This work presents the results of a study on the sensitivity of real power as a function of harmonic magnitudes and angles. These transducers are ones that are typical of those used, in combination with metering instruments, to measure the performance of variable-speed drives. Frequency response tests were performed on two different signal transducers (voltage and current) up to the 50th harmonic in accordance with Institute of Electrical and Electronics Engineers (IEEE) Standard 519-1992. Results from these tests gave maximum magnitude percentage errors of -2.2% (at 15th harmonic) and maximum phase shifts of +3.3 (at 50th harmonic) for voltage (at 120 V) and maximum percentage error of -1.9% (at 2d harmonic) and maximum phase shift of -4.3 for current (at 5 A). Using the percentage errors at 120 V and 5 A throughout the 60 Hz through 3000 Hz range, in harmonic magnitudes and harmonic angles exhibited by the transducers, mathematical analysis was done on two sets of distorted signals and a sensitivity analysis was done to determine what would be the impact of the actual transducers' errors on the final calculation of real power. Variations of -1.48% and -1.60% were obtained. Then, assuming a maximum 5% change in current harmonic magnitudes and angles, a maximum real power percent variation of 0.02% and 2.47% were obtained for the same two sets of field-recorded voltage and current waveforms sets.