[en] Owing to small and inexpensive specimens the Charpy impact test is widely used in quality control and alloy development. Limitations in power reactor survellance capsules it is also widely used for safety analysis purposes. Instrumenting the tup and computerizing data acquisition, makes dynamic fracture mechanics data measurement possible and convenient. However, the dynamic effects (inertia forces, specimen oscillations) in the impact test cause inaccuracies in the recorded load-time diagram and hence diminish the reliability of the calculated dynamic fracture mechanics parameters. To decrease inaccuracies a new pendulum type of instrumented impact test apparatus has been developed and constructed in the Metals Laboratory of the Technical Research Centre of Finland. This tester is based on a new principle involving inverted test geometry. The purpose of the geometry inversion is to reduce inertia load and specimen oscillation effects. Further, the new impact tester has some other novel features: e.g. the available initia impact energy is about double compared to the conventional standard (300 J) impact tester allowing the use of larger (10 x 20 x 110 mm) bend specimens than normal Charpy specimens. Also, the rotation asix in the three point bending is nearly stationary making COD-measurements possible. An experimental test series is described in which the inertia effects and specimen oscillations are compared in the conventional and new impact tester utilizing Charpy V-notch specimens. Comparison of the two test geometries is also made with the aid of an analytical model using finite element method (FEM) analysis. (author)