臺灣博碩士論文加值系統

本文利用剛體-彈簧連桿模型來模擬I.C.封裝中銲線成形的過程，並藉此希望減少金線成形後的弧高和下陷現象。本文以建立一套模擬鋼嘴在上升過程與下降過程的分析模式，並在其中將加速度的影響考慮進去。由於加速度的作用而產生的慣性力以及其在不同應變率的情況下所產生的材料性質的改變在論文中也將一併探討。
同時，針對銲線成形的設計導入田口式實驗計劃法。並利用變異數分析找出各個控制因子對銲線成形的影響程度，並利用各個因子的最佳水準來作最佳化的實驗。最後，吾人分別利用正向的類神經網路與逆向的類神經網路來得到所欲達到金線外形的成形參數。而藉著此一方法，吾人可以針對大幅縮短時間在金線成形的設計上，以期望達到即時設計的目的。

The linkage-spring model has been used to determine the design rules to reduce the loop heights and the sagging of gold wire in the package. The purpose of this thesis is to simulate the capillary trajectory from the first bond to the second bond stages by a modified linkage-spring model including a looping speed factor. Also, the looping speed effects on the wirebond profiles due to the inertia mass and the change of material properties in different strain-rates have been studied in this thesis.
At the same time, Taguchi method was used for the robust design of wire-bonding profile for I.C. packaging. Using the Analysis of Variance finds out major factors which affects quality characteristics. The object is to find out the optimal assembly of control factors and levels. Finally, we presents forward and inverse neural network models to predict the loop profile parameters respectively. With the present methodology, how to design loop profile parameter would take a considerable amount of time. The design process can be accelerated in real time by using a neural network.

目錄
頁次
中文摘要 i
英文摘要 ii
致謝 iii
目錄 iv
圖目錄 vii
表目錄 ix
第一章 緒論
1.1 前言 1
1.2 研究動機 2
1.3 研究目的 2
1.4 文獻回顧 3
1.5 章節瀏覽 5
第二章 分析模式
2.1 銲線程序 7
2.2 上升過程分析模式 8
2.3 下降過程分析模式 9
2.3.1 基本假設 9
2.3.2 剛體彈簧模式 9
2.3.3 金線材料特性與彈簧彈性係數之關係 12
2.4 銲線速度對金線成形之影響 14
2.4.1慣性質量的影響 15
2.4.2 應變率對材料性質之影響 16
2.5結論 17
第三章 田口品質設計方法
3.1 田口品質設計方法簡介 27
3.2 品質工程 28
3.3 田口實驗規劃 29
3.3.1 成形三角形 30
3.3.2 成形梯形 31
3.4 預估最佳水準組合與驗證 32
3.4.1 成形三角形loop height 33
3.4.2 成形三角形sagging 33
3.4.3 成形梯形loop height 35
3.4.4 成形梯形sagging 35
3.5 結論 35
第四章 類神經網路在金線成形分析上的應用
4.1 類神經網路簡介 43
4.1.1生物神經元與人工神經元 43
4.1.2 類神經網路系統 45
4.2 倒傳遞類神經網路 .45
4.2.1 網路架構 46
4.2.2網路演算法 46
4.2.3網路參數設定 52
4.2.4學習結果評估 54
4.2.5 類神經網路優缺點 55
4.3類神經網路在金線成形分析上的應用 56
4.3.1成形三角形正向網路模組 56
4.3.2成形三角形逆向網路模組 57
4.3.3成形梯形正向網路模組 58
4.3.4 成形梯形逆向網路模組 59
4.4結論 60
第五章 綜合討論與未來發展
5.1 綜合討論 76
5.2 未來發展 77
參考文獻 79

參考文獻
Burden R. L. and Faires J. D., Numerical Analysis, PWS-KENT Publishing Company, USA., pp.553-557, 1993.
Calmidi V. V. and Mahajan R. L. “Optimization for Thermal and Electrical Performance for a Flip-Chip Using Physical-Neural Network Modeling,” Electronic Components and Technology Conference, pp.1163-1168, 1997.
Dennis E. Pachucki Iocco, “Environmental Stress Testing Experiment Using the Taguchi Method,” IEEE, CPMT part A, Vol. 18, NO.1, pp. 3-9, 1995.
Groover R., Shu W. K.and Lee S. S., “Wire Bond Loop Profile Development for Fined Pitch-Long Wire Assembly ,” IEEE Transactions on Semiconductor
Holdgrafer W. J., Levine L. R. and Gauntt D. L., “Method of Making Constant Clearance Flat Link Fine Wire Interconnections,” US Patent Number ：5,156,323, 1992.
Li Y., Mahajan R. L. and Tong J., “Design Factors and Their Effect on PCB Assembly Yield-Statistical and Neural Network Predictive Models,” IEEE, CPMT part A, Vol. 17, NO.2, pp. 183-191, 1994.Manufacturing, Vol. 7, pp. 393-399, 1994.
Mertol Atila., “Application of the Taguchi Method on the Robust Design of Molded 225 Plastic Ball Grid Array Packages,” IEEE, CPMT part B,Vol. 18, NO.4, pp. 734-743, November 1997.
Nakamura S., Applied Numerical Methods with Software, Prentice-Hall, Inc., USA., pp.73-76, 1991.
NSK Technical Journal＃659, “Development of Linear Motor for Wire Bonder,”
Ohno Y., Ohzeki Y., Aso T. and Kitamura O., “Factors Governing the Loop Profile in Au Bonding Wire,” 42nd Electronic Component and Technology Conference, May 18-20, San Diego, USA, pp. 899-902, 1992.
Okura T., Kanai M., Ogata S., Takei T. and Takakusagi H., “Optimization of Solder Paste Printability with Laser Inspection Technqiue,” IEEE/CPMT Int''l Electronics Manufacturing Technology Symposium, pp. 361-365, 1995.
Ravi, K.V. and Philofsky E., “The Structure and Mechanical Properties of Fine Diameter Al - 1pct Si Wire,” Metallurgical Transactions, Vol. 2, pp. 711-717, 1971.
Ryu Y. K. and Cho H. S., “Visual Inspection Scheme for Use in Optical Solder Joint System,” Proceeding of the 1996 IEEE International Conference on Robotics and Automation Minneapoils, Minnesota, April 1996.
Saint-Martin Xavier, Bull S.A., Les Clayes Sous Bois, France. Yuan Chollet, E.N.S.M.M., Besancon, France., “Wire-Bonding Machine Optimization Using Taguchi Method,” IEEE/ISHM Symposium, pp. 74-83, 1992.
Tay A. A. O., Seah B. C. and Ong S. H., “Finet Element Simulation of Wire Looping During Wirebond Fabrication,” ASME, Journal of Electrical Packaging , pp.93-101, 1998.
Toshiba Mechatronics Co., LTD., Wire Bonder Technology Manual, 1998.
Vikram S. and Nguyen L., “Neural Network-Based Real Time Thermal Design Expert System,” IEEE/CPMT Int''l Electronics Manufacturing Technology Symposium, pp. 443-449, 1998.
William Y. Fowlkes and Clyde M. Creveling, “Engineering Methods for Robust Product Design-Using Taguchi Methods in Technology and Product Development,”Addison-Wesley Publishing Company, 1995.
Yamazaki N., Hasegawa T., Enomoto J., Terkado Y. and S Kumazawa., “Method of Wire Bonding,” US Patent Number：4,932,584, 1990.
陳昭亮, 林業超,郭柏宏 “銲線機金線結球製程參數分析,” 中國機械工程學會第十六屆全國學術研討會論文集, 第三冊, pp. 27-34, 1999
何天露, “銲線成形設計之最佳化, ”碩士論文,國立成功大學機械工程學系, 1998.
蘇朝墩, “產品穩健設計”,中華民國品質學會, 1997.
葉怡成, “應用類神經網路”,儒林圖書, 1997.