The maximum explosion overpressure and the maximum rate of pressure rise, which characterize the dust explosion severity, are commonly measured in apparatuses and under specific conditions defined by international standards. However, those standards conditions, designed for micropowders, may not be fully adapted to nanoparticles. Investigations were conducted on different nanopowders (nanocellulose, carbon black, aluminum) to illustrate their specific behaviors and highlight the potential inadequacy of the standards. The influence of the sample preparation was explored. Various testing procedures were compared, focusing on the dust cloud turbulence and homogeneity. Dust dispersion experiments evidenced the importance of the characterization of the dust cloud after dispersion, due to the fragmentation of agglomerates, using metrics relevant with nanoparticles reactivity (e.g. surface diameter instead of volume diameter). Moreover, the overdriving phenomenon (when the experimental results become dependent of the ignition energy), already identified for micropowders, can be exacerbated for nanoparticles due to their low minimum ignition energy and to the high energy used under standard conditions. It was evidenced that for highly sensitive nanopowders, pre-ignition phenomenon can occur. Finally, during severe explosions and due to a too long opening delay of the fast acting valve', the flame can go back to the dust container. (C) 2020 Institution of Chemical Engineers. Published by Elsevier B.V. All rights reserved.