Module quality losses

The Module quality loss is a parameter that expresses your confidence in the actual module performance relative to the manufacturer's specifications.

It is at your entire disposal: you can put it at any value  (for example for keeping some reserve on the production warranty, etc).

Default value

By default, PVsyst initializes the "Module Quality Loss" based on the PV module manufacturer's tolerance specifications.

PVsyst uses one-quarter of the difference between these values. For example, with ±3%, it will be 1.5%, and with positive sorting (0...+3%), it will be −0.75% (a negative loss value, representing a gain).

NB: This one-quarter value is a PVsyst choice. We typically use a conservative approach (i.e., modules will not exceed announced specifications). There are no other specific reasons for this choice.

During the simulation, this factor will induce a loss on the Array Pmpp production, constant (in percentage) over all operating conditions.

In the past, it was well-known that most PV module series did not meet manufacturer nominal specifications. The actual module behavior relative to specifications was one of the major uncertainties in PV system performance evaluation.

Now, with "guaranteed" power claims and increased availability of independent certifications, the situation appears to be improving.

At the factory, each PV module is flash-tested and placed in the power series corresponding to its measured STC power. This is the basis for the "tolerance" definition.

Now we have to know:

• According to flash-test manufacturers, flash-test measurements cannot be guaranteed to better than ±3% accuracy (except for high-quality laboratory instruments).
• During the first days of operation, crystalline PV modules (p-type wafers only) may undergo LID degradation of about 1–3%, depending mainly on crystal manufacturing quality (oxygen traces). The LID is now explicitly specified as a loss in PVsyst.

You can find some references about values reported by different authors in Thevenard (2010)¹, p.19.