以實驗設計探討太陽能電池焊帶的最佳焊接條件

Optimizing the welding conditions for solar-cell bus bar through experimental design

曾馨綿(Hsin-Mien Tseng)

指導教授 ： 邱裕方

中原大學/電機資訊學院/工業與系統工程研究所/碩士(2019年)

https://doi.org/10.6840/cycu201900584

摘要

本研究是以T太陽能公司為研究對象，針對T太陽能公司所面臨到焊接不良導致拉力不足而產生的客訴問題，當次拉力客訴最終衍生出高昂的賠償金。反思到工廠端如何由尚未量產階段，來透過實驗設計找出最佳銲帶的焊接條件，進而提升公司產出的良率。由於太陽能模組的單價較高，且有不易重工的特性。因此若造成整片模組的報廢將會讓公司帶來不小的損失。而現今半導體產業晶圓片的單價更是高昂，因此無論太陽能產業還是現今半導體產業甚至其他電子產業，都適用於此實驗設計，來減少製程上無謂的耗損，提升產出與企業競爭的優勢。
本研究透過實驗設計挑選出前三大實驗因子:烙鐵焊接溫度、加熱底板溫度與焊接時間，並將其每個因子，三個水準做成本實驗設計架構，觀察其相關分佈圖是否為常態分佈，可驗證此模型分析所得到的結果是否具有參考之價值。並分析其三元因子間的交互作用，再分析其二元因子間的交互作用與其相關的顯著性，最後再分析其主因子間的顯著性，後續進行實驗分析來找出最佳焊接條件。最後採用簡單迴歸分析模型，將改善前與改善後，拉力值與實際模組產出之良率進行迴歸分析。將其研究結果進行彙整，提出相關的改善與建議。對於未來太陽能產業之發展與參考

關鍵字

太陽能 ； 焊接條件 ； 實驗設計 ； 產出良率 ； 迴歸分析

並列摘要

The object of this research is based on T Solar Cell Company. T Solar Cell Company encountered the customer complaints resulted from insufficient peeling force caused by inappropriate welding process and then this peeling force complaint has finally caused high compensation. After this customer complaint, the company started to consider how to figure out the optimized bus bar welding condition through Design of experiment (DOE) of cell ribbon before mass production then production line can further enhance output yield after product launched. Since the solar cell module possessed the characteristics of high unit price and is not easy to re-work, it will cause high loss for company if the whole solar cell module was scrapped. In the meanwhile, the wafer and die unit price are extraordinarily high in semiconductor industry, thus DOE methodology is applicable no matter for solar cell or semiconductor field to reduce process waste, improve output and promote corporate competitiveness.
This research chose the top three process factors, says iron welding temperature, heating plate temperature and welding time, for DOE analysis and each factor was set three levels as DOE construction. Through the statistical analysis, it could be observed if it belongs to normal distribution and if it has referable value by verifying the result via such analytic model. The significance of three-factor interaction was clarified at the first place, then two-factor interaction clarification was executed in the following step. At last, all three independent main factors were also identified if it is critical to the output peeling force respectively. Then the optimized welding process condition was figured out by DOE analysis. Furthermore, regression model was also applied to prove the correlation between peeling force and solar cell module yield for both before and after improvement data. Based on the research conclusion, the improvement suggestion was summarized and proposed to the corporate for future development and reference in solar cell industry.

並列關鍵字

Solar cell ； Welding condition ； DOE ； Output yield ； Regression analysis

參考文獻

1. Atkinson, A. (1987), Plots, Transformations, and Regression: An Introdution to Graphical Methods of Diagnostic Regression Analysis. Oxford Science Publications. Clarendon Press, 35(1), 147-156.

2. Beneking, C., Rech, B.,&Wieder, S.(1999), Recent developments ofsilicon thin film solar cells on glass substrates,Thin Solid Films,351, 241-246

3. Chellaih, T., Kumar, G., and Prabhu, K. N., (2007), "Effect of thermal contactheat transfer on solidification of Pb–Sn and Pb-free solders." Materials & Design, 28, 1006-1011

4. Coleman, David E. & Montgomery, Douglas C. (1993). A Systematic Approach to Planning for a Designed Industrial Experiment. Technometrics, 35(1), 1-12.

5. Domb, E. (1998),The 39 Features of Altshuller Contradiction Matrix, The TRIZ Journal, 16(3), 179-193