Structure Shading Factor for Bifacial Modules being used with Fixed Tilt racking system in Portrait Orientation

[Shivya]

NMMS GA 2.pdfMMS GA 2.pdfeed your guidance in ascertaining Structure Shading Factor for Bifacial Modules being used with Fixed Tilt racking system in Portrait Orientation. Arrangement shown in the pic below. 

 

This would have two mounting rails (pink color) running across the module width, thereby crossing all the internal strings of the module. Mounting rail and back side cells will have typical separation gap of 35-40mm (module frame thickness). 

 

Following are our specific queries:

1. What should be the Structure Shading Factor to be considered for above arrangement
2. What would be the correct interpretation of the mechanics covering half a cell or full cell in the above design

[Shivya]

Requesting response from PV syst team

[Shivya]

Requesting reply from PV Syst Team

[Ben R]

Can PVsyst please comment on this post?  I have the same question and want to add that there are a lot of sources that recommend calculating the structure shade factor simply by taking the ratio of the width of the racking (or torque tube) on the rear side by the length of the module (assuming it's in portrait orientation).  My concern is that this is not taking into account the electrical effect of shading a row of cells that is in series with the rest of the cells, i.e. the current will be reduced in the entire module if you shade just a single row of cells.  

So the calculation should be the width of the row of cells divided by the width of the racking or torque tube, correct?  This is in line with PVsyst's instructions, "the current in a string is the current in the worst cell, if this mechanics covers  x%  of one cell, the shading effect will be  x% for the concerned string. This loss may be reduced by a factor of 2 if the mechanics is covering half a cell (i.e., the shade is distributed on 2 cells).  "

In summary, following these instructions, assuming the racking above shades 40% of the row of cells behind it, even if it's only 10% of the length of the module, would the Structure Shading Factor be 40%?  Or 10%?

[Linda Thoren]

Dear Shivya and Ben,

This is difficult question and we do not have a direct answer.

As a first approximation, the structure shading could be estimated as the ratio of the area covered by structures to the photovoltaic area. In this approximation, the effect of structures is therefore their projection on the rear plane. Therefore, it neglects the fact that light may arrive on the rear side from multiple directions, akin to diffuse light. This approximation is hence likely an underestimate of the shading loss, in full generality. Indeed, even if structural elements are not directly in front of the rear side of modules, they may cause significant shading. A better approximation is beyond the present possibilities of PVsyst.

Then there is the the variable for the mismatch loss factor, caused by a heterogenous irradiance distribution on the rear side. Any source of heterogeneous irradiance will cause electrical mismatch effects. Such is the case of shading cast by structural elements (but also differences in location relative to the environment and the ground). At present, PVsyst has no easy way to make an accurate estimation for these effects. Based on comparisons with measurement data, we have determined a default value (10%) for a general case. With beams covering the module width directly in front of the rear side of modules, the mismatch losses may increase drastically. In such a case, one can roughly estimate the mismatch effect as follows:
as the current in a string is the current in the worst cell, if the structure covers x% of one cell, the shading effect will be x% for the concerned string. For example, the loss may be reduced by a factor of 2 if the structure covers half a cell (or two half cells).