Experience with turn insulation failures in large 13.2 kV synchronous motors

Experience of dealing with a very high rate of failures in a population of 21000 HP motors is recounted. Special testing done to identify the failure mechanism and the conclusions that were drawn are described. Various recommendations to prevent premature winding failures proved ineffective. It is shown how an engineering analysis indicated that the accepted view of the failure mechanism was untenable. A theory of partial discharge in the end turn of the stator winding that is compatible with all known facts is proposed.<
>