Calibration and uncertainty estimation and minimisation of Loughborough University’s array spectroradiometer and standard lamp

The main aim of ENG55 PhotoClass project is to facilitate the classification of PV devices according to their energy performance at different conditions, referred to as energy rating. To enable this rating, accurate spectral irradiance measurements are required. These measurements are relevant to spectral angular dependence, spectral responsivity measurements using polychromatic filter and even estimating the uncertainty of standard spectral responsivity measurements. Furthermore, spectral irradiance measurements are essential for mismatch factor correction, which is a major source of uncertainty.

Spectral irradiance measurements are often performed via array spectroradiometers due to their convenience and speed allowing to capture the full spectrum measurement in a single lamp flash. They, however, have issues associated with temperature stability, stray light, non-linearity with integration times and irradiance levels, angle of incidence and others. Because of these limitations there is an excessively large uncertainty in absolute measurements that propagates in the measurements listed above. These are evident in inter-comparison results, even between devices calibrated against the same standard lamp.

The purpose of the Research Mobility Grant - RMG6 was exchange expertise in spectral irradiance measurement uncertainties and to perform the necessary measurements at Physikalisch-Technische Bundesanstalkt (PTB) to better understand and characterise the uncertainty sources listed above for the Loughborough University’s (LU) spectroradiometer. All sources mentioned above were comprehensively characterised and corrections were applied and/or the procedures changed where appropriate to minimise the associated uncertainties. Both the LU array spectroradiometer and standard lamp were calibrated and the results were validated via a cross calibration which was well within the uncertainties.