On the importance of considering the incident spectrum when measuring the outdoor performance of amorphous silicon photovoltaic devices

Conventional measurement practice for the outdoor performance evaluation of solar cells does not make use of the complete spectrum, relying instead on the total irradiance as measured, say, with a pyranometer. In this paper it is shown that this can result in significant errors for solar cells having wide band gaps, in particular, for amorphous silicon solar cells. Two effects are investigated. The first relates to quantifying the typical errors associated with instantaneous measurements; what one might term the calibration of devices. The second relates to quantifying the impact of neglecting variations in the spectrum on the estimation of the annual energy production. It is observed that the fraction of the spectrum falling in the spectrally useful range for amorphous silicon can vary by as much as +10% to -15% with respect to standard test conditions at the test site used in this study, which translates directly into performance variations of similar magnitude. The relationship between changes due to spectral variations as opposed to variations in device temperature is also investigated. The results show that there is a strong case for investigating spectral effects more thoroughly, and explicitly including the measurement of the spectral distribution in all outdoor performance testing.