Abstract :
Part 1 discusses the effects on the application and short-time rating of traction motors of their limited thermal capacity; a method is developed by which these effects can be considered, more correctly and conveniently than hitherto, in motor application studies. A survey of motor duties, heating and rating leads to the conclusion that equations containing two exponential terms are the most appropriate for representing motor temperature/time curves. This is verified by analysis of several published curves and of experimental results obtained on a small series motor. The development and use of a resistance-capacitance analogue based on these equations to simulate to a smaller time scale the armature slot-copper heating and armature iron heating of an actual motor on its normal run are then described. This analogue gives results differing by only about ?2% from those calculated from the same equations, whilst automatic recordings with errors within ?4% can also be obtained. It is concluded that motor application studies will be more reliable if they are made by using equations containing two exponentials, and that such studies will be facilitated by using analogues of the type described to solve the equations in continuous form. In Part 2 a method is described by which allowance can be made for the effect of the temperature rise of ventilating air upon the temperature/time equation for the armature copper. The method is applied to an equation containing two exponential terms. Consideration of the temperature/time relation for a ventilated body shows that this allowance introduces a factor which modifies only the final body-to-air heat-transfer coefficient and does not affect the form of the equation. The method is verified experimentally, and a practical application, which involves an approximation, is given. The approximation is shown to be valid by a dimensionless study of temperature/time equations containing two exponentials. Time can be saved by the use of dimen- sionless coefficients by which the exact equation may be derived from an approximate one more easily obtained.
