High strain rate tensile experiments

High strain rate tensile experiments

The stress-strain response of engineering materials at high strain rates is typically determined based on uniaxial compression experiments on Split Hopkinson Pressure Bar (SHPB) systems (Kolsky, 1949, [96]). In the case of sheet materials, it is difficult to perform a reliable dynamic compression experiment due to limitations associated with buckling. Different techniques have been developed in the past for the high strain rate tensile testing of materials. The key challenges lie in the generation of the tensile loading pulse and the attachment of the specimen to the Hopkinson bars. The specimen gage section is always subject to a tensile pulse, but it is worth differentiating between techniques where the specimen boundaries are subject to tensile loading and those where a compressive loading is applied to the specimen. In the case where the specimen is subject to compressive loading, the loading pulse is inverted within the specimen. When a tensile loading is applied to the specimen boundaries, one may differentiate further between techniques where a tensile loading pulse is generated in the input bar and those where a compressive loading pulse is generated. Throughout our discussion of SHB systems, we will always assume a left-to-right positioning of the specimen, the input bar and the output bar.

Harding et al. (1960, [73]) proposed a technique where the specimen boundaries are subject to tensile loading while a compressive loading pulse is generated in the input bar. They placed a round inertia bar (output bar) inside a hollow tube (input bar). The right specimen shoulder is then connected to the right end of the tube, while the left shoulder is connected to the right end of the inertia bar; both connections are established through mechanical joints (Fig. 3-1a). A rightward travelling compressive loading pulse is generated in the tube which loads the specimen under tension. Only the elastic waves propagating in the output bar are measured in this experiment, therefore requiring two tests with the same loading pulse (one with a specimen, one with the output bar directly connected to the mechanical joint) to measure the force and velocities applied to the specimen boundaries. Nicholas (1981, [135]) and Ellwood et al. (1982, [54]) developed a tensile experiment based on a SHPB system, in which an axisymmetric tension specimen is attached to the bar ends and surrounded by a collar initially in contact with the right end of the input bar and the left end of the output bar (Fig. 3-1b). The generated compressive loading pulse is transmitted to the output bar through the collar, without inducing any plastic deformation in the specimen. The loading pulse is then reflected into a tensile pulse at the free right output bar end which then loads the specimen under tension.

As an alternative to inverting the compressive loading pulse within the bars prior to loading the specimen, different techniques have been proposed where the loading pulse is inverted at the specimen level. Lindholm and Yeakley (1968, [112]) developed a hat-shaped specimen (Fig. 3-1c) which is put over the right end of the input bar (input bar diameter equals the hat’s inner diameter), while the hat’s rim is connected to a tubular output bar that partially overlaps with the input bar (inner output bar diameter equals the hat’s outer diameter). Four gage sections are machined in the hat’s cylindrical side wall to guarantee a uniaxial stress state at the specimen level. Mohr and Gary (2007, [123]) proposed an M-shaped specimen to transform a compressive loading at the specimen boundaries into tensile loading of the two gage sections (Fig. 3-1d). The M-shaped specimen is used for transverse plane strain tension, which allows shorter gage sections and therefore higher strain rates, and has been validated for strain rates higher than 4,000/s. The key advantage of their technique is that there is no need to attach the specimen to the bars.

For sheet materials, Tanimura and Kuriu (1994, [171]) proposed a non-coaxial SHPB system (Fig. 3-1e). The input and output bars are laterally shifted while a pin joint is used to attach the sheet specimen to the sides of the bars. Mouro et al. (2000, [129]) used a co-axial SHPB system to perform high strain rate tension experiments on sheet metal. They modified the shape of the bar ends such that a two-dimensional multi-gage section hat-shaped sheet specimen can be inserted between the input and output bars of a SHPB system. The specimen must be bent into the hat geometry prior to testing which introduces additional geometric inaccuracies. Haugou at al. (2006, [74]) adopted a configuration where four specimens are bonded to the external part of two threaded sleeves initially in contact and screwed to the input and output bars (Fig. 3-1f), thus permitting to use optical strain measurement methods.