Announcement

Collapse
No announcement yet.

What's up with these panels?

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

  • What's up with these panels?

    Quick background: Bought an off-grid cabin in Dec 2020 in the Pacific NW with an existing DIY solar power system. It works fine for our minimal needs as a weekend/vacation place (our well pump runs, our phones stay charged, our little refrigerator stays cold). Since the sun has come back, I've started exploring it with my multimeters mostly to educate myself on how it works, but also so that eventually I can plan out some upgrades so we can spend more consecutive days out here (especially once the sun leaves again for the winter). While working on the array, I ran into a situation I didn't understand and am curious what y'all think about it.

    The array has a combined nameplate rating of about 2000W, but it's composed of a variety of modules of different ratings. They are rather mismatched, and I've never gotten more than about 800 to 900 watts out of it. Here are the nameplate specs:
    Panel # 12 11 10 9 8 7 6 5 4 3 2 1
    Brand Sharp ?? ?? ?? ?? ?? BP Solar Evergreen Solar Evergreen Solar BP Solar MX Solar USA MX Solar USA
    Model NT180U1B tyns-195S6 tyns-195S6 tyns-195S6 tyns-195S6 tyns-195S6 BP590 EC-102-GL EC-102-GL BP585 MX 60 MX 60
    Pmax 180 195 195 195 196 195 90 102 102 85 235 235
    Vmp 35.86 25.05 25.14 25.08 24.97 25.19 18.5 16.7 16.7 18 29.3 29.3
    Imp 5.02 7.81 7.76 7.78 7.86 7.77 4.86 6.11 6.11 4.72 7.84 7.84
    Voc 44.08 33.14 33.25 33.18 33.09 33.29 22.3 21.3 21.3 22.03 36.8 36.8
    Isc 5.6 8.63 8.6 8.57 8.62 8.62 5.2 7.32 7.32 5 8.36 8.36
    These were all wired in parallel EXCEPT 6+5 are in series and 4+3 are in series, I assume to get their combined voltage up enough to be useful. These all fed into a Midnite Classic 150 which is charging a 24V 400AH lead acid battery bank.

    One thing I observed is that when the batteries are full and the Midnite is in float, the array voltage rises to the 33-35V range. At this point, the Midnite is only sending about 1A through the batteries, but my DC clamp meter shows that the higher voltage panels are still producing several amps, most of which is flowing backwards through the tyns panels (because the array voltage is higher than their open circuit voltage?). I imagine this is not good for them, but I wasn't sure how serious a problem this is.

    While exploring that, I completely disconnected all the panels to measure their Voc and Isc, on a moderately sunny day where temps were in the mid 50s. The tyns panels all measured about 6-7A Isc, and Voc was around 31V (as opposed to the 33V nameplate). Is this within the normal variation from the lab conditions on the nameplate, or indicative that the tyns panels are damaged?

    Finally, as a second experiment on the tyns panels, I tried connecting four of them into a 2x2 array, and connected this to the Midnite (all other panels were still disconnected). At the time I did this, my batteries had not yet completed bulk charging, so the Midnite began charging in MPPT mode. It saw that the open circuit voltage of this 2x2 array was 62V as expected, but still found that the maximum power point was about 26V -- I think this is the lowest it could go while charging the batteries -- yielding about 350W. When I measured the voltage across each individual panel, one in each string had dropped to 2-3V, with the other panel in the string carrying most of that 26V. I didn't understand this at all, but it didn't seem good either. Are there other ways I can check these out to understand what they're doing?

    Since then, I've split the 5 tyns panels onto a Trace C40 charge controller that the previous owners left, to keep them from dissipating power from the other panels (which are still on the Midnite). They seem to be able to help out with bulk charging this way, but gradually drop off as the battery voltage rises.

  • #2
    Depending on how those 12 panels were wired I am concerned that the lower Vmp rated ones are hurting the performance of the higher Vmp rated ones.

    Comment


    • #3
      With a Midnite 150 controller, that array should be wired as series strings of three to five panels. They probably started out as a few panels on a PWM controller and just never wanted to change. You need to series them up so the strings more or less match but stay less than 150Voc. Sorry that it is so many different panels - that will always be a compromised situation...
      BSEE, R11, NABCEP, Chevy BoltEV, >3000kW installed

      Comment


      • #4
        Yeah, I think it did start out as a few panels wired into a PWM controller.

        OK, that's a more aggressive rewiring than would've occurred to me. After playing around in Excel a little bit, I came up with the following, sizing each string to be comfortably under 150 volts, and estimating each string's expected current as the minimum Imp on each string.

        String 1: Panel 10+9+8+7, 132.81 Voc, 100.38 Vmp, 7.76 amps
        String 2: Panel 12+6+5+4+3, 131.01 Voc, 105.75 Vmp, 4.72 amps
        String 3: Panel 11+2+1, 106.74 Voc, 83.65 Vmp, 7.81 amps

        So then assuming the entire array would run around the lowest Vmp, I'd estimate roughly 83.65 volts at 20.29 amps, or around 1700 watts, give or take.

        I can get the voltages on the strings to match more closely, but it spreads the low current panels out more, which seems to hurt the total power more than this configuration. at least by this way of estimating.

        Does this make sense to actually attempt?

        Comment


        • #5
          I'd try that configuration out. As long as no low amp panels are in series with higher amp panels.
          Powerfab top of pole PV mount (2) | Listeroid 6/1 w/st5 gen head | XW6048 inverter/chgr | Iota 48V/15A charger | Morningstar 60A MPPT | 48V, 800A NiFe Battery (in series)| 15, Evergreen 205w "12V" PV array on pole | Midnight ePanel | Grundfos 10 SO5-9 with 3 wire Franklin Electric motor (1/2hp 240V 1ph ) on a timer for 3 hr noontime run - Runs off PV ||
          || Midnight Classic 200 | 10, Evergreen 200w in a 160VOC array ||
          || VEC1093 12V Charger | Maha C401 aa/aaa Charger | SureSine | Sunsaver MPPT 15A

          solar: http://tinyurl.com/LMR-Solar
          gen: http://tinyurl.com/LMR-Lister

          Comment


          • #6
            As Mike said, try that configuration out. The all parallel configuration (except
            3, 4, 5, and 6) might have been a decent panel matchup, except perhaps it
            could not deliver enough voltage to charge 24V at full power.

            The Voc and Isc is a decent sanity test, but not enough to assure a panel is
            fully functional. Checking panel voltage in a string under load is a good test.
            When you saw a panel drop to a couple volts, it is unable to keep up (deliver
            as much current as the other panels), and defaulting to conducting circuit
            current through the bypass diodes instead. Here that seems most likely to
            happen at max sun and load current. Sometimes panel voltage will drop to
            2/3 or 1/3 of normal voltage, because 1 or 2 bypass diodes (typically there
            are 3 panel sections each with a diode) are conducting. This I call a defective
            panel, as opposed to a poor array configuration. Check them out.

            Using the standard PV cell circuit model, sun provides current, which may be
            drawn off by a lower voltage load. If the voltage is allowed to rise, the diode
            will begin conducting the current, increasing to all internally generated current
            at Voc. If load is small and panels increase toward Voc, some current from a
            higher voltage panel may flow over to a lower voltage panel to join its internal
            current. This would be normal in an unloaded array (which never happens
            here with net metering), I would expect no damage if the current stays below
            a panels normal output max current. Bruce Roe

            CelMod.png

            Comment


            • #7
              Thanks Bruce (and all), to check my understanding on the PV cell model (last EE class was 20 years ago and only concerned "ideal circuit elements" ), that "diode" should be able to handle twice the current that its associated cell can produce, so an example panel with a 7A Isc should be able to tolerate having 7A pushed backwards through it by a stronger part of the array. I guess in this case I don't know how to reason about the 25V difference in Voc between string 3 and the other two as t0 how much backwards current might flow.

              Also, in this case, these are connected up through a combiner box -- I recall reading somewhere that DC circuit breakers have a polarity and don't like to be run backwards. If current ends up flowing from the high strings backwards through the breaker to the low string when the array is unloaded, is that a cause for concern (at least at the potential currents we're talking about)? I guess that is happening some today in the mostly-parallel configuration anyway.

              In any event, we are heading out there again this weekend, so if I get a chance where I can take the array offline for a couple of hours, I'll try this out..

              Comment


              • #8
                OK, I got a few minutes over the weekend to experiment on the array, here are the results. In each of these cases, I connected just this one string to the Midnite, and it was in Float MPPT mode from our loads. I watched for the Midnite to do a sweep to pick the max power point. We had some high, thin clouds, but sun was pretty good and no shade on the array.

                1. Panel 10+9+8

                Voc: 88.9V
                Vmp: 29.1V
                Imp: 6.7A

                While operating, panel 10 measured 8.2V, panel 9 3.5V, and panel 8 17.4V

                2. Panel 10+9+8+7

                Voc: 119.2V
                Vmp: 39.0V
                Imp: 6.7A

                Panel 10 2.9V, Panel 9 3.2V, Panel 8 20.4V, Panel 7 12.5V

                3. Panel 8+7

                Voc: 59.7V
                Vmp: 27.0V
                Imp: 6.9A

                Panel 8 9.1V, Panel 7 17.9V

                Given these results, it didn't seem worthwhile to try running them as a parallel string with any other reconfiguration of the array, so I stopped there. Also, my combiner box breakers are only rated to 125VDC so the 130V configurations are no-go anyway.

                Seems like these panels probably just need to be replaced if I want more power.

                Comment

                Working...
                X