Solar Panels - Series or Parallel ?

[colemj]

We installed a new Morningstar MPPT controller three months ago which allowed me to do some experimentation with our 4 Kyocera 120W panels. They are mounted on the bimini 2 on each side. These panels see random partial shading at certain times of day due to the SSB antenna, the mast and the radome. I move the boom out of the way to prevent shading from that.

Wired in parallel they average 125Ahr/day. Wired in series/parallel (2 panels in series, 2 banks in parallel) they average 150Ahr/day.

So Nick is wrong - we are only seeing a 20% improvement and not 25%.

I have not wired all 4 of them in series yet because of the PITA of doing so coupled with the fact I had other fish to fry over the past few months, but I plan on doing that experiment now that we are back out cruising with wide open days.

Mark

[s/v Jedi]

Originally Posted by colemj

So Nick is wrong - we are only seeing a 20% improvement and not 25%.

I have not wired all 4 of them in series yet because of the PITA of doing so coupled with the fact I had other fish to fry over the past few months, but I plan on doing that experiment now that we are back out cruising with wide open days.

Well, we all know where that 5% is hiding then

But you must be wrong because I have lost all perspective of reality with my claims that series connection works more efficient than parallel

p.s. Did you discover the thread I started about your "reckless behavior"?

p.p.s. Don't forget to disengage the integrated autopilot while at anchor. It might start your engines and power you right onto the rocks where the Sirens are singing their song of complacency!

ciao!
Nick.

[prroots]

Originally Posted by colemj

We installed a new Morningstar MPPT controller three months ago which allowed me to do some experimentation with our 4 Kyocera 120W panels. They are mounted on the bimini 2 on each side. These panels see random partial shading at certain times of day due to the SSB antenna, the mast and the radome. I move the boom out of the way to prevent shading from that.

Wired in parallel they average 125Ahr/day. Wired in series/parallel (2 panels in series, 2 banks in parallel) they average 150Ahr/day.

So Nick is wrong - we are only seeing a 20% improvement and not 25%.

I have not wired all 4 of them in series yet because of the PITA of doing so coupled with the fact I had other fish to fry over the past few months, but I plan on doing that experiment now that we are back out cruising with wide open days.

Mark

Your setup is quite similar to mine except that I have only two panels and my total is 420 watts vs your 480 watts. I also get random shading at different times of day from wind generator and radar radome.

Do you have the 45 or 60 Amp version of the controller? Are you using MSView to monitor daily output from the panels? How have you come up with the parallel vs parallel/series results given the variable effects of weather? The only way I can imagine it is that you average the results of several days to, hopefully, filter out the weather variable. That would seem to be a very difficult thing to do given the variability of weather ie, cloud cover. An alternative is to wait for a cloudless day and then switch back and forth between series and parallel. That would give the percent difference, but not the average daily output. This is what I plan to do as soon as that illusive day arrives
Pete

[colemj]

Originally Posted by prroots

Your setup is quite similar to mine except that I have only two panels and my total is 420 watts vs your 480 watts. I also get random shading at different times of day from wind generator and radar radome.

Do you have the 45 or 60 Amp version of the controller? Are you using MSView to monitor daily output from the panels? How have you come up with the parallel vs parallel/series results given the variable effects of weather? The only way I can imagine it is that you average the results of several days to, hopefully, filter out the weather variable. That would seem to be a very difficult thing to do given the variability of weather ie, cloud cover. An alternative is to wait for a cloudless day and then switch back and forth between series and parallel. That would give the percent difference, but not the average daily output. This is what I plan to do as soon as that illusive day arrives
Pete

We have the 60A version. I use MSView to monitor the output. The weather and sun here in San Blas Panama is very consistent this time of year. One day looks just like the next - sunny and windy with scarce high white clouds. I first averaged measurements for a month wired as parallel, then wired them as series/parallel and averaged the measurements for another month. This morning, I wired all 4 panels in series and will now measure the output from that for a couple of weeks.

Note that my panels are 17V ones, so I did not get any benefit from the MPPT when they were wired in parallel. Essentially, the controller simply directly connected them to the battery bank every day without doing any conversion or regulation until late in the day when the batteries would get near full.

Yours are already higher voltage panels, so will see a benefit from the MPPT even in parallel mode.

Mark

[prroots]

Originally Posted by colemj

We have the 60A version. I use MSView to monitor the output. The weather and sun here in San Blas Panama is very consistent this time of year. One day looks just like the next - sunny and windy with scarce high white clouds. I first averaged measurements for a month wired as parallel, then wired them as series/parallel and averaged the measurements for another month. This morning, I wired all 4 panels in series and will now measure the output from that for a couple of weeks.

Note that my panels are 17V ones, so I did not get any benefit from the MPPT when they were wired in parallel. Essentially, the controller simply directly connected them to the battery bank every day without doing any conversion or regulation until late in the day when the batteries would get near full.

Yours are already higher voltage panels, so will see a benefit from the MPPT even in parallel mode.

Mark

Sounds good. We've also spent time in San Blas and the rest of Panama as well. They were great. Yes, my panels give me some Voltage overhead even in parallel mode for low light conditions. Hopefully, we'll get a relatively cloudless day soon.
Pete

[prroots]

I just received a schematic of the bypass diodes on my Photowatt PW2050-210 panels. I posted it here:

http://www.cruisersforum.com/forums/...tml#post892252

Can anyone figure it out?
Pete

[prroots]

Originally Posted by ChuckSK

Hi Pete

I checked the other products produced by Photowatt and all other panels had 1 bypass diode for each 18 cells. I suggest that your panel of 54 cells has 3 bypass diodes and that the documentation has a misprint.

If that is the case then 2 Photowatt 210 watt panels in series would be equal to 3 Kyocera 130 watt panels in series. Both configurations would have 6 sub-panels.

Chuck

Good point. I finally received the schematic from Photowatt showing placement of bypass diodes. There are actually 4 bypass diodes. The panel is divided into 3 substrings of 18 cells each. Each of these substrings has its own bypass diode. The 4th bypass diode covers 2 substrings. In this manner if the panel were in series with other panels, one would never encounter more than a two diode drop.
Pete

[SailFastTri]

Just to throw in another data point: yesterday I rewired my 8 panels into series/parallel so now I have 4 pairs, and saw more than a 10% increase in output, measured in amps. I attribute this to my long cable runs and small diameter conductors.

[wdkester]

Originally Posted by colemj

"I have not wired all 4 of them in series yet because of the PITA of doing so coupled with the fact I had other fish to fry over the past few months, but I plan on doing that experiment now that we are back out cruising with wide open days."

Mark

Mark did you ever get around to connecting all panels in series?

[colemj]

Originally Posted by wdkester

Mark did you ever get around to connecting all panels in series?

Yes I did. The shading killed me. The panels are located 2 on each side of a large bimini, so it is highly likely that one side may get shadows at certain times of the day.

Running panels on each side in series and the two series in parallel is the sweet spot for us.

Mark

[wdkester]

Originally Posted by colemj

Yes I did. The shading killed me. The panels are located 2 on each side of a large bimini, so it is highly likely that one side may get shadows at certain times of the day.

Running panels on each side in series and the two series in parallel is the sweet spot for us.

Mark

Darn! I thought I had it understood. I thot that if all the panels were in series they would contribute whatever power that they were able to produce no matter which ones were shaded. This must be wrong. Can anyone explain?

[s/v Jedi]

Originally Posted by wdkester

Darn! I thought I had it understood. I thot that if all the panels were in series they would contribute whatever power that they were able to produce no matter which ones were shaded. This must be wrong. Can anyone explain?

It sounds like the difference Mark saw was substantial. With four panels in series and two shaded, there are two diodes in the circuit that take some voltage off. This loss is prevented when you put two series-strings of two panels in parallel, but how big is that difference?

On the other hand, with partial shaded panels, it can change again. But the MPPT controller might be able to find a lower-voltage point where unshaded sections help again. This is possible because there are still two panels in series.

The only way to get reliable results on what exactly is happening, is indoors in a controlled setup.

[Pelagic]

Boy... somehow I missed this focused Thread and to disclose, I am a Solar Dummy with little support in the Philippines.

Like so many others, I had assumed from various past articles and sailors advice that if you expected shading then parallel was the best solution and buying panels with the largest VOC for a 24v house bank was recommended.

So I went ahead and committed to buying the following:
4 x Sunpower E20/327 Watt panels…. VOC 64.9
1 x Outback Max 80…. PV Open Circuit Voltage (VOC) 150 VDC absolute maximum coldest conditions / 145 VDC start-up and operating maximum


My large Bimini is now modified and mounts are made for a fixed horizontal installation, but the 4 panels are not mounted or wire connectors purchased yet.

So opportunity for me to wire intelligently based on the (evolving) consensus of others and to order the remaining hardware that is needed.

Attached is sketch and mock-up photos to illustrate that I will always have some shading which is the reality of a schooner

I labeled the panels and obviously S-1 and P-1 will get more mast and aft stay shading at anchor.

Boom at anchor is stored and centered low so not a shading issue but the topping lift still is.

12m wiring run to MPPT in pilothouse from where I show a combiner

(I can place a combiner anywhere)

Nick/Chuck would you still go ALL Series?
1 ... If yes how would you wire this?
2.....Use a combiner or individual panel cables to breaker by MPPT?
3 Then not sure if my one MPPT enough or is it practical to buy a 2nd?

Colejmi…. Your experience suggests 2 x Series in Parallel… then same questions
1.... If yes how would you wire this?
2 .... Use a combiner or individual panel cables to breaker by MPPT?
3 .... Then is my one MPPT enough or do I need a 2nd?

Obviously, any advice is at my risk, but I do need guidance on connectors and cable purchases to finish this off and I trust my fellow CF council over anything I can find in the Philippines.

Thanks in advance.

[noelex 77]

First thanks Mark for reporting your results.

My interpretation is that panels experiencing disparate conditions did better wired in parallel where panels experiencing the same conditions did better wired in series.

That makes sense.

The series/parallel combination is probably a good solution for many boats and at least hedges the bets between series/parallel when the ideal answer is not clear.

Originally Posted by s/v Jedi

It sounds like the difference Mark saw was substantial. With four panels in series and two shaded, there are two diodes in the circuit that take some voltage off.

Nick I think you are assuming there is a bypass diode across the whole panel. This is not normally installed in the solar panel. An extra bypass diode can can be added if you want to wire the panels in series, and it think this would be a good idea, or at least worthy of some experimentation. (Have you added these diodes?)

With the installed bypass diodes with two shaded panels the current has to pass through all the installed bypass diodes in the panel. So typically 6 each with a 0.5v voltage drop.

The performance of series and parallel panel is very difficult to mathematically model especially in the common situation where we have partial shading of the panel but the cells are still capable of producing some power.

[sailinglegend]

Originally Posted by noelex 77

....My interpretation is that panels experiencing disparate conditions did better wired in parallel where panels experiencing the same conditions did better wired in series.

I think your post sums up the problem and the misunderstandings very well.

Not all panels have the same number or combination of bypass diodes, so all the theory in the world cannot take this into account and simulate the shadow effect on different boats, which can sometimes be considerable.

My suggestion is that every solar owner does their own tests in parallel and series - and a combination of both. An MPPT controller may give you all the information you need, like the Ah per day or per week, or the current flowing from the controller. This needs to be done under the same conditions which is not always easy.

I did some tests 12 months ago and it very clearly showed that parallel is dramatically better than series - certainly with my panels. I have very thick wires to the regulator so there was no difference in the charge current in series coming out of the controller.

My results were rejected by Jedi as "flawed". So I repeated the tests the next day and when both panels were connected in Series the output charging current fell to zero with the wrong kind of shadow on just one panel. In Parallel the same shadow dropped the charging current by half.

For my new Solara semi-flexible panels Parallel clearly wins over Series.

I suspect that if I had fitted an extra bypass diode ACROSS each panel there may have been no difference between Series and Parallel connection.