AISI-M42高速鋼十字沖頭擠鍛成形製程之研究

Study on Extrusion-Forging for Philips Punch of AISI-M42 High Speed Steel

彭御芳(Yu-Fang Peng)

指導教授 ： 許源泉

國立虎尾科技大學/工程學院/創意工程與精密科技研究所/碩士(2009年)

摘要

十字沖頭用於沖製螺絲頭之十字型槽，對於產量及品質要求高；而十字沖頭於擠鍛生產時，易產生充填不足及表面裂紋等缺陷，而影響生產效率及產品品質。對於現場技術人員憑藉個人經驗和試誤法已無法滿足未來技術研發的需求；因此，必須對此製程建立科學化的分析，以提升生產技術及產品品質。
本研究係針對十字沖頭擠鍛製程進行探討，先以AISI-M42進行材料性質分析；再經由壓縮試驗結合FEM求得材料之破裂值，並利用材料之應變硬化情況建立等效應變及硬度的相關性，以便於預測鍛件成形之破裂及應變硬化狀況；而後進行十字沖頭擠鍛傳統製程模擬分析與實驗，並相互比對以驗證其之間的差異性。接著對十字沖頭擠鍛之鍛件破裂值、等效應變及等效應力進行分析，以得更為詳細之製程分析資料。另一方面，本文也針對擠鍛製程中的介面摩擦因子、胚料預成形角度、母模圓角半徑及頂桿球面形狀等參數對成形的影響性進行探討；並結合田口法作最佳化分析，以求得最佳製程參數；最後應用類神經網路建構出製程參數對成形負荷與充填品質之預測模式。
經由研究結果發現，本文使用的破裂準則中，以Normalized Cockcroft & Latham可較準確的預測鍛件的破壞，而鍛件肋部表面之破裂值皆高於壓縮試驗所得之破裂值。十字沖頭經擠鍛成形後有明顯的加工硬化現象，而等效應變-硬度預測值高於實際鍛件，但在等效應變0.4-1之間具有相差約38HV之相對應性。本文並歸納出製程參數中，介面摩擦因子越大，成形負荷越大，充填狀況則越差；胚料預成形角度越大，成形負荷越小，充填狀況亦越好；而母模圓角半徑越小，十字沖頭肋部與凹角處的破裂值則越大。利用田口法分析分別得到成形負荷與充填品質之最佳製程參數，並以變異數分析了解各製程參數中，影響成形負荷以摩擦因子與胚料預成形角度佔有較大比重，而對於充填品質的影響則是以摩擦因子佔有較大比重。最後利用類神經網路所建構出製程參數對成形負荷與充填品質的預測模式，其平均誤差分別為4.73%與6.24%，故此模組用於擠鍛製程的預測具有相當的準確性。透過本文對AISI-M42十字沖頭擠鍛製程之研究結果，建立完整分析資料，以期提供於此相關鍛造製程設計之參考。

關鍵字

AISI-M42 ； 十字沖頭 ； 擠鍛 ； 有限元素分析

並列摘要

The Philips punch forming the cross groove of screw’s convex-head is a mass-produced and high-quality product. The defects such as unfilled cavity and surface fissures, found effortlessly in extrusion-forging process for producing the Philips punch, would influence the production efficiency and the products quality. Relying on experiences from the technicians and the trial-and-error approach can not fulfill the requirements of the future technological research and development. Therefore, it’s necessary to apply the scientific analyzing methods to promote the production technology and the products quality in the extrusion-forging process.
The current study probed into the extrusion-forging for Philips punch. It firstly analyzed the material characteristics of AISI-M42 High Speed Steel. To predict the fracture and strain hardening of the forgings, the compression test combined with FEM was employed to get the damage value of material, and the condition of strain hardening was considered to construct the relationship between the effective strain and hardness. Next, analysis was made through finite element simulation and experiment on the conventional extrusion-forging for Philips punch. Comparison result was used to verify their differences too. Then, the distributions of damage value, effective stress, and effective strain on the forgings were analyzed to get more detailed analytic data of extrusion-forging for Philips punch.In addition, the study also explored the influence of various parameters on forming, such as friction factors of die/workpiece interface, angle of the preformed workpiece, shoulder radii of the upper die, and geometries of the ejector pin. To obtain the optimal process parameters, Taguchi method was also applied to make an optimum analysis. Finally, Abductive network was employed to construct the prediction model of the influence of process parameters on the forming load and the die cavity filling quality.
Results showed that the Normalized Cockcroft & Latham was the most accurate fracture criteria to predict defect, and the damage values of ribs surface on Philips punch were higher than those obtained by the compress testing. An obvious work hardening was found in the Philips punch formed by extrusion-forging, and the effective strain-hardness prediction value was higher than the real forging sample; but the difference was about 38HV in the effective strain between 0.4 and 1. It can also be concluded that the higher the friction factor of die/workpiece interface, the higher the forming load will be, but the less the die cavity filling ability will be. When the angle of the preformed workpiece is larger, the forming load will be smaller, and the die cavity filling ability will be better. When the shoulder radius of the upper die is smaller, the damage value of the concave on Philips punch will be bigger. In the current study, the optimal process parameters of the forming load and filling quality were obtained by Taguchi method respectively. Through the variance analysis, it has been found that the friction factor and the preform angle were the main process parameters affecting the forming load. The predictive model of the process parameters to the forming load and the die cavity filling quality were also constructed by Abductive network, and the average error of the forming load and the filling quality were 4.73% and 6.24% respectively. Therefore, a conclusion can be drawn that the model constructed in this study is accurate to predict the extrusion-forging processes. Through this exploration, a detailed analytic data of AISI-M42 Philips punch has been developed, and hopefully it can provide the industry with the guidelines in the process design of the extrusion-forging for Philips punch.

並列關鍵字

AISI-M42 ； Philips punch ； Extrusion forging ； Finite element analysis

參考文獻

[1]M. S. J. Hashmi, and F. B. Klemz, Axisymmetric extrusion forging: effects of material property and product geometry, International Journal of machine tool design and research, 26, (1986) 157-170

[2]G. Maccarini, C. Giardini, and A. Bugini, Extrusion operations: F.E.M. Approach and experimental results, Journal of material processing technology, 24, (1990) 395-402.

[3]K. P. Rao, S. M. Doraivelu, and K. Sivaram, Physical modelling studies using spike forging to verify analytical predictions, Journal of materials processing technology, 28, (1991) 295-306.

[4]G. Maccarini, C. Giardini, G. Pellegrini, and A. Bugini, The influence of die geometry on cold extrusion forging operations: FEM and experimental results, Journal of materials processing technology, 27, (1991) 227-238.

[5]W. Hu, and M. S. J. Hashmi, Study of metal flow in extrusion forging of rectangular billets, Journal of materials processing technology,43, (1994) 51-59.