Partitioning strings on single modules

[kautzer]

I have a near shading scene where each module is represented by its own individual PV table. I would like to run the simulation according to module strings. In order to partition the strings, I have to mark each module as 1 string in width by 1 string in height (modules are in landscape, 18 modules/string). I know this is not technically true, but is this the most accurate way to model it? I have also thought of using 3 strings in height, 1 in width due to the module having 3 diodes. Any thought would be appreciated. Thanks.

[André Mermoud]

Defining each table-module as one "rectangle" is the correct way.

If the tables are subject to shades from surrounding (not-regular), this will under-evaluate the real electrical shading losses.

If the modules are in shed-like arrangement (regular shadings on each row), and you have more than one string per inverter input, this way will give correct results.

Indeed, as as soon as 1/3 of the sub-modules are shaded in a string, the string will be completely inactive (for beam component).

See the explanation on How to evaluate the effect of by-pass diodes in shaded arrays?

[dtarin]

Andre,

When you say "table-module", do you mean one one module? It seems to me that when we perform detailed calculation according to module layout, it performs the calculations as if there were 3 sub-modules. Is this correct? The reason I say this is that I am working with a shade scene now, where the difference between the according to strings method (100% elec effect), and detailed calculation according to layout method is around 2.3%, which is quite significant. This is if I partition the modules as the size of 1 string, 1 in height, 18 in length. The layout is 2 modules in landscape. If I model according to strings and set the number of partitions in height to 6 (modeling the submodules), my losses are very close to what results in the detailed calculation according to module layout.

Using version 6.49.

So is the according to module layout the most accurate method for calculating shading losses?