A study under natural sunlight

Providing comprehensive rating information for photovoltaic modules

By Philippa Martin-King

A study by Arizona State University Photovoltaic Reliability Laboratory (PRL) based on IEC 61853-1 provides insights into the difficulty of assessing photovoltaic performance testing. Excerpts are published with the permission of Solar ABCs (Solar America Board for Codes and Standards).
 Link to full report

More comprehensive information needed

Manufacturers typically rate PV modules at STC (standard test conditions). The STC rating involves only one temperature (25° C), one irradiance (1000 W/m²), and one sunlight spectrum (AM [air mass] 1.5G [global]). However, the actual energy production of field installed PV modules is a result of a range of operating temperatures, irradiances, and sunlight spectra. Therefore, there is an urgent need to characterize PV modules at different temperatures and irradiances to provide more comprehensive rating information.

Recognizing this issue, the IEC Technical Committee 82 Working Group 2 (TC 82/WG 2) has been developing an appropriate power and energy rating standard. This IEC 61853 standard titled "Photovoltaic Module Performance Testing and Energy Rating" consists of four parts.

The first part of this four-part standard requires the generation of a 23-element maximum power (Pmax) matrix at four different temperatures and seven different irradiance levels. The Pmax matrix can be generated using an indoor solar simulator method or outdoor natural sunlight method. The outdoor test method introduces little/no spectral mismatch error and is much less expensive than the indoor test method because it avoids the use of very expensive solar simulators. However, obtaining an accurate and repeatable Pmax matrix using the outdoor method over time (several months or years) would be extremely challenging.

Energy rating standard tested

The Arizona State University PRL (Photovoltaic Reliability Laboratory) ran a study (prepared by (Mani) Govindasamy TamizhMani, Karen Paghasian, Joseph Kuitche, Meena Gupta Vemula, Ganesh Sivasubramanian) related to the power rating of PV modules using natural sunlight. Study title: Photovoltaic Module Power Rating IEC 61853-1 Standard: A Study Under Natural Sunlight.

Objectives of the study

The objectives of this study were to:

  • identify measurement repeatability issues with a non-standardized test setup,
  • standardize the measurement setup,
  • verify the device linearity per IEC 60904-10,
  • generate the power (Pmax) matrix per IEC 61853-1, and
  • validate four different current-voltage translation/interpolation techniques of IEC 60891 and the National Renewable Energy Laboratory Method (NREL Method).

Study methodology

The study included two rounds of outdoor measurements. During round one (2009) measurements were carried out to identify the repeatability issues of the non-standardized test setup. In round two (2010) measurements were carried out to standardize the test setup, check the PV device linearity, generate the Pmax matrix, and validate four different current-voltage (I-V) translation/ interpolation techniques of IEC 60891 and NREL Method.

Based on the round one 2009 study, the authors concluded that the repeatability issues can be minimized to an acceptable level by an appropriate selection of outdoor measurement conditions and test equipment. Outdoor measurement conditions should be clear sunny days (> 90 % direct normal irradiance) at lower air mass values (< 2.5). Test equipment should include:

  • a manual or automatic two-axis tracker for tracking the sun;
  • a fast (< 1 second) I-V curve tracer to minimize issues related to changing irradiance, spectrum, and module temperature;
  • a matched technology (or matched spectral response) reference cell to practically eliminate spectral mismatch error; and
  • a pre-cooled test module to naturally change the module temperature while exposed to sunlight (to obtain data at the temperature of 75°C, a thermal insulating foam can be used to cover the backside of test module).

The 2010 measurements with the standardized test setup (2-inch module-to-screen gap and matched reference cells kept outside the screen) were used to verify the device linearity and to demonstrate that the 23-element Pmax matrix of IEC 61853-1 standard can be generated using 8 to 23 measured reference I-V curves and four translation/ interpolation procedures of IEC 60891 and NREL Method. The study used 23 reference curves for the first two procedures (called translation procedures 1 and 2), and 8 to 10 reference curves for the other two procedures (called interpolation procedures 3 and 4). Procedures 1 and 2 required the use of five mesh screens, and procedures 3 and 4 required only three or four mesh screens to generate the entire 23-element Pmax matrix of IEC 61853-1. The report presents the power rating results for four commercial module technologies—crystalline silicon (c-Si); amorphous silicon (a-Si); cadmium telluride (CdTe); and copper indium gallium diselenide (CIGS). The authors also present a detailed investigation on the validation of the four translation/interpolation procedures for all four technologies.

The results of this study were recently published in Solar ABCs (extracts reprinted with permission).  


Major study conclusions:

  • Outdoor measurement repeatability issues: The repeatable power rating measurements at various irradiance levels under natural sunlight within an acceptable deviation limit of 2 % could not be achieved when uncalibrated mesh screens were placed directly (0 inch distance) on the test module and reference cell.
  • Standardization of measurement setup: A standardized measurement setup was established with reference cells kept outside the calibrated mesh screens, which were placed at a 2 inch distance above the test modules.
  • Verification of device linearity per IEC 60904-10: The linearity requirements of open-circuit voltage, short-circuit current, and maximum power versus temperature are met by all four test technologies. Similarly, the linearity requirement of open-circuit voltage versus the logarithm of irradiance is also met (with two minor exceptions presumably due to some experimental errors) by all four test technologies. For the short-circuit current versus irradiance, the devices met the linearity requirement (2 % deviation limit) for irradiance levels above 200 W/m2 but they surprisingly showed a higher deviation for the irradiance levels below 200 W/m2. This higher deviation was objectively attributed to some minor experimental errors related to the calibration of low transmittance mesh screens under natural sunlight.
  • Generation of Pmax matrix per IEC 61853-1: The required 23-element Pmax matrix of IEC 61853-1 was successfully generated for all four module technologies using the four translation/interpolation procedures of IEC 60891 and NREL Method. Unless the data processing personnel pay extreme attention or commercial test laboratories automate the data processing, the translation procedures (1 and 2) are more prone to human error than the interpolation procedures (3 and 4). However, procedures 1 and 2 would work extremely well if multiple narrow irradiance ranges are used with individual sets of correction values for each narrow irradiance range or multiple sets of correction values are used for a single wide irradiance range.
  • Validation of procedures of IEC 60891 and NREL Method: An extensive validation analysis of the four translation/interpolation procedures, at both narrow and wide irradiance ranges indicates that all four procedures are remarkably accurate within an average error of 3 % and a root mean square error (RMSE) of 4,5 %.
Standardized rating essential for manufacturers and end-users Standardized rating essential for manufacturers and end-users
PV modules Source DuPont. PV modules Source DuPont.