Forces on turbogenerator end windings

Abstract

The complete distribution of the forces acting over the whole of a typical, 2-pole turbogenerator end winding of the involute type is studied and presented in a general form. Further, the way in which these results can be applied to calculate the forces for any other (2-pole) winding of involute type, without restriction as to size, is described. The basic results are presented in such a form that they can be used to determine the point-by-point distribution of forces for any condition of current loading. Additional results give the total forces experienced by individual coils of the winding. The method of calculation, which is based upon expressions for the force between any two short, straight, current-carrying conductors, is described. It is very simple and flexible and can be used to determine the distribution of force in any machine end winding or any other complex arrangement of conductors. The effects that changes in winding design have on the force distribution are studied. It is shown that the cone angle, the spacing between involutes and the number of stator coils (for fixed electrical loading) all have a considerable influence on the forces. In the cases of the involute spacing and the coil number, simple empirical equations describing their effects are developed. With regard to accuracy of representation, it is shown that the effects of air-gap magnetisation are unimportant; that errors due to the neglect of coil cross-section are of fixed sign and about 5?10% of the true force values; and that errors due to neglect of the machine end surface are generally ?10%. The influences of the stator casing and rotor shaft and retaining ring are briefly considered. The forces due to rotor winding currents are examined, and it is found that, whilst the peripheral and axial components are generally unimportant, the radial component can be as large as 50% of that due to stator currents, and is directed outwards. Brief consideration is given to certain eddy-current effects, and, in conclusion, it is indicated that, owing to these effects, forces computed using values of the subtransient reactance, or indeed actual currents flowing at the instant of short-circuit, will be too high.