Effects of air entrainment on response of squeeze film dampers

Funded by National Science Foundation (1999-2002) and TAMU Turbomachinery Research Consortium (1998-2002)

High performance turbomachinery demands the largest power to weight ratios at ever increasing speeds and light-flexible rotors. These requirements accentuate the two most commonly recurring problems in rotordynamics, namely excessive steady-state synchronous vibration and sub harmonic rotor instabilities. Squeeze film dampers (SFDs) are virtually the only means to introduce damping in aircraft jet engines and commercial compressors. SFDs have been successfully used to solve these problems, stabilizing otherwise unstable units. As with most hydrodynamic bearings, the classical Reynolds equation for thin film lubrication is generally used to model squeeze film dampers. However, researchers and users have recurrently reported important discrepancies between theory and practice.

Our work in the Laboratory focused on understanding the complex phenomena occurring in SFDs.

A SFD comprises of a cylindrical housing, a journal and a thin lubricant film. In SFDs the journal is typically mounted on ball bearings and does not spin, only whirls with the shaft. As a result, the hydrodynamic (damping) forces generated are reactions to the whirling journal velocity. The figure shows a typical experimental pressure distribution in a SFD as measured in a laboratory test apparatus.

The pressure profile shown lacks symmetry due to the presence of air in the lubricant, and which is generally acknowledged to be one of the main sources of discrepancy between theory and practice. It is generally accepted that, under high speed operation with high vibration levels, a damper drags air into the film or releases air that is in solution within the oil. This phenomenon produces a bubbly mixture of oil and air that replaces the (pure) lubricant, and is generally misnamed as gaseous cavitation. The resulting complex squeeze flow is prevalent in SFDs used in aircraft engines.


Current models for the design of SFDs are extensions of the theory developed for journal bearings. The classical Reynolds equation is solved and boundary conditions for the definition of a vapor cavitation zone are applied to the computed pressures. The left figure above shows a measured pressure field that would agree with such models. In this case, the only effect of the cavitation by lubricant vaporization is the occurrence of a cavitation zone in which the pressure is constant and equal to the lubricant vapor pressure at the operating temperature. The vapor cavitation zone is clearly delimited and repeatable, thus rendering a predictable cyclic pressure field. Unfortunately, to obtain this kind of behavior some unusual requirements should be fulfilled. For the measurements presented, the SFD had to be operated fully submerged in oil and with a relatively high supply pressure.

Unfortunately, most SFDs operate with low levels of external pressurization and are open to the ambient on the sides, thus allowing the formation of a bubbly mixture in the film as mentioned before. Operation under these conditions results in completely different pressure profiles, as shown in the figure to the right. The most noticeable difference is that now the pressure field does not repeat itself for every cycle of journal motion. Random fluctuations in the peak-to-peak magnitude of the pressures are observed and the flat zone at the minimum pressure value is not present. Some sort of constant pressure zone (the gaseous cavitation zone) is developed, but at atmospheric pressure instead of the vapor pressure, and a minimum peak of sub atmospheric pressures is reached before the constant pressure zone.

The figures below show the stationary cavity developed by vapor cavitation in a steadily loaded journal bearing and the bubbly mixture formed in an orbiting SFD. Click on the pictures to see more information about them.

bly mixture formed

 Vapor cavitation on a journal bearing                    Bubbly Lubricant in a SFD

TRC-SFD test rig is set up for the study of gaseous cavitation in SFDs. The damper journal is mechanically constrained to perform circular centered orbits. The dimensions of the open-end damper tested are, Diameter=129 mm, Length=31 mm, Clearance=0.343 mm. The apparatus is instrumented with four piezoelectric pressure transducers, two strain gage pressure transducers, two eddy-current displacement sensors, one optical tachometer, one thermal air mass flow meter, one oil gear flow meter, and several thermocouples and pressure gauges. The damper is fed with controlled mixtures of oil and air produced in a sparger that allows the generation of mixtures with void fractions (air volume ratio) ranging from zero (pure oil) to one (pure air). The damper is operated a constant speed (16.67 Hz) and its outlet is fully submerged in a bath of the same mixture that is fed to the film lands so that no external air can be dragged.

The experimental results correlate the appearance and extent of a zone of constant pressure, or null pressure generation, with the introduction of air in the lubricant (void or volume fraction)as shown in the figuresbelow. A 3D representation of these test results shows a different perspective, and makes evident how the occurrence of the gaseous cavitation zone results in a reduction of the squeeze film pressure generation.

The measured pressure depicted on the graphs is absolute and makes evident how the zone of constant pressure takes the value of the pressure at the damper exit (discharge plane). The contour plot at the right correlates the squeeze pressure with the local film thickness as a function of time for different mixture (void fraction) conditions, and shows that the flat pressure zone develops about the point of maximum film thickness.

The major objectives of the research were to determine experimentally the effect that air entrainment has on the performance of the SFDs and to develop a model quantifying these effects so as to enhance our understanding of squeeze film dampers. The tests and model aid in the design of new dampers and modification of old ones. Ultimately, the damper hydrodynamic forces exerted on the journal become the most important factors to be studied, since these forces directly affect the dynamics of a rotor-bearing system employing SFDs. The experiments have shown that the presence of air in the lubricant results in reduction of the damping forces as shown below.

Experimental radial and tangential forces vs air volume content

The research completed the analysis of the pressures and forces to develop a semi-empirical theoretical model that accounts for the dynamics of the bubbles within the lubricant. Modifications to the test rig to perform flow visualizations were completed: See research on air entrainment for digital video clips

To learn more, order/read our Publications

The support from National Science Foundation (NSF) and the Turbomachinery Research Consortium (TRC are gratefully acknowledged.