How many feet of wire for a 48v solar system? Is 350ft too far? What gauge of wire would work?

[curiouscarbon]

thanks for sharing your experience

i was quoting victron official documentation regarding 5V minimum

your perspective is appreciated?

i too try to design to have PV voltage significantly higher than 5V over pack voltage

 

[Brettw]

Let's say you wire 10 standard 250W grid-tie panels in series at your designated location. Here's the voltage drop for 10 gauge wire running at 300VDC. I used this voltage drop calculator. https://www.calculator.net/voltage-drop-calculator.html

With 10 panels in series, you are getting less than 2% loss. With only 8 panels in series, it's still less than 3% loss. What you can do with the calculator is play with the parameters and see what you get.

You may find it's more cost-effective to just add a panel or two to a series string, rather than purchasing thicker, more expensive wire. With a 450Voc limit, you have lots of flexibility. Just make sure you account for your winter's lowest temperature. https://www.midnitesolar.com/sizingTool/index.php Midnight's string calculator predicts that this ten panel string will reach 450Voc at about 30 degreesF below zero. You may be happy with just a string of 8 or nine panels.
View attachment 72976

View attachment 72979

Utilizing an eight panel string of the panels you mention above, you'll reach 450Voc at about zero F. What are your winter lows like?
View attachment 72981

Coldest Temps by average are below 0°f at night for 1 day per year. 3 months of below 32°f at night. These are statistical averages for my area.

 

[Brettw]

Let's say you wire 10 standard 250W grid-tie panels in series at your designated location. Here's the voltage drop for 10 gauge wire running at 300VDC. I used this voltage drop calculator. https://www.calculator.net/voltage-drop-calculator.html

With 10 panels in series, you are getting less than 2% loss. With only 8 panels in series, it's still less than 3% loss. What you can do with the calculator is play with the parameters and see what you get.

You may find it's more cost-effective to just add a panel or two to a series string, rather than purchasing thicker, more expensive wire. With a 450Voc limit, you have lots of flexibility. Just make sure you account for your winter's lowest temperature. https://www.midnitesolar.com/sizingTool/index.php Midnight's string calculator predicts that this ten panel string will reach 450Voc at about 30 degreesF below zero. You may be happy with just a string of 8 or nine panels.
View attachment 72976

View attachment 72979

Utilizing an eight panel string of the panels you mention above, you'll reach 450Voc at about zero F. What are your winter lows like?
View attachment 72981

I'm thinking I will need 12 to 16 panels, do you think this will work?

 

[Hedges]

Coldest Temps by average are below 0°f at night for 1 day per year. 3 months of below 32°f at night. These are statistical averages for my area.

Use lowest temperature ever recorded in history for your location.

Averages can be used for performance, but absolutes to avoid damage.

 

[MichaelK]

I'm thinking I will need 12 to 16 panels, do you think this will work?

With 16 panels, no. 12, and 14 would be better for the panel you mention. Looking at the second solar string prediction in my post above, you'll see that 8 panels in series will reach ~450Voc around zero F. So, two strings of 8 (written 8S2P) might fry your electronics. Since your winter lows dip below zero, a string of 8 is a no-no. Two parallel strings of 6 or 7 would give you a high, but still safe voltage that won't exceed your electronic's limits. You could even wire 21 panels in 7S3P, because each parallel string would be producing up to 9A of power, so 9A + 9A + 9A =27A. Use the link I provided and plug in your parameters yourself. Playing with the calculator will give you lots of insight into your design.

But, can you afford that much copper wire? That would require 2100 feet total of 10 gauge wire, which is going to cost a pretty penny. But, even just 14 of those panels would be >5500W of power, and you haven't mentioned anything yet about what you actually want to power.

Thoughtful design starts out with an itemized list of what you want to power, and then you select the components to accomplish that goal. Just one parallel string would likely power a house with near normal consumption patterns. Two would be 5500+W. What do you want to power that needs more then 5000W of solar? Making a list first is the single best thing you can do at this point in your design.

 

[Hedges]

Two strings of different orientation, connected in parallel to one run of 10 awg.
That uses wire wire for more hours, at lower current.
And it's OK to have higher IR loss, that can be made up for with more panels. For instance, 14 vs. 12 panels is 17% more power, makes up for 17% higher losses. Compare cost of Si to cost of Cu.
Loss in wire goes as I^2, so drops of rapidly when current is less than maximum. Only under ideal conditions would you lose much (and you can afford to). Times of less light, less loss but the extra panels still make 17% more.

 

[Brettw]

Two strings of different orientation, connected in parallel to one run of 10 awg.
That uses wire wire for more hours, at lower current.
And it's OK to have higher IR loss, that can be made up for with more panels. For instance, 14 vs. 12 panels is 17% more power, makes up for 17% higher losses. Compare cost of Si to cost of Cu.
Loss in wire goes as I^2, so drops of rapidly when current is less than maximum. Only under ideal conditions would you lose much (and you can afford to). Times of less light, less loss but the extra panels still make 17% more.

So basically you're saying to put 8 panels facing south and then another 8 panels facing west and I should be good on one run of wire without over doing the electronics or over doing the wire?

 

[Brettw]

With 16 panels, no. 12, and 14 would be better for the panel you mention. Looking at the second solar string prediction in my post above, you'll see that 8 panels in series will reach ~450Voc around zero F. So, two strings of 8 (written 8S2P) might fry your electronics. Since your winter lows dip below zero, a string of 8 is a no-no. Two parallel strings of 6 or 7 would give you a high, but still safe voltage that won't exceed your electronic's limits. You could even wire 21 panels in 7S3P, because each parallel string would be producing up to 9A of power, so 9A + 9A + 9A =27A. Use the link I provided and plug in your parameters yourself. Playing with the calculator will give you lots of insight into your design.

But, can you afford that much copper wire? That would require 2100 feet total of 10 gauge wire, which is going to cost a pretty penny. But, even just 14 of those panels would be >5500W of power, and you haven't mentioned anything yet about what you actually want to power.

Thoughtful design starts out with an itemized list of what you want to power, and then you select the components to accomplish that goal. Just one parallel string would likely power a house with near normal consumption patterns. Two would be 5500+W. What do you want to power that needs more then 5000W of solar? Making a list first is the single best thing you can do at this point in your design.

My temps drop below zero at night when the panels aren't being used, is the below zero still a problem in this scenario?

 

[Hedges]

So basically you're saying to put 8 panels facing south and then another 8 panels facing west and I should be good on one run of wire without over doing the electronics or over doing the wire?

Yes, except it looks like 8 panels in series would be to many. About 400Voc under nominal conditions. Is max input 450V? 12.5% probably isn't enough margin, need to get temperature coefficient of Voc from panel data sheet and record coldest temperature of your location to be sure.

7 facing South and 7 facing West (90 degrees between them) or SE and SE, would reduce Imp from 19.4A to about 0.7x that, 13.6A. Loss from resistance in copper is cut in half (but power will stay at that level more hours.)

The temperature of panel only matters when there is light on it, but don't count on day being much warmer than night. Just use record cold temperature to calculate. Various panels have different specs, so some fit a given inverter and situation better than others.

 

[Brettw]

Yes, except it looks like 8 panels in series would be to many. About 400Voc under nominal conditions. Is max input 450V? 12.5% probably isn't enough margin, need to get temperature coefficient of Voc from panel data sheet and record coldest temperature of your location to be sure.

7 facing South and 7 facing West (90 degrees between them) or SE and SE, would reduce Imp from 19.4A to about 0.7x that, 13.6A. Loss from resistance in copper is cut in half (but power will stay at that level more hours.)

The temperature of panel only matters when there is light on it, but don't count on day being much warmer than night. Just use record cold temperature to calculate. Various panels have different specs, so some fit a given inverter and situation better than others.

Thank you