THERMAL DC TEST AND ANALYSIS OF A STATOR MADE WITH RESIN TRICKLE IMPREGNATION

Abstract

As the demand for higher power density and higher efficiency for electric motor grows, thermal behaviour is becoming more and more critical to the performance. Previous studies using motorette tests on single wound teeth of segmented stators demonstrated varying impregnation goodness. It is the result of a stochastic resin trickle impregnation process. This paper presents results of a DC test on a complete stator made with the same settings of the resin trickle impregnation process. The stator was instrumented with a number of temperature sensors at various locations. A large variation in winding temperature from tooth to tooth is shown. It correlates to the variation in the impregnation goodness of the segments as expected. However, the calibrated lumped parameter thermal network model from the motorette tests failed to predict accurately the transient behaviour of the stator during the DC tests. An improved thermal model is presented accounting for the resin heat capacity and the actual resin weight in the stator during the calibration, leading to more accurate transient simulations.