Towards Using PV Reference Cells for Solar Resource ‒ “PV-Resource”
Historically, the true broadband solar resource represented as either field deployed radiometric measurements or satellite-based estimates is used as the input to PV plant models to estimate energy production. However, PV modules respond differently to the solar resource in that they have: 1) limited spectral range 2) spectral response dependent on silicon-based cell construction 3) angle of incidence effects and 4) temperature dependence.
Thus, the actual solar energy available to a module, the “PV-resource” has become of interest to the PV community to lower uncertainty in plant production. Following its framework for reference cell standards for PV resource applications  NREL is expanding its effort to calibrate, deploy and evaluate various PV reference cells along with co-located pyranometers and spectroradiometers in typical outdoor PV plant fixed and tracking orientations.
In the first phase of this work multiple calibration methods were evaluated: factory calibration, NREL pulsed simulated sunlight Cell Lab calibration and natural sunlight Broadband Outdoor Radiometer Calibration (BORCAL) - originally designed for pyranometers. The assumption and benefit of using BORCAL to calibrate PV reference cells is that they become traceable to the world radiometric reference (WRR) under outdoor irradiance conditions where they are deployed instead of corrected indoor standard test conditions.
The preliminary results for 39 tested reference cells representing six different manufacturers and 10 different models show that between factory and NREL Cell Lab there is a general positive bias of 0.8% in the reported responsivity for Cell Lab calibrations over factory calibrations which does fall within the stated uncertainty of about 1%. Inter-comparing NREL Cell Lab and outdoor BORCAL calibrations, responsivities are consistently about 2% below NREL Cell Lab. Also, repeatability of calibration is an important requirement for any calibration procedure. Here, BORCAL has found to return repeatable responsivities within 0.5% for any given instrument.
Based on the results presented here, the WRR traceable, BORCAL method has been shown to be capable of repeatable and similar responsivities as other methods. With additional analysis we hope to show that reference cells have the potential to be both characterized and calibrated with this method while also reducing the total expanded uncertainty in the reported responsivity.
 A. Habte, A. Driesse, C. Gueymard, S. Wilbert, and F. Vignola. 2018. Developing a Framework for Reference Cell Standards for PV Resource Applications. Golden, CO: National Renewable Energy Laboratory. NREL/TP-5D00