No announcement yet.

Monitoring Panels in Strings

  • Filter
  • Time
  • Show
Clear All
new posts

  • Monitoring Panels in Strings

    If a system is using a SMA inverter (ex. 6000TL-US arranged in 3 arrays 7,7,5) and you want to keep an eye on the panels for premature failure.
    Which monitoring equipment would you need to add and how would it present? I had my reasons for not going with micro-inverters.

    When my dealer/installer brought out the panels, they did not bring them laying down flat on a pallet in a truck. They stood them up on end in a pickup truck and strapped them against
    the rack or window. There was some kind of padding, so I was told (probably opened flattened cardboard boxes on the bed and around the window).

    Long story short they installed half the panels, then felt the need to move the truck a little closer to the edge of the roof. They were in my driveway, so you would think 2ft wouldn't be that difficult.
    Of course the panels were unstrapped by now and when he stepped on the brake after rolling back the 2ft. the remaining 9 panels kept going and dominoed onto the floor of the truck.

    I of course don't know any of this until they had to show up an additional day because the first panel to hit the floor actually shattered the glass in a corner and they took it back to the warehouse and had to bring out another the next day (completely different serial #). I'm always at work so I wasn't seeing any of this, neither was she.

    Anyway the concern is either the fall or the 60min highway trip caused micro cracks in the solder that will reveal themselves. But when, 2-3-5-7-24yrs?
    I complained all during the addition 5 months of the install. Failed inspection 4x's and had even more trouble with having their elec. subcontractor to actually come out and relocate/upgrade the panel and run wiring.

    Still trying to keep a long story short... complained to no end, must have-no exaggeration, just under 100 emails to installer owner and panel manufacturer. They stood their ground and said the panels were tough (hey they must be as I found a foot print on one after I got up there-not even smart enough to wipe it off). Everything fired up fine but I still demanded they individually test each panel and just not a system test (3 arrays 7,7,5). All they did was a multi-meter test of the voltage and gave me an csv file (panel #1 64V, etc...)

    This is when I learned a dirty little secret of the solar industry, "degradation" and how you can't monitor it. I asked how can I monitor the degradation they boast about on their sell sheets with the software they provide me. They said I can't. They assure you they are monitoring at their office with their sophisticated software that will send them alerts if anything is amiss and the manufacturer receives a feed as well. The dealer said they will likely notice something before I will and will come out to change any panel. Righhhhhhhhht!
    They said the only other thing I could do if I felt the panel was no performing was to remove it myself (I purchased my system) and send it to a lab for STC testing and pay for it all upfront of course. And if the panel failed they would reimburse me for everything. But until their software tells them something is wrong they can't do anything because it is performing.

    My concern is that it degrades a little faster then normal because of my unique install experience. I think 3-4 panels could fall below the ANNUAL degradation specs but never be seen, especially if they are on separate strings. 6 panels could easily hide a barely sub-par 7th panel. Then after the warranty complete failure, no 30-40yrs. for me. Obviously complete failures early on will be easier to see even on a string, though probably rather difficult with the generic consumption/production software they give you.

    So any ideas on monitoring or are we going with they're built to withstand hail storms so they should be fine.
    I was looking at the Sunny Sensor Box but they have discontinued it and I figured there was a reason; either it was difficult to keep calibrated or they didn't want you to know. That then sent me down the rabbit hole of how do I prove the Sunny Sensor Box accuracy if I'm filing a panel claim. What if the closest official weather station near me, Long Beach, CA experienced different irradiance on the day in question, etc...

    Maybe checking quarterly with a thermal camera for hot spots might be something to consider?

    Enough of my late night ramblings, any input appreciated.
    Last edited by LLB; 05-15-2018, 07:10 AM.
    19 SPR X21 345 / SMA 6000TL-US22

  • #2
    The good news is that Degradation has nothing to do with how your modules were handled.
    The handling will cause them to fail or not or partially fail (internally the module is broken up into sub strings so you could get a partial failure). These failures are easily detected and different from degradation.

    Degradation can't be measured without sophisticated test equipment and cleaning.

    Further your system does not have optimizers so there is no way to know about individual PV module output.
    You have some indication of the string level production but likely would be hard to see if you lost 1/3 of a single PV modules output at that level.

    SMA purchased Tigo optimizers so you could retrofit the entire system with Tigo (now SMA) optimizers which would give you some monitoring at the PV module level. Though you are likely to have some lose just in all the reconfiguring of the array for that.
    OutBack FP1 w/ CS6P-250P


    • #3
      Experience here is weak panels are most likely to show up at max sunlight, I keep an eye
      on strings with a DC clamp on ammeter. To find a suspected section going into bypass
      mode, some of these taps are inserted along the string at night. Notice the 3rd port is
      kept plugged by a dummy MC4 with the wire input plugged. These test plugs have the
      latches ground off to save time.

      With strong sun and a tap in the middle, check the voltage of each half of the string. The
      voltage per panel should be nearly the same. If a group is bypassing, the voltage across
      those panels will be lower, likely by 1/3 or 2/3 the voltage of a panel, if your panels have
      3 bypass groups. Move all your taps (overnight) into the low voltage area to increase
      resolution, eventually to the individual panel.

      This plan may require some long DC meter leads, my extensions have MC4 connectors
      on the ends (latches ground off). Don't get careless, string voltage can be very dangerous.
      A volt meter may still draw a very small arc, don't even think of disconnecting a connector
      which is carrying output current. Bruce Roe



      • #4
        I'd think that the "stressed" panels would show signs of failure within a couple years, well within warranty period, if the warranty covers rough handling.

        Do you get any information about your array at all, string volts, string amps ??. If you can compare strings, A to B, that makes things easy . Otherwise, log the numbers you get at mid day for a sunny week, and then you have a reference for the future. You will notice, that the daily readings will never match, one day voltage may be 438V @ 15.2A and the next may be 441V @ 15.1A. That's all due to atmospheric changes you can't see, but the panels respond to. Might take 5 years to start to spot some trending, or things can be so variable you will never spot anything till a panel fails.
        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



        • #5
          I use a DC ammeter, simple and effective. But the strings have to have the same orientation, same power characteristics, same shading, etc. to make accurate comparison. Your 5 panel string should indicate 71% of your 7 panel strings.......given of course that all other variables are equal.


          • #6
            Originally posted by DanS26 View Post
            I use a DC ammeter, simple and effective. But the strings have to have the same orientation, same power characteristics, same shading, etc. to make accurate comparison. Your 5 panel string should indicate 71% of your 7 panel strings.......given of course that all other variables are equal.
            The current out of the 5 panel string should be almost identical to the other strings. Voltage, not current, changes with string length.


            • #7
              Agreed, I didn't make clear I knew they were separate (degradation and handling). It was because I wasn't getting anywhere with them giving me new panels that I started looking for other ways to have them replaced within the 25yrs, but hey, read a lot - learned a lot.

              I will look into this a bit more and probably be in touch.

              That's what I'm hoping and I have pictures and emails regarding the foot print on the panels as well as head and forearm prints and yes, the SP warranty even specifically covers installation and "being stepped on". It's just if they dig their heels in how much am I willing to spend on a lawyer.

              I can log into my account and have monthly reports sent showing 15min intervals of the whole install. No per array info. My (2) 7 panel arrays are both South and a East 5 panel array. West has trees. But yes, flying blind. I was hoping to come across a SMA guy running Sunny Sensor Box or Sunny WebBox to see if it would give me helpful info. Perhaps a call to the TED Pro (The Energy Detective) people as their system uses DC amp clamps.

              Thanks for the 2nd on the DC amps. Today was 80 degrees and the panels were 99 at the bottom and 106 at the top and 107 in the middle (roof 130). I have good clearance and there was a slight breeze today. (digital laser thermometer)

              I noticed no one commented on using a thermal camera like the FLIR or FLIR attachment for your cell phone. Supposedly the commercial guys attach this to a drone and fly over the arrays to find the bad panels faster. ??

              Thanks everyone... hopefully the thermal picture attaches correctly, otherwise there are some on the FLIR site or a million more on google images.
              Attached Files
              19 SPR X21 345 / SMA 6000TL-US22


              • #8
                Do you know how the array was divided up into strings? Because evaluating an array's output is so hard because of all the factors and variations, we always try (if possible) to match the strings so you can compare them. String A and string B both the same number of panels etc. If they are matched, its easy to see if one panel goes bad as the string volts and amps will show a big difference.
                BSEE, R11, NABCEP, Chevy BoltEV, >2000kW installed