| 摘要: | 金屬粉末射出成型(MIM)的製程包括金屬喂料混和、喂料射出、保壓、冷卻、頂出脫模、脫脂及燒結工藝，MIM需要遵循這些步驟流程才能製作出金屬零件。現今產業的競爭，迫使射出成型的產品需要高品質、更快速、更便宜的生產零件，並盡可能地避免缺陷，例如最小化翹曲、避免裂痕等，因此射出成型的參數對產品的影響需要更深入的探討。在射出成型中，模流分析軟體可以滿足探討射出成型參數影響品質特性的這個需求，Moldex3D為商業模流分析軟體，利用三維模擬分析技術解決射出成型的問題，並且結合參數優化的方法，例如田口法及反應曲面法，可以更有效率的節省時間及金錢上的成本。
本研究利用Moldex3D對隨形水路加隔板水路鎖類零件的MIM製程進行模流分析，經由探討參數對於翹曲的影響，並將參數優化以最小化產品的翹曲。產品的網格建立是利用Rhinoceros 3D進行網格建模，並將其進行網格獨立性測試，挑選出最佳的網格作為後續參數優化的分析。參數優化的分析中挑選出的控制因子，分別為保壓壓力四階段(保壓壓力1、保壓壓力2、保壓壓力3、保壓壓力4)、保壓時間四階段(保壓時間1、保壓時間2、保壓時間3、保壓時間4) 、喂料溫度、模具溫度及充填流率，共11個控制因子進行分析。利用參數優化的方法，田口法配合方差分析(ANOVA)篩選並優化對翹曲影響顯著的控制因子，田口法優化後，利用一半法則得到三個對翹曲影響最顯著的控制因子，分別為喂料溫度、模具溫度及保壓壓力1。後續結合反應曲面法將這三個控制因子更進一步優化分析，藉由這些因子的優化，本研究將鎖零件的翹曲降低了27.7%，達到參數優化並最小化翹曲的目的。最終本研究的結果可以作為工程師或研究人員一套參數優化流程的參考。
關鍵字：金屬粉末射出成型、翹曲、參數優化、DOE、ANOVA、Moldex3D
;In this study, Moldex3D was used to perform mold flow analysis on the metal injection molding process of conformal cooling channel in lock parts. By discussing the influence of parameters on warpage, the parameters were optimized to minimize product warpage. Rhinoceros 3D, a CAD software, was used for the generation and analysis of the model grid. A grid independence test was performed before conducting the mold flow simulation for subsequent parameter optimization analysis. The control factors selected in the in the study of parameter optimization included the four stages of analysis of the four stages of packing pressure, the four stages of packing time, material temperature, mold temperature, and injection flow rate, which amounts to 11 control factors. The Taguchi method combined with the Analysis of Variance (ANOVA) was adopted to screen the control factors that have significant effects on warpage. Using the half rule, we obtained the three most essential control factors for warpage, which were the material temperature, mold temperature, and the first phase packing pressure. The subsequent optimization and analysis of these three control factors were performed using the response surface method, which obtained the warpage reduction of the lock parts by 27.7%. The results can serve as a reference for engineers or researchers in the community of mold flow analysis and optimization.
Keywords: Metal injection molding, Warpage, Parameter optimization, DOE, ANOVA,
Moldex3D |
