Background: Multiaxial dynamic loadings occur in many industrial cases and multiaxial dynamic test development is thus a crucial issue. Objective: To meet this challenge, a biaxial compression Hopkinson bar set-up is designed. Methods: The set-up consists of a striker, an input bar, an internal output bar and a co-axial external output tube (surrounding the internal bar). The internal output bar measures the axial loading of the cross sample whereas the external output bar measures the transverse one via a mechanism. This mechanism uses two intermediate parts with inclined sliding surfaces. Results: Gauges on the bars enable for force measurements in the set-up, and the sample displacement field is obtained by digital image correlation. Simple compression tests on cuboid samples inserted between the input bar and the internal output bar give the sample material behavior. Then, to determine the friction at the mechanism sliding surfaces, identical samples are inserted between the input bar and the external output bar, and are compressed. Conclusions: Finally, the consistency of the measurements obtained during a biaxial compression test on a cross sample can be checked from the previously measured parameters and from numerical simulations.

Shape memory alloys (SMA) undergo a solid-solid phase transformation called martensitic transformation, involving a "high temperature" phase (austenite) and a "low temperature" phase (martensite). The stress-strain pseudo-elastic behavior of a nickel-titanium under equi-biaxial dynamic compression is measured thanks to a new home-made impact testing set-up using split Hopkinson bars. The use of thermal and optical cameras allows strain and heating sources fields to be identified. The stress field is estimated by the combination of the strain gauges information placed on bars, and a finite element analysis of the specimen. Experimental average stress-strain behavior and thermal emission are finally compared to the results of a finite difference axisymmetric model where the constitutive law is given by a fully coupled stochastic multi scale model.