A capacitor-run motor is often necessary where silent running, low starting current, high efficiency and power factor and overload capacity are required. When a capacitor motor is used in control or similar applications a knowledge of the variation in performance over a range of supply frequency is often important, and this has been investigated by previous workers using analyses of varying validity. For example, in early studies it was common to base performance calculations on the revolving field theory, when a general lack of agreement was found between calculated and test values. This occurred particularly in the starting wlnding under running conditions, and was attributed usually to saturation and to space harmonics of mmf. The crossfield theory provides a useful alternative to the revolving-field theory for analyses of single-phase motors, and when applied to the capacitor motor it is found that a much more accurate prediction of the machine performance is obtained. This thesis investigates the behaviour of a small singlephase capacitor-run induction motor with relation to variable frequency supplies and different capacitor values. The performance of the capacitor motor is calculated, using an equivalent circuit developed from basic cross-field theory considerations. However, before calculating the performance of the motor, the experimental determination of the parameters involved is necessary. From the equivalent circuit for no-load and locked rotor conditions, a set of voltage/current equations was obtained. Use of these equations requires a very complicated set of mathematical procedures to determine the required parameters, and no solution could be found. This difficulty was overcome by the extensive use of a digital computer, enabling the parameters to be obtained by the Newton-Raphson method of numerical iteration. Using these parameters, the predicted performance of the machine was compared with that obtained from an extensive experimental investigation. The agreement which was found to exist confirms both the validity of the theoretical analysis and the accuracy of the measurement from which the parameters were derived.