Multi-electrode geoelectrical imaging has become very popular and is used for many different purposes. For some of these, the inclusion of IP data would be desirable as it would allow the interpreter to distinguish between, e.g. sand formations with saltwater infiltration and clay formations or help delineate landfills. However, present-day IP measuring techniques require the use of nonpolarisable potential electrodes and special wire layout and are thus cumbersome and expensive. In this paper, we suggest making IP measurements with multi-electrode cables and just one set of steel electrodes. The polarisation potentials on the potential electrodes are corrected for by subtracting the polarisation potential measured when no primary current and no IP signal are present. Test measurements indicate that the polarisation potentials vary slowly and that the correction procedure is feasible. At two sites in southern Sweden, we have compared measurements with only stainless steel electrodes and measurements with both stainless steel and Pb -PbCl nonpolarisable electrodes using one or two sets of multicore cables, respectively. Almost no difference between the two data sets was observed. At one site, the charge-up effect on the potential electrodes was not important, while at the other site, the correction procedure was crucial. Though only two sites have been studied so far, it seems that time-domain IP imaging measurements can be taken with only steel electrodes and ordinary multicore cables. Coupling in the multicore cables has not presented any problems at the investigated sites where grounding resistances were moderate, making the coupling effect small. High grounding resistance sites have not yet been investigated. D 2002 Published by Elsevier Science B.V.