Uncommon solar panel parallel series hookup question.

[Mattb4]

Given 4-100w of same manufacture but can only have the voltage from 3 in series to meet already established system. It occurred to me that 2 of the panels could be hooked up in parallel and than the combined panel (double the amperage but the same voltage as 1) could be hooked in series with the other 2 for the combined voltage of 3. From my understanding the combined 2 panels in parallel would have its output amperage limited to the lowest amperage from the other panels when hooked into series or in essence there would be no benefit over just running 3 panels in series. However is this the entire story?

Would the fact of the two 100w panels in parallel added to the other 2 in series help boost production of total watt-hours even if max watts was the same as only 3 panels? Has anyone explored this in practice?

 

[FilterGuy]

I have not tried it, but it seems to me that the double current would not be allowed through the other panels.

 

[efficientPV]

I have thought this would be a wise thing to do if there was a section of an array which had regular partial shading, two panels in parallel and the rest in series. In bright sun. Considering how cheap and available panels have become. The power point voltage would raise in full sun so there would be an increase in power. Even shade from a small antenna wire has been shown to dramatically reduce power. My system has excessive shade with 70% of panels shaded at any time. Good news is I can run basic system on just ambient light.

 

[Mattb4]

I have not tried it, but it seems to me that the double current would not be allowed through the other panels.

That is why the current is limited to the lowest rating in uncommon hookups in series. However a lot of the day you do not reach max current production of 1 panel. So if the 2 in parrallel help raise the amperage of the string for a longer period it might increase the total watt-hours.

 

[0truck0]

Consider adding another charge controller and just having your odd panel be a separate charging system to the same battery. You can have multiple solar panel/charge controllers feed into the same battery(s) and still be fine. volts plus volts plus volts of the 3 panels feed the battery, then completely other solar/charger system to feed the battery also.

I dont quite understand your question. Are you trying to get volts or amps of 3 panels? Like 60 volts from three 20 volt panels (aka 12 volt panels)?
You can, and I think u know already, run two parallel, two parallel and those pairs then in series. This give 20+20=20 volts, plus 20 volts in series to equal 40 volts with higher amps too. or 18+18=36, or 12+12=24.... whatever your panels say on the sticker..

 

[SamDeleted]

Look at this calculator:

Solar panel #1 is two separate 100w panels

Solar panel #2 is your 2x 100w in parallel, double the amps (acting as one panel for the calculation)

In series you loose 25% , which is one panel , the one you put in parallel

Solar Panel Series & Parallel Calculator - Footprint Hero

Use our solar panel series and parallel calculator to easily find the wiring configuration that maximizes the power output of your solar panels.

footprinthero.com

 

[Mattb4]

Consider adding another charge controller and just having your odd panel be a separate charging system to the same battery. You can have multiple solar panel/charge controllers feed into the same battery(s) and still be fine. volts plus volts plus volts of the 3 panels feed the battery, then completely other solar/charger system to feed the battery also.

I dont quite understand your question. Are you trying to get volts or amps of 3 panels? Like 60 volts from three 20 volt panels (aka 12 volt panels)?
You can, and I think u know already, run two parallel, two parallel and those pairs then in series. This give 20+20=20 volts, plus 20 volts in series to equal 40 volts with higher amps too. or 18+18=36, or 12+12=24.... whatever your panels say on the sticker..

Yes the traditional approach of creating a new array and SCC to the same battery bank is one approach to odd solar panels combinations. However not really worthwhile for just one additional panel.

In terms of my question it is not that you can exceed amperage at peak for 3 in series by adding the 2 in parallel before series but can you hold a higher amperage in average over time. The volts of course will be the sum of 3.

 

[Mattb4]

Look at this calculator:

Solar panel #1 is two separate 100w panels

Solar panel #2 is your 2x 100w in parallel, double the amps (acting as one panel for the calculation)

In series you loose 25% , which is one panel , the one you put in parallel

Solar Panel Series & Parallel Calculator - Footprint Hero

Use our solar panel series and parallel calculator to easily find the wiring configuration that maximizes the power output of your solar panels.

footprinthero.com

That calculator does not work for this proposed hookup.

 

[LakeHouse]

That is why the current is limited to the lowest rating in uncommon hookups in series. However a lot of the day you do not reach max current production of 1 panel. So if the 2 in parrallel help raise the amperage of the string for a longer period it might increase the total watt-hours.

I don't think there would be any change in current with this arrangement. Like @FilterGuy said, the 'extra' wouldn't flow through the other two panels, or at least not through the solar cells. Maybe through a bypass diode, but then the other panels wouldn't add any voltage so total power would be much less.
But, what you should expect to get is at least some increased voltage from the string compared to the normal single-single-single arrangement. Consider that your current will be limited by the two singles, then each of the parallels will push (about) half that amount of current. But since they're being limited on current, they'll provide it at a higher voltage.

For example (using my 355W panels, cause it's the curves I have handy), let's say it's a sunny day and your panels are receiving full 1000W/m2 insolation, your MPPT should find the max current point (Imp) of about 10.3A. The two series panels provide that current at the Vmp of 34.6V. But the two parallels are limited to half that current, 5.1Amps, which they provide at closer to 39V. So instead of 10.3A * (3 * 34.6V) = 1070W, you get 10.3A * (39V + 2 * 34.6V) = 1115W.

 

[Mattb4]

I don't think there would be any change in current with this arrangement. Like @FilterGuy said, the 'extra' wouldn't flow through the other two panels, or at least not through the solar cells. Maybe through a bypass diode, but then the other panels wouldn't add any voltage so total power would be much less.
But, what you should expect to get is at least some increased voltage from the string compared to the normal single-single-single arrangement. Consider that your current will be limited by the two singles, then each of the parallels will push (about) half that amount of current. But since they're being limited on current, they'll provide it at a higher voltage.

For example (using my 355W panels, cause it's the curves I have handy), let's say it's a sunny day and your panels are receiving full 1000W/m2 insolation, your MPPT should find the max current point (Imp) of about 10.3A. The two series panels provide that current at the Vmp of 34.6V. But the two parallels are limited to half that current, 5.1Amps, which they provide at closer to 39V. So instead of 10.3A * (3 * 34.6V) = 1070W, you get 10.3A * (39V + 2 * 34.6V) = 1115W.
View attachment 153502

In your example, if I understand you, there is a gain in production. 1115w is more than 1070w.

 

[LakeHouse]

In your example, if I understand you, there is a gain in production. 1115w is more than 1070w.

Correct, there is a small increase in power. The gain is all from the increase in voltage that the SCC gets. 10.3A at 108.2V instead of 10.3A at 103.8V.

I haven't tested this, but it at least seems compatible with the conventional wisdom.

 

[Mattb4]

Correct, there is a small increase in power. The gain is all from the increase in voltage that the SCC gets. 10.3A at 108.2V instead of 10.3A at 103.8V.

I haven't tested this, but it at least seems compatible with the conventional wisdom.

I might run a test since I have the odd panel. I need to first see how the other 3 are producing over a day (average) before I can tell if there is any gain from doing it.

 

[LakeHouse]

I might run a test since I have the odd panel. I need to first see how the other 3 are producing over a day (average) before I can tell if there is any gain from doing it.

I'd be interested to see the results.
On further thought, there actually might be a significant advantage in the case where one of the parallel panels is shaded. i.e., total power might not drop much at all. Here's a further test for a sunny day if you can monitor power from the string while you're at the panels:

• Shade one of the series panels (just cast a shadow with your hand or body). The string production should drop to something very low, close to zero.
• Shade one of the parallel panels (with the other three in full sun). I think one of two things happens here:
  1. There's a momentary small dip in power while the MPP finds the new optimal voltage, but power output remains high, close to the same as it was with all panels in full sun. OR
  2. Power drops to effectively zero, as the shaded parallel panel acts like a short for unshaded one, pulling the whole string down.

My guess is #1, but I don't have high confidence on that. If you installed blocking diodes on the two parallels, it should definitely be #1, but that might be more work than it's worth.

 

[Mattb4]

I'd be interested to see the results.
On further thought, there actually might be a significant advantage in the case where one of the parallel panels is shaded. i.e., total power might not drop much at all. Here's a further test for a sunny day if you can monitor power from the string while you're at the panels:

• Shade one of the series panels (just cast a shadow with your hand or body). The string production should drop to something very low, close to zero.
• Shade one of the parallel panels (with the other three in full sun). I think one of two things happens here:
  1. There's a momentary small dip in power while the MPP finds the new optimal voltage, but power output remains high, close to the same as it was with all panels in full sun. OR
  2. Power drops to effectively zero, as the shaded parallel panel acts like a short for unshaded one, pulling the whole string down.

My guess is #1, but I don't have high confidence on that. If you installed blocking diodes on the two parallels, it should definitely be #1, but that might be more work than it's worth.

I have no way to monitor panel output outdoors at the strings while shading but I suppose some form of test could still be arranged.

ETA: I mounted the 4 panels (JJN 9BB 100w panels) and will be able to do some testing as soon as I get the cable and parallel connectors.

 

[Mattb4]

Very preliminary testing today I am watching how having the left two panels in parallel and than in series with the next two are functioning. Due to a cellular antenna shading I can observe how that affects the input. When it was over the two parallel panels the wattage production for the string was heavily impacted. However at the time of this image the shadow had moved to only be over one of the parallel set. The wattage jumped up considerably. So it looks like your scenario #1 is accurate, Lakehouse.

 

[LakeHouse]

Very preliminary testing today I am watching how having the left two panels in parallel and than in series with the next two are functioning. Due to a cellular antenna shading I can observe how that affects the input. When it was over the two parallel panels the wattage production for the string was heavily impacted. However at the time of this image the shadow had moved to only be over one of the parallel set. The wattage jumped up considerably. So it looks like your scenario #1 is accurate, Lakehouse.

Cool! Thanks for sharing that. Isn't it nice when a theory actually aligns with what happens in practice?

 

[Mattb4]

In theory I agree.

In terms of peak power today on the panels. Not the best day of the year for output due to Summer soltice (panel angle)and the heat (92F). I started with my previous setup which were 2 - 1200w arrays (#2 array rated at 1190w) before adding these 4-100w uncommon parallel-series connection to array #1. It had been in a 3S4P arrangement. About all I can do to see how doing this changed total production was to look at peak wattage (occurred at 2:45pm in full sun) for each array and see how they compared. At this time wattage from array #1 was at 945w. Array #2 was at 710w.

945w/1600w = 58.75%
710w/1200w = 59.17%

I would say they tracked pretty well with each other allowing for slightly different placement.

If the odd P&S string had limited the #1 array to 1500w the numbers would be
945w/1500w = 63.00%