In this paper the results of experimental investigations of the effects of radial taper on mechanical face seals are presented and compared to theory. The previously published theory considers the effects of thermal taper caused by a temperature gradient in the seal rings; mixed friction in the case where load support is shared between hydrostatic support and partial contact of the seal faces; surface roughness, which affects both load sharing and leakage; and wear, which alters the radial profile. Fifteen tests were run using a 100 mm diameter carbon versus tungsten carbide seal at 1800 rpm and 3.45 MPa in water. Test duration was up to 100 hr. Varying amounts of radial taper were used. Tests were run at balance ratios of 1.00 and 0.75. Initial and final surface profiles were recorded. Seal torque, leakage, and face temperatures were recorded as functions of time. Results show that theory predicts initial torque and leakage as functions of initial taper quite well, given knowledge of seal surface characteristics. Predicted equilibrium thermal taper as a function of torque for a balance ratio of 1.0 is good. For a seal having a balance ratio of 0.75, predicted equilibrium thermal rotation shows some agreement but more experimental data are needed. The results of 1.00 balance ratio tests suggest that after a long period of operation, any initial taper will be worn away and the seal would continue to operate as a parallel face seal. Results from long-term tests indicate that the wear coefficient is not a constant. While the experimental results support the basic concepts of the model, the results show where further work must be done to better understand the role of surface roughness and wear processes in mechanical face seals.

This content is only available via PDF.
You do not currently have access to this content.