Question about unequal strings in one MPPT

Collapse
X
 
  • Time
  • Show
Clear All
new posts
  • frmky
    Junior Member
    • Feb 2016
    • 15

    Question about unequal strings in one MPPT

    We recently had solar designed and installed by Sullivan Solar in southern California. Limited by roof space, they designed it as a 15 panel south facing array and a 9 panel west facing array using Kyocera 270W panels and one SMA SB6000TL-US-22 inverter. After installation, I discovered that the south array was split into two strings, one with 7 panels and the other with 8 panels, connected in parallel to one MPPT input while the west array is connected as one string to the second MPPT input. From the research I've done, connecting unequal strings in parallel isn't recommended, but how bad is it? How much efficiency am I losing from this configuration?

    I have asked them to exchange the inverter for a SB6.0-1SP-US-40 inverter, which was recommended by SMA for this configuration, but they refuse. They state that it's currently exceeding production estimates for February and they will monitor it going forward.

    Last edited by frmky; 02-24-2016, 04:02 PM.
  • inetdog
    Super Moderator
    • May 2012
    • 9909

    #2
    The installer is marginally incompetent.
    You are probably losing as much as 5% of the production of the west array.
    Interestingly enough, it would be better to split the South array into 6 and 9 and put the two 9 arrays in parallel into one input and the south 6 into the other.
    The south and west strings will not produce max current at the same time but their voltages will match well.
    A good installer would know that. There are papers by SMA on the subject.
    Production estimates are not an indication whether it is working well.
    SunnyBoy 3000 US, 18 BP Solar 175B panels.

    Comment

    • sensij
      Solar Fanatic
      • Sep 2014
      • 5074

      #3
      Originally posted by inetdog
      Interestingly enough, it would be better to split the South array into 6 and 9 and put the two 9 arrays in parallel into one input and the south 6 into the other.
      That is a clever solution. With a Vmp of 31.0 at STC, or 27.9 at NOCT, a 6 panel string would be operating at around 170 V. That is within the inverter's DC range, but out of the MPPT range, so there would probably be some efficiency hit on that string. However, the net result should still be better than what the OP has now, as long as there isn't regular shade on the short string that might knock it out of the DC range completely.
      CS6P-260P/SE3000 - http://tiny.cc/ed5ozx

      Comment

      • frmky
        Junior Member
        • Feb 2016
        • 15

        #4
        The installer is strongly discouraging a 6-panel string since it is below the MPPT range of the inverter. He is worried it will be very inefficient on a hot day. They have offered to move a panel from the south to the west roof, but that does not sound like a good option.

        Edit: There is no shading of the panels, so that is not a concern.
        Last edited by frmky; 02-24-2016, 08:10 PM.

        Comment

        • solar pete
          Administrator
          • May 2014
          • 1816

          #5
          Originally posted by frmky
          We recently had solar designed and installed by Sullivan Solar in southern California. Limited by roof space, they designed it as a 15 panel south facing array and a 9 panel west facing array using Kyocera 270W panels and one SMA SB6000TL-US-22 inverter. After installation, I discovered that the south array was split into two strings, one with 7 panels and the other with 8 panels, connected in parallel to one MPPT input while the west array is connected as one string to the second MPPT input. From the research I've done, connecting unequal strings in parallel isn't recommended, but how bad is it? How much efficiency am I losing from this configuration?

          I have asked them to exchange the inverter for a SB6.0-1SP-US-40 inverter, which was recommended by SMA for this configuration, but they refuse. They state that it's currently exceeding production estimates for February and they will monitor it going forward.

          http://www.solarreviews.com/installe...power-reviews/
          Hmmm..sounds a bit weird, I met with the guys from Sullivan last year and I was quite impressed, so keep talking to them, I am sure they will want to get it right, good luck keep talking to them.

          Comment

          • frmky
            Junior Member
            • Feb 2016
            • 15

            #6
            I forgot to link earlier... I'm uploading the data from their monitoring to pvoutput.org. You can see the system production here:
            PVOutput.org - share, compare and monitor live solar photovoltaic output data

            Comment

            • frmky
              Junior Member
              • Feb 2016
              • 15

              #7
              After a couple more conversations they decided that a 6-panel string would be ok. Sullivan came out this morning and rewired it as two 9-panel strings (one west-facing and one south facing) in parallel on one input and a 6-panel south-facing string on the other. Thanks for your advice!

              Comment

              Working...