Linearity of photovoltaic devices: quantitative assessment with N-lamp method

Abstract

The short-circuit current output of photovoltaic (PV) reference device is typically used to determine the incident irradiance of natural or simulated sunlight. Normally the PV reference device is calibrated at standard test conditions and other irradiances are calculated based on a proportionality assumption (termed linearity) between short-circuit current output and incident irradiance. Here the linearity of PV devices is newly defined including a quantitative correction for non-linearity (NL) when measuring incident irradiance. Linearity can be determined experimentally by the flux addition principle such as in the two-lamp method. The latter provides information about linearity between two irradiance levels which differ by a factor of two, but no information on linearity inside this interval. Here this concept is extended to the N-lamp method. It is shown that this provides more detailed linearity information with low uncertainty. Measurements were made with an 11-lamp steady-state solar simulator and showed a NL deviation of 2% in the irradiance range from 100 W m−2 to 1100 W m−2 for the PV reference cell tested. The method is easily implemented, provides detailed quantitative linearity assessment at low cost and can be considered a primary method for linearity assessment, as it does not require any reference device.