Optimizer question

[Funkeechild]

I'm wondering if I should have my installers move some stuff around or not. Here is what is going on.
I have Trina Solar smart modules. 22x300 Watt panels and a Fronius Primo 6.0.1 Inverter. Panels are 72 cell instead of 60, total of 6600 watts and they have Tigo optimizers on them from the factory. I wanted to input from experts so I'm posting in this forum. The panels are laid out on 3 different sections of my roof. They were supposed to split them into two strings of 11 ea, they ended up going with 3 strings, 7 7 and 8. Anyhow one of the strings extends into a panel on a different section of my roof and at a different tilt and that panel hasn't been producing well at all. It seems like it is limited by the string it is attached to, as that section is shaded until afternoon. Here is a screenshot from my tigo monitoring pg you can see the top right panel of the left side section is under producing.

http://1drv.ms/1OGE7AR

You can kind of tell just by looking at them. The 7 on the right side face south and produce great. The others are at a SW orientation and produce pretty good. The 6 in the middle are along the ridge of my roof. You can see my house is the background of the Tigo screenshot so you can see how they are laid out. The 9 on the left sit flatter on my loft. The top right panel of the left section doesn't produce well until about noonish, then it catches up. In other words the other panels in the same area of my roof with same tilt and no shade are producing like 100 watts and this one producing like 40 watts. I don't know if there is a threshold where the optimizer starts to work better or what? I'm wondering if I should have installers fix this. It doesn't seem like they did it right. I thought with the optimizers that the panels would be producing just as much as the others right next to it and it shouldn't matter that it's on a different string, but it is definitely affecting it. As I read more about optimizers it seems like they only help to produce more and don't maximize 100%, and since the rest of that string isn't seeing much light that one panel can't put out its full potential. Installers said they didn't want to extend the string too long and surpass 600 volts...which I understand. I'm still not sure why they didn't do 11 and 11 or just split them up by sections. I think they said that the inverter would work better having equal strings, which also makes sense. I know about solar panels but I'm not an expert. I feel like it would have been best to put 9 on that one string on the flatter part of the roof and have 6 on the middle string, and 7 on the string on the right. They are on different spots and produce differently throughout the day so it is hard to figure out what would've been the best way to install them Any help would be appreciated. Here is a pic of them producing early in the day, and the panel I mentioned is lagging badly compared to the panel right next to it....the layout of the panels on the screenshots is by layout on the house not by string or block. You can see the middle string extends to the left string one panel. AGain the background is my house and layout. The front of the house faces SW

http://1drv.ms/1Lmnx5v

Here is a pic of them producing well after about 1230 pm

http://1drv.ms/1LmnEhw

AGAIN ANY HELP IS GREATLY APPRECIATED. I HAVE A 10 YR LABOR AND MATERIALS WARRANTY AND I SPENT A LOT OF MONEY AND WANT TO MAKE IT BACK ASAP. LET ME KNOW IF THERE IS A BETTER WAY TO SET THESE UP. THX!!!

[foo1bar]

I'm wondering if I should have my installers move some stuff around or not. Here is what is going on.
I have Trina Solar smart modules. 22x300 Watt panels and a Fronius Primo 6.0.1 Inverter. Panels are 72 cell instead of 60, total of 6600 watts and they have Tigo optimizers on them from the factory. I wanted to input from experts so I'm posting in this forum. The panels are laid out on 3 different sections of my roof. They were supposed to split them into two strings of 11 ea, they ended up going with 3 strings, 7 7 and 8. Anyhow one of the strings extends into a panel on a different section of my roof and at a different tilt and that panel hasn't been producing well at all. It seems like it is limited by the string it is attached to, as that section is shaded until afternoon.

Is it limited by the string it's attached to?
Or is that corner of the 3x3 panels also in the shade?
Even a small amount of shade on part of the panel can affect it's performance significantly.
I would take a picture of what the sun/shade looks like and re-do the snapshot of the production at the same time you take the picture.
Then someone can look at it and see clearly whether there is a difference due to shade or if it's due to the string.

[Funkeechild]

Its not shaded...there are no large trees by my house. The 3X3 panels especially. The only shade they might see is from the ridge and that isn't the case with this scenario. I've watched them throughout the day. The 7 panels facing south, the string on the right bottom, there is a panel that is shaded by roofline in the morning. This definitely seems to be the other string that is causing the issue...or the fact that it is attached to the other string that isn't seeing much sun yet. Once the rest of the string gets sun that panel starts to perform well. I'm really just wondering why it was installed that way and if it would've been better to have all the ones in the same areas on the same strings. I think I would be getting more power that way, but then the strings might not be evenly distributing to the inverter....I'm not sure what affect that would have though...uneven watts coming from different strings? I was told they need to be balanced, which make sense. Right now its 7, 7, and 8 on strings. The way I'm wondering about would be 7, 6, and 9. 9 being the flatter roof and seeing sun for longer in the day, 7 the south facing on garage, and 6 on the ridge.

[sensij]

Slow down. You need to understand that an optimizer will help mitigate the consequence of shade on the rest of the system, but can do nothing to make the shaded panel perform well. Shaded panels will underperform significantly, no matter what you do. Also, it would be expected that the S and SW strings will outperform and array with a different azimuth, that is a natural result of the angle at which the sun will be hitting the sub-arrays.

Your 7, 7, 8 string construction is (hopefully), the two 7 combined going into one MPPT input of your inverter, and the 8 panel string going into the other. Unbalancing these would be a bad idea.

Your links aren't working for me, so I can't comment much more on the specifics.

[inetdog]

Your 7, 7, 8 string construction is (hopefully), the two 7 combined going into one MPPT input of your inverter, and the 8 panel string going into the other. Unbalancing these would be a bad idea.

Since I do not know how they operate, I have no idea whether the Tigo optimizers could correct for the different string length too. SolarEdge definitely can.

But, if you want to see any reasonable output from the panel which is facing in a different direction, I do not think that the Tigos can do that. The problem is that while the sun shining mainly on the single east facing panel the power available from the other panels in the same string is probably low enough that the optimizers simply cannot produce enough voltage at that current to make the string inverter happy.
There are limits to what an optimizer can do.
Specifically, an optimizer is great for fitting one or two low producing panels into a string of high producing panels.
They have a much harder time fitting a single high producing panel into a string of low producing panels.

[Funkeechild]

Since I do not know how they operate, I have no idea whether the Tigo optimizers could correct for the different string length too. SolarEdge definitely can.

But, if you want to see any reasonable output from the panel which is facing in a different direction, I do not think that the Tigos can do that. The problem is that while the sun shining mainly on the single east facing panel the power available from the other panels in the same string is probably low enough that the optimizers simply cannot produce enough voltage at that current to make the string inverter happy.
There are limits to what an optimizer can do.
Specifically, an optimizer is great for fitting one or two low producing panels into a string of high producing panels.
They have a much harder time fitting a single high producing panel into a string of low producing panels.

sensij and inetdog-You both have given me good info. Very much appreciated.

Sensij-the panels aren't shaded at all. I live in new subdivision with no trees, the roof line shades for a few mins a day on two of the panels but the rest are good. The 7 panels on the lower right are facing south, all of the rest of the panels face SW, but at two different tilts. I think you are correct that the 7 and 7 are run into one mppt and the 8 into the other. Or it could be the south 7 are going to one mppt since they produce so much and the other two strings into the other? I'm not sure but I will ask my installer.

integod-I do know that solar edge optimizers work differently, they are more complex and perform two functions instead of just the one. I should've done a bit more research, but I got a good deal on the Trina 72 cell smart panels. Half the price of what the higher priced companies had bid, and substantially less than the Winaico, LG and Sunpowers I looked into. I was going to go with a more efficient panel but I have lots of room so I just added two extra trinas and it seems to be producing well. I think you hit right on the money when you said:

There are limits to what an optimizer can do.
Specifically, an optimizer is great for fitting one or two low producing panels into a string of high producing panels.
They have a much harder time fitting a single high producing panel into a string of low producing panels.

I think that they should've had all 9 on the loft on the same string. That one panel produces crap all morning til the sun gets further over my ridge. FYI....the panels face S and SW, see explanation above about orientation. I changed the pics from png to jpeg and attaching below. Any info/assessment/analysis/criticism will be appreciated! Tow pics show the morning, the flatter panel on the loft producing badly....the background is my house to give you a better idea of layout. Also the layout of the panels on the tigo site is by physical location and not block or electrical layout

A9 not producing-2.jpgA9 producing-3.jpgModule A9 under producing.jpg

[sensij]

OK, thanks for the additional pics, I understand better now.

The Fronius Primo has an MPP range of 240 - 480 V. Strings should be designed be in that range.

The TrinaSmart 72 cell panels have a Vmp of 35.8 - 37 V. That means you need at least 7 of them in a string to stay in the inverter's mpp range.

Your system would naturally want strings of 9, 6, and 7 based on the sub-array orientations. If you want to experiment, I would be tempted to make the 9 panel string go into one of the MPPT inputs, and put the 6 and 7 panel subarrays into a 13 panel string into the other MPPT input. I don't really know the Tigo product, but it seems possible that it would treat the half of the 13 panel string with the less optimal sun angle as "shaded", and the optimizers would do their thing to allow the more optimal half of the array to fully produce.

Beyond that, I'm not sure what other options exist that will be better than what you already have. Putting a 6 and 7 panel string in parallel into one MPPT input isn't likely to be great... it might work well when the 6 panel string is low producing, operating cooler, and therefore the voltage difference between the strings would not be as high as you might think. Once you get to the middle of the day though, when you are producing the most, you'll see some loss.

In the grand scheme of things, a single panel that under produces for only part of the day in a 22 panel system is really not costing you much. If you do the math on what the system would produce with that panel generating to its full potential, relative to what you are getting now, you will probably find that it isn't worth the time or effort to make any changes at this point.

[inetdog]

OK, thanks for the additional pics, I understand better now.

The Fronius Primo has an MPP range of 240 - 480 V. Strings should be designed be in that range.

The TrinaSmart 72 cell panels have a Vmp of 35.8 - 37 V. That means you need at least 7 of them in a string to stay in the inverter's mpp range.

Your system would naturally want strings of 9, 6, and 7 based on the sub-array orientations. If you want to experiment, I would be tempted to make the 9 panel string go into one of the MPPT inputs, and put the 6 and 7 panel subarrays into a 13 panel string into the other MPPT input. ...

A string of 13 panels at 37V would bring it right to the 480V upper limit of the MPPT range. Which is fine.
Is the absolute high voltage limit on the unit high enough to withstand the 720V that might come from Voc and low temperarture? Or will the Tigo optimizers keep the output voltage down to 37 under all circumstances?

[sensij]

A string of 13 panels at 37V would bring it right to the 480V upper limit of the MPPT range. Which is fine.
Is the absolute high voltage limit on the unit high enough to withstand the 720V that might come from Voc and low temperarture? Or will the Tigo optimizers keep the output voltage down to 37 under all circumstances?

According to the data sheet, the TrinaSmart panels "limit the Voc" to 39.4 V. I don't know exactly what it means, but on the surface, it doesn't look like a 13 panel string would exceed the max voltage input rating.