ORIGINAL ARTICLE Process parameter definition with respect to the behaviour of complex kinematic machine tools Sylvain Pateloup & Hélène Chanal & Emmanuel Duc Received: 9 April 2013 / Accepted: 31 May 2013 # Springer-Verlag London 2013 Abstract The definition of machining processes with re- spect to complex kinematic machine tool behaviour involves the control of machine accuracy and kinematic perfor- mances. The aim is to propose process settings and tool paths which guarantee the required machining quality while max- imizing productivity. This article presents an experimental protocol which enables the determination of machine tool structure behaviours which have an influence on machining quality. In parallel, an experimental analysis of the different kinds of settings which can improve machining quality is carried out. Two kinds of settings appear: the first class of settings improves machining quality or machining time, and the second class has an antagonistic influence on machining quality and machining time. Thus, the definition of the second class of settings arises from an optimisation between first-order defects, second-order defects and machining time. The developed method is illustrated on a parallel kinematic machine tool, the Tripteor X7. Note that this study is a first step towards controlling machine tool behaviour during machining. Keywords Parallel kinematic machine tool . Machining process . Experimental measurements . Mechanical behaviour 1 Introduction In machining, the geometrical shape of the manufactured part is generated from the feed movement of the tool with regard to the part. This movement is realised through the machine tool structure and the actuated joints. Thus, machine architecture and motor performances have a great influence on the ability to generate a movement with a programmed feed rate [1]. In the same way, the geometrical accuracy of tool pose and mechanical machine tool structure behaviour have a direct influence on machining quality. Thus, geometrical, kinematic, static and dynamic behaviour has an influence on the quality of the machined part and its productivity [2]. Different kinds of machine tool architectures can be pro- posed. Complementary to three-axis high speed machine tools, five-axis machine tools, Parallel Kinematic Machine tools (PKM) and anthropomorphic robots have appeared. The structure of each machine tool presents different behav- iour characteristics and thus different performances during the machining of the same workpiece [3, 4]. With regard to the machine tool structure and loads gen- erated by the cutting process and tool path, performance optimization, in terms of quality and productivity, requires the development of specific methods based on the prediction of the machine tool structure behaviour during machining [5, 6]. This is the framework of the presented research activity. In his work, Pritschow proposes a classification of PKM behaviours which generate tool pose defects (Fig. 1)[2]. Tool pose defects are related to several aspects of PKM structure behaviour, separated into two main classes: static and dynamic behaviour. Machining defects on the workpiece are generated by all the phenomena associated with the static and dynamic behaviour of the machine tool even if static behaviour generates higher machining errors than dynamic behaviour (Fig. 1). Moreover, in the case of parallel structures, the influence of each phenomenon on machining defects is not easily quantifiable and compensable with simple models be- cause of their complex anisotropic behaviours. S. Pateloup Département GMP, UMR CNRS 5295, Institut de Mécanique et d’Ingénierie, IUT Bordeaux 1, 15 Rue Naudet, 33170 Gradignan, France e-mail: sylvain.pateloup@iut.u-bordeaux1.fr H. Chanal (*) : E. Duc IFMA, UMR 6602, Institut Pascal, Clermont Université, BP 10448, 63000 Clermont-Ferrand, France e-mail: helene.chanal@ifma.fr E. Duc e-mail: emmanuel.duc@ifma.fr Int J Adv Manuf Technol DOI 10.1007/s00170-013-5118-3