英文原文.pdf

Journal of Materials Processing Technology 139 2003 422 427 New processes to prevent a fl ow defect in the combined forward backward cold extrusion of a piston pin D J Leea D J Kimb B M Kimc a Department of Precision Mechanical Engineering Graduate School Pusan National University Pusan South Korea b Department of Mechanical Design Engineering Graduate School Pusan National University Pusan South Korea c Department of Mechanical Engineering Engineering Research Center for Net Shape and Die Manufacturing Pusan National University No 3 Janjeon Dong Kumjeong Ku Pusan 609 735 South Korea Abstract A fl ow defect of a piston pin for automobile parts are investigated in this study In the combined cold extrusion of a piston pin a lapping defect whichisakindoffl owdefect appearsbythedeadmetalzone Thisdefectisevidentinproductswithasmallthicknesstobepierced and is detrimental to dimensional accuracy and decrease of material loss The fl ow defect that occurs in the piston pin has bad effects on the strength and the fatigue life of the piston pin Therefore it is important to predict and prevent the defect in the early stage of process design The best method that can prevent the fl ow defect is removing or reducing dead metal zone through the control of material fl ow Finite element simulations are applied to analyze the fl ow defect This study proposes new processes which can prevent the fl ow defect by removing the dead metal zone Then the results are compared with the results of experiments for verifi cation These FE simulation results are in good agreement with the experimental results 2003 Elsevier Science B V All rights reserved Keywords Flow defect Piston pin Material fl ow control Forward backward extrusion Dead metal zone FE simulation 1 Introduction Cold forming is extremely important and economical pro cesses especially for producing parts in large quantities Because of advantages of cold forming such as high pro duction rates excellent dimensional tolerances and surface fi nish mechanical and metallurgical properties cold form ing is by far the largest application of industry for producing parts However cold forged parts are also used in manufactur ing aircraft motorcycles nuts and bolts 1 but it is possible for defects to occur in forged parts depending on the de formation history forming conditions and material fl ow pat tern etc The kind of defects are ductile fracture caused by the state of stress and the deformation history fl ow defects caused by unstable material fl ow and poor dimensional tol erances caused by inferiority of the die and friction condi tion Further defects in forged parts are classifi ed as internal defects and external defects 2 4 These defects have harmful effects on the quality of the product and an increase in the cost of production Therefore Corresponding author Tel 82 51 510 3074 fax 82 51 514 7640 E mail address bmkim pusan ac kr B M Kim it is important to predict and prevent defects in the early stage of process design Wifi et al 5 studied ductile fracture in bulk formed parts using different workability criteria by the fi nite ele ment method Kim and Kim 6 studied internal and exter nal defects of cold extruded products with double ribs and performed process design to prevent these defects In this study is examined a defect which occurs in produc ing a forward backward extrusion product a piston pin for an automobile part and new processes are designed to pre vent the defect by fi nite element method in the early stage of process design Then the results are compared with the results of experiments for verifi cation 2 Forming and defect occurrence analysis 2 1 Forming process The piston pin is an automobile components used in the transmission of power between the connecting rod and the crankshaft In the cold extrusion of a piston pin the design requirements are to keep the same height of the forward extruded part and the backward part Fig 1 without any defect in the forged product for use under high and repeated 0924 0136 03 see front matter 2003 Elsevier Science B V All rights reserved doi 10 1016 S0924 0136 03 00515 6 D J Lee et al Journal of Materials Processing Technology 139 2003 422 427423 Fig 1 Shape and dimension of the piston pin Fig 2 Photograph of a fl ow defect of a piston pin load The material used for the piston pin is AISI 4135H Fig 2 alloy steel with the following fl ow stress behavior 768 06 0 139 MPa Fig 4 Distribution of effective strain and fracture value Fig 3 Conventional forming process for a piston pin The lubricant used is phosphate coating and bond lube The friction factor m is assumed to be 0 1 which is confi rmed by the ring compression test The sequence of the conventional process for the piston pin is performed using a multi stage former Fig 3 The fi rst and second stages are pre upsetting to eliminate defects by the cropping process such as ovality and ec centricity of the billet for improvement of dimensional tolerances and die life whilst the third and forth stages are forward or backward extrusion for the forming of one di rection from the web and fi nal stage is the piercing process for the pin shape However the results of experiment for the conventional processdisplayedadefectinthewebpartformedearlyinthe third process Fig 3 Especially a nonuniform fl ow pattern is observed in part of the defect occurrence which looks like a fl ow defect similar to lapping with an undesirable fl ow pattern 2 2 Prediction of defects by FE analysis DEFORM is used which is commercial code of a rigid plastic FE program for forming and defect analy sis The diameter of the initial billet is 30mm and the height is 61mm the whole volume of fi nal product being 424D J Lee et al Journal of Materials Processing Technology 139 2003 422 427 Fig 5 Metal fl ow and velocity distribution where a defect occurs according to stroke 43 118mm3 The forming is simulated with a conventional process sequence The maximum fracture value that can estimate the occur rence of a crack 7 is small at 0 08 and is distributed in a position within the head part of the punch so that a de fect does not occur Thus this defect is not one due to duc tile fracture Fig 4 Then fl ow line tracking scheme that was proposed by Altan and Knoerr 8 is performed for de fect analysis According to the progress of the punch stroke severe variation of fl ow lines appears and discontinuity of velocity occurs in the part that a defect occurred in the ex periment Fig 5 Consequently the metal fl ows only in the backward di rection without fl ow to the forward direction in the fourth process and metal near the web part is pulled up in the rib part like a lapping defect Therefore the cause of the ini tiation and development of the fl ow defect that occurred in piston pin is the velocity discontinuity between backward and forward direction by the formation of a dead metal zone This appearance evidently occurs in products like a piston pin with a low thickness to be pierced for the dimen sional accuracy and the decrease of material loss A fl ow defect occurring in a piston pin has harmful effects on the strength and the fatigue life of a piston pin that has high and repeated load at high temperature Therefore it is necessary for a new process to prevent the fl ow defect 3 Process redesign and analysis for the prevention of defect The cause of the initiation and development of the fl ow defect is the restriction of metal fl ow by the dead metal zone For the elimination of the dead metal zone in the early extruded part 3rd process in the conventional process the forward or backward extrusion process is modifi ed to combined forward backward extrusion which is performed simultaneously in the two directions Because of the variety of extrusion ratios and lengths in the forward and backward directions the simultaneous completion of the material fl ow in the both directions is very diffi cult Consequently one of the directions is completed early then material fl ow stopped and dead metal zone appears in this part just like that in the conventional process Therefore in the case of piston pin forming the extrusion ratio and the length of both directions are the same at 1 89 and51mm First analysisofopendieforward backwardex trusion is performed for an investigation of extrusion lengths of the piston pin The difference of two extruded ribs is 24 9mm and the backward extruded rib is shorter than the forward extruded rib as shown in Fig 6 The metal fl ow of backward direction must be restricted compulsorily for the satisfaction of the design conditions and this means the occurrence of a dead metal zone There fore for the same extrusion length in both directions three Fig 6 Extrusion length in forward backward extrusion D J Lee et al Journal of Materials Processing Technology 139 2003 422 427425 Fig 7 Modifi ed process sequence for a multi stage former methods are proposed to control the metal fl ow without the compulsory restriction of metal fl ow 3 1 Change of preform shape To secure the same length of both directions from the center of web it is required that the backward extruded rib is performed by preform design as the difference of both direction lengths at 24 9mm from the above results before forward backward extrusion Fig 7 shows the mod ifi ed process sequence and Fig 8 shows the metal fl ow of the fi nal stage of forward backward extrusion in this case From the results of simulation the lengths of two extruded ribs are 51mm which is the dimension of the piston pin and satisfi ed the design condition In addition the metal fl ow is uniform in the defect zone where the fl ow defect occurred in the conventional process and there is not a discontinu ity of velocity in both extrusion directions This means that metal fl ows uniformly in the whole process without a dead metal zone by restriction of metal fl ow Fig 8 Metal fl ow of web in case of using preform Fig 9 Schematic diagram of the axially moving container die structure 3 2 Driving of extrusion container The driving extrusion container method 9 is used for metal fl ow control for the satisfying of the design condi tion This structure is that the extrusion container is moved in the counter direction to the early extruded one Fig 9 This has the effect of increasing the metal fl ow in the late extruded direction and restricting metal fl ow in the early ex truded direction In the case of the piston pin because of the early completion of backward extrusion the extrusion container is moved in the forward direction for the increase of metal fl ow to this direction In this process the princi pal process variables are the relative velocity ratio of the punch and the moving extrusion container and the friction condition between the material and the moving extrusion container In this study because the friction factor m is 0 1 be tween the material and container simulation is performed only according to the variation of the relative velocity ratio VC VP 0 1 0 25 0 5 0 75 1 0 If the relative velocity ratio is smaller than the optimum which can complete form ing simultaneously extrusion in the backward direction is completed earlier than in the forward direction and a fl ow defect occur in the same part as in the conventional process Otherwise if the relative velocity ratio is larger than the op timum one extrusion in the forward direction is complete earlier than backward direction and a fl ow defect occurs in the opposite part to where a defect occurs in the conven tional process Therefore for satisfaction of the design conditions the optimum relative velocity ratio is searched for by an opti mization technique the bisection method From the result the optimum relative velocity ratio is 0 48 Figs 10 and 11 show the deformation modality and metal fl ow according to the relative velocity ratio 0 1 0 48 1 0 for a punch stroke of 42 7mm respectively Fig 11 c shows the metal fl ow 426D J Lee et al Journal of Materials Processing Technology 139 2003 422 427 Fig 10 Deformation modality of the piston pin according to the relative velocity ratio Fig 11 Comparisons of metal fl ow according to the relative velocity ratio at the optimum relative velocity 0 48 where an improved fl ow pattern without a fl ow defect can be noted 3 3 Modifi cation of die structure A modifi cation of the die structure is proposed which can restrict the metal fl ow of backward direction which is deformed early for simultaneous completion of extrusion in both directions In this case for simultaneous completion and the same length in both directions the stripper which is Fig 12 Schematic diagram of die structure using stripper equipmentforpunchextractionfromproducts isredesigned If a fi xed stripper of conventional type is used a dead metal zone appears from the middle stage of backward forward extrusion by the restriction of material fl ow Therefore a structure is used that can delay the metal fl ow in the backward direction by spring force Fig 12 shows the die structure For this method it is very important to decide the proper spring force for simultaneous completion of forming Therefore the necessary spring force for this is calculated by FE simulation From the simulation result it was 5t to be applied load to stripper Fig 13 shows metal fl ow in this case The metal fl ow is similarly uniform at the defect zone without discontinuity of velocity in comparison with other modifi cation methods Fig 13 Metal fl ow of web in case of using stripper D J Lee et al Journal of Materials Processing Technology 139 2003 422 427427 Table 1 Comparison process for each of the proposed method Conventional methodUse of preformUse of stripperUse of moving container Maximum load t 97 296 396 184 0 Process of extrusion2 stage2 stage1 stage1 stage DefectExistNoneNoneNone 4 Results and experiment From the FE simulation the three proposed methods are proper to prevent a fl ow defect by metal fl ow control The characteristics of each process are as follows The fi rst method that uses a preform needs three stage processes pre forming forward backward extrusion piercing and has a simple die structure however the second method that uses a stripper and the third method that uses an axially mov ing container need two stage processes forward backward extrusion piercing and have a complex die structure In respect of the forming load the processes are similar to each other Especially themaximumformingloadissmallerthanthat of other processes by about 10t in the case of the axially moving container because the axially moving container in creases material fl ow in the direction punch movement It is compared with the proposed method for forming by a press in Table 1 In this study an experiment using a preform is performed and uses a multi stage former having 250t ca pacity for the verifi cation of simulation Etching for obser vation of metal fl ow is performed to examine for a fl ow defect for the piston pin before piercing Fig 14 shows the experiment result based on the fi rst proposed method changing the preform The experiment result shows that metal fl ow is uniform in the defect zone where the fl ow de fect had occurred and satisfi ed the simultaneous completion of forming and the same length in both extrusion directions This tendency is in good agreement with the simulation result Fig 14 The elimination of the fl ow defect by the fi rst proposed method 5 Conclusions In this study the fl ow defect that occurs in the manu facturing process of the piston pin is examined and a new process to prevent the defect is redesigned by FE analysis First the cause of the defect is invest