Module testing. Image: JA Solar.

Module testing. Image: JA Solar.

The current IEC and UL certification testing standards for PV modules have pass/fail criteria to help reduce the risk of early field (infant mortality) failures, however, in their article, ‘Why Do PV Modules Fail?’ (Energy Procedia, 2011) Claudio Ferrara and Daniel Philipp present reasons that demonstrate that these tests are not sufficient in assessing the relative reliability of a module’s performance over its service life. This limits the useful data available to solar PV investors, as well as manufacturers’ scope for product improvement. The potential for PV modules to fail before the end of their intended service life increases the perceived risk, and, therefore, the cost of funding PV installations.

Additionally, current testing standards do not include a protocol for relative durability assessment of different modules. Hence, financial models rely on a patchwork of methods to forecast relative durability in the absence of these benchmarks. Consequently, the uncertainty in quantifying which solar modules are best suited to a particular installation results in an increase of perceived risk, delayed financing and, ultimately, raises the cost of building PV power plants.

Photovoltaic Module Durability Initiative overview

A joint initiative between the Fraunhofer CSE and Fraunhofer ISE, the Photovoltaic Module Durability Initiative (PVDI), is an integrated lifetime assessment initiative to understand and improve PV module durability. The goal of the PVDI is to establish a baseline PV durability assessment programme that provides quantitative, independent, third-party assessment on a module’s long-term durability. Moreover, the PVDI enables comparison of the relative durability risk between different module designs, which the current test protocols lack.  The PVDI test sequences are summarised in Figure 1, and further explanation of the purpose of each group of tests is provided in Table 1, below.

Figure 1: The PV Durability Initiative test sequences.

Figure 1: The PV Durability Initiative test sequences.

GroupTesting protocolPurpose
1Potential-Induced DegradationAssesses a module’s ability to perform under the stress of high electrical potential. Potential-induced degradation (PID) refers to a class of degradation mechanisms caused by a high potential between internal and external components of a PV panel.
2Humidity Freeze and UVAssesses a module’s susceptibility to moisture in the presence of high temperatures as well as freezing caused by sub-zero temperatures after the module has been saturated by humidity, and at high levels of UV radiation.
3Static and Dynamic Loading, Thermal Cycling, and Humidity FreezeAssesses the effect of both static and dynamic loading on a module’s performance and package integrity.
4Thermal CyclingAssesses a module’s ability to withstand the effects of shade induced, diurnal, and seasonal temperature changes.
5Outdoor Energy PerformanceAssesses a module’s performance under real-world (non-accelerated) operating conditions at six-month intervals. The ultimate goals are to understand long-term wear-out, identify new failure modes, and determine the acceleration factors that correlate the accelerated test results to outdoor operating lifetime.

Table 1: PVDI testing protocol overview

PVDI goes beyond the pass/fail criteria of today’s testing protocols. Modules are repeatedly characterised as they pass through the assigned test sequence. For example, in Group 4, each module is characterised after every set of two hundred thermal cycles. At each interim test point, electrical performance, electroluminescence and infrared images are collected. In some instances, wet leakage current and insulation resistance are also measured.

Each module is rated based on its performance under PVDI’s robust testing protocol to generate a credible rating of PV modules based on their likelihood of performing reliably under different kinds of stress. All the results are summarised in a report that provides solar PV financiers, developers and other industry players with a widely available quantitative dataset to assess long-term durability.

The rating criteria are shown in Table 2, below.

RatingRating criteria
5P >= 0.95
40.88 <= P < 0.95
30.75 <= P < 0.88
20.50 <= P < 0.75
1P < 0.5
P = 0

PVDI test results to date

To date, three rounds of PVDI testing have been completed on ten commercial module types. Manufacturers of four modules have attached the identification to the results:

  • PVDI01* is the SunPower E20 module.
  • PVDI06* is the Aleo (Type S18) module.
  • PVDI09* is the First Solar Series 3 Black (FS-395-Plus) module.
  • PVDI10* is the First Solar Series 4 (FS-495/497) module.

The PVDI01*, PVDI09* and PVDI10* modules were tested at the Fraunhofer CSE/CFV Solar Test Laboratory and the PVD106* modules were tested at the Fraunhofer ISE.

In the first two rounds of PVDI, the PID test conditions were 85oC temperature and 65% relative humidity, for a total of 400 hours. The PID testing protocol was modified for the third round of PVDI testing based on the new draft IEC62804 TS. In this round the modules were subject to 288 hours of PID testing at 60oC temperature and 85% relative humidity and interim measurements were taken at 96, 192 and 288 hours and after a recovery step.

The third and latest round of PVDI testing was completed this year. The test protocols are revised as necessary to incorporate changes as the understanding of PV module durability grows. For example, in the first two rounds of PVDI testing no significant degradation was observed among any of the modules tested in the previous PVDI rounds of damp heat/UV testing sequence (Figure 3). This implies that the wear-out regime for these conditions was not reached and no conclusions can be drawn with regard to the relative susceptibility to damp heat and UV stress. For the third round of PVDI testing, the protocol was changed by replacing damp heat with humidity freeze and doubling the UV dose. Both the module designs tested showed some degradation after the second round of UV testing where the modules’ power output decreased by 5.1% in both cases. However, the power recovers to a certain degree after exposure to the second HF 10 testing.

PVDI strives to continuously improve the testing protocol based on results from past rounds and establish a comprehensive dataset to enable comparisons with rest of the field and upcoming modules. The results and durability ratings for leading PV modules subjected to PVDI testing are published regularly to enable PV system developers and financiers to make informed deployment decisions.

Module performance ratings

The module’s performance is based on the measured electrical performance at standard test conditions (STC). For the rating a mean of the weighted normalised module power is used:

Here n is the number of performance measurements within a test sequence, and Pn,i  is the mean power, normalised with regard to the initial measurement, of all modules in a test group at the measurement step i

Table 3, below, shows module performance ratings based on mean weighted normalised power measurements.

PVDI IDEnvironmental conditions
PIDDH/UVHF/UV/UV/HFDynamic loadStatic loadThermal cycling
0155N/A555
0245N/A555
0345N/A552
0425N/A452
0555N/A344
0655N/A455
0745N/A444
0835N/A323
095N/A4555
105N/A4555

There were five participants in the first round, followed by the addition of three more module designs in the second round and another two were tested in the third round.

Conclusion

Other PV testing protocols available today provide an assessment of a module’s quality based on pass/fail criteria, and are not robust enough to push the module to the point of degradation. Subjecting modules to qualification testing does not provide manufacturers with any information on the susceptibility of modules to degradation under different conditions, and failure mechanisms that might occur in the field. This inhibits further product development and innovation, which is essential to continuously improve module design and performance.

PVDI strives to provide accelerated tests to assess a module’s reliable (relative) performance over its service life, identify degradation mechanisms and provide information to accelerate innovation in PV modules. PVDI provides an assessment of module’s susceptibility to degradation under each testing group, as well as a relative performance rating to compare against other modules. This helps manufacturers determine the areas of weakness in their modules and strive to address those, improving their module’s performance. The rating system enables financers and investors to compare the relative performance of modules, and making informed decisions.

PVDI is an ongoing initiative and welcomes new participants. A detailed account of the findings from PVDI 3 will be featured in the next edition the quarterly technical journal PV Tech Power, which will be published on Monday 24 August 2015. To susbcribe for free, click here.

Figure 2a: Mean normalised performance degradation of all modules of a test group in PID testing under positive bias and (b) negative bias. To determine the PID rating, the final performance value after light soaking/conditioning is used.

Figure 2a: Mean normalised performance degradation of all modules of a test group in PID testing under positive bias and (b) negative bias. To determine the PID rating, the final performance value after light soaking/conditioning is used.

Mean normalised performance degradation of all modules of a test group in PID testing under negative bias. To determine the PID rating, the final performance value after light soaking/conditioning is used.

Mean normalised performance degradation of all modules of a test group in PID testing under negative bias. To determine the PID rating, the final performance value after light soaking/conditioning is used.

Figure 3: Normalised module power for all modules subjected to outdoor testing.

Figure 3: Normalised module power for all modules subjected to outdoor testing.

Figure 4: Mean weighted normalised module power of all modules of a test group (see eq.1) following damp heat and 100 W/m2 UV exposure for PVDI01*-PVDI08 and mean weighted normalised module power of all modules of a test group following humidity freeze and 200 W/m2 UV for PVDI09* and PVDI10*.

Figure 4: Mean weighted normalised module power of all modules of a test group (see eq.1) following damp heat and 100 W/m2 UV exposure for PVDI01*-PVDI08 and mean weighted normalised module power of all modules of a test group following humidity freeze and 200 W/m2 UV for PVDI09* and PVDI10*.

Comments