Jump to content

Connection of modules with different azimuths or inclinations.

Georgy PM

Recommended Posts

Dear PVSYST team. First of all, my thanks for the latest software updates, the improvements are very welcome. Now based in South Korea, I have very particular cases of systems with modules in series having different orientations and/or inclinations.  I insisted a lot to my colleagues that this would result in very high mismatch losses, due to the difference in current produced by each module. However, I was asked a very good question (which I think is perfectly valid): how far apart in orientation can you connect modules in series without incurring significant losses (less than 1%, for example)? Would it be 1 degree apart? or 2? 

Is PVSYST able to simulate a string of modules with 2 different orientations? I've been trying to do it for a while, unfortunately without success, so I'd be very grateful for your help. Thank you very much in advance. Bon début d'aneee ! 

Link to comment
Share on other sites

Dear Georgy,


I can't answer you because it depends on multiple variables.


In PVsyst, you can't set different orientations within a series of modules.


I suggest increasing the detailed losses for mismatch in your simulation. The idea is to simulate the worst-case scenario. That way, when you install your PV modules, the actual electrical production will likely be higher than the simulation.




Muhammed Sarikaya

Link to comment
Share on other sites

  • 5 weeks later...

Dear Muhammed, 

Thank you for your previous response,

I tried to convince everyone of avoiding mixing orientations, but more interesting cases are appearing. For instance, some clients are requiring more “stylish” solar carports, things like this:


Hence, it is harder to match module’s orientations and strings (like if solar installer’s life wasn’t already hard enough 🙄). Since in cases like this one we have to absolutely mix two different orientations, the big question that arises now is : how much appart the orientaiton can be ? Lacking availibility to think about complex formulas of irradiation and photo current, I tried to get the mismatching losses directly from PVSYST results. First, I made a simple simulation with 4 panels with 15degrees of inclination, then other 2 four modules systems with 5 and 10 degrees (same azimuths). Then, I exported the hourly results and “mixed” the Array’s outputs (for a maximum of 8 modules, only 2 orientations at time), follwing two simple rules: voltage of each group of 4 modules are added, and the current of the string is the smallest produced by any of the 2 groups of 4 modules. For instance, the results of the Array current for each group and the mixed results are shown below (Iarray5g and the mixed are the same):


Finally, in order to estimate the loss produced by the mismatching, I did: image.png.ecacd55d3d60f4bdd804adbda707422a.png

I mean, the loss would be the ratio between the he total energy that one would get by having each group connected to independents MPPT divided by the mixed E_Array. Since the almost linearity of the inverter’s output, the E_Grid should have a similar reduction. By mixing 5 and 10° orientations I got -1,98%, 10 and 15° -1,64% and 5-15° -3,54%. They seem big to me as expected, so I will try smaller orientation variations in order to get less than -1 or -0,5% of losses. Economically this would be acceptable since it is much more expensive not doing the installation 😊.

Before continuing, I wanted to have your opinion about this approach. Do you think that we can reproduce this procedure for calculating the mismatch losses for much more complex systems? For instance, having 1 module with one orientation and 7 with another, or varying the Azimuths (which will start to shift the output’s curves). Below I send you two links where you can download the project and the excel file.

Many thanks in advance for your feedback.

원주공항 Project v2.zip

원주공항 Project_VC5 Az0, 15 vs 10 g, 4PV.xlsx

Best regards

Link to comment
Share on other sites

Dear Gregory,

This seems like a very good approach, I see no inconvenient in applying it to other subdivisions. Of course, keep in mind that this is an approximation, but that will give you a good estimate of orientation mismatch within a string.

Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
  • Create New...