Load Support System Analysis-High Speed Input Pinion Configuration

Abstract

Structural assemblies which transmit power, helicopter gearboxes for example, require accurate evaluation of load and displacement distributions to ensure system design optimization. Particular performance parameters which depend on detailed understanding of the distributions are operating life and rigidity. The coupling complexity of system components; gears, shafts, bearings and the housing itself, prevents meaningful evaluation of performance from the study of isolated components. The entire load support system has to be examined in its interacting entirety. In this paper a theoretical study has been made to determine the performance of a load support system, consisting of a shaft and two taper roller bearings for the high-speed input pinion of an advanced helicopter transmission. SHABERTH, a computer program designed to perform thermo-mechanical analyses of arbitrarily configured rolling element bearing-shaft systems, was employed to analyze, both, a straddle arrangement where the spiral bevel pinion gear is located axially between the two bearings, and a cantilevered arrangement where the pinion is outboard of the bearings. The effects of preload, shaft wall thickness, bearing spacing and misalignment of bearing races were examined. Their influence on load distributions and bearing rating lives including simulation of bearing and shaft elastic deformations were evaluated for both support geometries. Additional effort was expended to detail the performance of the pinion designs over a range of shaft rotational speeds. Lubrication and friction effects were included. Particular attention was directed to local as well as global heat generation rates (HGR) to provide design information for proper lubrication of the bearings. The results provide guidance for improved design of transmissions and the load vector control within them. In particular, the relative merits of the straddle versus the cantilever design have been exposed so that their individual characteristics may be exploited to increase system survivability.