Concepts for hot sheet metal forming of titanium alloys

Sammanfattning: To increase the competitiveness of the Swedish aero engine industry alternative manufacturing methods for load carrying aero engine structures are desired, in order to reduce product cost and enable weight reduction and thereby fuel consumption. Traditionally, these structures mainly consist of large-scaled single castings of e.g. titanium- and nickel based super alloys. By fabrication, the structures are instead built from sheet metal parts, small ingots and simple forgings which are welded together and heat treated. The alternative approach implies the need for time and cost efficient evaluations of candidate manufacturing techniques, early in the product development process. One challenge in producing complete structures within shape tolerance lies in accurate predictions of springback and compensation for shape deviation which occurs in the different processes of the manufacturing chain. Finite Element (FE) simulations are used extensively in e.g. the sheet metal stamping industry where the technology has contributed to a better understanding of chosen forming processes and where the prediction capabilities has significantly reduced the time consuming, inexact and costly die tryouts. However, the reliability of the numerical simulations depends not only on the models and methods used but also on the accuracy and applicability of the input data. The material model and related property data must be consistent with the conditions of the material in the process of interest. In addition, creating as little deviance as possible between the FE model and the experimental setup is a prerequisite for the correlation between predicted and measured values. Naturally, difficulties regarding e.g. modelling and estimations of friction arise, among others. The objective of this thesis is to suggest possible hot and cold forming concepts based on FE analyses for the production of sheet metal prototype components in the titanium alloys Ti-6Al-4V and Ti-6242 together with the nickel based super alloy Inconel 718, respectively. The research activities are focused on material characterisation, evaluation of suitable constitutive models and its calibration, virtual tool design and manufacturing of sheet metal forming tools together with production of prototype components. The aim is to perform a direct-hit research and development work in which the lead time is short and the need for the manual die tryout can be kept at a minimum. The forming tests functions as validation tests in which predicted responses of global forming force, draw-in, temperature, strain localisation and shape deviation are correlated with predicted responses. Different yield criteria which include the anisotropy and strength differential (asymmetry in yielding between tension and compression) of the titanium alloy Ti-6Al-4V are compared with an isotropic assumption. Special emphasises are made to models and methods suitable for analyses in the medium temperature range, for evaluations early in the product development process. In paper A, compression tests on Ti-6Al-4V were performed at different temperatures in order to study the mechanical behaviour and create experimental reference data for identification of material model parameters of constitutive equations. Inverse modelling was used as a method for the parameter identification, in which three different equations were studied. At a temperature of 500°C, none of the studied constitutive equations were found able to satisfactory describe the flow behaviour. However, the method was found suitable for the purpose of identifying model parameters. Later on, the physically based constitutive equation developed by Nemat-Nasser et al. (2001) was found able to describe the flow behaviour of Ti-6242 [1]. In the work by Nemat-Nasser et al. the model has been shown to be able to describe the flow behaviour of Ti-6Al-4V at different temperatures and strain rates accurately. The equation was applied in FE analyses of a hot forming test, a U-bend test, of Ti-6242. The experimental study of Ti-6242, including U-bend tests,at different thermo-mechanical conditions performed in Paper B, revealed that the formability is increased and that the springback can be decreased with increasing temperature. However, it was also found that an increased temperature alone does not necessary imply a reduced shape deviation. In paper C, a short lead time methodology for the design, compensation and manufacturing of deep drawing tools in the nickel based alloy Inconel 718 is suggested. Rather than stating a new methodology, the work contributes to the idea that it is possible to perform a virtual direct-hit development work for the production of five different double-curved components within tolerance at an extremely short lead time. Compensation for the predicted shape deviation was performed in which the tool surfaces were over compensated by means of FE analyses. In paper D and F, the short lead time methodology was applied for the development of hot forming concepts to produce two different Ti-6Al-4V sheet metal components. The material characterisation, presented in paper E, provides with experimental reference data for calibration of three different yield criteria. Predicted responses such as punch force, draw-in and shape deviation show promising agreement with experimental observations when applying anisotropic yield formulations. The shape of the yield surface was found important for the prediction of shape deviation and the occurrence of strain localisation. Some issues of the FE-model suggest areas for further development. An interesting extension to the present work would be to include models for phase transformation and creep or stress relaxation and include the effect of strain rate for sheet metal forming in the higher temperature range. Further on, it is of interest to extent anisotropic yield criteria to function in coupled thermo-mechanical analyses and to include orthotropic elasticity. This, in order to increase the possibility to perform detailed studies of the temperature as an important process parameter for the prediction of shape deviation and studies of strain localisation limits.