Solar Panels - Series or Parallel ?

[TeddyDiver]

Originally Posted by sparrowhawk1

The video clearly shows that parallel is better than series when panels are being used where there might be shading, even partial shading..

That's true only with panels without bypass diodes, with a single diode it would be a draw. More than one bypass diode in every panel series wins.. Should be simple except I have seen only one brand announce the amount of diodes in their panels.

[socaldmax]

Originally Posted by noelex 77

37v panels are identical to two "12v" panels connected in series. The only difference is the higher voltage panel is housed in one frame.

If two "12v" panels are connected in series and placed together they will be identical to single larger 37v panel of the same wattage (assuming the other characteristics of the cells such as temperature coefficients etc are the same)

If parallel connection is best for "12v" panels it is even more important that higher voltage panels are connected in parallel.

All solar panels that a boat is likely to use will have (or should have, some of the very cheap no name panels have been known to leave them out)) bypass diodes. Their primary purpose is to protect the cells from overheating if there is partial shade, but they do increase the output especially in series. If you use series connection adding an extra bypass diode around the whole panel is sensible.

My internet connection is too slow to see the video, but they should indicate the properties of the panels they are testing. If there are no bypass diodes that will effect the results.

I still have an open mind, but I think most of the evidence points to parallel connection as superior to series connection on a boat providing the wiring is adequate. If this is true if also indicates, if all other things are equal, "12v" panels are superior to higher voltage panels. Of course things are not equal, high efficiency large 12v panels are not produced.

It is also possible that a combination of series and parallel connection of solar panels is best. It is worth rembering the individual cells in solar panels are connected in series so technically even if we wire the panels up in a parallel we actually have the solar cells arranged in a series/parallel configuration.

The question we should be really asking is what is the optimum number of solar cells that we should be connecting in series. It is unlikely that the correct answer is always 36 (as occurs in a "12v"panel). If we knew this ideal number of cells we could wire up both 12V and high voltage panels to be as close as possible to this optimum.

I started this thread almost 5 years ago and it surprising given the widespread use of solar on yachts that we still do not have a definitive answer.

It's too bad you can't see the video, because it does give a definitive answer: parallel is clearly superior in any shading scenario. They covered one single cell and in a 2 panel series system, it dropped the entire system to 1 amp or less. In a 2 panel parallel system, it dropped about 50%. They experimented with diagonal shading, partial shading of multiple cells, etc, but in all cases, parallel was either slightly better or much better than series.

I can't imagine the number of cells in a panel has any difference in the results. When they shaded the one corner cell, the panel acted as if it had an open in it, virtually no power flowed. At least in the parallel system, power was still flowing from the other panel.

The only time I foresee any different results would be with Sunpower panels, which are designed with a different cell connection scheme. The youtube video on that indicates only an 11% drop in output with one or two cells covered. That's 89% better than the competition, and definitely a feature to look for in a marine application.

[OceanPlanet]

Originally Posted by sparrowhawk1

The video clearly shows that parallel is better than series when panels are being used where there might be shading, even partial shading. Through my own experience of having 30 cell 33 cell and 36 cell panels I know 36 is the minimum requirement to get the voltage needed to charge batteries fully. For maximum output for a given space( where there's shading) smaller panels are better, but larger panels are a better buy. And as already stated when you get really large panels most are two smaller panels in one frame.( if they're rated for 12 volts).

Note that for 12v batteries, having a panel voltage above 18v or so doesn't make much difference EXCEPT when used with lower level MPPT controllers that require more "headroom" (delta) between the panel voltage and the battery voltage to work efficiently. With Genasun controllers we effectively charge 12v batteries from panels that are only 16v....

[socaldmax]

Originally Posted by OceanPlanet

Note that for 12v batteries, having a panel voltage above 18v or so doesn't make much difference EXCEPT when used with lower level MPPT controllers that require more "headroom" (delta) between the panel voltage and the battery voltage to work efficiently. With Genasun controllers we effectively charge 12v batteries from panels that are only 16v....

There's also the scenario where the Vmp of the panel is in the 17v range, the cable voltage drop plus panel heat coefficient combine to the point where you get less than 15v at the controller, and the batteries require 15v to hit full charge.

There's also a short period at the beginning and end of the day when a higher V panel will produce usable charging current and a lower V panel (which might also be lower current) won't. Every minute, every watt counts over the long run.

[sparrowhawk1]

Originally Posted by TeddyDiver

That's true only with panels without bypass diodes, with a single diode it would be a draw. More than one bypass diode in every panel series wins.. Should be simple except I have seen only one brand announce the amount of diodes in their panels.

Or Show videos where shading does not destroy their output.

[sparrowhawk1]

Originally Posted by OceanPlanet

Note that for 12v batteries, having a panel voltage above 18v or so doesn't make much difference EXCEPT when used with lower level MPPT controllers that require more "headroom" (delta) between the panel voltage and the battery voltage to work efficiently. With Genasun controllers we effectively charge 12v batteries from panels that are only 16v....

36 cell panels are rated at 18 volts. I'm saying that having more cells is not more productive. ( specifically because of shading) do you have any information on how 16 volt panels produce in shade with your controllers ?

[sparrowhawk1]

Originally Posted by socaldmax

There's also the scenario where the Vmp of the panel is in the 17v range, the cable voltage drop plus panel heat coefficient combine to the point where you get less than 15v at the controller, and the batteries require 15v to hit full charge.

There's also a short period at the beginning and end of the day when a higher V panel will produce usable charging current and a lower V panel (which might also be lower current) won't. Every minute, every watt counts over the long run.

And just a little shading can change 18 volts into 15. The balance is the more cells you have the more susceptible you are to shading.

[socaldmax]

Originally Posted by TeddyDiver

That's true only with panels without bypass diodes, with a single diode it would be a draw. More than one bypass diode in every panel series wins.. Should be simple except I have seen only one brand announce the amount of diodes in their panels.

Kyocera's published spec sheet indicates they have 8 bypass diodes, yet the performance in the video indicates it doesn't help.

http://www.kyocerasolar.com/assets/001/5640.pdf

[Maine Sail]

Originally Posted by socaldmax

There's also the scenario where the Vmp of the panel is in the 17v range, the cable voltage drop plus panel heat coefficient combine to the point where you get less than 15v at the controller, and the batteries require 15v to hit full charge.

There's also a short period at the beginning and end of the day when a higher V panel will produce usable charging current and a lower V panel (which might also be lower current) won't. Every minute, every watt counts over the long run.

On every boat I have data logged for series vs. parallel, where shading was involved, parallel 18V panels has beat series 18V panels....

[Eder]

Well this is weird...I have a 4 panel flexible total 400w set up with mppt controller.
After reading the last few posts I ran upstairs to test my panels due to shading. They are wired in series.

Right now full sun on all panels 10.1 amps (sun is not yet high enough for full output.

I put beach pail on one square of a panel...out put unchanged
I put pillow on 1/2 a panel...still 10 amps.
I just now centred my boom so partial shade to 2 panels...output up to 10.5 (most likely due to sun slowly getting higher.
I just put a beach towel completely covering 1 panel....11 amps now
Cover 1 panel and pillow on other amps dropped to 6 amps
just pillow on another panel....12 amps

So theres a lot more to it than the video shows I guess. Anyway I normally anchor....move boom out of way so I'll stay in series for now.

Maybe I'll retry in 4 hours when the sun is overhead

[SailFastTri]

Originally Posted by robert sailor

Ours are in series now, used to be in parallel, I think we make a little more power early in am and later in pm, using Morningstar controller.

Same here. I have long wiring runs and have seen a 25% (or more) improvement in overall charge performance by changing from parallel to series. In longer wire runs, higher voltage is less affected by voltage drop for equal wire diameter.

The panels are 12v and I have 4 on each side of my boat. They are wired as series-pairs, and the two pairs are connected in parallel back to the controller. Both sides are in parallel. The MPPT controller sees ~37v open voltage.

[socaldmax]

Originally Posted by Eder

Well this is weird...I have a 4 panel flexible total 400w set up with mppt controller.
After reading the last few posts I ran upstairs to test my panels due to shading. They are wired in series.

Right now full sun on all panels 10.1 amps (sun is not yet high enough for full output.

I put beach pail on one square of a panel...out put unchanged
I put pillow on 1/2 a panel...still 10 amps.
I just now centred my boom so partial shade to 2 panels...output up to 10.5 (most likely due to sun slowly getting higher.
I just put a beach towel completely covering 1 panel....11 amps now
Cover 1 panel and pillow on other amps dropped to 6 amps
just pillow on another panel....12 amps

So theres a lot more to it than the video shows I guess. Anyway I normally anchor....move boom out of way so I'll stay in series for now.

Maybe I'll retry in 4 hours when the sun is overhead

Did you notice how they were getting 15 A out of 240w of panels and 16A out of 280w of panels?

With 400w, you should be getting north of 20 A, at least. Either all of your panels are producing less than they should, or possibly the one panel you were shading wasn't producing anything to begin with. The 2nd scenario would more likely explain your results.

Think about it. A fully shade panel should produce nothing. If shading it didn't drop your output, then logically, it wasn't producing anything to begin with.

[socaldmax]

Originally Posted by Maine Sail

On every boat I have data logged for series vs. parallel, where shading was involved, parallel 18V panels has beat series 18V panels....

Your results match mine, they make perfect sense to me.

How does this relate to my post that you quoted?

[TeddyDiver]

Originally Posted by socaldmax

Did you notice how they were getting 15 A out of 240w of panels and 16A out of 280w of panels?

With 400w, you should be getting north of 20 A, at least. Either all of your panels are producing less than they should, or possibly the one panel you were shading wasn't producing anything to begin with. The 2nd scenario would more likely explain your results.

Think about it. A fully shade panel should produce nothing. If shading it didn't drop your output, then logically, it wasn't producing anything to begin with.

At seven to eight a clock in the morning? I'd say 10amps is pretty impressive so early

[socaldmax]

Originally Posted by TeddyDiver

At seven to eight a clock in the morning? I'd say 10amps is pretty impressive so early

Then why even post those results if they were not representative and without mentioning the time of day?

What is your normal max. output? I suppose your latitude would also help us determine how much sun you're getting vs someone near the equator.

That still doesn't explain the lack of change in output when you shaded the one panel.