M.Ing. / During mineral exploration a whole lot of different exploration techniques can be applied. In this thesis the instrumentation used for electromagnetic surveying, which is only one of the prospecting techniques, is inspected. In particular the thesis will be about the transmitter of the single transmitter wide band receiver airborne electromagnetic prospecting system. A new power electronic topology is proposed in order to improve this transmitter. Before this is done a reference is first established. This is done in the form of the exponential transition inverter, which is the inverter that is currently applied as transmitter in South-Africa. Another reason for looking at this circuit topology is because it is the simplest way to realize the transmitter. During the analysis of this inverter topology all the proporties of the transmitter as a whole is also investigated. The main component of the transmitter, namely the transmitter coil is also discussed. This is followed by a summary of the properties of the exponential transition inverter in which the main shortcomings of this transmitter is discussed. The greatest of which is surely that pliable frequency spectra cannot be obtained. Subsequently resonance is viewed as a solution to the shortcomings. The main contribution of this is that the pliability of the frequency spectrum can be increased, because the capasitor in this circuit can easily be changed. It is however shown that a purely resonant circuit on it's own, also is not the solution. This leads to the improved inverter topology for the transmitter, namely the Kwasi resonant inverter. It is shown in detail that the Kwasi resonant inverter is a much beter option as transmitter then the exponential transient inverter. Since this circuit also has a capacitor in it, it is shown that pliable frequency spectra can be obtained. It is futher shown how the pliability of the frequency spectrum can be improved even more by deviating from conventional methods of switch control. Mathematical analysis, as well as an experimental setup, of the different inverter topologies, are used to illustrate all of the above mentioned. As a conclusion the only drawback of the Kwasi resonant inverter, namely high voltage peaks, is addressed. A solution to this problem is proposed in the form of two different voltage clamps. The best one of the two is identified and also implemented.