Simulations of Brush Contacts of Homopolar Motor/Generators

Time-dependent distributions of electromagnetic force and heating were studied.

December 1, 2007

Naval Research Laboratory, Washington, DC

A computational- simulation study of distributions of electric current and temperature in brushes and slip rings in two model homopolar- motor/generator configurations was performed in support of the development, by the U.S. Navy, of a superconducting homopolar motor/generator (SCHPMG) machine for ship propulsion. Electrical-contact performance (more specifically, brush/slip-ring contact performance) is a limiting factor in the performance of an SCHPMG machine. The present study and similar studies are needed to gain understanding of brush/slip-ring contact performance in order to enable optimal design of brushes for homopolar machines.

Half Symmetric Finite-Element Models of eight-brush stators with electromagnet coils and slip rings were used in the computational simulations.

At the interface between a brush and a slip ring, where transfer of electric current occurs, factors that affect wear include the spatial distributions of electric current, magnetic field, temperature, and electromagnetic forces as the brush slides at a given velocity on an imperfect rotor (slip-ring) surface. There is experimental evidence that wear is not only highly asymmetric between anode and cathode brushes but is also nonuniform within the contact area of an individual brush. The asymmetry and nonuniformity of wear limits the life and efficiency of a brush assembly, inasmuch as the full volume of brush fiber material is not fully utilized. Moreover, debris from worn brushes can give rise to internal short circuits and/or other adverse effects on the operation of a machine containing a brush assembly.

Each of the two models in the computational- simulation study was a half-symmetric finite-element model representing eight stator brushes, an electromagnet coil, a slip ring, and buses (see figure). The two models differed in the aspect ratios of the brushes (1:2 versus 1:3). The brush contact area was kept constant at 2 cm2, a transport current of 325 A was considered to be applied to each brush, and the current applied to the electromagnet coil was chosen to produce a magnetic flux density of 0.4 T at the brush locations. Coupled electromagnetic and thermal behaviors (including bulk ohmic heating) were included in the simulations. Local adiabaticity and a specific form of the temperature dependence of electrical conductivity were assumed. Time-dependent spatial distributions of electric current, temperature, and electromagnetic force were calculated. The conclusions drawn from the numerical results include the following:

Circumferential components of electric current at leading and trailing edges of brushes could interact with the magnetic field to produce forces tending to destabilize the brushes, but these forces could be reduced through appropriate design.
A brush having a higher aspect ratio is subject to smaller destabilizing forces.
Power loss per brush due to bulk ohmic heating is about 0.74 W for an aspect ratio of 1:2 vs. 0.72 watts for an aspect ratio of 1:3.
Assuming a contact pressure of 2 psi (≈14 kPa) and a tip speed of 12 m/s, frictional heating is about 16.5 W per brush.
The power loss per slip ring due to bulk ohmic heating is about 0.98 W for an aspect ratio of 1:2 vs. 0.9 W for an aspect ratio of 1:3.

This work was done by Kuo-Ta Hsieh of the University of Texas at Austin for the Naval Research Laboratory.

NRL-0018