PV array efficiency: series vs. parallel

[greg_mustang]

Hi there,

I have a small solar power system for my garage. It consists of:
4x 100w Renogy monocrystalline panels
4x deep cycle 12v lead acid batteries
Renogy 2000w pure sine wave inverter
Renogy Rover 30A MPPT controller

I recently upgraded to this controller after having a cheap PWM one.
My panels are currently wired all four in parallel.
I noticed, if I understood well, that the Rover can take a higher voltage for PV input, therefor I wonder if it is a good idea to rewire the panels in series.
If so, all four series (~48V) or 2 series to 2 parallel (~24V)?
My main cabling is 10 guage, the Y-splitters are 12 guage.
I do not really have issues with partial shading, its on the roof and no big trees nearby.

Let me know any advice. Hoping to gain some efficiency.

Thnks,
Greg
Canada

 

[sunshine_eggo]

It's a trade off.

The big jump is the switch to MPPT. PWM shorts the panels to the battery thus forcing the panels to work at battery voltage. Assuming an average charge voltage of 14V and most 100W panels have a Vmp of 17-18V, that's a 3.5V/14V = 25% boost to performance simply by the MPPT allowing the panel to operate at Vmp vs. battery voltage. This difference diminishes as the panels heat up because the Vmp drops with heat. Conservatively, the switch from PWM to MPPT is an instant 15-20% boost in typical conditions.

MPPT lose efficiency as the PV voltage increases. However, increasing the PV voltage also decreases the losses due to wiring. In most cases they are a wash within < 1%. This becomes more significant as the distance between the PV and MPPT increases.

1. More parallel tends to perform better when shading is a factor.
2. More series tends to perform better when the cable run is longer.
3. More series allegedly tends to perform better in low light conditions: sunrise/sunset/diffuse lighting, etc., but I have NEVER seen documented proof of this, AND you're talking about very small benefits in low power conditions.
4. More series reduced fuse count and may eliminate fuses if no more than two strings are in parallel.

Your 4P panels should have a fuse on each panel.

Voc must never exceed the Voc limit of the controller, and you must allow for cold temperature effects on voltage - higher voltage with lower temperature. Multiply your calculated Voc by the factor in this chart based on your record low temperature:

PV array temperature correction table (NEC 2017)

There are calculators like this one made by @upnorthandpersonal which help you calculate PV array voltage and power for low temperatures based on the specific specifications of your panels. These are great tools and will give more precise...

diysolarforum.com

All that blah, blah, blah brings me back around to it all being a trade off.... thus you simply pick something in the middle without a compelling reason to go another way... 2S2P - a little improvement in wiring losses for a small loss of inefficiency converting from ~36V to 12V vs. 18V to 12V.

2S2P eliminates the need for fuses in the array, but you may still use them if you like.

 

[rhino]

To be more explicit to what @sunshine_eggo is saying with your equipment, you can not do 4 in series because it would be too high. You need to look at the open circuit voltage of your panels (Voc) from the spec sheet which shows it is 24.3V. If you multiplied that by 4 it would be 97.2V. Then look at the spec sheet for the controller to check maximum voltage it can handle and it shows 95V.

 

[greg_mustang]

Thanks guys for this info!
I will likely go with 2S2P

 

[cbfraser]

the issue of better performance with parallel when there is shading isn't clear to me - if there are two panels (say 38v, 9amps max) - one in full sun - one in full shade -with parallel one panel will still be sending 9amps @ 37v to controller - no problem; with series will the shaded panel actually drag down the voltage and/or amperage of the sunshine panel? -or just not add to it so that the series array is still sending 9amps @ 37v to controller?

 

[littleharbor2]

the issue of better performance with parallel when there is shading isn't clear to me - if there are two panels (say 38v, 9amps max) - one in full sun - one in full shade -with parallel one panel will still be sending 9amps @ 37v to controller - no problem; with series will the shaded panel actually drag down the voltage and/or amperage of the sunshine panel? -or just not add to it so that the series array is still sending 9amps @ 37v to controller?

In a series string, the lowest current producing panel will dictate the overall current. There are bypass diodes in panels that try to mitigate this issue but suffice to say the unshaded panels output will be reduced. Picture a 1" pipeline with a 1/4" section in the middle. The 1/4" section, regardless how short will affect the total output by restricting the flow (current in this case).