If you are in DFW, don't hesitate to reach out: 972 675 7725 ( call or text) or shoot me an email: info@solartimeusa.com If you are not, you can also reach out, I do solar consultations and quotes review all the time :) Leave some thoughts below! I try to respond to all comments! :)
Great test, although I've seen so many conflicting tests on these! Two things I'd like to see done for additional testing would be: 1) Use another popular brand of string inverter (maybe Sol-Ark?). 2) Use more real-world realistic softer shading by placing something on front of the panels, but maybe 5'-10' away to allow some ambient light to still hit those shaded panels. Thanks again, you earned a new subscriber!
Excellent and thorough testing. I know this took much time to plan, execute, and report, so many thanks. Optimizers cost ~50% of panel cost, so the gain of 10% is hard to pencil out, though many other considerations like other installation costs and limited space for panels.
Of my 3 quotes currently, I have been offered a SunPower microinverter system, an Enphase with IQ7A micro's Jinko system and finally a Power Edge with optimizers system. Still trying to decide which system to accept. Your testing would indicate that I should look at a normal string system and wear the minimal loss of power. This begs the question that the industry is (in cahoots?) with the manufacturers to maximise the return for each installation by only offering micro or optimized systems. The price difference of each system varies from about $3000 in Australian dollars. I haven't even costed a string inverter system. Thank you for your efforts in "educating" us mere mortals in the intricacies that we are quite ignorant of.
The rapid shutdown requirement in the US requires an extra safety disconnect at the panels. It is easiest to meet this with microinverters or optimizers. Not sure if Australia is requiring this yet. It's unfortunate since I have seen many installs with microinverters fail. Even if they are covered by warranty, it is a pain to remove, reprogram, and then install again. Less electronics on the roof will be more reliable.
Thank you for sharing. Just so you know Sunpower microinverters are whitelabeled enphase microinverters...:) So essentially its the same product. Or at least it was when sunpower was offering microinverters here in the states. It wouldn't hurt to try to get a simple string inverter quote. It all comes down to shading, and if not much is present i really don't see the reason to pay extra for sth that would essentially be not useful. BUt if shading is heavily involved micros or optimizers are the way to go for sure. Ps. I plan to start testing shade effects on microinverters here pretty soon! :)
This is really an amazing video. Thank you for actually performing tests and sharing real data. Regarding tests 10, 11 and 12 where the non optimized panels were higher on individual panels, I think this is not important and that the optimized panels are working together to produce overall more wattage with the same shade. Looking at just the shaded panels (wattage), Test 10: Optimizers 24+173+171=368 vs non-optimizers 0+195+100=295. Test 11 optimizers 170+191+191=552 vs non-optimizers 0+188+216=404, and Test 12 optimizers 174+196+194+193+193+173+193=1316 vs non-optimizers 0+213+201+198+213+0+105=930. For example, on test 10, 173+171=344 Watts vs 195+100=295 watts. You end up with more overall wattage because the 2 optimized panels matched each other even though looking at just one panel, 195>173.
Best regards from france and Spain !! Our doubt is that the 2 series of panels are connected with a TIGO system (one optimizes and the other does not), but who says that the TS4-R-S does not influence the system without optimizing it? Have you tested panels connected directly with an inverter, without going through any Tigo device? If not, an ideal test would be a series of panels connected to a classic inverter and the other series to Optimizers. Of course, the details of each panel would not be available, but the general production of the whole would provide more than interesting data. Thank you for everything, you are great and your way of speaking English helps us a lot in Europe to understand your language more easily!!
Great video, I would love to see a video on panel efficiency and MNOT Power Output, and how often you are able to produce more then MNOT. On this video we see what I am assuming is a REC395aa-Pure panel producing a 323 watts, where the MNOT Power Output for said panel is 301w(7.2% above MNOT). I would love to see a video on what a panels peak is, and how long does a panel produces above MNOT on any given day. I have gotten around 8 to 10 qoutes, all using Enphase micro inverters, pretty much all of them are under-sizing them. Using an REC405aa-pure(MNOT is 309w) panel matched with an IQ8+ which only operates at 300w output, you are leaving a fair bit of power on the table due to clipping. Obviously it's almost entirely based on your specific environment, but I would love to see a video on MNOT Power Output and how often and how much you actually see above MNOT values.
That is such a GOOD idea! Yes! For sure want to test that. And like you said most will depend on the environmental conditions. If you’re interested I can send you some screenshots of what rec 410s generate during certain time of the year and some other ones!
Modern panels are definitely so much better that optimisers are really questionable now. Being able to monitor individual panels and 10% are maybe worth it to some, but for me, I only have to worry about shading in the depth of winter, so 10% of a period that is not great (I'm in the UK), really doesn't bring enough to the table for the extra cost, but more importantly the extra complexity you are bringing to the roof. With the hot summers and cold winters, I want to keep it as simple as possible for the difficult to access roof and putting a bunch of extra electronics cabling and connections in an area that will see heat and cold for many years = a recipe for failures. But I can see why if you had easier access then it might be worth it.
Another very good informative video. Given the increased costs with MI/Optimizers and the extra components required for panel level monitoring and combiner boxes, which typically have only a 5 year warranty, would money be better spent on a couple more panels. I know, it depends... hahaha Curious about filtered shading such as tree branch or power line 20 ft away.
Yes! I definitely want to apply that sort of shading. And test with cloudy days etc. if you look at tests 6-9 they have the building shadow applied and both sides were pulling out same power outputs! Little sneak peak into future video with filtered shading!!! 🙏
Compared to a rapid shutdown system, the cost for the optimizers is marginal ($10/panel + $500 for the controller/monitor). Tigo optimizers also carry 25 year warranty, not 5. Mine work exceptionally with a chimney on the south side and a 2nd story dormer and some trees on the west side.
@@greg_takacs Yes, more components required for Tigo which increases costs. And Tigo components such as Cloud Connect are only 5 year warranty. Same as Enphase Envoy and combiner box, 5 year. And if using SMA inverter, SMA makes it clear they do not support other manufacturers products. Not saying Tigo is bad. Its obvious that most customers have no idea what they are buying and are unable to make a full comparison.
@18:08 - that blew my mind! We are being lied to by installers big time. I have been designing a solar system for my home over the past two weeks, and have spoken with multiple solar system installer companies and without fail each one of them said that even if one panel is shaded in a string then it will impact all the remaining panels in that string. I am astounded after seeing your video!
THANK YOU. You should look into the PW3, its a hybrid inverter and it does not have any panel level devices.... bc we do not need them. I am not a huge Tesla Fan but they are truly making the future of solar super simple and I am so excited about it.
On DIYSolar forum, a person was getting a whole string shutdown in a different shade environment. No optimizers on the string. After a lot of testing it appears like the bypass diodes were not working. Doing tests like yours, they did work. Although I absolutely applaud you trying to do objective tests, those of us with statistics background know about “lurking variables”. In this case, those lurking variables we know would be different panels, different bypass diodes, different shade profiles, etc. (but there are many other variables we don’t know about”.) Trees and filtered light will NOT work in the same way as your “leaf” and cardboard tests. Still, there was some learning here. May want to simulate real life partial shading on the next test.
Our doubt is that the 2 series of panels are connected with a TIGO system (one optimizes and the other does not), but who says that the TS4-R-S does not influence the system without optimizing it? Have you tested panels connected directly with an inverter, without going through any Tigo device? If not, an ideal test would be a series of panels connected to a classic inverter and the other series to Optimizers. Of course, the details of each panel would not be available, but the general production of the whole would provide more than interesting data. Thank you for everything, you are great and your way of speaking English helps us a lot in Europe to understand your language more easily!!
I think you are right, there is no possibility to have different amps through the system without optimizers. Difference in 0.05 amp could be just some measuring error, but 0 on one panel means no current flowing, so how the rest of panels is flowing? there is some kind of optimization in those "nonoptimizers". Simply decoupling PV panel from rest of the string still is optimization.
Hi Martyna, was the SMA shade fix switched off on the Tigo optimized panels, and was it switched on for the non-optimized panels? Towards the end of the video some production data is shown from the SMA monitoring platform but it is not explained which shade setup is being used for this production data
It's great doing this, but I'm still left wondering whether the output of a string is severely affected by shaded panels. Of course the output of one panel is not affected by the one next to it, but in a serial string does a shaded panel cause resistance that reduces the string more than just the missing output of the shaded one? You should show the output of the string as well as the individual outputs.
In a truly series circuit, the current is the same through all the panels. The voltage of each panel can vary, adding up to the array voltage, divided by the entire circuit resistance (panels and load) ultimately determines the amperage, but if not enough amperage is available the voltage plummets to the point where V÷R=A becomes true again. So for any one panel to have a lower power with full sun than normal, would have to mean that it's being over loaded down to the point it can't maintain the voltage. When that happens VxA=lessW. This showed that did not happen in the series array here as the inverter adjusted its demand to match amperage availability. But theoretically the same thing could happen with a micro inverter on that one panel with full sun, using enough power to draw the panel voltage down thereby lowering its output wattage. Although I would think that the inverter would just cut off before that would happen, when it was the only panel with full sun it would be doing most of the work. But what I think actually happened here is all the inverters shared evenly, which meant it only drew from the panel with full sun an equal amount that it could get from the other panels so that each inverter was working the same and not overloading any one in particular. It doesn't mean the panel produced less, it means the inverter used less than the panel made available.
Great work !!! Best regards from Spain. As you said, the first real test I found in internet. In Spain 80% of the people use Hibrid Inverters, and in France 70% use Microinverters... Why ? No idea.. If you don't mine , I will post your videos on European forums. Best regards
The discrepancy in the test results can be attributed to the fact that the devices under consideration measure the current leaving (and voltage) of the panel, rather than the current (and voltage) put on the string line. This is an important distinction, as the current on the string line is identical for all the panels, given that they are connected in series. In order to optimize the power output from the panel, it is necessary to reduce the current while increasing the voltage (or vice versa). This is achieved by approaching the maximum power point tracking (MPPT) operating point. However, in certain circumstances, it is not possible to increase or decrease the voltage sufficiently, resulting in a lack of optimal power generation for the specific shadowing condition. The shadowing of other panels affects the power produced by a panel due to the slow output current and the necessity for the optimizer to adapt its output current to the current of the string. This occurs while the optimizer attempts to counter the inverter's efforts to alter its apparent load in order to center the MPPT. It is crucial to note that an essential parameter was not measured: the voltage entering the inverter. When the voltage drops below a certain threshold, the inverter is unable to track the MPPT, leading to complete shutdown.
Great video - thanks! Just to clarify, I assume the voltage and current measurements are made on the panel side of the optimizers and the monitoring electronics. On the inverter side of those devices, the current in each unit has to be the same since they are connected in series.
Exactly! The units measure the current on the input of the optimizers/ rapid shutdown device. Hence the current has different values. Its like seeing all panels in isolation.
Thank you so much for the kind words. I so appreciate it. I am just starting my youtube venture, so I hope in the future youtube algorithm will pick things up more! :)
This test actually shows the advantage (also for safety) of using a simple bypass diode in parallel with the solar cell groups. in order for the solar module to be efficient all the cells in a series circuit need to be producing the same voltage/amperage. If they are not the cells that are lacking become more resistive and produce heat. Heat is wasted energy and it can damage the non producing cell (even burn). Non producing cells in a string also start to have more characteristics that of an LED then a solar cell. If you could get the non producing cell from a string under load in a dark place you would actually see it give off some light as well. this reverse flow (biasing) is what causes loss in the string. it is also not good for the cell. A bypass diode gives another path for the electricity to flow. The theory is that the voltage will flow first through the path of least resistance. As the non producing cells heat up, they get more resistive and eventually the diode becomes the more prominent path. But, also bypass diodes create some voltage loss as well. Schottky diodes are used for bypass diodes in solar modules because they have higher series resistance then other type diodes and keep the reverse voltages at safer levels. Another use for diodes in a solar circuit is a blocking diode which blocks an entire module or string. Bypass diodes are installed in parallel, blocking diodes are installed in a series circuit. Thanks for all that you do!
Fantastic. Thanks a lot! I just got my 18 panel installed with the tigos. Pretty sure there is not an actual ROI under my conditions. But the generated data and graphics are so cool :-)
Great video as always! I know that SMA highly touts their "shadefix" system that purports to do some amount of optimization to handle shading. I'm curious whether you would expect a different result using a different inverter on the non-optimized side?
Yesss! I need to turn off shade fix on one inverter and test it and see what it does exactly to the string! I also want to test other inverters. But Shadefix comes in as turned on for most inverters and it was on for these tests. But I will definitely see how it works. !! :)
Question: You refer to "Ever Since Bypass Diodes Became Used In Panels", since when was this? I cannot find this on the web. Your tests show that Panels with Bypass Diodes do not shut down the series string but what about older panels? I think it was is true with older panels, but how old is old?
What a great video, thank you very much for the transparency and honesty, The way you explained it was so easy to understand. Looking forward to watch your next vids.
I really want to test fronius ability to optimize string, a few people asked about it too! That would be a very good test! I am definitely starting lots of tests this month! :)
Very impressive test! Constructive criticism: if testing partially shaded panels, it would be far more reliable and scientifically correct to use a jig to accurately position the leaves so they cover exactly the same amount of the same cells in each panel. It may or may not affect the results, but at least you’ve eliminated a variable. Keep up the good work!
@@solartimeusa making tape “loops” and keeping the tape completely behind the obstruction would also be a good idea. Thanks for the reply, it’s great to see creators who engage with the comments! 👍
Thank you for this test. I recently watched another video suggesting far less of an impact on string shading, but their test method, presentation, and conclusions were already suspect to me.
Thank you! I plan on doing other tests with similar shading but also with different brand inverters. I also want to test more filtered shading like trees. :) I am curious if results will be similar or very different.
Again, this would be case specific. So what are the extra cost per panel if you include the optimizers ? If you have no more space on the roof..possible a true advantage if you calculate around 1.2KW/h per day in saved cost. But if you have enough space on the roof, what would be the result if you would have gone with a slightly larger system at parity cost?
Can you do it again with half-cut panels? Ideally once vertically and once horizontally mounted with a flagpole shadow and a wide diffuse tree shadow sweeping across a full module? I suspect half cut modules are better mounted horizontally so a horizontally moving thin shadow only shuts down half a panel at most. But a tree shadow would be way more interesting.
I bet, if you compare "with optimizers" array with "no optimizers or bypass diodes" array - results would be so much different. Since the bypass diode actually does its job as well, and it optimizes the power output of a whole string/array based on the switching off capabilities it has.
i've seen demonstrations that showed that the optimizers don't work at all, and actually produce less power. It seems it is because of the 'blind deployment" without CCA+TAP. It's also related to a particular firmware from 2021, and not sure if that issue still exists. I'm very interested to know whether you tested "blind" or with CCA+TAP, and whether you found the TS4 to misbehave in any situations.
Hi Martyna, how about shingled panels? Does there any reliable shingled panels that can perform better compred to half-cut with optimizer? How about the cost comparison with half-cut+optimizer and shingled panels?
Hey! ShadeFix is automatically turned ON on most inverters so this included shade fix, but I do wnat to test two separate inverters one with ShadeFix on and Off to see what really it does. I almost cant imagine a drastic change, because the bypass diode worked, so I am very curious myself how DC would be then conditioned. Definitely need to TEST IT SOON! :)
Thanks for the video, I think this is one of the best reviews of how power optimizers can be useful. Very helpful and shows how panels could be affected in the real world. but please don't use that "Text" in the middle of the screen feature (early in video)... it is one of the most annoying YT trends ever.
Hey Marc! Thanks for the feedback on the test and on the editing! I appreciate all info, so thanks for letting me know. I have to take it up with my editor ...:D :) Martyna
Great video! Really helped me understand all this! This past summer I had a PV system installed on my house in San Antonio. It has 25 panels with mini-inverters and an Enphase system. The house does get significant shading from trees. I am already wanting to expand it. I do not have battery back-up but am considering installing it eventually. Also, the original installer (whom I shall never use again or recommend!) said there was no way to get power out of the system during a grid failure without a battery system, but as I understand it, there actually IS a device which allows for this. Do you know anything about that? Finally, the roof on my house is a low pitch standing seam metal roof. The current panels are installed on the south facing side. If I install panels on the north facing side would there be anything to gain by using a rack that will prop the panels up to an optimum south facing angle, especially if those are bifacial? (Roof is plain silver color) Thanks again for your great, informative videos!
You need to "island" your system if you want it to work while the grid is down, the inverter needs the grid frequency to have power so you will have to have something to provide this. Other than a battery system I don't know whats available in the US to do this. Im in Australia and have sold solar since 2011 but inverters/ micro inverters work the same anywhere so thats the issue you face
Hi! I am so sorry for the delay in responding to you. So yes, there is an option to use power during the grid outage with enphase. I will include the link enphase.com/homeowners/sunlight-backup-user-guide?hsa_acc=1236416428&hsa_cam=20130257651&hsa_grp=&hsa_ad=&hsa_src=x&hsa_tgt=&hsa_kw=&hsa_mt=&hsa_net=google&hsa_ver=3&gad=1&gclid=CjwKCAjws9ipBhB1EiwAccEi1Ld_T4TDI6NVlLA4375BrzzSQ9S33Y1lcXp6b6okeMKnX7FlpZXgCBoCvmIQAvD_BwE The downside is it’s like having a battery set up ( cost of extra equipment) but no battery. I don’t think it’s worth the cost. It might cost over 4K maybe even close to 6k if there is a need for rewire. I suggest waiting till you get a battery becasue you will spend money once and have a good back up solution. As far as pitching panels, you can email me and I can check the roof and see if that would make sense. The struggle here is that when you pitch panels you have to have spacing between them since they will start shading one another so you’re limiting space, and by just installing them facing north might cost similar to just have more panels, than less but a pitch. Email me if you want me to review the address :) happy to help, again so sorry for my delayed response. Info@solartimeusa.com
Oh wow! You know you can email me directly and I can try to help you determine which one is most realistic! I do consulting for homeowners outside of my work area :) Info@solartimeusa.com
It would be interesting to do the shade-from-building testing with the panels in landscape mode. Most panels strip the regions along the long length of the panel for each area covered by a bypass diode. That means the panels will work better in horizontal placement as the sun goes up or down. (I don't know if the optimizers would make a difference though)
They did do that test, it's towards the middle of the video. Additionally you can read this thread with a lot of good info from me and my system regarding all kinds of shading my system experiences diysolarforum.com/threads/can-shading-be-dealt-with-by-parallel-strings.54317/
That is a very good point, I do wonder the same. Surprisingly panels that were shaded in half and didnt have optimizers, completely shut down, which makes the marketing for half cut cell panels BS? I really want to test that too.
The problem originally came from solar panels that do not have bypass diodes. So a leaf on a single panel would drop production on the whole string by that %. So the 20..30% drop would be for all panels due to the single leaf on a single panel. The bypass diodes prevent this as you have clearly demonstrated. Thanks! I do have 8 of these pre bypass diode panels, but they won't see leafs or antennas (230Wp), but my next panels might get shaded, but have bypass and are slightly more than 500Wp. And there will be about 40 of them instead of, because I hope that would cover my winter days instead of having to settle for a bio-diesel generator. (a 500Wp solar panel is around $125 these days, a bio diesel is about $1500 so in costs I can have 6 panels with 2 mppt's)
yes, exactly! Yet the industry keeps using that to sell their own products. Just want people to have proper facts! :) I actually want to test older panels too, we have some leftovers in the warehouse from 2015s and prior... could be a fun battle to see :D
Thanks! I can explain the worries about "one shaded panel affects the others". You can see the data with your own tests, but the measuring devices/timing are very imprecise, so it wasn't obvious. When batteries/panels are connected in series (string) and you put a load, _exactly_ the same current passes through all batteries. No exceptions. The power-meters here you can see are not good, reporting 7.93A on one panel but 7.99A on another and 8.05 on another. That is physically impossible. It's possible that they are precise but the measurement+reporting happened a few milliseconds apart in time instead of simultaneously. Anyway, when one panel can do only 3.99A 36V due to shading, you can't possibly get more or less current than 3.99A from any other panel. So that's why people are worried that suddenly the whole string would produce half the amps thus half the wattage. Instead, the MPPT tries to draw different amount of input current, to find the sweet-spot. It tries 8.01A, the voltage of the shaded section rapidly falls below 0.6V as it can't meet demand, its diode turns on and the remaining 2 sections of a panel now output 8.01A 24V together. All panels resume to run at 8.01A, and the total voltage is simply 12V less. So, with 10 panels you don't lose half the total wattage (getting 3.99 x 36 x 10 = 1436.4W instead of 2883.6W). You only lose 12x8.01 = 96.12W, you get 360-12=348V 8.01A instead of the full 360V 8.01A. You see this in all your tests: all panels produce around 7-8A, and some of the sections of panels get turned-off, producing 0V and 7-8A. Well actually each diode that turned-on eats 0.6V * 8 = 4.8W and turns it into heat. The reported voltages on fully-shaded panels seem incorrect. It shouldn't be 40V 0A but 0V 8A. Actually -1.8V 8A. It must be a software bug. The bypass diodes ensure that suboptimal sections of the panels are turned off. The MPPT does a good job of finding the optimal current to use from the remaining turned-on sections. The resulting voltage of each section depends on that current (e.g 9V at 8.6A or 12V at 8.01A). Solar panels come with optimizers built-in: those bypass diodes 😤. A future improvement would be: remove the bypass diodes, and connect a new kind of optimizer with 4 wires to the panel. Optimize each section independently. I don't see it happening any time, soon, though. Currently the optimizers get crippled by the bypass diodes.
Hey!!! Thank you so much for the thoughts and input! I should have explained in the video that the Tigo units dont measure the OUTPUT of the optimizers/rapid shutdown units. It is measuring the input from the panel itself, so we can see them all separately in isolation. So the actual current of the string is the equalized by the mmpt in the string inverter itself, so I can totally see how that seemed misleading. What I learned from Tigo ( as much as they could tell me) is the units measure the input not the output of the unit itself. But this LEADS me to ANOTHER TEST! Lets see if I put 8 panels with the TS4S ( no optimization) and just string inverter with no units under the panels, and compare the outcomes. I am very curious how the results will vary, and if they will. I do like your idea too!!! Hopefully one day! :) Martyna
@@solartimeusa You may be interested to learn that Tigo record both Vin and Vout for each panel, but for some reason Vout isn't made available directly in their charting application. Good news is that you can just edit the URL, adding '&suffixes[]=vout'. The data is also available via their API.
At roughly $100 per panel to install Power Optimiser, works out to $1000 for a string of 10 panels, for that same $1000 you could purchase 5 more panels at roughly $200 per panel, giving you a string of 15, the output then would blitz the output of the 10 power optimised string, subject of course if you have the room for 5 extra panels.
Hey this is amazing! I’m really struggling to pick between a local installer and Tesla here in upstate New York. For a same sized system Tesla is estimating. 6300kWh / year but the local installer is estimating 9478kWh/ year. Whose right! I even did a shading analysis in helioscope and the local installer was much closer it seems. Why is Tesla estimate so low? Is it cause of their string inverter vs local installers micros? Thanks!
Hey! I would say go local, if the company has good reputation, good online reviews and if they have been around for some time. Does the installer offer any production warranty? Tesla is undershooting so they don’t have to deal with complaints if underproduction later on. So they are playing safe. The biggest problem I see with Tesla is service. In DFW in texas I service Tesla systems weekly at a charge becasue of how long they wait for service from Tesla. So if you can choose an installer that will service your system in the future, that’s the way to go. Local company might have a more expensive price but you’re also supporting local over huge corpo where nobody cares about you. I am biased because I am a local business too. But I think you will be better off going local if company is reputable. I hope this helps 🙌🏼
nice videos. I would have made sense to spend a minute at the beginning showing the sub-strings layout on these panels, if they are half-cut cells and explain how many bypass diodes they have.
I don't undertstand: Are the panels in each string wired is series (in which case I expect then to all have the same current) or parallel (in which case I expect them to have the same voltage), or what?
How well do split panels do with this test? Looks like they have double the diodes and should be closer to using optimizers. Can you show us the costs for both systems? a 10% improvement with optimizers is probably not worth the 20% or whatever increase in cost.
I think the real test is to run them side by side for an extended period of time. Don’t forget that modules and components within the modules will deteriorate over time - this will have a huge impact on performance as time passes by. Not to mention cloud cover and soiling which are inevitable. The inverter’s role really starts to pay off not on day one but over the 15 to 20 year life span of the system. Nearly every test I’ve seen on the subject does not cover the subject of long term yield
Hi Martyna, or Martina? ;-) Nice video!! I only little bit don't what all these boxes are. You showed two smaller black boxes with some cables... Does it mean that each solar panel has one black box connected underneath (so there are 16 hidden boxes)? It would be nice to see wiring from back of panels. And you are speaking about diods - are they in those black boxes or are they built-in solar panels? It would be nice to see some theory first (for such beginners like me) - like wiring diagram of both configurations. (If black boxes are placed in both strings, then it would be nice to have third string without any black boxes - 5 panels in each string is enough, no need for 8.) And I'd love to see what each black box actually do. Now I maybe understand that first black box if only for measuring current and voltage but is passive - doesn't do anything. (if so, then 3 strings doesn't make any sense) And second with optimizer is basically acting like tranzistor which is connected in parallel to panel and if there is shadow (power drop down + increasing internal resistence of solar panel), it opens itself partially and allows bypassing current from other panels from the string? (Btw excellent clear English, I am CZ and can't hear almost any accent.)
Hey! Its Martyna ( Polish) but when pronouncing it in english it is pronounced like Martina. Yes so each panel has a small back box behind, that is built in, and it includes the bypass diodes. Then what I did is used 8 power optimizers ( extra black boxes) under one set of panels, and 8 Rapid shutdown devices ( extra black box) under the other eight. I would love to dig deeper into how they are wired, hopefully in the future video. But in an easyway speaking the panels have their connectors that connect to power optimizers ( male and female connectors) and then those then connect in series to other panels. There are two ways of connecting panels together either in series or in parallel. I definitely want to do a video on that, so make sure to subscribe :)! Thank you for leaving the comment!
THANK YOU SO MUCH! I appreciate you taking the time to leave a comment. Hope your business is doing well. I know EU is suffering from downtime in solar...
Interesting stuff. You're right, the repeated narrative over and over has been, shade one panel on a string and the rest only produce the amount of the shaded panel. Obviously, this isn't necessarily the case. Do the brand of panel and maybe age of panels make this hypothesis different? I would like to see a test of two systems with mixed panel wattages and see what that does on a string system. That's another thing that is always repeated as a no-no, never mix wattages. How would that look in this same test?
Thanks for the feedback, and yes that would be a really cool test, to mix brands and panel voltages. Like adding a 480W panel to a 400 w panel string! Thanks for the idea!!! :)
Another benefit of optimizers is the ability to mix different panels within a string (within reason). Note that the Tigo optimizers are buck converters not buck/boost converters therefore they can only turn higher voltage lower current into lower voltage and higher current, not the other way around. This generally is fine unless you're starting to look at and work with uneven size parallel strings going to the same inverter input and the likes ;-).
Cześć, bardzo dobrze widać że dodanie Tigo tylko w wersji monitoringu do panela HalfCut zupełnie zmienia jego działanie. Diody bajpasowe nie powinny zadziałać po zasłonięciu połowy panela tylko prąd modułu powinien spaść o połowę - niestety w takim przypadku prąd całego stringu też by spadł o połowę więc dodanie Tigo bez optymalizacji to właściwie poprawia działanie całego tego systemu bo załącza się w nim wbudowana opcja bajpasu. Moduły HC sprawdzają się tylko z mikroinwerterami lub optymalizatorami. Co da samego testu zacienienia to w przyszłości proszę postawić przed panelami cienką rurkę lub powiesić linkę symulującą przewody zasilania lub cienkie gałązki. Taki "soft shadow" ma druzgocący wpływ na pracę całego całego układu stringowego - cień jest na tyle mały że diody bajpasowe się nie aktywują, a spadek prądu całego stringu jest ogromny. W takich przypadkach widać dopiero zyski z optymalizacji.
Thank you so much for leaving a comment! Good luck with your project! the 10% difference is also based on those extreme shading conditions, the percentage would go down, as the shading diminishes.
panels have an efficiency, an amount they can generate for a given amount of sun. cover half a panel, IE give it half as much sun, you will get half as much production, no magic box you can buy to add onto a panel will or can change that. top say that covering half a panel and yet still get 3/4s the power, means your changing the efficency, and thats not really possible. when panels are in series then 'per panel bypass diodes' are the only option available to you, per cell bypass diodes cant be added after the panel is built. best cure for localised shading, micro-inverters to ac or individual mppt->dc->boost to a dc bus
I'm not sure what that per-panel monitoring box is doing, but it is not behaving as series string of panels without those boxes. Interesting test none the less, but I don't think it will apply to a string without electronics at each panel.
Hey! Thank you for your comment. The units are called tigo Shut down devices. Supposedly they do not “condition” the power simply provide rapid shutdown and monitoring ( reading data) feature. But! I do want to test them compared to simple string inverter to test whether a string behaves like that or not. I am curious myself. Thanks again for leaving your thoughts! I appreciate it
Yes definitely. I do see big benefit in them, just for the most cost effective solution, sometimes not needed. Thanks for sharing your thoughts, I appreciate the comment! :)
Hey! I used the Tigo Rapid shutdown devices, that offer monitoring and rapid shutdown feature, with supposedly no optimization :) Her is the link: www.tigoenergy.com/product/ts4-a-s
one thing is missing to me... The test was o so great... what about to take the same money what optimizers costs and buy more panels without them and try that comparison (what do you get for your money).
Yes very good point! I’m actually testing other equipment like microinverters now. But definitely want to test different shading type like filtered shading from trees etc. stay tuned ! And thanks so much for leaving a comment! It helps me so much!
I really think spending your $ on more photovoltaic surface are has a better ROI than spending $$ on extra subsystems like optimizers. A optimizer costs near the same as a 2nd panel for only a 10% gain, where 2 panels is a 100% gain for the same cost?
Not so concerned about individual panel knock downs, but total power output of whole string. I would like to see this repeated with two strings in parallel like Tesla Solar Loves to do. I would like to see how different azimuths and pitch’s affect daily production on a single string.
Greetings, I am currently residing in the state of Florida and employed in the field of construction. I am interested in exploring the possibility of collaborating on the construction project you mentioned. Would it be possible for us to have a discussion regarding this matter? I am currently exploring the option of replacing our electrical service with a battery power bank for our job site trailer.
You have just answer my question that I posted in another video of yours regarding the need of optimizers in set up with no shading issues. I guess that uneven panel degradation over time and / or dirt accumulation on the panels play no practical role on the question optimizers or no optimizers. That is, because we are talking long term. Regarding the existence of the Tygo panel level emergency shut down modules on the no optimizer side of the test panels, these modules are just switches, hence their cheaper prices. Tygo introduced this solution in order to serve consumers who have some Code to abide to but not necessarily have shading or monitoring concerns. Being , nothing more than switches, they introduce extra resistance in the circuit as opposed to a straight wire connection. Therefore, your test results are 110% valid and actually on the conservative side as far the performance of the no optimization panel side. If you remove these switches completely on that side, the panel output will be slightly higher than with the cutout switches on.
I am so sorry I missed your other comment. I really try to respond to them all! :( But thank you so much for leaving your questions and feedback and thoughts, it helps me a ton, and obviously, I love to know if I make any mistakes while we test those different products, so I can improve and keep learning.
The problem with these tests is they are not representative of a real system shading. Basically the panels are 2 independent 200 watts solar panels. If you have square shadows, this is a great test. But in reality shadows are cast over both halves of the panels and this causes some serious output problems that will not be solved by a optimizer. That is why alot of systems are wired series/parallel to isolate shaded panels. Test the panels where both halves are shaded by a oval shadow. The best you can do is add more panels to a system as the panels are cheap. Great to show the issues with shading!
Hey ! Thank you so much for the input, and yes these were pretty extreme conditions, and I plan to do more diffused shadow as well that will overcast multiple panels etc. :) Thanks again for leaving your thoughts, I really appreciate that and love hearing the feedback!
If you are using a string inveter with all 8 panels in 1 string, how is it possible for shaded panels to have less current than the full-sun panels? My system is strings of 6 panels, and I thought optimizers were required to prevent a shaded panel from reducing the output of the other 5 panels in the string?
SO the way the optimizers work is the try to find ( in mathematical term) best PRODUCT of the two (voltage and current). Sometimes they have to lower the current to increase the overall production outcome. Also the optimizers do provide the INPUT values, so whatever enters ine optimizer, not the outcome. So later on the string MPPT also equalizes the values to track the bets power point of production. I hope that made sense? Martyna
Tigo optimizers cost me $22/panel on a 44 panel system and produce a 6.4% improvement in generation for 1.5% increase in total system cost. ROI in the Tigo optimizers is 4.4 years while it's closer to 13 on the whole system.
Additionally, sometimes adding 6% extra panels is not just a roof space issue but a string configuration issue, ie. You can't just add another panel to a 2x12 string.
That is a very good question!!! And the problem is the answer is not super straightforward. This will definitely depend on the severity of shading. In some cases shading is so minimal that it wouldn't be worth it, sometimes space is limited so power optimizers are the better option.
It looks that, when more than 70% or 80% of the panels are covered by shadows in a string, the optimisers just try to work at a medium level of current and voltage for all the serie.
The panels were connected in two separate strings and in two separate inverters. Does that answer the question? That should not cause any changes in the tests as they were connected in an identical way.
I only have shading in AM/PM when sun moves into field of panels or out of field. My neighbor has string inverters with optimizers which cost more than I paid for each panel. I bought used. Adding more panels and another inverter for me or anyone with no shading is the way to go if they also have a all in one inverter like I do which is 2 5k inverters.
Thank you for sharing! Yes unless there is substantial shading I would not worry too much about the optimization whatsoever! Thank you for leaving a comment.
Where space is not constrained you can probably install extra solar panels with the money saved from not using micro inverters. This is especially relevant for ground mount systems.
The optimizers are extra work and based on the promotional blurb can add up to 20%. So give it to them, if its a 20% gain you are after compare the cost of adding optimizers to each panel to adding 20% more panels. The panels are cheaper. This works assuming there is room on the roof for extra panels.
Optimizers in my real life system give me 6.4% improvement for 1.5% of the cost of the system. It's not 20 but the optimizers still had better ROI than my entire system. Adding an extra 6% panels also cost more than just the panel cost, mounting and labor are not free.
Thanks for sharing your data too Greg! I am sure that depending on shading level the results will vary as well. I do want to test more diffused shading like trees, and maybe different orientation of panels etc. Will be cool to see and verify results.
The tests show that your money is better invested in more solar panels than optimizers. The panels will likely last for decades while anything electronic is looking for an opportunity to fail. Of course,ground mount is better regardless of optimizers or not because of ease of maintenance.
Very very fair point, lots of people feel same way, the only issue in usa is the rapid shutdown law.. that makes us use some sort of panel level device for rapid shutdown ; /
Great Video, so much to unpack. You are so smart and so cute :) Great to see the diodes actually work ! I was a little confused how the non-optimized panels were configured. Did they use microinverters or some other way to achieve MPP ? Couple of questions for you: 1. If there is NO shading, would optimizers produce more/less/same power than no optimizers ? Other than having a better MPPT algorithm I cannot see why they would produce more power. 2. If 8 panels were connected in series to a string inverter, which I believe is your preferred configuration, how would the results change ? My guess is this is where optimizers really pay off, but it's to compensate for the disadvantage of connecting panels in series. Of course there are advantages to series connecting panels like saving on copper wiring costs. 3. Wouldn't microinverters be the best overall choice for a lowest cost/best performing system ?
I'm not as smart or cute, but I'll try to answer ;-) 1) If there is NO shading there is no benefit for optimizers, other than they can balance the small difference between panels as not every panel will produce the same current even at the same solar radiation due to manufacturing and material differences. But optimizers, while very efficient, will cost some power loss in and of themselves so I'd say the two (manufacturing differences in panels vs. power consumption of optimizers) is a wash. Note that you still get the panel level monitoring, including alterts if there is an anomaly, which I think is worth it. 2) Connecting panels in series is not only beneficial from high voltage low current but it's also a requirement as whatever combination you decided to go with has to match the working window of the inverter. Since most inverters work between 100-500V and panel voltage is 30-40V depending on cells you need to put them in series to reach the minimum working voltage of the inverter. Additionally they are also current limited (generally under 30A per MPPT) so you can only do so many strings in parallel before you overrun the current limit ( usually 2, maybe 3 strings in parallel per MPPT max) 3) For a smaller system microinverters make sense from a cost perspective as the system cost is marginal ie. 2x the system 2x the price, 15x the system 15x the price. But for a well sized and optimized system from the ground up you're almost always better off with a properly sized string inverter and matching size panels. Additionally, microinver systems end up costing a lot more if you're also considering battery backup, which in 2023 everyone putting in a solar system should.
@@greg_takacs Thanks for your reply. Well, you may not be cute, but I think you are pretty smart. As I can tell there are 3 downsides to operating the panels as strings: 1. A shaded panel will limit the current in the entire string. 2. The MPPT controller in the string inverter will maximize the power, so you better have all the panels producing about the same. 3. A degraded panel will reduce the current similar to 1. BTW, I do have degraded panels on my microinverter system. The remedy is, of course to use optimizers for case 1 and 3, which are basically buck converters, keeping the current in the string as high as possible at the expense of the panel voltage. Pretty clever really. But you effectively get this capability already when using microinverters and you don't need a string inverter. You can mix and match all kinds of panels with microinverters, big/small/shaded/degraded etc, since all power in normalized to 240V split phase and paralleled together. This allows for split strings with panels in different orientations (south and west) to capture the sun later in the day when the tariffs are high. I am a bit puzzled why the "cute one" is so enamored with string inverters but I'm sure you'll set me straight. BTW, I do have a grid tied system using microinverter and an off grid (3x strings) with batteries for pumping water in the event of a utility failure. The biggest downside I see on the off grid system is once the batteries are full and I have a full tank of water, the array throttles back. Maybe time to upgrade to a full hybrid inverter.
@@ai6mk897 issue #2 is also solved with optimizers, and Tigo optimizers can even solve it between parallel strings going to the same MPPT as long as every panel has an optimizer and they're controlled/managed through their CCA (Cloud Connect Adnvanced) I can't speak to why the original poster loves string inverters, but I can tell you why I would never buy microinverters as the sole solution; They're simply inefficient, especially once you add battery to the equation. Additionally they're too expensive as every panel requires another inverter and Enphase's solution requires the same on the battery side, specifically additional number of inverters for the batteries as well, again adding more to the cost and padding Enphase's profit margin. In 2023 I would not consider a solar only installation, full stop. And once you're in the thick of it with having batteries added to your system, a hybrid inverter with proper DC coupled battery management and solution is just overall a better user experience. I personally have a Sol-Ark 15K system with 38 kWh of batteries and 19.4kW of panels with Tigo Optimizers. I don't think there is much else on the market that can give you what a 15kW hybrid system can for the same cost. I don't have critical load panels, I just have an inverter sitting between my meter and my main panel. I'm off-grid during the day and use free electricity from the grid at night. Couldn't do it with a micro setup. Having said that, with the Sol-Ark 15K I can add AC coupled solar directly to this system and it will manage it as well and utilize it in full island mode, and this is where a few extra panels with micro inverters just might make sense, although I'm still leaning towards getting a string inverter and Tigo Optimizers for this as well.
Hey! :) My turn. So to address the 3 points you mentioned. This test showed that with panels being connected in series, the shaded panel does not affect the rest of the panels, so that would not be a downside. All thanks to the bypass diode. But even if that was the case, using a string inverter in a smart way can yield similar results as microinverters. For example I have an 8 KW system with 14 panels facing east and 6 panels facing east. I built 3 separate series of panels ( 3 strings) without mixing the panels from west or east. I also did power optimizers even though I am really not benefiting from it. :D I don't have any shading. Essentially both options are extremely good, Greg made a good point ( #3) about adding batteries to the system, and with hybrid inverters that is the most efficient way. But with your microinverters you can easily add ac coupled battery too. It might just be more expensive and a bit less efficient. I did make a video on why I do prefer string inverters. I hope you will give it a listen. Its too long to type :D ruclips.net/video/4K_q7hGFLuo/видео.html Let me known what you think after you watch it. I am very curious! Here is a link to my personal system ( got disconnected from wifi, but I will reconnect it.)But you can view some on it :) ei.tigoenergy.com/p/w5wjwUT3spa_
It was obvios and very easy to make the calls of the production even without the testing because you used hard shadows you are forcing bypass diodes to get activated. When it would get interesting it is with soft shadows let say 8 modules 3-4 covered with a soft shadow reducing the current output of the shadowed modules 80%, then the string inverter I bet it would produce 20% less in the whole system, because the other option for the MPPT of the string inverter is to activate all the bypass diodes in the 3-4 modules shadowed.
If you are in DFW, don't hesitate to reach out: 972 675 7725 ( call or text) or shoot me an email: info@solartimeusa.com
If you are not, you can also reach out, I do solar consultations and quotes review all the time :)
Leave some thoughts below! I try to respond to all comments! :)
Great test, although I've seen so many conflicting tests on these! Two things I'd like to see done for additional testing would be: 1) Use another popular brand of string inverter (maybe Sol-Ark?). 2) Use more real-world realistic softer shading by placing something on front of the panels, but maybe 5'-10' away to allow some ambient light to still hit those shaded panels.
Thanks again, you earned a new subscriber!
Yes! I definitely plan on testing many different products and with more filtered shading as well! Thank you so much for subscribing 🙏
One of the most comprehensive and detailed explanation I have seen on this matter. This was beyond valuable. Thank you so much for sharing!
Thank you so much for saying that! I appreciate it!
Probably the best test example on YT using latest up to date premium components.
Thank you so much for saying that :)!!!
Excellent and thorough testing. I know this took much time to plan, execute, and report, so many thanks. Optimizers cost ~50% of panel cost, so the gain of 10% is hard to pencil out, though many other considerations like other installation costs and limited space for panels.
Of my 3 quotes currently, I have been offered a SunPower microinverter system, an Enphase with IQ7A micro's Jinko system and finally a Power Edge with optimizers system. Still trying to decide which system to accept. Your testing would indicate that I should look at a normal string system and wear the minimal loss of power. This begs the question that the industry is
(in cahoots?) with the manufacturers to maximise the return for each installation by only offering micro or optimized systems. The price difference of each system varies from about
$3000 in Australian dollars. I haven't even costed a string inverter system. Thank you for your efforts in "educating" us mere mortals in the intricacies that we are quite ignorant of.
The rapid shutdown requirement in the US requires an extra safety disconnect at the panels. It is easiest to meet this with microinverters or optimizers. Not sure if Australia is requiring this yet. It's unfortunate since I have seen many installs with microinverters fail. Even if they are covered by warranty, it is a pain to remove, reprogram, and then install again. Less electronics on the roof will be more reliable.
Keen to know which you decide and why
Thank you for sharing. Just so you know Sunpower microinverters are whitelabeled enphase microinverters...:) So essentially its the same product. Or at least it was when sunpower was offering microinverters here in the states.
It wouldn't hurt to try to get a simple string inverter quote. It all comes down to shading, and if not much is present i really don't see the reason to pay extra for sth that would essentially be not useful. BUt if shading is heavily involved micros or optimizers are the way to go for sure.
Ps. I plan to start testing shade effects on microinverters here pretty soon! :)
This is really an amazing video. Thank you for actually performing tests and sharing real data. Regarding tests 10, 11 and 12 where the non optimized panels were higher on individual panels, I think this is not important and that the optimized panels are working together to produce overall more wattage with the same shade. Looking at just the shaded panels (wattage), Test 10: Optimizers 24+173+171=368 vs non-optimizers 0+195+100=295. Test 11 optimizers 170+191+191=552 vs non-optimizers 0+188+216=404, and Test 12 optimizers 174+196+194+193+193+173+193=1316 vs non-optimizers 0+213+201+198+213+0+105=930. For example, on test 10, 173+171=344 Watts vs 195+100=295 watts. You end up with more overall wattage because the 2 optimized panels matched each other even though looking at just one panel, 195>173.
awesome test. HUGE-HUGE Thanks!
Thank you so much for taking the time to leave a comment, and share your thoughts.
Best regards from france and Spain !! Our doubt is that the 2 series of panels are connected with a TIGO system (one optimizes and the other does not), but who says that the TS4-R-S does not influence the system without optimizing it? Have you tested panels connected directly with an inverter, without going through any Tigo device? If not, an ideal test would be a series of panels connected to a classic inverter and the other series to Optimizers. Of course, the details of each panel would not be available, but the general production of the whole would provide more than interesting data. Thank you for everything, you are great and your way of speaking English helps us a lot in Europe to understand your language more easily!!
Great video, I would love to see a video on panel efficiency and MNOT Power Output, and how often you are able to produce more then MNOT. On this video we see what I am assuming is a REC395aa-Pure panel producing a 323 watts, where the MNOT Power Output for said panel is 301w(7.2% above MNOT). I would love to see a video on what a panels peak is, and how long does a panel produces above MNOT on any given day. I have gotten around 8 to 10 qoutes, all using Enphase micro inverters, pretty much all of them are under-sizing them. Using an REC405aa-pure(MNOT is 309w) panel matched with an IQ8+ which only operates at 300w output, you are leaving a fair bit of power on the table due to clipping. Obviously it's almost entirely based on your specific environment, but I would love to see a video on MNOT Power Output and how often and how much you actually see above MNOT values.
That is such a GOOD idea! Yes! For sure want to test that. And like you said most will depend on the environmental conditions. If you’re interested I can send you some screenshots of what rec 410s generate during certain time of the year and some other ones!
@@solartimeusa Yeah if you could that would be amazing.
Great video! Really helps solar customers understand the impact of shading on a system with or without optimizers.
I am so grateful and happy it was helpful! Thank you!
Thanks so much for all the effort and continuously producing content of such high value!
I am so so glad it was helpful! I appreciate leaving the comment so much! :)
THANK YOU for posting this! I’d no idea until y’all explained it to me in person. Optimizers are a much better option than microinvertors.
Glad it helped! Thanks for the comment!
Modern panels are definitely so much better that optimisers are really questionable now. Being able to monitor individual panels and 10% are maybe worth it to some, but for me, I only have to worry about shading in the depth of winter, so 10% of a period that is not great (I'm in the UK), really doesn't bring enough to the table for the extra cost, but more importantly the extra complexity you are bringing to the roof. With the hot summers and cold winters, I want to keep it as simple as possible for the difficult to access roof and putting a bunch of extra electronics cabling and connections in an area that will see heat and cold for many years = a recipe for failures. But I can see why if you had easier access then it might be worth it.
Another very good informative video. Given the increased costs with MI/Optimizers and the extra components required for panel level monitoring and combiner boxes, which typically have only a 5 year warranty, would money be better spent on a couple more panels. I know, it depends... hahaha Curious about filtered shading such as tree branch or power line 20 ft away.
Yes! I definitely want to apply that sort of shading. And test with cloudy days etc.
if you look at tests 6-9 they have the building shadow applied and both sides were pulling out same power outputs! Little sneak peak into future video with filtered shading!!! 🙏
Compared to a rapid shutdown system, the cost for the optimizers is marginal ($10/panel + $500 for the controller/monitor). Tigo optimizers also carry 25 year warranty, not 5. Mine work exceptionally with a chimney on the south side and a 2nd story dormer and some trees on the west side.
@@greg_takacs Yes, more components required for Tigo which increases costs. And Tigo components such as Cloud Connect are only 5 year warranty. Same as Enphase Envoy and combiner box, 5 year. And if using SMA inverter, SMA makes it clear they do not support other manufacturers products. Not saying Tigo is bad. Its obvious that most customers have no idea what they are buying and are unable to make a full comparison.
How about showing something more common such as a cloudy day or a overcast day. This would be interesting as well.
And filtered shading such as tree branch(s) or utility line 20 ft away...
YES! Great idea. These were definitely super extreme conditions. Curious to test that! :) Thanks for leaving the idea!
YES! I really want to test that next! These were super extreme conditions... :D
Cloud or overcast has no impact as those are global sharing and impact every panel the same. Percentages will remain the same.
@@greg_takacs This was my assumption based on their 6-9 test where te building partly shaded the panels.
Would be awesome to see the same test with Solar Edge. Also would be nice to see performance of various/popular string inverters
@18:08 - that blew my mind! We are being lied to by installers big time. I have been designing a solar system for my home over the past two weeks, and have spoken with multiple solar system installer companies and without fail each one of them said that even if one panel is shaded in a string then it will impact all the remaining panels in that string. I am astounded after seeing your video!
THANK YOU.
You should look into the PW3, its a hybrid inverter and it does not have any panel level devices.... bc we do not need them. I am not a huge Tesla Fan but they are truly making the future of solar super simple and I am so excited about it.
On DIYSolar forum, a person was getting a whole string shutdown in a different shade environment. No optimizers on the string. After a lot of testing it appears like the bypass diodes were not working. Doing tests like yours, they did work.
Although I absolutely applaud you trying to do objective tests, those of us with statistics background know about “lurking variables”. In this case, those lurking variables we know would be different panels, different bypass diodes, different shade profiles, etc. (but there are many other variables we don’t know about”.) Trees and filtered light will NOT work in the same way as your “leaf” and cardboard tests. Still, there was some learning here. May want to simulate real life partial shading on the next test.
Our doubt is that the 2 series of panels are connected with a TIGO system (one optimizes and the other does not), but who says that the TS4-R-S does not influence the system without optimizing it? Have you tested panels connected directly with an inverter, without going through any Tigo device? If not, an ideal test would be a series of panels connected to a classic inverter and the other series to Optimizers. Of course, the details of each panel would not be available, but the general production of the whole would provide more than interesting data. Thank you for everything, you are great and your way of speaking English helps us a lot in Europe to understand your language more easily!!
I think you are right, there is no possibility to have different amps through the system without optimizers. Difference in 0.05 amp could be just some measuring error, but 0 on one panel means no current flowing, so how the rest of panels is flowing? there is some kind of optimization in those "nonoptimizers". Simply decoupling PV panel from rest of the string still is optimization.
Hi Martyna, was the SMA shade fix switched off on the Tigo optimized panels, and was it switched on for the non-optimized panels? Towards the end of the video some production data is shown from the SMA monitoring platform but it is not explained which shade setup is being used for this production data
It's great doing this, but I'm still left wondering whether the output of a string is severely affected by shaded panels.
Of course the output of one panel is not affected by the one next to it, but in a serial string does a shaded panel cause resistance that reduces the string more than just the missing output of the shaded one? You should show the output of the string as well as the individual outputs.
In a truly series circuit, the current is the same through all the panels. The voltage of each panel can vary, adding up to the array voltage, divided by the entire circuit resistance (panels and load) ultimately determines the amperage, but if not enough amperage is available the voltage plummets to the point where V÷R=A becomes true again. So for any one panel to have a lower power with full sun than normal, would have to mean that it's being over loaded down to the point it can't maintain the voltage. When that happens VxA=lessW. This showed that did not happen in the series array here as the inverter adjusted its demand to match amperage availability. But theoretically the same thing could happen with a micro inverter on that one panel with full sun, using enough power to draw the panel voltage down thereby lowering its output wattage. Although I would think that the inverter would just cut off before that would happen, when it was the only panel with full sun it would be doing most of the work. But what I think actually happened here is all the inverters shared evenly, which meant it only drew from the panel with full sun an equal amount that it could get from the other panels so that each inverter was working the same and not overloading any one in particular. It doesn't mean the panel produced less, it means the inverter used less than the panel made available.
Great work !!! Best regards from Spain. As you said, the first real test I found in internet. In Spain 80% of the people use Hibrid Inverters, and in France 70% use Microinverters... Why ? No idea..
If you don't mine , I will post your videos on European forums. Best regards
Being an Admin of 16K members FB Solar PV Solutions group
I wish to share this presentation.
Thanks
The discrepancy in the test results can be attributed to the fact that the devices under consideration measure the current leaving (and voltage) of the panel, rather than the current (and voltage) put on the string line. This is an important distinction, as the current on the string line is identical for all the panels, given that they are connected in series.
In order to optimize the power output from the panel, it is necessary to reduce the current while increasing the voltage (or vice versa). This is achieved by approaching the maximum power point tracking (MPPT) operating point. However, in certain circumstances, it is not possible to increase or decrease the voltage sufficiently, resulting in a lack of optimal power generation for the specific shadowing condition. The shadowing of other panels affects the power produced by a panel due to the slow output current and the necessity for the optimizer to adapt its output current to the current of the string. This occurs while the optimizer attempts to counter the inverter's efforts to alter its apparent load in order to center the MPPT.
It is crucial to note that an essential parameter was not measured: the voltage entering the inverter. When the voltage drops below a certain threshold, the inverter is unable to track the MPPT, leading to complete shutdown.
Great video - thanks! Just to clarify, I assume the voltage and current measurements are made on the panel side of the optimizers and the monitoring electronics. On the inverter side of those devices, the current in each unit has to be the same since they are connected in series.
Correct.
Exactly! The units measure the current on the input of the optimizers/ rapid shutdown device. Hence the current has different values. Its like seeing all panels in isolation.
Great content Martyna. Keep the videos coming. Thank you
Thank you sir for leaving a comment and words of encouragement. :) Appreciate it!
good job, very informative. deserves more views and top search results
Thank you so much for the kind words. I so appreciate it. I am just starting my youtube venture, so I hope in the future youtube algorithm will pick things up more! :)
This test actually shows the advantage (also for safety) of using a simple bypass diode in parallel with the solar cell groups. in order for the solar module to be efficient all the cells in a series circuit need to be producing the same voltage/amperage. If they are not the cells that are lacking become more resistive and produce heat. Heat is wasted energy and it can damage the non producing cell (even burn). Non producing cells in a string also start to have more characteristics that of an LED then a solar cell. If you could get the non producing cell from a string under load in a dark place you would actually see it give off some light as well. this reverse flow (biasing) is what causes loss in the string. it is also not good for the cell. A bypass diode gives another path for the electricity to flow. The theory is that the voltage will flow first through the path of least resistance. As the non producing cells heat up, they get more resistive and eventually the diode becomes the more prominent path. But, also bypass diodes create some voltage loss as well. Schottky diodes are used for bypass diodes in solar modules because they have higher series resistance then other type diodes and keep the reverse voltages at safer levels. Another use for diodes in a solar circuit is a blocking diode which blocks an entire module or string. Bypass diodes are installed in parallel, blocking diodes are installed in a series circuit. Thanks for all that you do!
Thank you so much for sharing that information! Really valuable knowledge! Thank you again! :)
Fantastic. Thanks a lot! I just got my 18 panel installed with the tigos. Pretty sure there is not an actual ROI under my conditions. But the generated data and graphics are so cool :-)
Thats great to hear! HAPPY PRODUCING IN 2024! :)!!!
Great video as always!
I know that SMA highly touts their "shadefix" system that purports to do some amount of optimization to handle shading. I'm curious whether you would expect a different result using a different inverter on the non-optimized side?
Yesss! I need to turn off shade fix on one inverter and test it and see what it does exactly to the string! I also want to test other inverters. But Shadefix comes in as turned on for most inverters and it was on for these tests. But I will definitely see how it works. !! :)
The voltage should be 0 or negative when the bypass diodes conduct...
Or does the tigo disconnect it from the string?
Good question, but I agree.
Question: You refer to "Ever Since Bypass Diodes Became Used In Panels", since when was this? I cannot find this on the web. Your tests show that Panels with Bypass Diodes do not shut down the series string but what about older panels? I think it was is true with older panels, but how old is old?
What a great video, thank you very much for the transparency and honesty,
The way you explained it was so easy to understand.
Looking forward to watch your next vids.
Thank YOU so much for saying that. Means so much to me it was helpful!!!
Great video!
It would have been great to include Fronius - it is known for its ability to handle shades better than optimizers.
I really want to test fronius ability to optimize string, a few people asked about it too! That would be a very good test! I am definitely starting lots of tests this month! :)
Very impressive test!
Constructive criticism: if testing partially shaded panels, it would be far more reliable and scientifically correct to use a jig to accurately position the leaves so they cover exactly the same amount of the same cells in each panel. It may or may not affect the results, but at least you’ve eliminated a variable.
Keep up the good work!
Thank you so much for the feedback! I do agree and appreciate it! I will try to be super detailed with tests coming out this summer. :)\
@@solartimeusa making tape “loops” and keeping the tape completely behind the obstruction would also be a good idea.
Thanks for the reply, it’s great to see creators who engage with the comments! 👍
Thank you for this test. I recently watched another video suggesting far less of an impact on string shading, but their test method, presentation, and conclusions were already suspect to me.
Thank you! I plan on doing other tests with similar shading but also with different brand inverters. I also want to test more filtered shading like trees. :) I am curious if results will be similar or very different.
Again, this would be case specific. So what are the extra cost per panel if you include the optimizers ? If you have no more space on the roof..possible a true advantage if you calculate around 1.2KW/h per day in saved cost. But if you have enough space on the roof, what would be the result if you would have gone with a slightly larger system at parity cost?
Can you do it again with half-cut panels? Ideally once vertically and once horizontally mounted with a flagpole shadow and a wide diffuse tree shadow sweeping across a full module?
I suspect half cut modules are better mounted horizontally so a horizontally moving thin shadow only shuts down half a panel at most. But a tree shadow would be way more interesting.
Fajny test. Otwiera oczy szeroko zamknięte 😊 pozdrawiam z PL
Bardzo dziekuje! Pozdrawiam i SMacznego jajka! :)
I would love for you to do this with the Enphase system vs string inverter vs power optimizers.
Thank you for this!
❤ from Malaysia!
You're welcome 😊
I bet, if you compare "with optimizers" array with "no optimizers or bypass diodes" array - results would be so much different.
Since the bypass diode actually does its job as well, and it optimizes the power output of a whole string/array based on the switching off capabilities it has.
That’s the most amazing real life pv test on YT. I was expecting perhaps 2-3% if anything at all.
Would you compare optimisers to microinverters?
Have you seen it yet? Just posted that this weekend!! :)
i've seen demonstrations that showed that the optimizers don't work at all, and actually produce less power.
It seems it is because of the 'blind deployment" without CCA+TAP. It's also related to a particular firmware from 2021, and not sure if that issue still exists. I'm very interested to know whether you tested "blind" or with CCA+TAP, and whether you found the TS4 to misbehave in any situations.
Awesome and very useful testing.
Thank you so much! :) I am very glad!!
Hi Martyna, how about shingled panels? Does there any reliable shingled panels that can perform better compred to half-cut with optimizer? How about the cost comparison with half-cut+optimizer and shingled panels?
So glad I found this channel!
I am so so glad it is helpful! Thank you for leaving a comment too! :)
Hi
Thank you for this brilliant presentation - All Real time performance comparison practical demonstration.
Keep smiling
Martyna super film. Dziękuję i pozdrawiam 😊
Ja dziekuje za komentarz! Bardzo serdecznie pozndrawiam :)!!!
Was SMA Shadefix turned on with this test? Have you tested to see if production is better with Tigo Optimizers and Shadefix or just Tigo Optimizers?
Hey! ShadeFix is automatically turned ON on most inverters so this included shade fix, but I do wnat to test two separate inverters one with ShadeFix on and Off to see what really it does. I almost cant imagine a drastic change, because the bypass diode worked, so I am very curious myself how DC would be then conditioned. Definitely need to TEST IT SOON! :)
Thanks for the video, I think this is one of the best reviews of how power optimizers can be useful. Very helpful and shows how panels could be affected in the real world.
but please don't use that "Text" in the middle of the screen feature (early in video)... it is one of the most annoying YT trends ever.
Hey Marc! Thanks for the feedback on the test and on the editing! I appreciate all info, so thanks for letting me know. I have to take it up with my editor ...:D :)
Martyna
The best panel video ever.
Thank you so so much!
Great video! Really helped me understand all this! This past summer I had a PV system installed on my house in San Antonio. It has 25 panels with mini-inverters and an Enphase system. The house does get significant shading from trees. I am already wanting to expand it. I do not have battery back-up but am considering installing it eventually. Also, the original installer (whom I shall never use again or recommend!) said there was no way to get power out of the system during a grid failure without a battery system, but as I understand it, there actually IS a device which allows for this. Do you know anything about that? Finally, the roof on my house is a low pitch standing seam metal roof. The current panels are installed on the south facing side. If I install panels on the north facing side would there be anything to gain by using a rack that will prop the panels up to an optimum south facing angle, especially if those are bifacial? (Roof is plain silver color)
Thanks again for your great, informative videos!
You need to "island" your system if you want it to work while the grid is down, the inverter needs the grid frequency to have power so you will have to have something to provide this. Other than a battery system I don't know whats available in the US to do this. Im in Australia and have sold solar since 2011 but inverters/ micro inverters work the same anywhere so thats the issue you face
Hi! I am so sorry for the delay in responding to you.
So yes, there is an option to use power during the grid outage with enphase.
I will include the link
enphase.com/homeowners/sunlight-backup-user-guide?hsa_acc=1236416428&hsa_cam=20130257651&hsa_grp=&hsa_ad=&hsa_src=x&hsa_tgt=&hsa_kw=&hsa_mt=&hsa_net=google&hsa_ver=3&gad=1&gclid=CjwKCAjws9ipBhB1EiwAccEi1Ld_T4TDI6NVlLA4375BrzzSQ9S33Y1lcXp6b6okeMKnX7FlpZXgCBoCvmIQAvD_BwE
The downside is it’s like having a battery set up ( cost of extra equipment) but no battery. I don’t think it’s worth the cost. It might cost over 4K maybe even close to 6k if there is a need for rewire. I suggest waiting till you get a battery becasue you will spend money once and have a good back up solution.
As far as pitching panels, you can email me and I can check the roof and see if that would make sense. The struggle here is that when you pitch panels you have to have spacing between them since they will start shading one another so you’re limiting space, and by just installing them facing north might cost similar to just have more panels, than less but a pitch.
Email me if you want me to review the address :) happy to help, again so sorry for my delayed response.
Info@solartimeusa.com
Great video.....very useful. next video idea is panel facing planning. I have 5 quote and they layout are all different....how to make decision
Oh wow! You know you can email me directly and I can try to help you determine which one is most realistic!
I do consulting for homeowners outside of my work area :)
Info@solartimeusa.com
It would be interesting to do the shade-from-building testing with the panels in landscape mode. Most panels strip the regions along the long length of the panel for each area covered by a bypass diode. That means the panels will work better in horizontal placement as the sun goes up or down. (I don't know if the optimizers would make a difference though)
They did do that test, it's towards the middle of the video. Additionally you can read this thread with a lot of good info from me and my system regarding all kinds of shading my system experiences diysolarforum.com/threads/can-shading-be-dealt-with-by-parallel-strings.54317/
That is a very good point, I do wonder the same. Surprisingly panels that were shaded in half and didnt have optimizers, completely shut down, which makes the marketing for half cut cell panels BS? I really want to test that too.
The problem originally came from solar panels that do not have bypass diodes. So a leaf on a single panel would drop production on the whole string by that %. So the 20..30% drop would be for all panels due to the single leaf on a single panel.
The bypass diodes prevent this as you have clearly demonstrated. Thanks!
I do have 8 of these pre bypass diode panels, but they won't see leafs or antennas (230Wp), but my next panels might get shaded, but have bypass and are slightly more than 500Wp. And there will be about 40 of them instead of, because I hope that would cover my winter days instead of having to settle for a bio-diesel generator. (a 500Wp solar panel is around $125 these days, a bio diesel is about $1500 so in costs I can have 6 panels with 2 mppt's)
yes, exactly! Yet the industry keeps using that to sell their own products. Just want people to have proper facts! :) I actually want to test older panels too, we have some leftovers in the warehouse from 2015s and prior... could be a fun battle to see :D
Thanks! I can explain the worries about "one shaded panel affects the others". You can see the data with your own tests, but the measuring devices/timing are very imprecise, so it wasn't obvious. When batteries/panels are connected in series (string) and you put a load, _exactly_ the same current passes through all batteries. No exceptions. The power-meters here you can see are not good, reporting 7.93A on one panel but 7.99A on another and 8.05 on another. That is physically impossible. It's possible that they are precise but the measurement+reporting happened a few milliseconds apart in time instead of simultaneously. Anyway, when one panel can do only 3.99A 36V due to shading, you can't possibly get more or less current than 3.99A from any other panel. So that's why people are worried that suddenly the whole string would produce half the amps thus half the wattage. Instead, the MPPT tries to draw different amount of input current, to find the sweet-spot. It tries 8.01A, the voltage of the shaded section rapidly falls below 0.6V as it can't meet demand, its diode turns on and the remaining 2 sections of a panel now output 8.01A 24V together. All panels resume to run at 8.01A, and the total voltage is simply 12V less. So, with 10 panels you don't lose half the total wattage (getting 3.99 x 36 x 10 = 1436.4W instead of 2883.6W). You only lose 12x8.01 = 96.12W, you get 360-12=348V 8.01A instead of the full 360V 8.01A. You see this in all your tests: all panels produce around 7-8A, and some of the sections of panels get turned-off, producing 0V and 7-8A. Well actually each diode that turned-on eats 0.6V * 8 = 4.8W and turns it into heat. The reported voltages on fully-shaded panels seem incorrect. It shouldn't be 40V 0A but 0V 8A. Actually -1.8V 8A. It must be a software bug.
The bypass diodes ensure that suboptimal sections of the panels are turned off. The MPPT does a good job of finding the optimal current to use from the remaining turned-on sections. The resulting voltage of each section depends on that current (e.g 9V at 8.6A or 12V at 8.01A).
Solar panels come with optimizers built-in: those bypass diodes 😤. A future improvement would be: remove the bypass diodes, and connect a new kind of optimizer with 4 wires to the panel. Optimize each section independently. I don't see it happening any time, soon, though. Currently the optimizers get crippled by the bypass diodes.
Hey!!! Thank you so much for the thoughts and input! I should have explained in the video that the Tigo units dont measure the OUTPUT of the optimizers/rapid shutdown units. It is measuring the input from the panel itself, so we can see them all separately in isolation. So the actual current of the string is the equalized by the mmpt in the string inverter itself, so I can totally see how that seemed misleading. What I learned from Tigo ( as much as they could tell me) is the units measure the input not the output of the unit itself. But this LEADS me to ANOTHER TEST! Lets see if I put 8 panels with the TS4S ( no optimization) and just string inverter with no units under the panels, and compare the outcomes. I am very curious how the results will vary, and if they will.
I do like your idea too!!! Hopefully one day! :)
Martyna
@@solartimeusa You may be interested to learn that Tigo record both Vin and Vout for each panel, but for some reason Vout isn't made available directly in their charting application. Good news is that you can just edit the URL, adding '&suffixes[]=vout'. The data is also available via their API.
At roughly $100 per panel to install Power Optimiser, works out to $1000 for a string of 10 panels, for that same $1000 you could purchase 5 more panels at roughly $200 per panel, giving you a string of 15, the output then would blitz the output of the 10 power optimised string, subject of course if you have the room for 5 extra panels.
I like that idea too! You just have to see if you have the space and least amount of shading, but very very good thinking!
Hey this is amazing! I’m really struggling to pick between a local installer and Tesla here in upstate New York. For a same sized system Tesla is estimating. 6300kWh / year but the local installer is estimating 9478kWh/ year. Whose right! I even did a shading analysis in helioscope and the local installer was much closer it seems. Why is Tesla estimate so low? Is it cause of their string inverter vs local installers micros? Thanks!
Hey! I would say go local, if the company has good reputation, good online reviews and if they have been around for some time.
Does the installer offer any production warranty?
Tesla is undershooting so they don’t have to deal with complaints if underproduction later on. So they are playing safe.
The biggest problem I see with Tesla is service. In DFW in texas I service Tesla systems weekly at a charge becasue of how long they wait for service from Tesla. So if you can choose an installer that will service your system in the future, that’s the way to go. Local company might have a more expensive price but you’re also supporting local over huge corpo where nobody cares about you.
I am biased because I am a local business too. But I think you will be better off going local if company is reputable.
I hope this helps 🙌🏼
nice videos. I would have made sense to spend a minute at the beginning showing the sub-strings layout on these panels, if they are half-cut cells and explain how many bypass diodes they have.
Very fair point, thank you so much for the tips. I am working on a ton of tests coming out this summer, so I will make sure to be much more detailed.
I don't undertstand: Are the panels in each string wired is series (in which case I expect then to all have the same current) or parallel (in which case I expect them to have the same voltage), or what?
What should we test next!?
best solar companies based on customer service and accreditted best service??
Diffused shading from trees that moves across the panel array would be neat to see
@@notexpected ....yes, do this !!!
How well do split panels do with this test? Looks like they have double the diodes and should be closer to using optimizers.
Can you show us the costs for both systems? a 10% improvement with optimizers is probably not worth the 20% or whatever increase in cost.
BougeRV solar pales.
I just created a mini nuclear reactor in my garage so all is well.
You bought an EV😂
@@sirjohng1 you didn’t 😂
No you didn't
For test one, on the non-optimized side, was the leaf placed on the panel at the beginning of the series, or at the end of the series of that string?
I think the real test is to run them side by side for an extended period of time. Don’t forget that modules and components within the modules will deteriorate over time - this will have a huge impact on performance as time passes by. Not to mention cloud cover and soiling which are inevitable.
The inverter’s role really starts to pay off not on day one but over the 15 to 20 year life span of the system. Nearly every test I’ve seen on the subject does not cover the subject of long term yield
Hi Martyna, or Martina? ;-) Nice video!! I only little bit don't what all these boxes are. You showed two smaller black boxes with some cables... Does it mean that each solar panel has one black box connected underneath (so there are 16 hidden boxes)? It would be nice to see wiring from back of panels. And you are speaking about diods - are they in those black boxes or are they built-in solar panels? It would be nice to see some theory first (for such beginners like me) - like wiring diagram of both configurations. (If black boxes are placed in both strings, then it would be nice to have third string without any black boxes - 5 panels in each string is enough, no need for 8.) And I'd love to see what each black box actually do. Now I maybe understand that first black box if only for measuring current and voltage but is passive - doesn't do anything. (if so, then 3 strings doesn't make any sense) And second with optimizer is basically acting like tranzistor which is connected in parallel to panel and if there is shadow (power drop down + increasing internal resistence of solar panel), it opens itself partially and allows bypassing current from other panels from the string? (Btw excellent clear English, I am CZ and can't hear almost any accent.)
Hey! Its Martyna ( Polish) but when pronouncing it in english it is pronounced like Martina.
Yes so each panel has a small back box behind, that is built in, and it includes the bypass diodes. Then what I did is used 8 power optimizers ( extra black boxes) under one set of panels, and 8 Rapid shutdown devices ( extra black box) under the other eight. I would love to dig deeper into how they are wired, hopefully in the future video. But in an easyway speaking the panels have their connectors that connect to power optimizers ( male and female connectors) and then those then connect in series to other panels.
There are two ways of connecting panels together either in series or in parallel. I definitely want to do a video on that, so make sure to subscribe :)!
Thank you for leaving the comment!
Hi your videos are excellent!Installer in ireland.Thanks mg.
THANK YOU SO MUCH! I appreciate you taking the time to leave a comment. Hope your business is doing well. I know EU is suffering from downtime in solar...
Interesting stuff. You're right, the repeated narrative over and over has been, shade one panel on a string and the rest only produce the amount of the shaded panel. Obviously, this isn't necessarily the case.
Do the brand of panel and maybe age of panels make this hypothesis different?
I would like to see a test of two systems with mixed panel wattages and see what that does on a string system. That's another thing that is always repeated as a no-no, never mix wattages. How would that look in this same test?
Thanks for the feedback, and yes that would be a really cool test, to mix brands and panel voltages. Like adding a 480W panel to a 400 w panel string! Thanks for the idea!!! :)
Another benefit of optimizers is the ability to mix different panels within a string (within reason). Note that the Tigo optimizers are buck converters not buck/boost converters therefore they can only turn higher voltage lower current into lower voltage and higher current, not the other way around. This generally is fine unless you're starting to look at and work with uneven size parallel strings going to the same inverter input and the likes ;-).
Cześć, bardzo dobrze widać że dodanie Tigo tylko w wersji monitoringu do panela HalfCut zupełnie zmienia jego działanie. Diody bajpasowe nie powinny zadziałać po zasłonięciu połowy panela tylko prąd modułu powinien spaść o połowę - niestety w takim przypadku prąd całego stringu też by spadł o połowę więc dodanie Tigo bez optymalizacji to właściwie poprawia działanie całego tego systemu bo załącza się w nim wbudowana opcja bajpasu. Moduły HC sprawdzają się tylko z mikroinwerterami lub optymalizatorami.
Co da samego testu zacienienia to w przyszłości proszę postawić przed panelami cienką rurkę lub powiesić linkę symulującą przewody zasilania lub cienkie gałązki. Taki "soft shadow" ma druzgocący wpływ na pracę całego całego układu stringowego - cień jest na tyle mały że diody bajpasowe się nie aktywują, a spadek prądu całego stringu jest ogromny. W takich przypadkach widać dopiero zyski z optymalizacji.
Is the boost with it $/kWh vs just spending the optimizer $$$ on hanging more cheap solar panels?
A bifurcated panel test would be interesting. I have a customer who wants ground mounted but doesn't have the space.
oh yes!!! I really want to get that done soon! Thank you for the idea. I have some in stock too that are 480W QCELLS.
Very, very helpful my next project will be without optimizer but with panel ID device
Thank you so much for leaving a comment! Good luck with your project! the 10% difference is also based on those extreme shading conditions, the percentage would go down, as the shading diminishes.
panels have an efficiency, an amount they can generate for a given amount of sun. cover half a panel, IE give it half as much sun, you will get half as much production, no magic box you can buy to add onto a panel will or can change that.
top say that covering half a panel and yet still get 3/4s the power, means your changing the efficency, and thats not really possible.
when panels are in series then 'per panel bypass diodes' are the only option available to you, per cell bypass diodes cant be added after the panel is built.
best cure for localised shading, micro-inverters to ac or individual mppt->dc->boost to a dc bus
I'm not sure what that per-panel monitoring box is doing, but it is not behaving as series string of panels without those boxes. Interesting test none the less, but I don't think it will apply to a string without electronics at each panel.
Hey! Thank you for your comment. The units are called tigo Shut down devices. Supposedly they do not “condition” the power simply provide rapid shutdown and monitoring ( reading data) feature. But! I do want to test them compared to simple string inverter to test whether a string behaves like that or not.
I am curious myself.
Thanks again for leaving your thoughts! I appreciate it
A ground mount system, without tree shading, but in a cold climate where snow might shade a panel, optimizers should be a benefit.
Yes definitely. I do see big benefit in them, just for the most cost effective solution, sometimes not needed. Thanks for sharing your thoughts, I appreciate the comment! :)
Just curious. How did you measure the output of the non-optimized panels? Nice video.
Hey! I used the Tigo Rapid shutdown devices, that offer monitoring and rapid shutdown feature, with supposedly no optimization :)
Her is the link:
www.tigoenergy.com/product/ts4-a-s
What an informative video! Thank you so much!
Thank you for saying that. I truly appreciate it :)!!!
one thing is missing to me... The test was o so great... what about to take the same money what optimizers costs and buy more panels without them and try that comparison (what do you get for your money).
Yes very good point! I’m actually testing other equipment like microinverters now. But definitely want to test different shading type like filtered shading from trees etc. stay tuned ! And thanks so much for leaving a comment! It helps me so much!
Awesome job!
Thank you! :)
I really think spending your $ on more photovoltaic surface are has a better ROI than spending $$ on extra subsystems like optimizers. A optimizer costs near the same as a 2nd panel for only a 10% gain, where 2 panels is a 100% gain for the same cost?
Not so concerned about individual panel knock downs, but total power output of whole string.
I would like to see this repeated with two strings in parallel like Tesla Solar Loves to do.
I would like to see how different azimuths and pitch’s affect daily production on a single string.
I cant wait to do some of those tests with newest PW3....:D
Greetings, I am currently residing in the state of Florida and employed in the field of construction. I am interested in exploring the possibility of collaborating on the construction project you mentioned. Would it be possible for us to have a discussion regarding this matter?
I am currently exploring the option of replacing our electrical service with a battery power bank for our job site trailer.
You have just answer my question that I posted in another video of yours regarding the need of optimizers in set up with no shading issues. I guess that uneven panel degradation over time and / or dirt accumulation on the panels play no practical role on the question optimizers or no optimizers. That is, because we are talking long term.
Regarding the existence of the Tygo panel level emergency shut down modules on the no optimizer side of the test panels, these modules are just switches, hence their cheaper prices. Tygo introduced this solution in order to serve consumers who have some Code to abide to but not necessarily have shading or monitoring concerns. Being , nothing more than switches, they introduce extra resistance in the circuit as opposed to a straight wire connection. Therefore, your test results are 110% valid and actually on the conservative side as far the performance of the no optimization panel side. If you remove these switches completely on that side, the panel output will be slightly higher than with the cutout switches on.
I am so sorry I missed your other comment. I really try to respond to them all! :(
But thank you so much for leaving your questions and feedback and thoughts, it helps me a ton, and obviously, I love to know if I make any mistakes while we test those different products, so I can improve and keep learning.
The problem with these tests is they are not representative of a real system shading. Basically the panels are 2 independent 200 watts solar panels. If you have square shadows, this is a great test. But in reality shadows are cast over both halves of the panels and this causes some serious output problems that will not be solved by a optimizer. That is why alot of systems are wired series/parallel to isolate shaded panels. Test the panels where both halves are shaded by a oval shadow. The best you can do is add more panels to a system as the panels are cheap. Great to show the issues with shading!
Hey ! Thank you so much for the input, and yes these were pretty extreme conditions, and I plan to do more diffused shadow as well that will overcast multiple panels etc. :)
Thanks again for leaving your thoughts, I really appreciate that and love hearing the feedback!
great Video,THANKS !
Thank you!
If you are using a string inveter with all 8 panels in 1 string, how is it possible for shaded panels to have less current than the full-sun panels? My system is strings of 6 panels, and I thought optimizers were required to prevent a shaded panel from reducing the output of the other 5 panels in the string?
SO the way the optimizers work is the try to find ( in mathematical term) best PRODUCT of the two (voltage and current). Sometimes they have to lower the current to increase the overall production outcome.
Also the optimizers do provide the INPUT values, so whatever enters ine optimizer, not the outcome. So later on the string MPPT also equalizes the values to track the bets power point of production. I hope that made sense?
Martyna
Are the costs of the optimizers more, or less expensive than adding 10% more modules?
Tigo optimizers cost me $22/panel on a 44 panel system and produce a 6.4% improvement in generation for 1.5% increase in total system cost. ROI in the Tigo optimizers is 4.4 years while it's closer to 13 on the whole system.
Additionally, sometimes adding 6% extra panels is not just a roof space issue but a string configuration issue, ie. You can't just add another panel to a 2x12 string.
@@greg_takacs You seem to know very well the ins and outs of these system!
Awesome test, taught me a lot. Please get rid of the ding noise very annoying.
Hey!!! Thank you for sharing the feedback, thats the only way we can improve! :)
Can someone help me understand why natural shade has no effect but artificial blockage does?
I don't have solar yet but trying to learn. So my question here is. Would it be more effective to buy 8 Optimizers or 1 more solar panel?
That is a very good question!!! And the problem is the answer is not super straightforward. This will definitely depend on the severity of shading. In some cases shading is so minimal that it wouldn't be worth it, sometimes space is limited so power optimizers are the better option.
It looks that, when more than 70% or 80% of the panels are covered by shadows in a string, the optimisers just try to work at a medium level of current and voltage for all the serie.
How were the panels connected and how did that affect your test
The panels were connected in two separate strings and in two separate inverters. Does that answer the question? That should not cause any changes in the tests as they were connected in an identical way.
I only have shading in AM/PM when sun moves into field of panels or out of field. My neighbor has string inverters with optimizers which cost more than I paid for each panel. I bought used. Adding more panels and another inverter for me or anyone with no shading is the way to go if they also have a all in one inverter like I do which is 2 5k inverters.
Thank you for sharing! Yes unless there is substantial shading I would not worry too much about the optimization whatsoever! Thank you for leaving a comment.
Where space is not constrained you can probably install extra solar panels with the money saved from not using micro inverters. This is especially relevant for ground mount systems.
YES! so so true!
YES! so so true!
The optimizers are extra work and based on the promotional blurb can add up to 20%. So give it to them, if its a 20% gain you are after compare the cost of adding optimizers to each panel to adding 20% more panels. The panels are cheaper. This works assuming there is room on the roof for extra panels.
Optimizers in my real life system give me 6.4% improvement for 1.5% of the cost of the system. It's not 20 but the optimizers still had better ROI than my entire system. Adding an extra 6% panels also cost more than just the panel cost, mounting and labor are not free.
Thanks for sharing your data too Greg! I am sure that depending on shading level the results will vary as well. I do want to test more diffused shading like trees, and maybe different orientation of panels etc. Will be cool to see and verify results.
The tests show that your money is better invested in more solar panels than optimizers.
The panels will likely last for decades while anything electronic is looking for an opportunity to fail.
Of course,ground mount is better regardless of optimizers or not because of ease of maintenance.
Very very fair point, lots of people feel same way, the only issue in usa is the rapid shutdown law.. that makes us use some sort of panel level device for rapid shutdown ; /
Great Video, so much to unpack. You are so smart and so cute :) Great to see the diodes actually work ! I was a little confused how the non-optimized panels were configured. Did they use microinverters or some other way to achieve MPP ?
Couple of questions for you:
1. If there is NO shading, would optimizers produce more/less/same power than no optimizers ? Other than having a better MPPT algorithm I cannot see why they would produce more power.
2. If 8 panels were connected in series to a string inverter, which I believe is your preferred configuration, how would the results change ? My guess is this is where optimizers really pay off, but it's to compensate for the disadvantage of connecting panels in series. Of course there are advantages to series connecting panels like saving on copper wiring costs.
3. Wouldn't microinverters be the best overall choice for a lowest cost/best performing system ?
I'm not as smart or cute, but I'll try to answer ;-)
1) If there is NO shading there is no benefit for optimizers, other than they can balance the small difference between panels as not every panel will produce the same current even at the same solar radiation due to manufacturing and material differences. But optimizers, while very efficient, will cost some power loss in and of themselves so I'd say the two (manufacturing differences in panels vs. power consumption of optimizers) is a wash. Note that you still get the panel level monitoring, including alterts if there is an anomaly, which I think is worth it.
2) Connecting panels in series is not only beneficial from high voltage low current but it's also a requirement as whatever combination you decided to go with has to match the working window of the inverter. Since most inverters work between 100-500V and panel voltage is 30-40V depending on cells you need to put them in series to reach the minimum working voltage of the inverter. Additionally they are also current limited (generally under 30A per MPPT) so you can only do so many strings in parallel before you overrun the current limit ( usually 2, maybe 3 strings in parallel per MPPT max)
3) For a smaller system microinverters make sense from a cost perspective as the system cost is marginal ie. 2x the system 2x the price, 15x the system 15x the price. But for a well sized and optimized system from the ground up you're almost always better off with a properly sized string inverter and matching size panels. Additionally, microinver systems end up costing a lot more if you're also considering battery backup, which in 2023 everyone putting in a solar system should.
@@greg_takacs
Thanks for your reply. Well, you may not be cute, but I think you are pretty smart.
As I can tell there are 3 downsides to operating the panels as strings:
1. A shaded panel will limit the current in the entire string.
2. The MPPT controller in the string inverter will maximize the power, so you better have all the panels producing about the same.
3. A degraded panel will reduce the current similar to 1.
BTW, I do have degraded panels on my microinverter system.
The remedy is, of course to use optimizers for case 1 and 3, which are basically buck converters, keeping the current in the string as high as possible at the expense of the panel voltage. Pretty clever really.
But you effectively get this capability already when using microinverters and you don't need a string inverter. You can mix and match all kinds of panels with microinverters, big/small/shaded/degraded etc, since all power in normalized to 240V split phase and paralleled together. This allows for split strings with panels in different orientations (south and west) to capture the sun later in the day when the tariffs are high.
I am a bit puzzled why the "cute one" is so enamored with string inverters but I'm sure you'll set me straight.
BTW, I do have a grid tied system using microinverter and an off grid (3x strings) with batteries for pumping water in the event of a utility failure. The biggest downside I see on the off grid system is once the batteries are full and I have a full tank of water, the array throttles back. Maybe time to upgrade to a full hybrid inverter.
@@ai6mk897 issue #2 is also solved with optimizers, and Tigo optimizers can even solve it between parallel strings going to the same MPPT as long as every panel has an optimizer and they're controlled/managed through their CCA (Cloud Connect Adnvanced)
I can't speak to why the original poster loves string inverters, but I can tell you why I would never buy microinverters as the sole solution; They're simply inefficient, especially once you add battery to the equation. Additionally they're too expensive as every panel requires another inverter and Enphase's solution requires the same on the battery side, specifically additional number of inverters for the batteries as well, again adding more to the cost and padding Enphase's profit margin.
In 2023 I would not consider a solar only installation, full stop. And once you're in the thick of it with having batteries added to your system, a hybrid inverter with proper DC coupled battery management and solution is just overall a better user experience.
I personally have a Sol-Ark 15K system with 38 kWh of batteries and 19.4kW of panels with Tigo Optimizers. I don't think there is much else on the market that can give you what a 15kW hybrid system can for the same cost. I don't have critical load panels, I just have an inverter sitting between my meter and my main panel. I'm off-grid during the day and use free electricity from the grid at night. Couldn't do it with a micro setup.
Having said that, with the Sol-Ark 15K I can add AC coupled solar directly to this system and it will manage it as well and utilize it in full island mode, and this is where a few extra panels with micro inverters just might make sense, although I'm still leaning towards getting a string inverter and Tigo Optimizers for this as well.
Thank you Greg!! You beat me to my answers :) I do agree with everything you said. :)
Hey! :) My turn.
So to address the 3 points you mentioned. This test showed that with panels being connected in series, the shaded panel does not affect the rest of the panels, so that would not be a downside. All thanks to the bypass diode.
But even if that was the case, using a string inverter in a smart way can yield similar results as microinverters.
For example I have an 8 KW system with 14 panels facing east and 6 panels facing east. I built 3 separate series of panels ( 3 strings) without mixing the panels from west or east. I also did power optimizers even though I am really not benefiting from it. :D I don't have any shading.
Essentially both options are extremely good, Greg made a good point ( #3) about adding batteries to the system, and with hybrid inverters that is the most efficient way. But with your microinverters you can easily add ac coupled battery too. It might just be more expensive and a bit less efficient.
I did make a video on why I do prefer string inverters. I hope you will give it a listen. Its too long to type :D
ruclips.net/video/4K_q7hGFLuo/видео.html Let me known what you think after you watch it. I am very curious!
Here is a link to my personal system ( got disconnected from wifi, but I will reconnect it.)But you can view some on it :)
ei.tigoenergy.com/p/w5wjwUT3spa_
It was obvios and very easy to make the calls of the production even without the testing because you used hard shadows you are forcing bypass diodes to get activated. When it would get interesting it is with soft shadows let say 8 modules 3-4 covered with a soft shadow reducing the current output of the shadowed modules 80%, then the string inverter I bet it would produce 20% less in the whole system, because the other option for the MPPT of the string inverter is to activate all the bypass diodes in the 3-4 modules shadowed.