This paper presents a comparison of predicted and measured axial stiffness of an oil-lubricated thrust collar (TC) for use in an integrally geared compressor (IGC). The test rig used to measure the stiffness closely resembles a TC that would be used in a production IGC. Separate motors spin two shafts at predetermined spin speeds, and a pneumatic loader applies an axial load to the pinion shaft. The load is transmitted from the pinion shaft through the TC and onto the bull gear shaft. The relative deflection between the faces of the TC is measured as the axial load is increased. Tests were conducted at pinion spin speeds of 5, 7.5, and 10 krpm, and the axial load was varied from 75 to 2,600 N (1 to 32 bar specific load). A custom curve fit describes the load-deflection plot. The slope of the equation is the stiffness coefficient, k_z. The measured k_z values increase versus axial load, and decrease versus spin speed. Measured and predicted values agree in trend and magnitude. Predicted k_z values slightly overestimate the measured results. The power loss from the lubricated area is measured from VFD power outputs. Power loss increases with spin speed and the measured power loss tends to be twice as high as predictions. These results are useful in validating the predictions made by Cable et. al.[l] The results are also useful to IGC original equipment manufacturers (OEMs) or end-users who desire to model their machines.
展开▼
