Engineering topics
Publications and source records attributed to Fleming, D. P..
Research on bearings, gears, seals, and rotor dynamics (specifically high speed balancing and dampers) is presented. The research pertains to problems in both aircraft turbine engines and helicopter transmissions.
Radial stiffness of annular (ring-type) gas path seals is calculated for both constant-clearance designs and tapered designs for which the inlet clearance is larger than the outlet clearance. Under some conditions a constant-clearance seal can have a negative stiffness; this undesirable property can be completely eliminated by use of tapered seals. Leakage rates are only moderately higher in tapered seals
An annular seal is analyzed in which the inlet clearance is larger than the outlet clearance; the flow path may be either stepped or tapered. This design produces radial stiffness 1.7 to 14 times that of a constant clearance seal having the same minimum clearance. When sealing high pressure fluids, such a seal improves rotor stability and can be used to shift troublesome critical speeds to a more suitable location.
An annular seal is analyzed in which the inlet clearance is larger than the outlet clearance; the flow path may be either stepped or tapered. This design produces radial stiffnesses 1.7 to 14 times that of a constant-clearance seal having the same minimum clearance. When sealing high-pressure fluids, such as a seal can improve rotor stability and can be used to shift troublesome critical speeds to a more suitable location.
A flat sector shaped pad geometry for gas lubricated thrust bearings is analyzed considering both pitch and roll angles of the pad and the true film thickness distribution. Maximum load capacity is achieved when the pad is tilted so as to create a uniform minimum film thickness along the pad trailing edge. Performance characteristics for various geometries and operating conditions of gas thrust bearings are presented in the form of design curves. A comparison is made with the rectangular slider approximation. It is found that this approximation is unsafe for practical design, since it always overestimates load capacity.
A rigid vertical shaft was operated with known amounts of unbalance at speeds up to 30,000 rpm and gas supply pressure ratios to 4.8. From measured amplitude and phase angle data, dynamic stiffness and damping coefficients of the bearings were determined. The measured stiffness was proportional to the supply pressure, while damping was little affected by supply pressure. Damping dropped rapidly as the fractional frequency whirl threshold was approached. A small eccentricity analysis overpredicted the stiffness by 4 to 55 percent. Predicted damping was close to measured at low speeds but higher at high speeds
A rigid vertical shaft was operated with known amounts of unbalance at speeds to 30,000 rpm and gas supply pressure ratios to 4.8. From measured amplitude and phase angle data, dynamic stiffness and damping coefficients of the bearings were determined. The measured stiffness was proportional to the supply pressure, while damping was little affected by supply pressure. Damping dropped rapidly as the fractional frequency whirl threshold was approached. A small-eccentricity analysis overpredicted the stiffness by 20 to 70 percent. Predicted damping was lower than measured at low speeds but higher at high speeds.
A single mass flexible rotor analysis was used to optimize the stiffness and damping of a flexible support for a symmetric five-mass rotor. The flexible support attenuates the rotor motions and forces transmitted to the support bearings when the rotor operates through and above its first bending critical speed. An oil squeeze-film damper was designed based on short bearing lubrication theory. The damper design was verified by an unbalance response computer program. Rotor amplitudes were reduced by a factor of 16 and loads reduced by a factor of 36 compared with the same rotor on rigid bearing supports.
The experimentally proven range of application of the influence-coefficient balancing method, especially the least-squares procedure, is extended to include the case of multiple bending critical speeds within the operating range of a test rotor. Tests were conducted on a laboratory quality machine capable of speeds up to 18,000 rpm in order to investigate several distinct practical aspects of flexible-rotor balancing. These include: (1) balancing for operation through four bending critical speeds, (2) balancing of rotors mounted in both rigid and flexible bearing supports, (3) balancing of rotors with various amounts of measured vibration-response data and different numbers of correction planes, and (4) balancing of rotors with different arbitrary initial-unbalance configurations. The results show that a lightly damped, flexible rotor can be balanced systematically and efficiently through four bending critical speeds. It is concluded that the influence-coefficient method should be equally applicable to rotors or shafts having more than four bending critical speeds in their operating-speed ranges.
A program for the development and verification of drive-train dynamic technology is described along with its basis and the results expected from it. A central feature of this program is a drive-train test facility designed for the testing and development of advanced drive-train components, including shaft systems, dampers, and couplings. Previous efforts in designing flexible dynamic drive-train systems are reviewed, and the present state of the art is briefly summarized. The design of the test facility is discussed with major attention given to the formulation of the test-rig concept, dynamic scaling of model shafts, and the specification of design parameters. Specific efforts envisioned for the test facility are briefly noted, including evaluations of supercritical test shafts, stability thresholds for various sources and types of instabilities that can exist in shaft systems, effects of structural flexibility on the dynamic performance of dampers, and methods for vibration control in two-level and three-level flexible shaft systems.
A single-mass flexible-rotor analysis was used to optimize the stiffness and damping of a flexible support for a symmetric five-mass rotor. The flexible, damped support attenuates the amplitudes of motions and forces transmitted to the support bearings when the rotor operates through and above its first bending critical speed. An oil squeeze film damper was designed based on short bearing lubrication theory. The damper design was verified by an unbalance response computer program. Rotor amplitudes were reduced by a factor of 16 and loads reduced by a factor of 36 compared with the same rotor with rigid bearing supports.
This paper describes recent developments in the flexible rotor balancing technology area, with particular emphasis on methods for the addition and removal of correction weights. The currently existing multiplane-multispeed balancing procedure permits one-step balancing of final shaft-bearing assemblies simultaneously in a number of planes and at a number of speeds. Temporary addition of trial weights to the rotor, and the addition or subtraction of permanent corrections, are presently performed manually in the balancing process. The addition of a computer-controlled laser device to the balancing system shows promise of eliminating direct operator contact with the rotor in the balancing process, and thus could provide a considerable increase in the precision level at a critical step in the procedure.
Groove parameters were determined to maximize the stability of herringbone-grooved journal bearings. Parameters optimized were groove depth, width, length, and angle. Optimization was performed by using a small-eccentricity, infinite-groove analysis in conjunction with a previously developed Newton-Raphson procedure for bearings with the smooth member rotating or with the grooved member rotating at low compressibility numbers, and a newly developed vector technique for bearings with the grooved member rotating at high compressibility numbers. The design curves in this report enable one to choose the optimum bearing for a wide range of operating conditions. Compared with bearings optimized to maximize load capacity, bearings optimized for stability allow a thousandfold increase in bearing-supported mass in some cases before onset of instability, and lose no more than 77 percent of their load capacity in any case studied. Stability is much greater when the grooved member rotates.
Groove parameters were determined to maximize the stability of herringbone grooved journal bearings. Parameters optimized were groove depth, width, length, and angle. Optimization was performed using a small eccentricity, infinite groove analysis in conjunction with: (1) a previously developed Newton-Raphson procedure for bearings with the smooth member rotating or with the grooved member rotating at low compressibility numbers; (2) a newly-developed vector technique for bearings with the grooved member rotating at high compressibility numbers. The design curves in this report enable one to choose the optimum bearing for a wide range of operating conditions. These include: (1) compressibility numbers from 0 to 80; (2) length to diameter ratios from 1/4 to 2; and (3) smooth or grooved member rotating. Compared to bearings optimized to maximize load capacity, bearings optimized for stability allow a thousandfold increase in bearing-supported mass in some cases before onset of instability. Stability is much greater when the grooved member rotates.
The stability of herringbone grooved journal bearings was increased by optimizing the groove depth, width, length, and angle. Optimization was performed on the basis of a small-eccentricity infinite-groove analysis in combination with a previously developed Newton-Raphson procedure for bearings with the smooth member rotating or with the grooved member rotating at low compressibility numbers. A newly-developed vector technique for bearings with the grooved member rotating at high compressibility numbers also was used in the analysis. The design curves obtained are suitable for choosing the optimum bearing for a wide range of operating conditions (compressibility numbers from 0 to 80, length to diameter ratios from 1/4 to 2, and smooth or grooved member rotating). Compared to bearings optimized for maximum load, bearings optimized for maximum stability are shown to provide a drastic increase in bearing-supported mass. Bearings with grooved member rotating are substantially more stable than bearing with smooth member rotating, particularly at high compressibility numbers.
Curves span wide range of operating conditions, including: lubricant compressibility numbers from 0 to 80, bearing length-to-diameter ratios from 1/4 to 2, and either rotating or stationary grooved members.
Groove parameters were determined to maximize the stability of herringbone grooved journal bearings. Parameters optimized were groove depth, width, length, and angle. Optimization was performed using a small eccentricity, infinite groove analysis in conjunction with: (1) a previously developed Newton-Raphson procedure for bearings with the smooth member rotating or with the grooved member rotating at low compressibility numbers, and (2) a newly-developed vector technique for bearings with the grooved member rotating at high compressibility numbers. The design curves enable one to choose the optimum bearing for a wide range of operating conditions. These include: (1) compressibility numbers from 0 (incompressible) to 80, (2) length to diameter ratios from 1/4 to 2, and (3) smooth or grooved member rotating. Compared to bearings optimized to maximize load capacity, bearings optimized for stability: (1) allow a thousandfold increase in bearing-supported mass in some cases before onset of instability (the most dramatic increase are for bearings with small L/D operating at high compressibility numbers), and (2) lose no more than 77-percent of their load capacity in any case studied. Stability is much greater when the grooved member rotates.
A small eccentricity analysis was performed to predict load capacity and stability. Numerical results were obtained for a range of feeding parameter, pressure ratio, groove length and orifice recess volume for compressibility numbers from 0 to 50. These results were obtained from a digital computer program. Results showed that the addition of herringbone grooving to an externally pressurized bearing increases stability, but reduces load capacity at low compressibility numbers. A fully-grooved bearing is more stable than a partially-grooved bearing. Orifice recesses reduce stability, especially at high compressibility numbers.