Rotordynamics & Squeeze Film Dampers

Funded by National Science Foundation (1994-97) and TAMU Turbomachinery Research Consortium (1992-to date)

Squeeze film dampers (SFDs) provide viscous damping to rotating structures, allowing for reduction in vibration amplitudes and providing safe isolation from or of other structural components. SFDs are customarily used in aircraft jet engines, where rolling element bearings provide little damping to the rotor-bearing system, and in high performance compressors as retrofit elements in series with tilting pad bearings to soften bearing supports, reduce critical speeds, and allow for an extra margin of system stability. Most aircraft gas turbine engines employ at least one squirrel cage supported damper.

Squeeze film dampers derive their behavior from a lubricant being squeezed in the annular space between a non-rotating journal and a bearing housing. The journal, typically mounted on the outer race of rolling element bearings, whirls due to the forces exerted on the rotating shaft. The squeeze film action generates hydrodynamic pressures and damping forces at the film locations where the instantaneous gap (film thickness) is decreasing.

Squirrel cage supported dampers are the most commonly employed SFD design. Most large aircraft gas turbine engines use at least one, and in many instances, two or three dampers in one engine. The most distinctive feature of this damper configuration is the relatively large axial space required in comparison to the bearing hydrodynamic length.

OBJECTIVES: Funds allowed construction of a fully instrumented test rig for measurement of the imbalance response of a three disk rotor supported on SFDs (see figures below). The objectives of the research are:

(a) to provide reliable imbalance response measurements in a rotor-SFD configuration similar to that of an aircraft engine,

(b) to develop an empirical model to predict the forced dynamic performance of SFDs operating with air entrainment leading to a bubbly air/oil mixture, and

(c) to develop a non-linear SFD-structural model should the test results from (a) evidence deviations from linear behavior.

The first objective, fully completed, comprised the construction of the test apparatus and measurements of the test rotor in squirrel-cage supported SFDs and integral SFDs. The rotor-bearing system shows rigid body cylindrical and conical critical speeds below a top operating speed of 10 krpm. More than two hundred test measurements have shown the experimental rotor-SFD response to be linear even for large imbalance levels and off-centered damper journal operation. The experimental results allow the identification (and analytical validation) of the damping capability of integral squeeze film dampers and aid to determine the applicability of this novel technology to aircraft jet engines. These results have made the third objective irrelevant.

Other experiments conducted in a controlled orbit SFD rig have shown the effects of air ingestion on the performance of SFDs. An analytical model for performance prediction of SFDs with bubbly mixtures has also been completed.

ROTORDYNAMICS TEST RIG – ROTOR & SFDs

The Squeeze Film Damper (SFD) rig consistst of a three disk massive rotor (92 lb) supported on high precision angular contact ball bearings. The outer races of these bearings are supported on squeeze film dampers. The rotor is driven by a 10 HP DC motor and power supply. The supports are mounted on an isolated base attached to a table containing the motor and a protective cover. The rig is instrumented with six (X&Y) shaft displacement sensors, two support accelerometers, and one optical tachometer and keyphasor. A turbine-type flow meter, pressure gauges, and several thermocouples indicate the lubricant flow rate, pressure, and oil temperature in/out from the dampers.

Additional instrumentation includes four oscilloscopes, a FFT analyzer, and digital displays indicating rotor speeds and lubricant temperatures. The facility includes a 40 gallon oil tank, three gear pumps (one main oil supply pump and 2 return pumps), and 2 forced air convection coolers (for the lubricant and the drive motor). A Bentley Nevada ADRE for Windows DAIU collects and processes the test rig vibration measurements. The data processing software includes real time slow-roll subtraction, order-tracking and synchronous response filtering. An instrumentation console contains signal conditioners and digital displays of the operating rotor speed, flow rate, supply pressures and inlet/exit damper temperatures. The console includes the controls for operation of the lubrication pumps and the oil cooling and heating elements. Three oscilloscopes display the rotor orbits at the measurement locations. A fourth oscilloscope shows the bearing support housing accelerations, and a frequency analyzer depicts the FFT of selected vibration signals.

STATE OF THE ART TECHNOLOGY: INTEGRAL SQUEEZE FILM DAMPERS (ISFDs)

Modern technological advances in metal working allow the development of integral squeeze film dampers (ISFDs). This ingenious design is made possible by a wire Electrical Discharge Machining (EDM) process.

ISFDs are compact mechanical elements with a length no larger than the bearing itself, and comprised of arcuate pads attached to a bearing housing via thin wire-EDM webs. ISFDs can also be machined as split segments allowing rapid retrofit. Replacement of squirrel cage supported SFDs by integral dampers brings the following benefits to an aircraft gas turbine engine:

1. Reduced overall weight and length of the entire aircraft engine structure
2. Elimination of squirrel cage components ISFDs compact and with reduced number of parts
3. Ability to support axial thrust loads without locking the damper lateral motion
4. Accurate positioning (centering) by precise design and construction of the support web stiffness and pad film clearances
5. Split configuration which allows easier assembly and inspection than with any other damper design

A comprehensive study of the forced performance of Integral Squeeze Film Dampers is one of the main objectives of research and further development.

RESEARCH COMPLETED IN 1996/1997:
Measurements of the imbalance response of the test rotor supported on open ends, integral squeeze film dampers (ISFDs) have been completed. The dampers are compact with integral radial stiffness procured by wire EDM thin webs. The ISFDs have length and diameter equal to 3.8 inches (96.52 mm) and 0.91 inches (23.0 mm), with a clearance equal to 9 mils (0.229 mm). The tests are conducted with an ISO VG 10 oil at room temperature (73 F). The measurements include shaft speed, vibration displacements at six shaft locations, and two accelerations at the support housings. Other measurements include oil temperatures, feed pressures and flow rate.

Tests identifying the structural stiffness of each ISFD verify the design value (20 klb/in). Measurements of the synchronous rotor response with increasing imbalance masses are performed from coast-down tests. The measured vibration peak response at the rotor first critical speed is used to extract the system damping force coefficients and subsequent identification of the ISFD damping coefficients. The experiments show the open ended ISFDs to damp well the rotor response for the cylindrical modes of vibration, with peak vibration amplitudes proportional to the magnitude of the imbalances. Large rotor motions up to 80% of the nominal ISFD clearance are measured, and without shifts in the first critical speed denoting an absence of damper stiffness hardening. The test system damping coefficients increase slightly with the amplitude of rotor motion through the first critical speed. From these, the damping coefficients for the ISFDs are extracted and agree well with predictions from a full-film open ends, integral damper FEM model. This model is based on the solution of the classical Reynolds equation without fluid inertia effects for incompressible, isoviscous fluids flowing through the thin film land between the flexural pads and the damper housing. Given a specified damper journal position and instantaneous velocity, the program calculates the damper reaction forces and damping force coefficients in the (X,Y) directions.

RESEARCH COMPLETED IN 1997/98: SEALED INTEGRAL SQUEEZE FILM DAMPERS
Additional work on the experimental facility includes measurements of the test rotor-ISFD responses to couple mass imbalances and for ISFDs with end seals. The goal is to determine the effect of controlled end gap seals on the integral damper viscous force coefficients and their influence on the imbalance response of the test rotor. The measurements also include damper flow-rates and maximum temperature rise of the lubricant.

Measurements of the rotor synchronous response to couple imbalances exciting the conical mode of vibration further demonstrate the effectiveness of the integral SFDs to reduce rotor vibrations at this mode. Additional imbalance response measurements show the effect of controlled end gap seals on increasing the ISFDs damping coefficients while still allowing for cooling lubricant flow through the dampers.

The synchronous horizontal rotor (p-p) response for increasing levels of rotor imbalance is shown here for dampers with end seal gaps equal to 3 mils. The experiments show the sealed ends ISFDs to damp well the rotor response for the cylindrical mode of vibration and with peak vibration amplitudes proportional to the magnitude of the disk imbalances. Note that the rotor peak amplitude for the largest imbalance is nearly 90% of the damper radial clearance (0.230 mm). Damping coefficients extracted from the peak amplitudes are also shown below as a function of the peak rotor amplitude for various end gap seal clearances (3, 4 and 5 mils). Damping coefficients for the open ended dampers are also included. The damping values at zero rotor eccentricity correspond to the test results from impact response experiments without rotor spinning.

The paramount effect of the end seal gap clearance is clearly demonstrated from the experiments. Tighter end gap seals offer more damping, up to two times the magnitude obtained with the open ended dampers. However, the most notable finding is that the damper flow rate is not reduced as the end seal clearance decreases, thus allowing for the integral dampers to perform their function satisfactorily without lubricant overheating, as would be the case of a conventional damper with tight end seals.

RESEARCH COMPLETED IN 998-1999: SERIES TILTING PAD BEARING-AND INTEGRAL SF-DAMPER

High performance, high speed turbomachinery demands appropriate means to ensure structural isolation of components and stringent rotor vibration limits with tolerance to sudden imbalance loads due to blade loss events, shock, and maneuver actions. Squeeze film dampers are an effective mean to reduce vibrations and to suppress instabilities in high performance aero-engine systems. Integral squeeze film dampers (ISFDs) offer distinct advantages such as reduced overall weight and length of the damper structure with less number of parts, accuracy of positioning (centering), and a split segment construction allowing easier assembly, inspection and retrofit than with any other type of damper. Flexure pivot tilting pad bearings offer similar construction features as the ISFDs while minimizing assembly stack up tolerances and avoiding pivot wear and fretting. The series combination of a tilting pad bearing and a squeeze film damper has been implemented in numerous process compressors in the petrochemical industry to introduce flexibility and damping to the bearing supports. The proper design of these two mechanical elements allows for the optimum damping coefficient at the bearing support and accurate relocation of the (rigid mode) rotor bearing system critical speeds away from the operating speed range.

Measurements of imbalance responses of a test rotor supported on SFDs have been conducted since 1996. These experiments address to rotor-SFD configurations typical of aircraft gas turbines where safety and stability dictate the use of ball bearings instead of fluid film hydrodynamic bearings. In 1999 we are conducting measurements of the synchronous imbalance response of the test rotor supported on flexure pivot, tilting pad bearings and integral SFDs. The major objectives of the experiments are to determine the combined effect of the hydrodynamic bearings and SFDs on the location of critical speeds and effective logarithmic decrement, and to demonstrate the effectiveness of this bearing pair combination on reducing amplitudes of rotor vibration. The experimental results will allow benchmarking of predictive computational tools for estimation of force coefficients in both tilting pad bearings and squeeze film dampers.

THE EXPERIMENTS
Coast down rotor response measurements will be recorded for dry and wet dampers, i.e. without and with lubricant, for increasing mass imbalances located at the rotor middle disk. The programs SFDFLEXS and HYDROTRCM will be used to predict the force coefficients of the integral dampers and flexure pivot bearings over a frequency range. The XLTRC program will allow the prediction of the rotor synchronous response using equivalent impedance for the series SFD-tilting pad bearing at the bearing supports. The predictions will be correlated to the test measurements to evidence the effectiveness of the SFDs in suppressing rotor vibrations. A continuation project included the control of the damper stiffness based on the regulation of the feed pressure.

To learn more, order/read our Publications

ACKNOWLEDGMENTS
The support from National Science Foundation (NSF) and the Turbomachinery Research Consortium (TRC) is gratefully acknowledged. Thanks to Dr. F. Zeidan, KMC Bearings, Inc., for his assistance and support.