Announcement

Collapse
No announcement yet.

Wiring solar panels of slightly different voltages

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • Wiring solar panels of slightly different voltages

    I have a few solar panels I want to set up for an off grid system. I have two 260w, two 265w, and two 270w panels. They are from the same manufacturer but have slightly different volts/amps. What is the best way to wire these? Do the slight differences matter? I was Hoping to set up as 3s2p for a 48v system. Any advice is great appreciated!

  • #2
    Originally posted by Gingerjames View Post
    I have a few solar panels I want to set up for an off grid system. I have two 260w, two 265w, and two 270w panels. They are from the same manufacturer but have slightly different volts/amps. What is the best way to wire these? Do the slight differences matter? I was Hoping to set up as 3s2p for a 48v system. Any advice is great appreciated!
    If they are the same # of cells in all cases (they probably are) then not much of a big deal. I'd make them as symmetric as possible, so either two strings of 260-265-270 each or three strings of pairs (i.e. the 260's in one string, the 270's in the second, the 265's in the third) if that works for your voltages. If they are 72 cell panels they probably will work 2s, if they are 60 cell panels, probably not.

    Comment


    • #3
      In a 3S2P arrangement, I'd probably put 260-260-265 in one string and 265-270-270 in the other. The specs are all close enough that it won't matter much, but I think the impact of slightly different Vmp between two strings is less than the impact of a wider range of Imp within a string.
      CS6P-260P/SE3000 - http://tiny.cc/ed5ozx

      Comment


      • #4
        Just to state the obvious, You need a MPPT controller.
        2.2kw Suntech mono, Classic 200, NEW Trace SW4024

        Comment


        • #5
          Pretty simple to figure out, not rocket science.

          Assuming they all have the same cell count (60-cell) your best option with voltage matched perfectly with one of each wattage in a string. Example 3S2P 260 + 265 + 270. The only limitation, and it would be minimal is the current would be restricted by the panel with the lowest Imp rating which would be the 260 watt panel. In a nutshell it turns all your panels into 260 watts so instead of a potential 1590 watts you end up with 1560 watts.

          Let's just call it 1600 watts, you would need a 65 amp controller @ 24 volt battery or 35 amp controller at 48 volts. If it were me I would use a Morningstar TS-MPPT-45 and run at 48 volts. I doubt you can find a 35 amp controller for 48 volt battery. None I would want anyway. I can give you 125 Reason$ to use 48 volts. The 60 amp cost $600, and 45 amp cost $475. It will also save a big chunk of change on wiring.

          Be warned though it takes a 900 pound battery to handle 1600 watts of panels and that beast will cost you $1900 to $2500.

          One caveat here is if they are 72-cell panels again is super simple and you can keep that lost 30 watts. Go 3S2P one string of 260, another with 265, and the third with 270. All matched strings voltage and current matched with full 1590 watts. At 72-cell 2S3P still works with 48 volt battery. 60-cell will need 3S2P
          Last edited by Sunking; 05-22-2017, 08:06 PM.
          MSEE, PE

          Comment


          • #6
            Originally posted by Sunking View Post
            Pretty simple to figure out, not rocket science.

            Assuming they all have the same cell count (60-cell) your best option with voltage matched perfectly with one of each wattage in a string. Example 3S2P 260 + 265 + 270. The only limitation, and it would be minimal is the current would be restricted by the panel with the lowest Imp rating which would be the 260 watt panel. In a nutshell it turns all your panels into 260 watts so instead of a potential 1590 watts you end up with 1560 watts.

            Let's just call it 1600 watts, you would need a 65 amp controller @ 24 volt battery or 35 amp controller at 48 volts. If it were me I would use a Morningstar TS-MPPT-45 and run at 48 volts. I doubt you can find a 35 amp controller for 48 volt battery. None I would want anyway. I can give you 125 Reason$ to use 48 volts. The 60 amp cost $600, and 45 amp cost $475. It will also save a big chunk of change on wiring.
            This guy has lithium batteries, remember? ixnay on the charge controller recommendationay.

            I'll model the mismatch later tonight. I think we'll see that the consequence of voltage mismatch is less than the consequence of current mismatch within the string, so it would be better to not have the 260's and 270's in the same string.
            CS6P-260P/SE3000 - http://tiny.cc/ed5ozx

            Comment


            • #7
              Originally posted by sensij View Post
              This guy has lithium batteries, remember? ixnay on the charge controller recommendation.
              Does not change my CC recommendation as the 45 or 60 amp version can be made to work with LFP.

              As for voltage mismatch if all the panels are from the same manufacture assuming the are either 60 or 72 cell Voc and Vmp should be roughly equal. Only thing that changes slightly is Imp and the 260 watt panel has the lowest Imp of the three. So if you were to configure 2S32P you should have a perfect voltage match, and Imp match in each string. Catch is 2S3P would require 24 volt battery.

              With 3S2P, if you use one of each panel wattage you have a perfect voltage match, and only give up a small amount of current limited by the 260 watt panel.

              Example if they are 60-cell modules wired 3S2P:

              260 watt panel Vmp = 32 volts and Imp = 8.13 amps
              265 Vmp = 32 and Imp = 8.28 amps
              270 Vmp = 32, and Imp = 8.43 a,ps.

              Input to controller = 96 amps @ 16.16 amps 1550 watts

              If you wire 2S3P

              String 1 64 volts @ 8.13 amps = 520 watts
              String 2 64 volts @ 8.28 amps = 530 watts
              String 3 64 volts at 8.43 amps. = 540 watts.

              Input to controller = 64 volts at 24.84 volts = 1590 watts.
              MSEE, PE

              Comment


              • #8
                If there are no plans for expansion, a 30 A mppt CC is enough to handle the 1590 W array. Between real world conditions and panel fouling and degradation, that 33 A value calculated from STC ratings will not be reached often enough to matter.

                With respect to wiring, I modeled this using data from the Canadian Solar CS6P-260P, CS6P-265P, and CS6P-270P panels. All three are in the CEC database, and as polycrystalline panels, are well represented by the 6 parameter model used by the CDC (a refinement of the 5 parameter single diode model).

                Picking a clear spring day using a San Diego TMY file, with relatively conventional mounting on a clear spring day, I'd get the following at solar noon:

                260 W panel... Vmp = 28.87 V, Imp = 8.29 A, Pmp = 239 W
                265 W panel... Vmp = 29.06 V, Imp = 8.39 A, Pmp = 244 W
                270 W panel... Vmp = 29.55 V, Imp = 8.38 A, Pmp = 248 W

                If each panel was individually optimized in the scenario, the output would be 1461.5 W.

                To estimate how they will behave when wired in series and parallel strings, the IV curve shows how each panel would react as it gets pulled off its individual maximum power point as the controller seeks the maximum value for the entire array, IV Curve.JPG






                Let's consider the options.

                2S3P, in which matching panels are put into each string.
                All three strings will be at the same voltage, while each panel within a string will be operating at the same current.

                The maximum power generated in this configuration is 1460.2 W, with each string at 58.3 V and the currents at 8.20 A, 8.36 A, and 8.48 A respectively for the 260 W, 265 W, and 270 W strings.

                Only a 1.3 W penalty for the mismatch sounds good, but having three parallel strings isn't good because fuses are required, and heavier wire is needed to handle the higher current for the same power, relative to the 3S2P arrangements. The voltage might not be high enough to fully support the 48 V battery.

                For the 3S2P options, choices include 260-265-270 in both strings, 260-260-265 in one string and 265-270-270 in the other, 260-260-270 and 265-265-270, or 260-265-265 and 260-270-270.

                In the 260-265-270 configuration, the highest power available is 1457.6 W, with a string voltage of 88.86 V and a current of 8.20 A.

                In the 260-260-265 + 265-270-270 configuration, the highest power available is 1460.5 W, with a string voltage of 87.48 V and currents of 8.25 A and 8.44 A in the lower and higher power strings, respectively.

                In the 260-260-270 + 265-265-270 configuration, the highest power available is 1461.2 W, with a string voltage of 87.43 V and currents of 8.31 A and 8.41 A.

                In the 260-265-265 + 260-270-270 configuration, the highest power available is 1460.7 W, with a string voltage of 87.45 V and currents of 8.30 A and 8.40 A.

                So, in summary, no matter how they are wired, the cost is less than 0.3% relative to the individually optimized scenario, as long as the mppt is doing its job correctly. If you want to go for maximum, then go with 3S2P, keeping the 260 and 265 W panels together, and splitting up the 270's.

                (If your panels are not similar to the Canadian Solar panels used here, please post the actual panels being used)
                (It is possible this optimization is over-tuned for these specific circumstances, but the difference between the options is so small it is hard to go wrong)


                Last edited by sensij; 05-23-2017, 03:37 AM.
                CS6P-260P/SE3000 - http://tiny.cc/ed5ozx

                Comment


                • #9
                  Thank you for all the useful information!

                  They are 60 cell panels and close to the Canadian panels you referenced. They are Hyundai HiS-M260RG/M265RG/M270RG. I know the brand has a big name but do they make good panels? I suppose I shouldn't complain they were free.

                  Comment

                  Working...
                  X