Application of Tube Hydroforming in Square Cross-Section Die for Inverse Identification Method Validation

Abstract

Listen

Translate article

Tube hydroforming processes are an excellent way for manufacturing reduced weight parts with complex shapes in widespread fields. Accurate numerical simulation of tube hydroforming process is particularly based on precise material parameters deduced from experimental tests. The free bulge test is widely employed for the parameter identification of tubular material behavior models by means of analytical [1] and numerical methods [2]. In this context, an inverse identification methodology using free bulge tests was developed. These tests were carried out by means of a new home-designed and manufactured bulge forming machine. The objective of this work is the validation of the inverse identification method using tube hydroforming in square cross-section die. The analysis of this particular hydroforming process with respect to material parameters is performed. For this purpose, circular section tubes made of low carbon steel S235 and aluminum alloy AA6063-O are hydroformed against square-cross sectional die using our bulge forming machine. Afterwards, FE model is constructed to simulate square-sectional hydroformed parts. The influence of some parameters, such as strain hardening exponent, anisotropy parameter and friction coefficient, on numerical square cross-sectional hydroformed part thickness is analyzed. It permits to assess the sensitivity of the thickness relative to used material parameters in the FE model. In order to validate the inverse identification procedure for both materials, experimental thicknesses along the profile of cross-sectional hydroformed parts are measured and compared with the corresponding numerical thicknesses predicted by FE model. It is proven after analyzing the obtained results that the chosen response, i.e. thickness distribution along the profile of the tube hydroforming against the square cross-section die, used for the validation is sensitive to the identified material properties. Particularly, it is demonstrated for low carbon steel S235 that numerical thickness is in good agreement with experimental data. However, for aluminum alloy AA6063-O, a discrepancy between experimental and predicted thicknesses is noticed. Anyway, it is demonstrated that inverse identification approach leads to sufficiently accurate parameters used for numerical tube hydroforming simulations. Furthermore, it seems that Hill48’s yield criterion is more suitable for describing steels plastic behavior than aluminum alloys for tube hydroforming processes. Concerning aluminum alloy, certainly the choice of appropriate yield criterion is of paramount importance on the prediction of tubular plastic behavior in tube hydroforming. Consequently, it is shown that the use of simple tube hydroforming in square-section die is suitable for the validation of FE model which is identified by inverse method using free bulge test.