Shadow handling

Kommentare

15 Kommentare

  • Avatar
    Matthias Oliver
    Early Adopter

    Looks like B3 is optimizing to me. You have all but a sliver of sunlight on one side of that module. I bet all the bypass diodes are going crazy. The optimizers aren't magic; they can't create power where there isn't any to be made. I'd say you're lucky to get the 28 watts. 

    1
  • Avatar
    Marco Maljaars

    Hello Matthias Oliver thanks for your comment.

     

    I think you didn't look at the time laps video or at the graph in the video screen shot.

    There you can clearly see that at 14:00 the B3 panel is only covered by 10% (even less) of shade.

    But the output is only 35 Watts if you look at the Tigo web interface.

     

    So there is no optimization done by the Tigo.

    The B3 panel should definitely produce more at 14:00.

    It remains at an average of 30 Watt from 12:30 all the way up to 14:00. 

    Only when the B3 panel is in full sun without shade the Tigo comes to life, which does not make sense.

    0
  • Avatar
    Curtis Community Moderator

    Hello Marco Maljaars

    Thank you for contributing this great question to the Tigo Community!

    The images you have provided, shows a classic representation of the subject in this article: 
    Which orientation is best for PV Modules? (vertical or horizontal)

    Depending upon the panel's orientation (vs the internal diode layout), you may be experiencing a partial or complete shut down of the internal panel diode during extreme shade events. The production of the PV Module (when 1-2 diodes close), can be assisted by Tigo's Optimizers (depending upon the voltage allowed per diode). However, if 2 or more diodes are closed, it can present a barrier above the functional expectations of the product. 

    0
  • Avatar
    Marco Maljaars

    Hello @...

    Thanks for this information.

    But it's not truly the full case here as described in the article.

     

    The B3 panel delivers to the Tigo an input Voltage of 38V at 14:00.

    So the Tigo is activated, it gets enough voltage from the B3 panel to do it's job (the entire day).

    The B3 panel is equipped with three diodes (three sections)

    At full power at 14:38 the B3 panel delivers approximately 288W.

    As can be seen in the time laps video at 14:38:

    https://drive.google.com/file/d/12ytVt0i6AAjcQqfhY-67R10B2P_pGzhQ/view?usp=sharing 

     

    So with one third of shade (which is the case at 14:00) it generates approximately: 288/3 = 95W per diode section.

    For the unshaded part, this would be approximately 2x95W = 190W.

    Of course there are loses, this is basic calculation, but a considerable difference with the 35W output with a Tigo.

     

    So the Tigo is actually dragging the panel down for this particual kind of shade.

    The panel delivers 35W with a Tigo, which could be 190W.

    Because the Tigo supersedes the bypass diodes of the panel, so they can not engage.

    Which is okey as it's not (yet) possible to develope this.

    It does it job perfectly for shade from bottom to top (or vice versa) which can be seen for panel B2 from 10:30 till 10:50.

     

    But for hard shade from left to right (or vice versa) like panel B3 the Tigo has no additional value?

    For monitoring, warranty of the panel and complete string control it does, but not for this kind of shade and panel configurations.

    0
  • Avatar
    Matthias Oliver
    Early Adopter

    I watched the video and looked at your in-depth analysis, and although I appreciate this kind of monitoring, this one module, in the grand scheme of things is the least of your worries. Shade is difficult to analyze without other tools, but this array is all over the place, which causes mismatch losses, and several of the modules receive hard shade throughout the day. It seems that Curtis is spot on in the analysis.

    Before these optimizers came out, we didn't install modules in the shade, since they work better when sunlight hits the full face. I have seen this hundreds of times in the last 10 years with SolarEdge optimizers and Enphase microinverters. They aren't miracle devices. They will try to recuperate as much as they can, but they can only do so much. What is your yearly yield with this system? That is what I would focus on.

    2
  • Avatar
    Curtis Community Moderator

    Hello Matthias Oliver

    Thank you for your reply and valuable input. 

    Marco Maljaars, I am opening up a ticket with our Support Team, so they may fully review your system and provide any additional insight which may help your production. A Support Agent will be in contact very soon. 

    In the meantime, we thank you for your patience while they take a look. 

    0
  • Avatar
    Marco Maljaars

    Hello Matthias Oliver

    Thanks for your opinion. Of course shade can be difficult therefor it’s very interesting to see how a split cell panel behaves with a Tigo. The time laps together with the Tigo interface gives you a very clear picture on how this behaves. Especially if panel has shade left to right, Tigo doing nothing. Or shade bottom top, Tigo is in its element for a split cell panel.

    Let's be a little more nuanced, although the panels are spread out on string B, they all face west, 4 at 15 degrees and 3 at 45 degrees. The Tigo’s are doing their job for the tilt orientations as advertised. But not for a particular kind of shade which I explained earlier and the data shows.

    Again I do not consider Tigo as the “miracle devices”. Understanding their behavior in relation to the yearly yield is where I focus on. Especially if an optimizer is dragging one panel down on a daily base in the solar season which shouldn’t be necessary.

    If you read the article carefully and what I already explained in my earlier post it’s not fully related to a system with split cell panels and this orientation of the panel. I really appreciate the support of @... and Tigo in general, it’s a great community and product. But this article needs some update.

    So I removed the Tigo from B3. You can clearly see the diodes in the B3 panel are doing their job now (handling the shade) instead of being superseded by the Tigo and therefore not activated and dragging the panel down. The graph is in the next post which is pending approval...

    @... thanks for your support.

    2
  • Avatar
    Marco Maljaars

    In the evening on June 9th 2023 I removed the Tigo of panel B3.

    The total string B output was analyzed with the string B data from the Growatt inverter. With the results shown below.

    June 9th and 10th were stable day conditions. Same outside air temperature and no clouds with similar irradiance.

    B3 09:50 – 10:35

    In the morning B3 has shadow from bottom to top. As this is a Jinko Split Cell panel the Tigo is capable to optimize the shadow by generating power for the unshaded upper part of the panel (orange line). Once 50% is shaded the panel won’t deliver anymore as expected, Tigo are not miracle devices (blue line and orange line are similar now).

     

    B3 12:21 – 14:41

    The shade moves from left to right. With a Tigo the panel diodes are not activated. The Tigo supersedes  them so the panel is not delivering (orange line). With the Tigo removed (blue line) the panel diodes are activated. The B3 panel is generating power for the unshaded thirds. Instead of idling at an average of only 35 watts for almost 2.5 hours with a Tigo installed, until the shade is gone. This is where the orange line increase significantly.

    It's interesting to see that you can clearly see three steps in the blue line. Each time a diode is activated the B3 panel starts to deliver. As there are three diodes there are three steps in the blue line.

    Split Cell and a Tigo

    In the morning the B3 panel is losing without a Tigo. But during peak hours the B3 panel is gaining without the Tigo. This is just one panel, considering a system with multiple panels with this kind of shade, the small numbers will increase significantly over time.

    It’s not just a matter of putting a Tigo on a panel, you really have to look how shadow develops on your solar system.

    I removed the Tigo and leave it off.

    0
  • Avatar
    Curtis Community Moderator

    Hello Marco Maljaars

    The answer may be very simple:

    • The TS4-A-O datasheet shows a min/max voltage requirement of 16-80V.
      (optimizers receive their operational voltage of 16V from the PV Module)

    At 50% shade, the panel's 2nd diode may be experiencing enough resistance to shut down. If so, the PV Module's voltage output would be reduced to one diode's worth of voltage (11.61V), which is below the base operational voltage for the Tigo unit (so, it would shut off / removing itself from string). This reaction allows the string current to remain high, but the overall string voltage output is lowered by one panel (-34.84V).

    Optimization for current is considered an advantage because most shading events tend to be during morning or evening (when current is at it's lowest). However, your system appears to be experiencing a hard shade event during max irradiance (around mid-day), so the timing is not optimal for the height of your bell-curve.

    Is it possible to move the module, out of the range of the hard shade event? if so, you may see an increase in your power output for that time of day. 

    I also suggest that you continue to work with the Tigo Support Team to see if they can offer any other solutions. 

    0
  • Avatar
    Marco Maljaars

    Hello @...,

    Thanks for your valuable feedback. This seems the case for shadow bottom to middle in the morning. 

    The Tigo remains active and is doing it's job. Until the middle is reached and the Tigo shuts down. 

    It's below the 16V of the Tigo.

     

    I remain head scratching why the Tigo remains inactive on B3 in the afternoon with shade from left to right. 

    Even if 2/3 of the panel is in the sun. Only in full sun the Tigo activates.

    With the Tigo removed the B3 panel starts delivering once 1/3 of the panel is in the sun all the way up to 3/3.

    Since the Tigo is disconnected from the B3 panel the string B delivers more output.

    Despite the fact that it's loosing a bit in the morning. the overall performance is increased.

     

     

     

     

    0
  • Avatar
    Curtis Community Moderator

    Hello Marco Maljaars,

    Let's re-review some of the images you have supplied. 
     
    Image #1: Shows conditions that would cause both the lowered output of the B2 module, and the expected drop of the B3 unit (due to diode closure).

    Image #2: Taken later in the day (during full irradiance), the B3 is only under 1/3rd hard shade (but stays inactive until 14:00).

    The answer is simple: "1/3 > 1/4"

    Basically, 2/3rds of a Module is a greater differential than what the 25% Mismatch Rule allows for... Any unit that has experienced a hard-shade shutdown, will remain inactive until it has achieved a 75% (or greater) current match, to the output of other modules on the string. 


    🤔 FAQ's

     

    • "Is this a failure or a feature?"
      The 25% mismatch rule is definitely a Feature! It allows the Optimizer(s) that are experiencing high resistance to remove themselves, so that the module does not affect the current production of the entire string.
      Note: This is great for systems that are experiencing diode closure and mismatch issues during periods of low irradiance (mornings/afternoons), but can seem like a nuisance to the few systems that experience hard shade conditions at high noon.
    • "How can I get the B3 Optimizer to activate sooner?"
      The only way to get production to activate sooner, is to get more sun on that module, earlier in the day.
      Here are some suggestions: 
        1. Move the module to a less shady location/orientation.
        2. Provide an angle focused reflective surface (ie., mirror or zinc white), to redirect irradiation onto the module during early afternoon hours.

    • "Should I remove the Optimizer from the module?"
      Removing the optimizer from the B3 module allows it's resistance to negatively affect the entire string's current output (due to resistance and Ohm's Law). It may also cause safety and/or legal issues (depending upon your AHJ / local laws), as the Optimizer also provides module-level overcurrent protection, and rapid shutdown for first responders. 

    I hope that this sheds some light on the subject, and will help you with the design questions that make you scratch your head. 🤓 

    If you have more questions about your production, please feel free to reach out to the Customer Support Team.

     

     

    1
  • Avatar
    Marco Maljaars

    Hello @...

    Thanks a lot for your time and effort I really appreciate the time you (and your team) take to answer mine (and other members) questions.

    • Now the 25% mismatch feature makes sense. Indeed the 25% rule can be a feature but in systems with hard shade like B3 it is a nuisance, it bites the string. As what can be seen in the video and graphs. But still... how do you explain that at 10:51 the B2 panel is shaded for 45% but delivers 123W?
    • Regarding the removing of the Tigo from B3. That's interesting, that it affects the entire current output of the string. I don't see this in my daily production. If I look to the MPPT current of string B with a Tigo it's about 8.2 Amps with 187.3V (13:50). Removing the Tigo the current is about 8.4 Amps with 199.7V (13:50). With 2/3 of the B3 panel in the sun (1/3 shade). In total as you can see in my graph (see my previous post) the string B has a much better performance (over several days). It seems the diodes in the panel are now doing their job. In the Netherlands we don't have (yet) AHJ rules for safety etc.

     

     

     

     

     

    0
  • Avatar
    Curtis Community Moderator

    Hello Marco Maljaars

    Thank you for your reply. 

    Here are the answers to your questions: 

    • Data provided by the Tigo System (and possibly your inverter) is cloud-based and may include up to a 30 minute delay for data processing, so the timing may be a little off. However, Tigo's MLPE technology uses impedance matching to gain the most current possible out of the panel (when in shaded conditions). For this reason, you may see a greater yield than expected. 😉

    • The B3 panel should be reducing the module's voltage and 'bucking' for more current (so the results seem reversed). Did you record these on the same day? The reason I ask is: Solar irradiation can be a tricky thing to calculate in a non-scientific (organic) setting. Results that are recorded only a couple minutes later in the day (after removing the unit), can include stumbling blocks like slight differences in shading, temperature and string reaction to startup and shutdown periods. In a controlled (STC/NOCT) environment, the Tigo MLPEs consistently perform as expected. 

     

    0
  • Avatar
    Marco Maljaars

    Hello @...

    On a split cell panel yes the Tigo is gaining the most current possible for shadow bottom to middle. You can see this in the graph below.

    On a split cell panel a Tigo is not capable to gain the most current possible for shadow left to right. You can see it clearly in the graph below.

    It's doing nothing until the panel is completely unshaded. Then it starts regulating the panel it receives enough voltage now.

    Temperature and irradiance were similar on June 9th and 10th.

     

     

    So if there is shade on a Split Cell Panel and the vertical shadow is most of the day at the bottom of your Split Cell Panel(s) you definitely need a Tigo.

    If the shade is horizontally (left to right) the Tigo is dragging the panel down, the panel won't produce.

     

    If there is a shade combination over the day you have to look how long each shade is present and look at the overall day production.

    Which in my case the B3 is better of by removing the Tigo as my system has Split Cell Panels with hard horizontal shade in the middle of the day.

     

    Here some info on how the diodes influence the voltages and amperages.

    It would be beneficial to develop the Tigo further to deal with Split Cell Panels.

     

    0
  • Avatar
    Curtis Community Moderator

    Hello Marco, 

    Thank you for your developed insight!

    Tigo MLPEs function exactly the same with a Half-Cut modules vs Single Panel modules. The optimizer is connected to the cables of the junction box, so it is only affected by the accumulated output of the module. If there are any unique production or shading issues introduced by the PV Module design (itself), they would be the responsibility of the Module manufacturer (alone). 

    With that being said, I've reviewed the Half-Cut Module design, and the diodes are laid out in the same way as a standard Single Panel module (only, the paths of the 3 diodes have been split for efficiency). So, the concepts of vertical vs horizontal hard-shade (as discussed earlier in the conversation), would still apply: 

    Tigo optimization can help in most shade-scenarios (but not all). I suspect that if the B3 Module wasn't experiencing the multi-diode closure earlier in the day, the unit would remain active throughout the early afternoon and you would see a consistently higher string yield during the max irradiation period. 

    Simply put: 

    • Hard shade condition causing Tigo MLPE shutdown = Lowered string production
    • Removing Tigo from B3 = Slightly higher string production
    • Reducing or removing Hard shade event (instead of removing Tigo) = The highest string production of 3 scenarios!

    If you want to read more about Tigo's Impedance matching technology, see our article on:
    Optimization with Predictive IV and Impedance Matching

    0

Bitte melden Sie sich an, um einen Kommentar zu hinterlassen.