Studies on a thermal performance of totally enclosed fan cooled induction motors

Abstract

Out the three essential aspects of electric motor design-namely electromagnetic, mechanical, and thermal-the technology for thermal design happens to be the least developed, despite the fact that thermal performance determines the final rating of a motor. The present work is an attempt to improve the state of the art in induction motors with a particular type of enclosure (totally enclosed fan cooled type) that is very commonly used for smaller ratings. A mathematical model based on a three region boundary value problem has been formulated. Suitable boundary conditions have been used. A closed form solution has been obtained for the two dimensional equation to compute the temperature rise of the three regions-the rotor, the stator windings, and the body of the motor-using their physical dimensions and electromagnetic design data as input. Similarly, the cooling equation has also been solved. Somewhat simplified mathematical procedures have been developed for solving certain application problems such as duty cycling and ratings at different speeds. Extensive experimental work has been conducted to verify the temperature rises predicted for continuous running, duty cycling, and variable speed applications. A method has been devised for fairly accurate measurement of rotor temperature while in motion. Additional experimental work was undertaken to calculate one of the most essential parameters, namely the heat transfer coefficient (hrh_rhr) in the radial direction. This involved measurement of the air velocity discharged by the fan into trapezoidal channels (formed by the fins) at various sections along the length of the motor. These measurements were then used to derive empirical relations connecting air velocity, channel dimensions, and hrh_rhr, which varies along the axial length. Other parameters were obtained empirically or from published literature. The agreement between the measured and computed temperatures is found to be good. It is believed that the program developed will enable designers to use only design data to reasonably accurately predict the thermal performance of the motor. With further refinement of this program, the need to construct prototypes to determine thermal performance may be avoided. The present study involves one of the more commonly used induction motor types; however, the method employed is general enough to be extended to other types as well.