Existing practical laboratory tests for comparative strength determination of commercial explosives consist of the lead block test, the ballistic mortar test, the bubble test etc. These tests are carried out in material with quite different properties compared to rock. The theoretical performance potentials of commercial explosives are traditionally described by weight-strength (WS). Some explosive producers use the Langefors formula for weight-strength calculation, where also the gas volume is considered. Neither the theoretical nor the practical methods for determination of explosive performance such as the above, give good enough prediction of fragmentation and heave in rock blasting. For these reasons a project was started for the purpose of finding a small scale test method which predicts how different explosive parameters affect fragmentation and heave. Small scale tests were carried out by using 50 cm cubical blocks, mainly of concrete, but a few also of rock. All blocks were blasted in a steel cylinder with a diameter of 1.2 meters. The velocity of detonation was monitored, and the velocity of fragments was derived from their time of arrival at the steel wall, which was registered by accelerometers. After blasting all fragments were screened. The charge diameter was 22 mm and the amount of explosive 78 grammes, resulting in a specific charge of 0.62 kg/m3. Explosive properties such as energy, gas, VOD and density were systematically changed.The results presented in this paper show how fragmentation and heave systematically change when the explosive properties like energy, gas, VOD and density are systematically changed. The results from the small scale tests are compared to full scale tests carried out in tunnel blasting, and during summer 1990 also in bench blasting.