ABSTRACT--The split Hopkinson pressure bar (SHPB) technique is analyzed during the initial stages of loading by means of axisymmetric finite element simulations of dynamic compression tests. Limiting strains as functions of the test parameters such as the specimen diameter d and height h were found to ensure a one-dimensional stress state and axial stress homogeneity in specimens of elastic-perfectly plastic material. The one-dimensional stress state is necessary and sufficient for accurate test results for flat specimens (h/d <_ 0.5) and nonflat specimens, respectively, with diameters up to half of the bar diameter. Only very small values of the Coulomb friction constraint (/z ,~ 0.0l) seem to be acceptable. The significance of the determined limiting conditions to the more practical case of a rate dependent material is investigated using an elastic-viscoplastic material for the specimen. The stress and strain rate reconstructed from the calculated bar signals (according to the $HPB analysis) are compared with stresses and strain rates averaged over the cross section of the specimen. Well-known inertia corrections improve the results of the SHPB procedure, but errors remain for small strains and highly time dependent strain rates.KEY WORDS--Split Hopkinson pressure bar, high strain rate testing, dispersion correction, stress distribution, triaxialityThe split Hopkinson pressure bar (SHPB) technique is a well-established experimental method for high strain rate testing of materials. 1-3 Measurements of stress waves in two e]lastic bars on either side of a specimen are used to quantify the mechanical behavior of this specimen. The SHPB procedure is founded on a one-dimensional analysis of the test configuration. Hence, for valid results, the pressure bars and the specimen must be under a state of one-dimensional stress. Additionally, the stress and strain should be uniform throughout the specimen to ensure similar material behavior af: all positions of the sample. Both radial and axial inertia and frictional constraint at the specimen-pressure bar interfaces can violate these conditions. There is a long discussion about the optimum specimen geometry to reduce these effects. [4][5][6] It is well known that during the initial stages of loading (i.e., for small strains and highly time dependent strain rates), the U. Zencker is research associate Bundesanstalt fiir Materialforschung und -priifang, 12200Berlin, Germany. R. Clos is a Professor, Otto-von-Guericke Universittit, PO Box 4120, 39016 Magdeburg, Germany. Original manuscript submitted: January 23, 1998. Final manuscript received: June 22, 1999 assumptions for a conventional SHPB analysis are not violated. In this study, we determine whether and when the results of such an analysis are reliable. On this basis, the pros and cons of different specimen dimensions are considered.An investigation of these problems in real experiments is difficult because measurements are possible only at the specimen and bar surfaces. Therefore, two-dimensional axisymmetric nume...

No abstract

No abstract

No abstract

No abstract