DISCUSSION: PV Series vs. Parallel. Is parallel REALLY better for partial shading situations?

[sunshine_eggo]

3/23/23 EDIT: Will posted some interesting results:

Series string shading summed up in 4 pictures

Each row is a string. These two strings of panels are in parallel and connected to victron MPPT. You can see that the last panel is shaded: 420VDC. So plenty of voltage and the last panel in the string is shaded, but only producing 200W for both strings! Now wait 30 minutes.. The shading...

diysolarforum.com

I'm thinking series is better. Here's why:

Typical 60 cell panel with bypass diodes:

I think everyone can accept that this panel will lose 1/3 of its output when the cell is shaded. This isn't just a current loss, it's a voltage loss as well.

For easy numbers, let's say this panel is a 300W 60 cell with 36Voc/30Vmp and 10A Imp.

Thus this panel will drop to 24Voc/20Vmp and 10A Imp when the cell is shaded.

What happens if this is in parallel with another identical panel @ 30Vmp/10A?

Well, like batteries, panels have to be at the same voltage when in parallel, so the only way you'll get ANYTHING out of the shaded panel is if the array is below 24V with optimal at 20V.

Assuming max power, the moment the cell is shaded, power drops from 600W to 300W.

The MPPT tries to pull more current and finds that it can get max output of 400W at 20Vmp. Yes, this will actually be a LITTLE higher than 400W because the unshaded panel will output a little more than Imp, but I'm ignoring that since I don't know what the number will be, but certainly somewhere between Imp and Isc.

So, partial shading of ONE panel effectively kills the same amount on any parallel panel due to array voltage lose.

HOWEVER, if the two panels are in series, only the single section of the shaded panel is lost. So instead of getting 10A @ 60Vmp, you're getting 10A @ 50Vmp or 500W.

This suggests to me that series is better for partial shading than parallel.

Am I missing something? I ran some tests on my neighbors 2S3P array, and the results are consistent with ALL panels outputting as though they were shaded in the same way.

@RCinFLA
@Hedges

 

[timselectric]

This was always my thought as well.
But I never got around to testing it.
Glad you did.

 

[SignatureSolarJames]

parallel is better, sorry to say, have tried it a lot

 

[timselectric]

I know that it was better, before bypass diodes were added to panels. But, shading is not the problem it used to be. I'm sure that I will test it myself, one day.
Maybe when I build my ground mount. It would be easier to change the wiring around for testing.

 

[sunshine_eggo]

parallel is better, sorry to say, have tried it a lot

Can you explain why? Did you actually do measurements? The physics (P = I * V) and my own tests on my neighbor's 2S3P array say otherwise. I don't have hard data because I didn't record or document anything, but each time I fully shaded only two cells on a single panel in a single string, it cut the array output by 1/3rd when it should have only cut it by 1/9th.

After going through it in my head over the past several months, I expected it, but my neighbor was shocked to see 1/3rd of his output leave simply by shading two cells.

 

[brb58]

I did measurements, The voltage I saw was an average between the two panels with one cell shaded on one of the panels.

Same applies when to panels are at different angles to the sun...voltage averages.

Easy to test!

 

[sunshine_eggo]

I did measurements, The voltage I saw was an average between the two panels with one cell shaded on one of the panels.

In the example, at 25V (average of 20Vmp and 30Vmp), the panel with 1/3 of its cells blocked off can't deliver current since that's over the Voc of the unshaded panel.

 

[zanydroid]

I think this assumes the MPPT can locate that peak.

Also I think we would want to see a simulation of a lot of different shading scenarios before extrapolating a blanket statement from analyzing less than a dozen configs

 

[Hedges]

I believe series is better. If one string, so long as voltage remains within MPPT range.
If too many for max voltage, then series/parallel. I had 9s2p. I tried shading a single panel, and it appeared I only lost the output of a single panel, not two, because max power points of 8s and 9s are close enough in voltage.

I did measurements, The voltage I saw was an average between the two panels with one cell shaded on one of the panels.

About what I expect. A bit above Vmp for 8s, a bit below Vmp for 9s, is where the parallel combination delivers most power.

Shading 50% of one string when parallel with another, then you lose more than just the shaded panels.

parallel is better, sorry to say, have tried it a lot

What battery voltage, and what were the Vmp of series and parallel?
We expect greater loss with large voltage conversion, best efficiency with array voltage closer to battery voltage.
But not too close.

For a while, one of my inverters had 16s2p "12V" panels instead of 24s. (rearranged strings due to inverter out of service.)
In that case, just a little shade killed production because voltage got too low.
Series is good to have headroom for hot weather, and especially shaded panels.

 

[zanydroid]

What happens if there are multiple combinations of bypass diodes that can activate to provide close to the same chosen operating voltage, but different current?

 

[Hedges]

An MPPT sweep will find the peak power available from a given array. The is sometimes done once every few minutes. In between, the voltage is dithered to ride the peak.

Some array and diode configurations have local maxima, so only by sweeping can the higher ones be found.

What happens if there are multiple combinations of bypass diodes that can activate to provide close to the same chosen operating voltage, but different current?

That sounds like a passive non-linear circuit, but current finds the way. I think it will arrange itself how it desires. Possibly, depending on whether voltage approaches from higher or lower.

Do you have an example?

 

[EastTexCowboy]

All this math is giving me a headache. I'm just going to cut that damn tree down when my wife is gone to Walmart.

 

[zanydroid]

That sounds like a passive non-linear circuit, but current finds the way. I think it will arrange itself how it desires. Possibly, depending on whether voltage approaches from higher or lower.

Do you have an example?

Single panel, Impp of 10A. 1/3 at 10A, 1/3 at 6A, 1/3 at 1A.

If it picks ~2/3V (setting aside where the peak is). Which diode bypasses? The one that is less shaded- 6A, or more shaded - 1A? Ideally you want the latter.

 

[timselectric]

All this math is giving me a headache. I'm just going to cut that damn tree down when my wife is gone to Walmart.

These two trees are no longer an issue.
And I have extra fire wood for winter.
I only took a little off the top. lol

 

[brb58]

When I did my test, I shaded one panel completely and measured output on the other fully exposed panel.

I then shaded the other panel and half of the exposed panel....measured output.

I then measured the output of the fully exposed panel and the one that only had half its output. The sum was the exact of the measured output of both together.

It's an easy test...just do it with your own panels.

 

[Hedges]

At 40Voc, zero current.
At 36V, 1A, 36W
At 24V, 6A, 144W
At 12V, 10A, 120W

I think the cells with 6A worth of light go to 13V if 1A, 12V if 6A.
Once MPPT pulls voltage down to 24V, The 10A an 6A cells push current through the 1A cells' bypass diode.

Download LTSpice and make a subcircuit representing a PV cell.
Simulate.

These two trees are no longer an issue.
And I have extra fire wood for winter.
I only took a little off the top. lol

How much do you lose with the shade of that vent pipe?
And how did you make a round hole in tempered glass?

 

[EastTexCowboy]

These two trees are no longer an issue.
And I have extra fire wood for winter.
I only took a little off the top. lolView attachment 139934

Is that a vent pipe poking through a panel on the right side?

 

[timselectric]

How much do you lose with the shade of that vent pipe?

None, that's a bad panel filling a gap. (OCD)

And how did you make a round hole in tempered glass?

Round hammer. lol

Is that a vent pipe poking through a panel on the right side?

Yup, there's actually two of them. The other one is just out of the shot.
My OCD said, no gaps.

 

[sunshine_eggo]

All this math is giving me a headache. I'm just going to cut that damn tree down when my wife is gone to Walmart.

This FTW!

 

[EastTexCowboy]

None, that's a bad panel filling a gap. (OCD)

Round hammer. lol

Yup, there's actually two of them. The other one is just out of the shot.
My OCD said, no gaps.

I have to say that would have never occurred to me but it makes a lot of sense. Aesthetically it looks MUCH better than a big ugly gap. Fortunately for me mine are on the shop and there's no way to see them from the ground from any direction. But then, I don't have any vent pipes so I don't suppose that part matters.