This paper presents a composite measurement approach that capitalizes on complimentary strengths of 3 different test beds to significantly reduce uncertainty in I-V parameters for secondary module calibration. This approach addresses PV manufacturers' need for reduced uncertainty in the calibration modules that provide the basis for their module power ratings. This new method enables NREL to reduce uncertainty for secondary module calibration in commercial c-Si modules from +/- 3.2% to +/- 0.7% for I-SC and from +/- 3.3% to +/- 1.1% for P-MAX. This new module self-reference, or MSR procedure is based on the sensitivity of module voltage to temperature plus the uniformity of outdoor sunlight. By calibrating module V-OC in thermal equilibrium on a flash simulator we provide an accurate gauge of junction temperature for subsequent measurements. By calibrating I-SC outdoors in natural sunlight we enable the module to serve as its own reference device in setting the intensity of a continuous simulator, effectively eliminating the impact of spatial non-uniformity and spectral mismatch on the I-V measurement. These procedures significantly reduce measurement errors due to temperature uncertainty, spatial non-uniformity, and spectral mismatch.

This user's manual describes performance data measured for flat-plate photovoltaic (PV) modules installed in Cocoa, Florida, Eugene, Oregon, and Golden, Colorado. The data include PV module current-voltage curves and associated meteorological data for approximately one-year periods. These publicly available data are intended to facilitate the validation of existing models for predicting the performance of PV modules, and for the development of new and improved models. For comparing different modeling approaches, using these public data will provide transparency and more meaningful comparisons of the relative benefits.

A protocol was devised for preconditioning polycrystalline CdTe and CIGS cell technologies under bias with light exposure or forward-bias currents at elevated temperatures and the results reported.

The long-term performance data of copper indium diselenide (CIS) and gallium-alloyed CIS (CIGS) photovoltaic (PV) modules are investigated to assess the reliability of this technology. We study and report on numerous PV modules acquired from two manufacturers (A and B), deployed at NREL's outdoor test facility (OTF) in various configurations in the field: some are free standing, loaded with a fixed resistance and periodically tested indoors at STC; other modules are connected to data acquisition systems with their performance continuously monitored. Performance is characterized using current-voltage (IN) measurements obtained either at standard test conditions or under real-time monitoring conditions: the power parameters plus other factors relating to quality like diode quality factors or series resistance are analyzed for changes against time. Using standard diode analysis to determine the sources of degradation indicates that CIS modules can exhibit between moderate and negligible degradation, with the dominant loss mode being fill factor declines along with decreases in open-circuit voltage, for illumination intensities near 1-sun. At lower intensities, current losses can appear appreciable. The real-time performance data also indicate that fill factor loss is the primary degradation mode, generally as a result of increases in series resistance.

In a round robin outdoor exposure experiment carried out in three different climates, we have previously demonstrated that amorphous silicon (a-Si) PV modules reach higher stabilized performance levels in warmer climates. The four-year experiment involved three identical sets of thin-film a-Si modules from various manufacturers deployed outdoors simultaneously in three sites with distinct climates. Each PV module set spent a one-year period at each site before a final period at the original site where it was first deployed. The experiment aimed to determine the light-induced degradation and stabilization characteristics of a-Si regarding specific history of exposure, and to compare degradation rates in different climates. We propose that after the initial sharp degradation associated with the Stabler-Wronski effect (SWE) has passed, the subsequent stabilized performance levels attained will depend largely on light exposure and a characteristic temperature associated within a coherent time-scale. PV modules which were first deployed at the lowest-temperature site for one year, reaching a stabilized state, and were then further deployed at higher temperature sites for two more years, experienced considerable recovery in output parameters (Pmax and FF). However, when further deployed back at the original, lowest-temperature site, performance degraded back to the first year, original level.

We examine the status and question of long-term stability of copper indium diselenide (CIS) photovoltaic (PV) module performance for numerous modules that are deployed in the array field, or on the roof of, the outdoor test facility (OTF) at NREL, acquired from two manufacturers. Performance is characterized with current-voltage (I–V) measurements obtained either at standard test conditions (STC) or under real-time monitoring conditions, taken over the course of many years. We present and scrutinize I–V characteristics for degradation modes. Analysis yields that CIS PV modules can exhibit either moderate (2% to 4% per year) to negligible or small (less than 1% per year) degradation rates, and that the predominant loss mode appears to be fill factor diminution, often associated with increases in the series resistance in some of the modules. A secondary mode of degradation observed comprises metastable changes to the open-circuit voltage. The featured modules are deployed on three separate testbeds.

1National Renewable Energy Laboratory (NREL), USA 2Sandia National Laboratories (SNL), USA 3PV Testing Laboratory-Arizona State University (PTL-ASU), USA 4Florida Solar Energy Center (FSEC), USA 5 European Commission, Directorate General Joint Research Centre, (IES-RE-ESTI), European Union 6Laboratorio Energia Ecologia Economia (LEEE), Switzerland 7Immisssionsschutz und Energiesysteme GmbH Test Centre for Energy Technologies (TUV), Germany 8Institute of Solar for Energy Systems (Fraunhofer ISE),Germany 9Japan Electrical Safety & Environment Technology Laboratories (JET), Japan 10National Institute of Advanced Industrial Science and Technology (AIST), Japan

Most photovoltaic (PV) manufacturers trace their peak- watt rating through calibrations/measurements performed at recognized terrestrial calibration facilities. This paper summarizes the results of one such measurement performed by many different calibration facilities. The participants were selected from around the world based on their designation as a national PV calibration facility, prior participation in inter-comparisons, or as an ISO 17025-accredited PV module qualification or certification facility. Each facility was sent the same devices and was requested to treat them as a regular measurement. The modules were selected from newer thin-film manufacturers-ones that might stretch or exceed the current scope of capabilities of the different participants. A concentrator module was even included as part of the set. Short-circuit current (lsc) open-circuit voltage (Voc) fill factor (FF), and peak power (Pmax) results are reported

This paper investigates the feasibility of using fixed resistors instead of active maximum power tracking as electrical loads for long-term exposure testing of photovoltaic modules and arrays. The method of investigation was to compare resistive versus active loading on two modules for which historic current-voltage data over time were available. Also, a small amorphous silicon array was installed with a resistive load and the performance has been monitored versus time. The major conclusion of this work is that fixed resistive loading is an inexpensive and viable means of loading photovoltaic devices for exposure testing if the resistance value used is close to the ratio of the voltage to the current at the maximum power point under Standard Test Conditions. Copyright (C) 2006 John Wiley & Sons, Ltd.

The measurement of the photovoltaic (PV) performance with respect to reference conditions requires measuring the performance with respect to a given tabular reference spectrum, junction temperature, and total irradiance. This paper discusses the procedures implemented by NREL's PV Cell and Module Performance Characterization Group to achieve the lowest practical uncertainty. This paper describes the process of trusting and verifying software, hardware, calibrations, procedures, and results. As an ISO 17025 accredited calibration facility, the quality system that is in place is designed to assure customers that the results are valid within specified uncertainty limits and are traceable. The process of trusting performance claims but desiring an independent verification permeates the PV business and society.

The peak-watt rating is a primary indicator of PV performance. The peak power rating is the maximum electrical power that is produced when the PV device is continuously illuminated at 1000 Wm/sup-2/ total irradiance under International Electrotechnical Commission Standard 60904-2 reference spectrum, and 25 deg C cell temperature. Most manufacturers trace their peak-watt rating through calibrations performed at recognized terrestrial calibration facilities. Manufacturers typically perform intercomparisons among a set of their modules internally with other plants and among. Sometimes they have the same module measured at different calibration facilities to determine the differences in calibration. This intercomparison was to mimic this procedure and supply new thin film samples along with samples that could pose other problems. These intercomparisons sample the laboratories' everyday procedures better than a formal intercomparison where the laboratories' best procedures and data scrutiny are used.

By means of bilinear interpolation and four reference current–voltage (I–V) curves, an I–V curve of a photovoltaic (PV) module is translated to desired conditions of irradiance and PV module temperature. The four reference I–V curves are measured at two irradiance and two PV module temperature levels and contain all the essential PV module characteristic information for performing the bilinear interpolation. The interpolation is performed first with respect to open‐circuit voltage to account for PV module temperature, and second with respect to short‐circuit current to account for irradiance. The translation results over a wide range of irradiances and PV module temperatures agree closely with measured values for a group of PV modules representing seven different technologies. Root‐mean‐square errors were 1·5% or less for the I–V curve parameters of maximum power, voltage at maximum power, current at maximum power, short‐circuit current, and open‐circuit voltage. The translation is applicable for determining the performance of a PV module for a specified test condition, or for PV system performance modeling. Copyright © 2004 John Wiley & Sons, Ltd.

Recently, we presented the results of a 5-year solar weathering study of commercial crystalline-Si modules that found a linear relationship between the observed slow (0.2%-0.5% per year) short-circuit current (I/sub sc/) degradation and the total UV exposure dose. Consideration of previous literature reports of I/sub sc/ degradation against our results indicated that encapsulation browning cannot be identified as the cause. This paper reviews these results and outlines a new testing program recently initiated at NREL. The program will attempt to verify if UV radiation is responsible for the degradation and determine if the degradation is related to the rapid initial light-induced metastability caused by oxygen contamination in boron-doped Si solar cells.

We present an analysis of the results of a solar weathering program that found a linear relationship between maximum power degradation and the total UV exposure dose for four different types of commercial crystalline Si modules. The average degradation rate for the four modules types was 0.71 % per year. The analysis showed that losses of short-circuit current were responsible for the maximum power degradation. Judging by the appearance of the nondegraded control modules, it is very doubtful that the short-circuit current losses were caused by encapsulation browning or obscuration. When we compared the quantum efficiency of a single cell in a degraded module to one from an unexposed control module, it appears that most of the degradation has occurred in the 800-1100 nm wavelength region, and not the short wavelength region.

Following a proposal by BP Solarex to modify the standard module qualification sequence, we performed a forward-biased themal cycling on three types of commercial modules to evaluate the procedure. The total number of thermal cycles was doubled to 400 and maximum power measurements were made every 50 cycles.

The Photovoltaic (PV) Cell and Module Performance Characterization team at NREL supports the entire photovoltaic community by providing: secondary calibrations of photovoltaic cells and modules; efficiency measurements with respect to a given set of standard reporting conditions; verification of contract efficiency milestones; and current versus voltage (I-V) measurements under various conditions of temperature, spectral irradiance, and total irradiance. Support is also provided to in-house programs in device fabrication, module stability, module reliability, PV systems evaluations, and alternative rating methods by performing baseline testing, specialized measurements and other assistance when required. The I-V and spectral responsivity equipment used to accomplish these tasks are described in this paper.