Study on Long Term Reliability of Pv Modules and Analysis of Power Degradation Using Accelerated Aging Tests and Electroluminescence Technique

SiliconPV 2011 STUDY ON LONG TERM RELIABILITY OF PHOTO-VOLTAIC MODULES AND ANALYSIS OF POWER DEGRADATION USING ACCELERATED AGING TESTS AND ELECTROLUMINESCENCE TECHNIQUE I. SUMMARY To improve PV module’s lifetime and reliability, it is essential to understand the mechanisms and causes that degrade module performance over time under different climatic condition. In this study, our main aim is to correlate the degradation of module power with physical changes that are generated in the cells over time.

The multi C-Si 200W modules are subjected to long term accelerated tests (testing to be continued until failure) and the degradation mechanisms are characterized using EL technique. The characterization is done periodically to analyze the trend of physical changes occurring in the module under increased stresses which contribute in power degradation. More focus is given on investigating the changes in series resistance under different environmental conditions as it has a direct impact on module’s output power.

Iterative simulation based software and EL images have been used to estimate series resistance of each cell.

II. INTRODUCTION After the fabrication of a high power photovoltaic (PV) module, the most critical aspect is the duration for which the module can produce useful power. Reliability and lifetime of photovoltaic (PV) modules are key factors to system performance and warranty and thus are extremely important as far as PV business is concerned. Due to time constraints, accelerating aging test protocols established by organizations like IEC and UL have been adopted worldwide as official tests to evaluate module quality and reliability standards.

To improve the reliability and thus lifetimes of modules, it is extremely critical to analyze the effects of environmental conditions on the performance of modules.

Series resistance (Rs) is one of the most critical parameters to evaluate module’s quality. Changes in series resistance can substantially degrade module’s output power. Thus, as the module undergoes various environmental stresses, periodic measurement of series resistance will help in better analysis of the mechanisms that degrades module’s performance over time.

Electroluminescence (EL) imaging is fast becoming an important non-destructive technique to diagnose the cell level defects in the module. EL technique generates a defect map of each cell of the module which can be used to evaluate the degradation of cells.

Moreover, using EL images of a cell at two different voltages, O. Breitenstein et al [1] have developed a fast converging iterative method to calculate the local series resistance of the cell. Using the findings of Fuyunki et al [2], Breitenstein developed a relationship between intensity and series resistance.

Using an iterative method based simulation; series resistance values corresponding to each local intensity point (pixel) can be calculated. 1/3 SiliconPV 2011 III.

EXPERIMENT In this paper we will present Electroluminescence (EL) studies on power degradation mechanisms in 200 W poly-crystalline silicon modules as they undergo thermal cycle (TC 200 cycles between +85oC and -40oC with a dwell time of 30 minutes at the temperature extremes. . The 200W modules are fabricated under a control trial. The cells used in modules are taken from three different bins (low current, medium current & high current).

Before subjecting to accelerating testing, each module is characterized using EL at three different voltages (1V, 35V & 38V) as suggested in [1].

Three modules (S1, S2 & S3), one of each type (with low, medium & high current cells), are subjected to the TC200 test cycle. Again, post test, EL is done for each module at three different voltages. A simple algorithm has been used to extract series resistance maps from the EL pixel intensity information at the three different voltages. IV. RESULTS The results post TC200 are summarized below in table I.

TABLE I. Percentage Change in Electrical Parameters Post TC200 Test Power Change Isc Change Voc Change (%) (%) (%) S1 with low power cells -1. 68 -0. 8 -0. 85 S2 with medium power cells -1.

36 -0. 56 -1. 02 S3 with high power cells -1. 84 -0. 29 -1.

14 Sample FF Change (%) -0. 04 No change -0. 42 Post TC200 test, all the samples showed a degradation in output power by more than 1. 3% but less than 2%, which is not substantial and easily passes the IEC qualification test (passing criteria is less than 8% power degradation).

However, as mentioned earlier, in order to improve the module’s durability, it is essential to understand the cause behind performance degradation and work upon them. The EL and Rs estimation is done preciously for that and the results are displayed in figures I and II, and table II.

The fill factor for sample 2 shows no change post test, however, due to measurement related errors in the instrument, this doesn’t reflects a true picture thereby making estimation of series resistance a lot more critical to analyze degradation mechanisms.

The dark areas in EL image represent an electrically inactive area. In Rs map images; lighter areas implies higher resistance. A direct correlation between the changes observed in EL images and Rs maps can be observed. The dark areas in EL image correspond to light areas in the Rs map which signifies higher series resistance areas. The increase in series resistance is probably due to loss of contact between the Ag top contact and the silicon material or due to soldering related stresses.

Table II displays the changes in global Rs values (calculated using Rs map) of the two cells. A significant increase in Rs has been observed post TC200 test and validates the observations found in above figures. The defects caused by accelerated test have led to an increase in Rs of 2/3 SiliconPV 2011 the cells. This increase in Rs of each cell is adding up and causing the module’s overall series resistance to increase which is contributing in the degradation of output power. FIG I.

EL images and corresponding Rs maps of a cell (C1) of sample S2 a) Pre Test b) Post Test FIG II.

EL images and corresponding Rs maps of a cell (C2) of sample S2 a) Pre Test b) Post Test TABLE II. Percentage Change in Series Resistance of two cells of sample S2 Cell Rs Change (%) C1 5. 61 C2 7. 22 V. CONCLUSION EL images clearly show significant physical changes in the cells after temperature cycling test.

More often the defects are related to localized cracks or metallization defects which result in increase of series resistance of the module.

The estimation of series resistance of the cell using Breitenstein method validates this observation. Thus, increase in series resistance can be established as one of the causes of module degradation post TC200 test and in order to improve module’s reliability, work needs to be done to improve cell metallization. TC200 test will be further continued to observe the trend of changes in defects and series resistance. Also, more analysis is under progress to find other causes of module degradation.

Similar studies are under progress with other accelerating tests (DH1000 and HF10). VI. REFERENCES [1] O. Breitenstein, A. Khanna, Y.

Augarten, J. Bauer, J. M. Wagner, and K. Iwig, “Quantitative evaluation of electroluminescence images of solar cells” (Rapid Research Letters 4, No. 1, 7–9, 2010) [2] Takashi Fuyuki, Hayato Kondo, Yasue Kaji, Akiyoshi Ogane, and Yu Takahashi, “Analytic findings in the electroluminescence characterization of crystalline silicon solar cells”, (Journal of Applied Physics, 101, 2007) 3/3