having my solar array and all components in the good sun spot I have. The structures that will need the power are 500 to 600 feet away from that spot in various directions.
 
I assume the wires will connect to a breaker box at each structure, but is there any other component required between the Multiplus II and the wire to the structures?

You should get an electrician involved if you have to ask - since you need to get the grounding and bonding details right.

That said, the general idea would be that you would have a "main" breaker box at the building near the Multiplus which has breakers for each circuit in that building plus breakers for the circuits going to your other buildings. Each of those buildings would have a "sub" breaker box which has breakers for each of the circuits in that building.

I assume the Multplus II will be providing the power to the wires, and if so, what voltage and amperage does it supply power at?

Yes, through the main breaker box - if you use a pattern like I described above.

The voltage can be set on the Multiplus within the range that it supports - e.g. something between 115 - 125v for the 120v nominal version.

The amperage will depend on the version - you need to check the specifications.
I am vaguely familiar with this wire size calculator, but I'm not sure how to calculate the voltage and amps that will travel through the wires.

That calculator is for low voltage DC power. For the AC side, you need to use something like this one. To determine the amps of the load you would need to add up the amperage of all the loads that might be turned on at a single time - perhaps with a bit of buffer.

Am I correct that connecting the structures to the power means trenching electrical wires from each structure to the Victron Multiplus II?

Yes, power won't get there magically. :) One thing to consider is that in some cases it makes sense to have one sub breaker box downstream of another, so that could potentially reduce the total amount of cable/ditching - at the cost of needing to size up some of the upstream runs of cable.
 
What I'm envisioning is having my solar array and all components in the good sun spot I have. The structures that will need the power are 500 to 600 feet away from that spot in various directions.
Do not do this.
 
You should get an electrician involved if you have to ask - since you need to get the grounding and bonding details right.

That said, the general idea would be that you would have a "main" breaker box at the building near the Multiplus which has breakers for each circuit in that building plus breakers for the circuits going to your other buildings. Each of those buildings would have a "sub" breaker box which has breakers for each of the circuits in that building.



Yes, through the main breaker box - if you use a pattern like I described above.

The voltage can be set on the Multiplus within the range that it supports - e.g. something between 115 - 125v for the 120v nominal version.

The amperage will depend on the version - you need to check the specifications.


That calculator is for low voltage DC power. For the AC side, you need to use something like this one. To determine the amps of the load you would need to add up the amperage of all the loads that might be turned on at a single time - perhaps with a bit of buffer.



Yes, power won't get there magically. :) One thing to consider is that in some cases it makes sense to have one sub breaker box downstream of another, so that could potentially reduce the total amount of cable/ditching - at the cost of needing to size up some of the upstream runs of cable.

Okay, this makes sense to me. I need to dig into the Multiplus II specs to see what volts and amps it will supply AC power at on those long lines so I can determine the needed wire size.

The calculator I was using was for DC power which is what caused the confusion. The DC power runs will be ten feet or so and thus shouldn't require crazy huge wire. But I need to figure out the AC power sizing. Many thanks an my apologies to all for the confusion!
 
The run is too long. The voltage drop to great. The wire to expensive. Maybe run high voltage DC to the structure 500 feet away (where the inverter will live).

Edit-sorry for the blunt and rushed response. I think you will find a 500 foot AC run to be too much. The higher voltage your long run, the lower the amps. I was thinking a high voltage DC run to your living structures might be much more affordable. As high as you can safely manage and your inverter can handle.

You will know more after you price the copper for the ac run.
 
Last edited:
Do not do this.
Why not?

I have a PV array, and wouldn't I want a short run from it to the SCC and the Multiplus II and the battery bank? Where the Multiplus could then convert it to AC power and deliver it the long distances to the various structures that need the power?
 
The run is too long. The voltage drop to great. The wire to expensive. Maybe run high voltage DC to the structure 500 feet away (where the inverter will live).
This isn't jiving with other advice I got on a different thread where I initially brought up having multiple inverters at the various structures and power delivered that way.

I need to look at the AC wire sizing calculator and then maybe I will better understand where you're coming from.
 
Because of basic physics.
I have a PV array, and wouldn't I want a short run from it to the SCC and the Multiplus II and the battery bank? Where the Multiplus could then convert it to AC power and deliver it the long distances to the various structures that need the power?
No, because it's cheaper (cable-wise) to do a long high-voltage / low amp DC run than a lower voltage / high amp AC run.

There's also the topic of keeping your batteries warm enough, which requires them to be within a heated enclosure. This is easily done inside a building. Ditto your SCC and other components don't like temperature extremes.

If you don't understand these concepts, you should find a contractor who does.
 
Thank you for that.

The distance from the PV to the SCC and Multiplus II and 48V battery bank is small. The distance from the SCC, Multiplus II, and PV to the various structures that needs the power is large (500-600 feet).

I'm still confused on what is supplying power to the 500 foot trenched wire. My thought was the PV arrays go to the SCC which goes to the Multiplus II and that would supply the AC power to the 500 foot or so wire to the structure. Is that correct? And if so, what volts and amps does it use to supply power to the line?
Consider using a 480v inverter with transformers at your point of use to drop it back down to 120v. Your complexity (and costs) will go up significantly but it’s probably cheaper than running cable that costs $30/ft.
 
@Wibla You should check out OP's related thread. They were trying to figure out load sharing of the solar between SCC' at multiple buildings which would at best be needlessly complex.

Overall, this won't be a cheap project any way they do it, but it seems simplest and most conventional to centralize their panels/SCC/inverter infrastructure given their constraints.
 
Because of basic physics.

No, because it's cheaper (cable-wise) to do a long high-voltage / low amp DC run than a lower voltage / high amp AC run.

There's also the topic of keeping your batteries warm enough, which requires them to be within a heated enclosure. This is easily done inside a building. Ditto your SCC and other components don't like temperature extremes.

If you don't understand these concepts, you should find a contractor who does.
I'm admittedly light on the physics knowledge. One way being potentially cheaper makes sense to me.

Again I need to look at the AC wire sizing and the volts and amps that the Multiplus II will supply so that I can figure out the size of the AC wiring and thus the cost. Then I'll better be able to understand your recommendation of four separate SCCs and three DC wire runs versus what I am envisioning with a single SCC and three AC wire runs to see which one costs less.

The batteries I may have failed to mention, will be buried in a root cellar underground structure near the PV, single SCC, and Multiplus II as I'm envisioning it with a small wood structure overtop so keeping them warm shouldn't be an issue.

I agree I should find a contractor who understands these things and hopefully I'll be better able to discuss with them and ensure I get the right contractor after these discussions on this DIY forum. I appreciate your efforts to help me along that path.
 
Last edited:
If you're really in Alaska I suspect you're being awfully optimistic in your minimum output predictions. It looks like you're putting up about 12 kw of panels. Right now, here in Oregon, for the last two weeks I've been seeing 2-3 kWh per day of production from my 10 kw of panels, with a couple of days less than 1 kWh per day. That's the combination of low sun angle and rain. It may be that you don't get cloudy days in the winter where you are, but in Alaska that would generally mean you're even further north, making the sun angle worse.

Just offering my experience, I've had to upsize my battery system significantly to plan for a week without significant solar output, but you'll have a generator if you need it.

One tip that might save you a few bucks with the long run you have ( I have 250' or so). I use UDF direct burial rated cable (the grey stuff) and use 10/3 with ground for my panels. The 10/3 plus ground gives me 2 pairs of wires, each good for 35 amps, so I bond them together and can handle 70 amps combined. I found it was a lot cheaper than going with the larger sized wire and it's rated for direct burial.
 
If you're really in Alaska I suspect you're being awfully optimistic in your minimum output predictions. It looks like you're putting up about 12 kw of panels. Right now, here in Oregon, for the last two weeks I've been seeing 2-3 kWh per day of production from my 10 kw of panels, with a couple of days less than 1 kWh per day. That's the combination of low sun angle and rain. It may be that you don't get cloudy days in the winter where you are, but in Alaska that would generally mean you're even further north, making the sun angle worse.

Just offering my experience, I've had to upsize my battery system significantly to plan for a week without significant solar output, but you'll have a generator if you need it.

One tip that might save you a few bucks with the long run you have ( I have 250' or so). I use UDF direct burial rated cable (the grey stuff) and use 10/3 with ground for my panels. The 10/3 plus ground gives me 2 pairs of wires, each good for 35 amps, so I bond them together and can handle 70 amps combined. I found it was a lot cheaper than going with the larger sized wire and it's rated for direct burial.
Thanks for that wiring info, I definitely need to look into that for sure before I do this trenching.

As far as output predictions for my spot, you may well be right. I may have to upsize the PV in the future. I'm just going off of what PVWatts spit out for my location using whatever weather model it uses with the panels straight up and down (to keep snow off) and facing directly south. We definitely get fog because we're 1000 feet up and clouds roll over us not infrequently. I've confirmed visually that on 21 Dec the spot for the PV ground array will get 4 hours of direct sun with no terrain or tree shading. Plus we'll be using bifacial panels so the reflection from the snow should be a bonus and apparently they do better than most panels on foggy days. Gotta start somewhere and it will be interesting to see what it really produces. We are planning to have 20KW of lithium in a heated structure with the system and a Honda EU7000 tied in to charge the bank as required.
 
Last edited:
That calculator is for low voltage DC power. For the AC side, you need to use something like this one. To determine the amps of the load you would need to add up the amperage of all the loads that might be turned on at a single time - perhaps with a bit of buffer.

The wife and I went to this AC calculator and came up with 1 AWG copper buried wire using a cable run of 500 feet, voltage of 230 (which is what the larger Victron Multiplus II produces), and 43 amps, which assumes a 10kw draw at our cabin. We plugged in just 500 feet wire run although somebody had told me before that when using the DC calculator to double the length for "there and back" so I plugged in 1000. We didn't double in this calculation that produced 1 AWG, not sure if we should have or not. Priced it out at around $2000 for the wire which is about what we priced out for using the DC calculator. So I'm not seeing larger costs associated with running AC through the wire vs DC, but then again if I should have doubled the distance for there and back then that would change dramatically I'm sure.

The two things that have me leaning toward AC power currently are firstly, saving on the cost and complexity of multiple SCCs, multiple battery banks, and multiple inverters at each place we're trying to power and secondly, some discussion I've seen about AC power being safer in case there is arcing. I can't pretend to understand the safety aspect but if one power source is less likely to start a fire in our heavily forested area, I'm all for that.

If the Victron can supply 230 volts +-3 then I *think* that should be sufficient for our usage. Our biggest draw will be a ground source heat pump and we've seen some models that operate in that range and lower, so we think we can make that work without having to use a transformer but perhaps I'm wrong on that.

Thanks again for all who have tried to help a dumb kid understand better what smart kids already know.
 
Consider using a 480v inverter with transformers at your point of use to drop it back down to 120v. Your complexity (and costs) will go up significantly but it’s probably cheaper than running cable that costs $30/ft.
We must have done our wire sizing incorrectly if you're showing the wire would cost us $15,000 to go 500 feet. That would definitely be a huge reason to go to the DC run with multiple inverters, batteries, and charge controllers OR to go with the transformer.

When I double the 500 feet in the calculator to 1000 feet (which I still don't understand if I need to do this or why, since it's a 500 foot run) then it comes up with 4/0 cable which is around the $30 a foot you mentioned. So I assume not doubling that 500 foot run was my error.
 
Last edited:
Back
Top