The application of intense pulsed electric fields (PEF) in foods is intended to be a non-thermal method to inactivate microorganisms. However, it is well known that an increase in temperature is present in this process due to ohmic heating, where the pulsed electric field energy input is transformed into heat. The aim of this study was to investigate the computer modeled temperature increase in the outflow for different flow-through PEF treatment chamber designs. Given equal experimental conditions, the temperature increase is indicative of the PEF dose, and a more uniform temperature profile is thus indicative of a more homogeneous PEF treatment. The radial distribution of the temperature increase was simulated in computer models of four different chambers. The temperature increase was found to be more homogeneous in the treatment chambers making use of a decrease in the insulator diameter, i.e. a design letting the insulators and electrodes intersect at angles close to 90°. The maximum temperature increase was found close to the wall, where the flow velocity is low. Cooling of the electrodes and electric insulators is recommended to avoid too high a temperature increase. The minimum temperature increase found was 29% of the calculated average in the worst case studied here. The minimum PEF dose to which the food was subjected would thus is less than the intended dose, since the food clearly was not subjected to the intended electric field strength during the intended exposure time. This is an important result in terms of food safety in the sense that a minimum PEF treatment should be guaranteed. The microbiological inactivation was experimentally evaluated using two of the treatment chamber designs. The result is consistent with the simulations and shows a small increase in inactivation and less needed energy input giving less average temperature increase for the chamber implementing a contraction of the diameter of the insulating spacer. © 2002 Elsevier Science Ltd. All rights reserved.