Two parallel strings with different voltage

[bwkemp]

I know this belongs in the water pump category but figured people would be able to answer my question and see it here quicker.

On a grundfos solar pump we are wanting to gain more hours in the day by facing the existing two panels south west, and adding two more panels south east. I have done this before with panels that were the same, but in this case the old panels are 225 watt Vmp29.6v Voc 36.7v and the new are 265 Vmp 31.4 voc 38.7

They are wired in series and i will add another set of series panels wired in parallel with the existing one.

I'm not to worried about the watts difference as it works fine in the other instances even when half are shaded and the other half are in full sun.

I could rearange the panels so each set has one new and one old panel so they match but i would rather not have to just for ease of installation and mounting differences.

Let me know what you think
Brian

[inetdog]

I know this belongs in the water pump category but figured people would be able to answer my question and see it here quicker.

On a grundfos solar pump we are wanting to gain more hours in the day by facing the existing two panels south west, and adding two more panels south east. I have done this before with panels that were the same, but in this case the old panels are 225 watt Vmp29.6v Voc 36.7v and the new are 265 Vmp 31.4 voc 38.7

They are wired in series and i will add another set of series panels wired in parallel with the existing one.

I'm not to worried about the watts difference as it works fine in the other instances even when half are shaded and the other half are in full sun.

I could rearange the panels so each set has one new and one old panel so they match but i would rather not have to just for ease of installation and mounting differences.

Let me know what you think
Brian

If you are not considering shading, the rule of thumb is that you should not put two strings in parallel if the Vmp of the strings differs by more than 5%.
What this means is that if the difference is more than 5% the total power will be noticeably less than the sum of the two string power values. But if they are even within 10% or 20% the result will be usefully greater than either string alone.

The difference between 29.6 and 31.4 is slightly more than 5% but in my opinion is low enough that it will probably not be worth splitting the existing panel string to "match" the panels.
The effect of paralleling two strings with different shading near peak sun hours into one MPPT input, on the other hand, may be extreme.

When the two strings face different directions, as different as SE and SW, you can pretty much ignore the 5% or 6% difference. It will not hurt you much at all in terms of the total power for the day, and it will give you the extended pump hours you are looking for.

[bwkemp]

Thanks that is the answer i was looking for!

Hopefully this will keep my customer from getting behind as bad (2 gpm well for livestock) We think this should gain him 4 hours or so of run time so aprox 500 gal a day.

You can move this thread to the solar water pump forum if you want to.

[bwkemp]

I would also like to add that in respect to shading one of the panels (or set of two) while wired this way, almost any shade seems to take away the power output from the whole panel (or set of two) So you need to make sure and have enough panels to run the pump on just the one (or two) that are in the sun.

[Raul]

I was always under the impression that diferent orientation will require separate MPPT's . If the one in the shade outputs 30w and the one in the full sun is paralleled to it , then the output of the 2 combined will be 60w . The one in the sun will only contribute to the level of the shaded one even if its potential is greater.

[inetdog]

I was always under the impression that diferent orientation will require separate MPPT's . If the one in the shade outputs 30w and the one in the full sun is paralleled to it , then the output of the 2 combined will be 60w . The one in the sun will only contribute to the level of the shaded one even if its potential is greater.

You have to be careful about the wording here, since there is as big difference in effect between partial shade (with some objects casting shadows on the panels) and low insolation from the sun hitting different panels at different angles.

Partial shade will reduce the current from some section of a panel or panels and that will cause the bypass diode(s) to be active and the Vmp of that panel, and therefore the string, will be reduced from the nominal value. If two strings have a different Vmp because of shading you cannot run them into a single MPPT input without some loss of output.
But if all of the panels in a string are producing less power because the sun is not hitting them head on (i.e panel facing SW while sun is only at SE at the moment) that will reduce the current but will not significantly change the Vmp of the string.
So two unshaded strings facing in different directions can be combined into a single MPPT input. Theory says so and at least two CC or GTI manufacturers have confirmed this with practical experiments and published the results.

If you have more than one MPPT input available, there is no reason not to use them even if you do not have to.

[Raul]

Got it , so different orientation but shade free will not affect the output since vmp is stable , right? once shading is involved then vmp will be affected wich dictates the output contribution?

[inetdog]

Got it , so different orientation but shade free will not affect the output since vmp is stable , right? once shading is involved then vmp will be affected wich dictates the output contribution?

Yup.

[bwkemp]

I'm not sure if it is the same kind of mppt as a grid tie inverter or not but here is what we have done in the past. With a lorentz solar pump (i presume grundfos would be the same, but i dont think we have any of them running this way so not real sure) we plugged in one panel, measured gpm output at 40psi If you plug in another panel in parallel, covered with a peice of cardboard compleatly there was no noticeable change to gpm or pressure, even plugging/unplugging while running no noticeable change (i would say less than 2% but definatly less than 10%). If it wasnt covered and in sun the gpm would increase as expected, if the extra panel was pointed directly away from the sun had little effect compared to covered with cardboard.

The lorentz rep says the engineers say it shouldnt work when one panel is shaded, but obviously it does, that has been a few years back not sure if the engineers figured out why it works yet or not.

We have aprox 4 or so pumps running this way up to 6 years old or so with no problems. (they all have matched panels)

Brian

[inetdog]

It works for two reasons:

1. A single panel is producing more power than the pump needs for your head and flow rate.
2. All panels these days contain bypass diodes so that the current from the working panel just bypasses (!) the shaded panel and it works as if you only had the one panel.That is if you are putting them in series. If you are putting them in parallel the shaded panel just does not contribute current but it does not drag the voltage of the other down.

That indicates to me that you could put two panels in parallel and aim one close to west and the other close to east you could probably keep the pump going 8 or more hours in summer. If it drops out in mid morning and mid afternoon you might need to put two panels aimed in each direction to really extend the running time.

[bwkemp]

I agree that is how it works. the interesting thing to me is that it seemed to have no or not noticeable negative effects by adding an extra panel in parallel to the first even completely shaded in the test. I was surprised that the bypass diodes didnt draw the voltage down a bit when wired in parallel. unshaded it did add to the gpm at same pressure so the pump was not completely overpowered by one panel.

I also agree that you need enough panels to run the pump at desired head (psi) and flow rate (gpm) with each set of panels pointed in different directions.


For your #2.
We also tried this test wired in series and anytime you added shade to the second panel the flow rate was much lower than if you just had one panel in sun (i presume the bypass diodes just draw the volts too low for the pump to run due to its min voltage requirements)




Now the only reason you would do this (point panels in two directions) is if you need to gain hours per day, if your supply tank and well are both good enough you would gain more by pointing all the panels south (with a properly sized pump). But if your well cant yield that much than you need more run time even if it means not getting the most total watts out of your panels because one set of them are shaded 1/3 of the day.

It has the same effect as if you had a tracker rack, only for our relatively small (1-3 180-280 watt panel) systems it is cheaper to buy the extra panels than a tracker rack and you gain a lot of available watts during the mid day hours so if it is cloudy it can still run while half that many on a tracker rack wouldn't. (also no moving parts like a tracker rack)


Thank you for answering my original question inetdog, the rest here is just for some nice reading material.

[inetdog]

Bypass diodes in one panel will not "draw down" a panel connected in parallel with it. They only become active when panels are in series or when all of the parallel panels are partially shaded.

[Guest]

I have two rows of Solar Panels. Each row has 10 panels of 325w each. Both rows are placed one over the other. All panels are attached in series of 10 panels each.Thus forming two strings of 10 panels each. My problem is that the two strings are giving different output, with a difference of up to 250w DC during peek. Can someone please guide me to this reason of difference in output in spite of identical strings.

[SunEagle]

I have two rows of Solar Panels. Each row has 10 panels of 325w each. Both rows are placed one over the other. All panels are attached in series of 10 panels each.Thus forming two strings of 10 panels each. My problem is that the two strings are giving different output, with a difference of up to 250w DC during peek. Can someone please guide me to this reason of difference in output in spite of identical strings.

Hello Vivek Gupta and welcome to Solar Panel Talk

While an 8% difference (3000w / 3250w) doesn't seem like much, there maybe a few different issues with your system.

1) There is a wiring problem with the string that produces less wattage
2) There is a shading problem on the string that produces less wattage
3) There is less sunlight getting to the string that produces less wattage

Besides the above issues there may be something wrong with your inverter and how the strings are wired to it.

[Guest]

Thank you SunEagle for a quick reply.

However on a closer scrutiny I found that in morning string 1 is producing higher than the string 2 and in afternoon or early evening string 2 takes the lead. So it must be the sun playing tricks along its path. In case I find something else then will post it in next few days.

Thanks.