Original scientific article Received: Mar 07, 2017 Accepted: Apr 20, 2017 DOI: 10.1515/rmzmag-2017-0006 Reliability of system for precise cold forging Zanesljivost sistemov za precizno preoblikovanje v hladnem Vid Krušič1*, Tomaž Rodič2,3 1MAHLE Letrika d.o.o., Polje 15, 5290 Šempeter pri Gorici, Slovenia 2University of Ljubljana, Faculty of Natural Sciences and Engineering, Aškerčeva cesta 12, 1000 Ljubljana, Slovenia 3C3M d.o.o., Centre for Computational Continuum Mechanics, Tehnološki park 21, 1000 Ljubljana, Slovenia *vid.krusic@si.mahle.com Abstract The influence of scatter of principal input parameters of the forging system on the dimensional accuracy of product and on the tool life for closed-die forging process is presented in this paper. Scatter of the essential input parameters for the closed-die upsetting process was adjusted to the maximal values that enabled the reliable production of a dimensionally accurate product at optimal tool life. An operating window was created in which exists the maximal scatter of principal input parameters for the closed-die upsetting process that still ensures the desired dimensional accuracy of the product and the optimal tool life. Application of the adjustment of the process input parameters is shown on the example of making an inner race of homoki-netic joint from mass production. High productivity in manufacture of elements by cold massive extrusion is often achieved by multiple forming operations that are performed simultaneously on the same press. By redesigning the time sequences of forming operations at multistage forming process of starter barrel during the working stroke the course of the resultant force is optimized. Key words: cold foring, process reliability, product accuracy, tool life, FE simulation Povzetek Prispevek podaja vpliv raztrosa glavnih vhodnih parametrov preoblikovanega sistema na natančnost izdelka in vzdržljivost orodja pri hladnem preoblikovanju v zaprti matrici. Za postopek nakrčevanja v zaprti matrici je bila izvedena prilagoditev glavnih parametrov procesa na maksimalne vrednosti, ki zagotavljajo zanesljivo proizvodnjo preciznega izdelka in optimalno vzdržljivost orodja. Kreirano je bilo operacijsko okno z maksimalnim raztrosom vhodnih parametrov za postopek nakrčevanja v zaprti matrici, ki zagotavlja želeno natančnost izdelka in optimalno vzdržljivost orodja. Prilagoditev vhodnih parametrov je prikazana na izdelavi kroglaste glave homokinetičnega zgloba v masovni proizvodnji. Visoko produktivnost izdelkov lahko dosežemo z več stopenjskim preoblikovanjem na isti stiskalnici. Z rekonstrukcijo časovnega zaporedja poteka preoblikovalnih operacij pri večstopenjskem preoblikovanju pesta zaganjalnika lahko izvedemo optimizacijo poteka rezultante sil. Ključne besede: hladno preoblikovanje, zanesljivost procesov, natančnost izdelka, vzdržljivost orodij, simulacija z MKE 3 Open Access. © 2017 Krusic V., Rodic T., published by De Gruyter Open. | feci m.'l.g.TI This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 LicensBrought to you by | National & University Library Ljubljana Authenticated Introduction When designing the cold forging manufacturing process, the forging process can be taken as a complex system that includes implicit interactions between the workpiece, tool and press influenced by tribological and environmental conditions. The centre of the system is the forging process that enables to shape the work-piece of simple initial shape into final product of complex shape and accurate dimensions next to good surface conditions. Response of the system is desired system function or the product of desired geometry and dimensional accuracy, respectively, the tool life, etc [1,2]. It must be emphasized that parameters of random and systematic deviations enter the forging process, and they consequently cause deviations of the product dimensional accuracy and of the tool life, and thus they cause the unreliability of process [3,4]. Reliable production of mechanical components by cold forging can be achieved with designing such a manufacturing process that enables production of components with required quality at minimum production costs which are in cold forging most often related to the service life of tools. Figure 1 schematically presents the forging system into which the parameters enter with random and systematic deviations. The system response represents a desired function or, in our case, the process operating window where the quality of product and the tool service life are in the limits of desired and permissible scatter, respectively [5]. High productivity in manufacture of elements by cold massive extrusion is often achieved by multiple forming operations that are performed simultaneously on the same press. By the help of FE we can control the material flow and load for individual stage already in the planning stage and in this way try to optimize the process. However, we have no insight into the entire forging system, where during the process elastic deflections and rotations of the press ram occur due to eccentric load, which is caused by the resultant force of forging operations during the working process. The main motivation is to plan a reliable multistage process that assures production of accurate cold forged parts. By redesigning the time sequenc- es of forming operations at multistage forming process during the working stroke the course of the resultant force is optimized. The article analyses the influence of scatter of principal input parameters of the forging system on the dimensional accuracy of products and on the tool life for closed-die forging process. The input parameters that most strongly influence the scatter of process response have been identified. By the help of the adjustment of basic input parameters of the production process to maximal values of deviations the production process is positioned into operating window that enables reliable achievement of the desired system function. Application of the adjustment of the process input parameters is shown on the example of making an inner race of homokinetic joint from mass production. An example from the regular mass production is represented to demonstrate prediction of tool loads and optimization of the resultant force. Reliability and robustness of the process One of the main goals of the forging process is to produce components with required quality. The quality of the production process should be designed already in the stage of development and designing new products. In industrial production process, periodic control of the production process is necessary. Monitoring and control of the production process is achieved by statistical process control (SPC) that determines the process capability to produce quality products. Product is supposed to be of good quality if its dimensional tolerances are in a certain tolerance interval, determined by the upper (USL] and the lower (LSL] specification limit (Figure 2]. Important standpoint of the SPC is to determine the process capability indices that show the capability of the manufacturing process to produce quality products. Statistical distribution of measured product dimensions is represented by normal frequency distribution curve - Gaussian distribution. Position and spread of the normal distribution is defined by the mean value, / , and standard deviation, a. The process capability indices, Cp and Cpk are defined as: Tool life Figure 1: Scatter impact of input parameters of the forming system on the accuracy of products and tool life. Cp = and USL - LSL 6a Cpk = mm USLLSL 3a 3a (1) (2) where Cp index gives the ratio between the size of tolerance range and the 6a tolerance interval of the production process, while Cpk index gives the position of the tolerance interval of the production process or the centricity of the process, respectively. Full line in Figure 2 represents the 3a process, for which the values of both indices, C and C,, p pk are unity. The 3a process represents the interval in which 99.73 % of all the results is expected to be, i.e. 2700 defective products per million. Dashed line in Figure 2 represents the robust 6a process with 99.9999998 % efficiency, i.e. 3.4 defective components per million, for which C = C, = 2 [6]. The main aims of the 6a method p pk are the following: to increase the satisfaction of customer, to improve the process capability, to improve the efficiency of company, to achieve advantage against rival companies, to operate with the least defects, to increase the yield, and to reduces the variability of processes. Implementation of the 6a method can be achieved by various methodologies, two of them are widely acknowledged and standardized: DMA-IC (Define-Measure-Analyse-Improve-Control), and DFSS (Design For Six Sigma) [7]. The main difference between the two methods is that the first one is used for improving the existent products and processes, while the second one Figure 2: 3o and 6o process. is used in designing and development of new products and processes. Optimization of the production process based on the reliability and the robustness usually takes place along the path shown in Figure 3. At first, it is assured that the distribution curve position of our process is inside the ± 3a interval. In the optimization, based on the reliability, the mean value of the distribution function is shifted towards the centre, i.e. optimal values for a certain objective function are sought. The reliability is related to the probability of occurrence of a defect. The aim of the reliability based optimization is that response of our system is in the region in which the probability of a defect occurrence is minimal. Figure 3: Process optimization based on the reliability and robustness. Figure 4: Closed-die upsetting process, value of relative deformation e = 0.45. Press Stress ring Upper die insert Preform Workpiece Lower die insert Stress ring Figure 5: FE model of closed-die upsetting process. In the Taguchi's robust process optimization, as described Yang and Haik [7], first the variability of the system response is reduced. This is made in such a way that the variability of those input parameters are reduced which influence the distribution scatter Lower variability of the response means higher quality of products and lower production costs. The next step in the robust optimization is to shift the mean value of the distribution with reduced variability towards the desired mean value, T (Figure 3). Focussing on those process parameters that have important influence on the distribution mean value and the minimal influence on the distribution scatter enables to shift the distribution towards the desired value, T. Created an operating window for closed-die upsetting process Closed-die upsetting process was analysed to evaluate stochastic interactions (Figure 4). Workpiece mass and flow stress, friction as process parameter, and press stiffness were chosen as the principal parameters of the form- 900 MPa 700 Of 500 400 300 200 100 0 . —.— —.— —.— —,— —I 0 0,2 0,4 0.6 0,3 1 1,2 1.67 was assured on the one hand and the tool life for 42,000 manufactured pieces as an average number on the other hand. Change of the preformed-piece mass has influence on the scatter of product height and the fullness of the outer shape. Preformed pieces with the mass concentrated on the upper level give products with the height on the upper tolerance limit and vice versa. Flow stress has not such a pronounced influence on the product dimensional accuracy. For reliable achievement of desired product height accuracy, also elastic deformations of die had to be taken in account. 100 Figure 16: An example of a two-stage forming process. Die was made of the ASP 23 steel with the hardness of 60 + 1 HRc (Figure 13 (c)). Suitable pre-stressing of the die, and forging in a 10 MN vertical mechanical press with vertical dynamic stiffness of 2.1 MN/mm enabled achievement of the specified tool life. Permanent plastic deformation and later crack appeared on the die bottom, as shown in Figure 13 (d). The adjustment of scatter of mass and flow stress parameters, the selection of suitable press stiffness, the design of the optimal die interference, and the selection of suitable material and of heat treatment process in making die assured reliable production of inner race for the whole series. Optimization of the resultant force course for a multi-stage cold forging process High productivity in massive production is often achieved by simultaneous implementation of more forming operations on the same press. The problem, which occurs in such situations, is schematically shown in Figure 16. In the illustrated case, first the forming process starts on the left side of the press axis, so that the resultant force equals the forming force of the first process. Later on, the second process is included, the maximum force of which is bigger than that of the first one. Consequently, during the process the resultant force moves to the other side of the press axis. The changing force torque causes relative rotations between the ram and the press table and performs asymmetry after Figure 17: Clutch barrel of a starter. ÇT ÇT Figure 18: A multi-stage forming process for manufacture of a starter clutch barrel. Figure 19: Three-stage tooling for barrel production. the first process. These faults can increase in the stage of the process when the force torque changes the sign and the ram is unstable due to the clearance in the guiding elements. The example in Figure 18 gives an analysis of a three-stage forming process used for manufacture of a starter clutch barrel (Figure 17), where operations E, F, and G are formed on the three-stage tool (Figure 19) at the same time. I Déplacement Figure 20: Basic course of forming forces. Dispiaoemerrt ( Figure 21: Optimum course of forming forces. 250 200 ê £ i> 150 if) O £L 100 c = =3 50 n> OL 0 - —- EFC" - / / ETC / ■ _ / / Clearance press region 1 S ■J 1 . 0 5 10 15 20 Too! Displacement (mm) Figure 22: A comparison of the resultant force course for a basic and optimized process. 25 Operation F is positioned in the centre of the press, whereas operations E and F are 250 mm left and right from the press centre. The loading for individual operation is shown in Figure 20. The biggest loading appears in the last operation G, where the procedure of simultaneous backward and forward extrusion is performed. The resultant force course is shown in Figure 22, the curve EFG. By the help of reconstruction of the beginning of the course of forming operations the forming force course and consequently the resultant force are changed [10]. A preform (Figure 18, operation E] is changed in a way that the beginning of the forming process E* (Figure 21] is delayed with reference to the course of E shown in diagram Figure 20. In this way, the force required for the process E* is always smaller than the force required for the process G*. Therefore, the course of the resultant force is always on the same side of the press centre (Figure 22, curve EFG*]. Consequently, the resulting torque has the same sign during the entire movement of the ram. In this way we avoid a double ram Figure 23: A relative displacement between the upper tool and press bolster. transition through the unstable area that is affected by clearance in the press guiding elements. Thus we increase the reliability of the process considering endurance of vital tool parts and improve the possibility to reach narrower tolerances of a product. This method of resultant force optimization prevents the large portion of the horizontal displacement of the press ram in the clearance region (Figure 23] and, therefore, increases the dimensional accuracy of the clutch barrel as well as the tool- and press-service life. Conclusion A numerical-experimental approach for optimization of forming parameters for closed-die process in an operation window was presented, which ensures reliable achievement of the desired accuracy of a product at the optimum tool life. Application of the adjustment of the process input parameters is shown on the example of making an inner race of homokinetic joint from mass production. The press stiffness is an important parameter in the formed piece-tool-press parameter system, which increases accuracy of products and unfavourably influences the tool life, so it has to be taken into consideration when planning technologies for precise cold forging. The majority of the processes for manufacture of cold-extruded parts are carried out by a multi-operation forming on the same machine. For such forming process the resultant of forces of a multi-stage process was optimized which consequently reduced the variations of elastic displacements and rotations between the ram and the press bolster. In this way, the reliability and accuracy of a multi-stage forming process was considerably increased. Acknowledgements The article was published in Conference Proceedings: New Developments in Forging Technology, Editor Prof. Dr. M. Liewald MBA, Institute for Metal Forming Technology, University of Stuttgart, Fellbach, Germany, 2011. [6] Breyfogle III, F. W. (2003): Implementing Six Sigma; Smarter solutions using statistical methods. New York: John-Wiley and Sons, Inc, 1187 p. [7] Yang, K., El-Haik, B. (2003): Design for Six Sigma; A road map for product development. New York: McGraw-Hill Education, 768 p. [8] Kuzman, K., Pepelnjak, T., Gantar, G., Erjavec, M., Sve-tak, D. (2007): Evaluation of the material and product development in the frame of the Polycentric Technological Centre-Subproject: Stability control of the cold forging system- Phase 2: Final Report. Faculty of Mechanical Engineering Ljubljana, Forming Laboratory, Ljubljana. [9] Krusic, V., Arentoft, M., Rodic, T. (2005): Impact of mechanical press on accuracy of products. In: Proceedings of the 8th ICTP, Verona, pp. 381-382. [10] Krusic, V., Arentoft, M., Masera, S., Pristovsek, A., Rodic, T. (2011): A commbined approach to determine workpiece-tool-press deflections and tool loads in multistage cold-forging. Journal of Materials Processing Technology, 211, pp. 35-42. References [1] Lange, K., Cser, L., Gaiger, M., Kals, J.A.G. (1992): Tool life and tool quality in bulk metal forming. CIRP Annals - Manufacturing Technology, 41:2, pp. 667-675. [2] Altan, T., Ngaile, G., Shen, G. (2005): Cold and Hot Forging: Fundamentals and Applications. ASM International, 341 p. [3] Lange, K. (1985): Handbook of Metal Forming. New York: McGraw-Hill Book Company. [4] Engel, U. (1995): Reliability analysis of cold forging tools. In: Proceedings of the 9th International Cold Forging Congress, Solihull, Standring, P. (ed.). FMJ International: England, Redhill, pp. 103-112. [5] Krusic, V., Masera, S., Pritovsek, A., Rodic, T. (2009): Adjustment of stochastic responses of typical cold forging systems. Journal of Materials Processing Technology, 209, pp. 4983-4993.