Analysis of harmonic parameters and detection of foreign frequencies in diagnostic signals, which are most often interpreted as fault results, may be problematic because of the spectral leakage effect. When the signal contains only the fundamental frequency and harmonics, it is possible to adjust its spectral resolution to eliminate any distortions for regular frequencies. The paper discusses the influence of resampling distortions on the quality of spectral resolution optimization in diagnostic signals, recorded digitally for objects in a steady state. The method effectiveness is measured with the use of a synthetic signal generated from an analog prototype whose parameters are known. In order to achieve low values of harmonic amplitude errors in the diagnostic signal, a high quality resampling algorithm should be used, therefore the analysis of distortions generated by four popular reasampling methods is performed. Errors are measured for test signals containing different spectral structures. Finally, the results of the test of the analyzed method in practical applications are presented.