FöRFATTARE
Kajberg, Jörgen

DATUM
2003-11-05

INSTITUTION
Tillämpad fysik, maskin- och materialteknik / Hållfasthetslära

SAMMANFATTNING
In many technical processes, material is deformed under conditions involving large deformations (strains) and/or high strain rates. Examples of such processes are collisions, impact, penetration, metal forming, powder compaction and crack propagation. For description of these kinds of situations a variety of constitutive models, based on both physical foundations and empirical considerations, is available. Common for all models is that they contain material parameters, which have to be estimated by utilising experimental methods.

For material characterisation under quasi-static conditions standardised tension tests of uniaxially loaded specimens are commonly used. With these tests stress-strain relations are obtained up to moderate strain values, whereupon the onset of strain localisation, so-called necking, restricts their validities. Correction methods have been developed to compensate for the onset of necking (e.g. Bridgman's correction method for round bars).

The Taylor impact test and the split Hopkinson bar arrangement are frequently used methods for the investigation of incompressible (volume conserving) materials in the high strain rate regime. Typically, the specimens are short and stubby cylinders, which ideally facilitate a homogeneous state of loading necessary for a simple interpretation of the experimental results.

In this thesis a methodology is suggested for characterisation of materials subjected to large deformations and high strain rates, where neither homogeneously loaded specimens nor incompressible behaviour are necessary. Experimental methods similar to standardised tension tests and split Hopkinson bar arrangement are complemented with an optical method, digital speckle photography (DSP), for in-plane point-wise displacement and strain measurements. By using a common digital camera in the former tests and a high-speed camera with a CCD-unit (Charged Coupled Device) in the latter tests digitised images are obtained for the method of DSP.

An inverse method is used to estimate the material parameters in constitutive models. Three-dimensional numerical simulations of the specimens are performed by the finite element method (FEM). By adjusting the parameters to give a best fit between experimental and numerical results (displacements and strains) in least-square sense optimal values are obtained.

In the quasi-static tension tests true strain values up to 0.8 were obtained for a hot-rolled steel. The mild steel specimens in the high strain rate tests were subjected to strain rates of magnitudes 10^2-10^3 1/s.

ISSN 1402-1544 / ISRN LTU-DT--03/35--SE / NR 2003:35