A magnetostrictive brake has been designed as a more energy-efficient alternative to a magnetic fail-safe brake in a robot. (In the specific application, “fail-safe” signifies that the brake is normally engaged; that is, power must be supplied to allow free rotation.) The magnetic fail-safe brake must be supplied with about 8 W of electric power to initiate and maintain disengagement. In contrast, the magnetostrictive brake, which would have about the same dimensions and the same torque rating as those of the magnetic fail-safe brake, would demand only about 2 W of power for disengagement.

In the Magnetostrictive Brake, a large braking force would be generated by the permanent magnet. When power was supplied to electromagnet coils (not shown) surrounding the magnetostrictive cylinders in the solenoid assemblies, the resulting magnetostrictive strain would be converted to a force and displacement that would oppose the braking force.

The brake (see figure) would include a stationary base plate and a hub mounted on the base plate. Two solenoid assemblies would be mounted in diametrically opposed recesses in the hub. The cores of the solenoids would be made of the magnetostrictive alloy Terfenol-D or equivalent. The rotating part of the brake would be a ring-and-spring-disk subassembly. By means of leaf springs not shown in the figure, this subassembly would be coupled with the shaft that the brake is meant to restrain.

With no power supplied to the solenoids, a permanent magnet would pull axially on a stepped disk and on a shelf in the hub, causing the ring to be squeezed axially between the stepped disk and the hub. The friction associated with this axial squeeze would effect the braking action.

Supplying electric power to the solenoids would cause the magnetostrictive cylinders to push radially inward against a set of wedges that would be in axial contact with the stepped disk. The wedges would convert the radial magnetostrictive strain to a multiplied axial displacement of the stepped disk. This axial displacement would be just large enough to lift the stepped disk, against the permanent magnetic force, out of contact with the ring. The ring would then be free to turn because it would no longer be squeezed axially between the stepped disk and the hub.

This work was done by Myron A. Diftler and Aaron Hulse of Lockheed Martin Corp. for Johnson Space Center. MSC-23629-1