Announcement

Collapse
No announcement yet.

Strange dip in production on a single panel every day. (not shadow)

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

  • Strange dip in production on a single panel every day. (not shadow)

    EDIT - Issue resolved (sort of) its a firmware issue. More info in post below - https://www.solarpaneltalk.com/forum...966#post421966

    Hi all,

    Ive had this setup for about a year. 15 Panasonic panels with Enphase microinverters. Today I noticed that one panel was generating a bit less than others for the day. I went back day by day and noticed the same pattern going back 3 months to early October.

    I looked at the timelapse of generation, and it appears that this one panel dips to about half of what panels around it are outputting for anywhere between 15 to 30 minutes. Always around the same time of day (around 10am). After the dip the panel goes back in line with other panels. On several days a dip happened in the afternoon as well.

    First instinct was shadows. But it cant be, this panel is surrounded by other panels that dont have this issue. A shadow would have to be cast from the southern direction as thats the only place that has trees around the house (and even their shadows never reach the array, not even close), this panel is on the north side of the array, so if it got a shadow, everything under it would have shadow as well, but those panels are fine. I would also expect a shadow to change its location over course of 3 months and hit other panels. And as far as cloud shadows, this seems way too consistent for that.

    I climbed on the roof to check it out, there was a splatter of bird poop on this panel. I cleaned it off and will see how it does tomorrow. But can bird poop do this? Like specific angle of the sun or something at that time of day cause the generation to drop in half, then recover to full? The splatter was about the size of half of one cell. Im dubious for one major reason, behavior is since October, we have had at least 2 rains since then. I dont think the bird poop is 3 months old. I was able to wash it off without scrubbing with a hose in 1 minute, I dont think sustained 6 hour rain would have had an issue.

    Here is a video of the generation timelapse, 3 random weeks between October and today (but every day is like this). Sorry for recording on phone off screen, I dont have a good way to record a video of the screen right now. Time of day is bottom left, dates top left, bottom of the screen is south. Panel in question is top row 2nd from the right.



    Has anyone seen anything like this before? Is the panel or microinverter going bad? What could this behavior signify? Also a technical question. When using a single inverter all panels can be brought down to the generation of the weakest one. Does each panel internally work the same way? Like if a cell is completely blocked off on a panel, does generation of all other cells also go down to whatever the lowest cell generates?

    Here is a comparison of the problematic panel's graphs vs the panels to either side of it:

    https://imgur.com/IwVvizI

    IwVvizI.png
    Last edited by Duxa; 01-05-2021, 08:36 PM.

  • #2
    It might be interesting to setup a camera on the panels to see if the same bird roosts on that panel each time.
    4400w, Midnite Classic 150 charge-controller.

    Comment


    • #3
      The pattern goes back 3 months.
      Any stink vents or local obstructions such as roof peaks etc. near the panel that has the erratic behavior ?
      What's the array's tilt with respect to the horizontal ?
      Have you tried swapping out the optimizer on the panel with the erratic behavior ?

      Most importantly, what did the installer say when you notified them of the situation ?
      If you didn't notify the installer, why not ? A year old array is under warranty.

      Comment


      • #4
        Originally posted by J.P.M. View Post
        The pattern goes back 3 months.
        Any stink vents or local obstructions such as roof peaks etc. near the panel that has the erratic behavior ?
        What's the array's tilt with respect to the horizontal ?
        Have you tried swapping out the optimizer on the panel with the erratic behavior ?

        Most importantly, what did the installer say when you notified them of the situation ?
        If you didn't notify the installer, why not ? A year old array is under warranty.
        My own install, so I am the installer. Just noticed this today, so starting the process of troubleshooting. Cleaned off the panel today, will see how it goes tomorrow. No vents or any obstructions near that panel. Tilt is 19 degrees, azimuth is 220. If it behaves the same way tomorrow after cleaning Ill try and see if anything is landing on that panel around that time of day. But I dont know that a bird would be that regular for 3 months straight.

        Next step if there is no obvious answer is to contact Enphase and replace the microinverter.

        My goal posting here was to see if anyone has seen this kind of behavior before. This is my first time with solar, so not sure in what ways components can fail (if this is one of the symptoms of a failing component).

        Comment


        • #5
          Very unusual. Either a panel that has an intermittent solder joint or the microinverter has a problem with its mppt algorithm. The pattern is not particularly heat related so I suspect the electronic device. Thing to do is swap microinverters with another panel and see whether the problem follows the panel or the inverter....
          BSEE, R11, NABCEP, Chevy BoltEV, >3000kW installed

          Comment


          • #6
            Verified, no bird

            Talked to Enphase, pulled up some logs.

            It actually seems to dip and then completely drop to 0 production for 5 to 15 minutes before recovering. Here is comparison of it versus another panel.

            Enphase rep said it seems like a software issue. And he will escalate it to SW team. Whats strange to me is, why would a software bug affect one out of 15 identical panels that are all operating under exact same conditions.

            https://i.imgur.com/QA1c3uT.png

            Untitled.png
            Last edited by Duxa; 01-05-2021, 08:36 PM.

            Comment


            • #7
              Ok, got it figured out.

              After chatting more with Enphase and them looking at the Micro, they confirmed this is a known issue that has been going on for over a year with their micros. They have a software fix for it, but due to the scale they are currently rolling it out to only those that reported the issue (with the goal of eventually updating all units regardless if they show the issue or not).

              The reason it is only happening on one of my micros out of 15 is due to the fact that the issue only affects units in specific serial number range. Anything over xx1923xxxxxx serial number can be affected (not all are affected). My other 14 are under that serial. They also said that my case is mild, often the issue causes extremely low production.

              Im guessing that there was a hardware component change starting with that serial number.

              They said that the update will roll out to my unit in 2 to 3 weeks. No way to verify if you got the update, so Im just going to contact them again if its still happening by February.

              Comment


              • #8
                Originally posted by Duxa View Post
                Ok, got it figured out.

                After chatting more with Enphase and them looking at the Micro, they confirmed this is a known issue that has been going on for over a year with their micros. They have a software fix for it, but due to the scale they are currently rolling it out to only those that reported the issue (with the goal of eventually updating all units regardless if they show the issue or not).

                The reason it is only happening on one of my micros out of 15 is due to the fact that the issue only affects units in specific serial number range. Anything over xx1923xxxxxx serial number can be affected (not all are affected). My other 14 are under that serial. They also said that my case is mild, often the issue causes extremely low production.

                Im guessing that there was a hardware component change starting with that serial number.

                They said that the update will roll out to my unit in 2 to 3 weeks. No way to verify if you got the update, so Im just going to contact them again if its still happening by February.
                Thank you for the update. I'll be interested to see how long the fix takes.

                Comment


                • #9
                  Give Enphase another 10 years and I'm sure they have a good product eventually....
                  BSEE, R11, NABCEP, Chevy BoltEV, >3000kW installed

                  Comment


                  • #10
                    My monthly stroll out to my production meter to record my total output is looking better and better as it appears that the vast majority of issues with new PV systems are communications issues versus actual generation issues. All my inverters are dumb so I aggregate the total output and log into a website to report the production for my SREC sales.

                    Comment


                    • #11
                      Update: Today the dips didnt happen, so there is a chance they have pushed the update. The panel did take longer to ramp up than the ones around it, but it peaked higher and generated a bit more overall in the end. Top one is the one that was having issues.

                      https://imgur.com/h7C26Pq

                      h7C26Pq.png

                      Comment

                      Working...
                      X