How important is it to maintain identical panel angle for MPPT?

[solarsocal]

Hi,

Plans are being finalized for my house for a roughly 20.4kw solar system, currently made up of 69-70 295W panels and probably 3 SMA 8000US inverters. Due to the flat roof with a variety of skylights, other obstructions, and fire dept restrictions, the panels would be located in a lot of different groupings (no less than 7 arrays), some of which are restricted to lower mounting angles than others for various reasons (big arrays get too tall, one array could shadow another, etc).

The installer is a good guy but newer to the business and still going through some learning curve. The latest plans submitted to me would blend groupings of different angles onto at least one of the SMA inverters. Ie, one inverter that has 24 panels assigned could have 20 panels mounted @ 10 degrees + 4 more at a different angle (20 degrees). Or if that example fails with regards to string practices without even taking MPPT into account (?), then use "16 @ 10degrees & 8 @ 20degrees" for the example.

I've seen a couple comments on the forum from experienced folks here saying (if I understood correctly) that all solar panels assigned to a single central inverter that has 1 MPPT must be mounted at the same angle or it will confuse the MPPT system. Questions:

1) Is this "confusion" significant -- is it likely to cause a bigger loss of efficiency than lowering all the panel mounting angles per inverter to the lowest common denominator?

2) I did a few different searches on the web and couldn't find any information anywhere that would give me something solid to show the installer that mixing angles per inverter is a bad thing. I think I even checked an SMA install manual and didn't see a mention. Why is info so scarce on this? Can someone point me to anything definitive?

Much appreciated!

[pvpanels]

confused

The invertor may be confused , but this would depend on the difference of the angle , it may cause the output to be reduced at certain times of the day , when one panel gets more sun than the other or shading occurs . Any panels on the same angle could simply be put on a different string , using one suitable invertor

Most manufactures of Invertors should be able to give you better information

Hope this helps a little

[solarsocal]

Update: my installer called SMA today and apparently SMA said it's fine for the strings going to the inverter to have different panel tilts (pitch?). So, keep all panels on a given string identical, but apparently if all the panels on String1 are at 10 degrees and all the panels on String2 are at 20 degrees, no issues. ??

Maybe I misunderstand a couple of forum msgs stating not to mix panel orientation on the same inverter where I took "orientation" to mean tilt when maybe Mike90250 meant the longitude(?). But Naptown seemed to clearly call out tilt as something that shouldn't be mixed on the same inverter. Here's one of the threads:

http://www.solarpaneltalk.com/showthread.php?3798-Max-of-these-panels-I-can-safely-put-on-SMA-8000US-s&p=22569&viewfull=1#post22569

All my panels would be racked to face the exact same longitude, it's how much they are tilted upwards that would vary and right now would result in different upwards tilts (10, 20, etc) being on the same inverter.

Thanks

[Naptown]

Here is a response to an orientation question that was gleaned from the SMA Q&A site
Hi Casey, I see no flaws in your reasoning, but I would not yet dismiss the idea of a larger array, even if it means splitting it up. It is weird that you are getting conflicting info from the same company. That would certainly make me think twice about things. This is from the Sunny Boy FAQ at :

Q: What is the best panel configuration to use with my Sunny Boy inverter?
A: Ideally, every PV panel connected to one Sunny Boy inverter should be installed in the same plane-of-array orientation (facing the exact same direction). You should not place PV panels in different orientations. Every module should also be installed in a location to avoid shading, no matter how small. If the orientation is mixed or partially shaded, the weakest panel(s) in the system will limit the output of the entire PV array. Other orientation considerations should be made to optimize time-of-day and/or time-of-year performance.
So, for multiple planes and/or shading situations, it would appear that multiple inverters are called for. If I were in your shoes, I would call up the folks at SMA-America and explain the situation to them. The more info you have about how much time each of the roof faces are in the sun, the better they will be able to advise you. Their reputation for helping customers is among the best in the industry.

The question is still how much will a different tilt on a single inverter change the mppt tracking. I have always used the 10 Degree rule that all modules on an inverter should be with 10 degrees tilt from each other and would use separate inverters on different tilts and azimuth. ( One advantage to microinverters is this is a non issue)
Changing tilt would be the same as changing orientation The panels will have a different angle of incidence and the inverter will track to the lowest common denominator. There will be some loss of performance regardless in this situation how much is anyone's guess.

[solarsocal]

The question is still how much will a different tilt on a single inverter change the mppt tracking. I have always used the 10 Degree rule that all modules on an inverter should be with 10 degrees tilt from each other and would use separate inverters on different tilts and azimuth. ( One advantage to microinverters is this is a non issue)
Changing tilt would be the same as changing orientation The panels will have a different angle of incidence and the inverter will track to the lowest common denominator. There will be some loss of performance regardless in this situation how much is anyone's guess.

The plan is still changing but the current one should allow 2 of the 3 inverters to be paired with panels that all have the exact same orientation (azimuth & tilt).

For the third inverter, all panels would have the same azimuth, but because they are located on 2 separate panel arrays, eight panels would be at 25 degrees tilt and thirteen would be at 15 degrees tilt. Sounds like it meets your "10 degree rule"?

As far as string sizing goes for that third inverter, that would mean three strings of 7:
1 string of 7 at 25 degrees.
1 string of 7 15 degrees.
1 string with 6 panels @15 degrees & 1 panel @ 25 degrees.

I assume the last string's single 25 degree panel will yield as if it was 15 degrees (lowest common denominator). My main concern is confirming it shouldn't really hurt my overall efficiency/MPPT operation beyond that single panel as best as anyone can tell?

Changing tilt would be the same as changing orientation The panels will have a different angle of incidence and the inverter will track to the lowest common denominator. There will be some loss of performance regardless in this situation how much is anyone's guess.

When the inverter tracks to the lowest common denominator, is that within a single string, or within all strings? In other words, if I have 2 equal strings on an inverter, one with panels at 15degrees, the other with panels at 25degrees, do I get the benefit of 25 degrees for that 1 string or do ALL panels on the inverter act as if they are at 15 degrees when it comes to output?

Thanks!

[Naptown]

The plan is still changing but the current one should allow 2 of the 3 inverters to be paired with panels that all have the exact same orientation (azimuth & tilt).

For the third inverter, all panels would have the same azimuth, but because they are located on 2 separate panel arrays, eight panels would be at 25 degrees tilt and thirteen would be at 15 degrees tilt. Sounds like it meets your "10 degree rule"?

As far as string sizing goes for that third inverter, that would mean three strings of 7:
1 string of 7 at 25 degrees.
1 string of 7 15 degrees.
1 string with 6 panels @15 degrees & 1 panel @ 25 degrees.

I assume the last string's single 25 degree panel will yield as if it was 15 degrees (lowest common denominator). My main concern is confirming it shouldn't really hurt my overall efficiency/MPPT operation beyond that single panel as best as anyone can tell?



When the inverter tracks to the lowest common denominator, is that within a single string, or within all strings? In other words, if I have 2 equal strings on an inverter, one with panels at 15degrees, the other with panels at 25degrees, do I get the benefit of 25 degrees for that 1 string or do ALL panels on the inverter act as if they are at 15 degrees when it comes to output?

Thanks!

Have you confirmed strings of 7 with SMA for trigger voltage?

[solarsocal]

Have you confirmed strings of 7 with SMA for trigger voltage?

Not personally (and haven't learned about trigger voltage yet), I figure the installer would take care of things like that. They are CSI 295W panels, voltage specs are here: http://www.solarhub.com/pv-modules/5...Canadian-Solar

I think the 7-string (21 panel) setups would likely be on SB 7000US's and the 9 string (27 panel) setup would be on an SB 8000US. Specs on the 2nd page here: http://www.wholesalesolar.com/pdf.fo...000US-spec.pdf

Does that generally sound ok?

Thanks

[Mike90250]

when strings are off-axis, the harvest will decrease. How much is unknown, because this is not a normal condition, and can vary by panel mfg, angle differences, heat.

Just impossible to predict except it will be less than optional.

[Naptown]

Not personally (and haven't learned about trigger voltage yet), I figure the installer would take care of things like that. They are CSI 295W panels, voltage specs are here: http://www.solarhub.com/pv-modules/5...Canadian-Solar

I think the 7-string (21 panel) setups would likely be on SB 7000US's and the 9 string (27 panel) setup would be on an SB 8000US. Specs on the 2nd page here: http://www.wholesalesolar.com/pdf.fo...000US-spec.pdf

Does that generally sound ok?

Thanks

No start voltage on that inverter is 300V VOC is 45
45x7=315
Once you add a load the voltage will drop
36.3 is optimum load x7=256V
That is at NOCT in hot weather the voltage will be lower.

[solarsocal]

Update: I gave up on trying to keep an area of the roof clear of solar (for a future roof deck that I'll probably never build) and now have the space to get the yield we are after using more common 230 watt panels (still Canadian Solar, CS6P series) instead of those 295w panels.

All panel tilts are uniform at 15 degrees now (and all panels continue to have the same azimuth).

Current string plan has 3 x 12 panels on inverter1, 3 x 11 panels on inverter2, and 2x10 panels on inverter 3. Hopefully that sounds more workable.

Thanks!