Estimating Inducer Blade Damping in Water with On-Blade Strain Measurements
No abstract available
Mulder, Andrew↗
Engineering topics
Skelley, Stephen
Publications and source records attributed to Skelley, Stephen.
Estimating Inducer Blade Damping in Water with On-Blade Strain Measurements
This paper describes the testing, methodology, and results for estimating critical damping ratios of inducer blade structural natural frequencies. A sub-scale inducer water flow test was performed with on-blade strain gauge measurements. Data were acquired in quiescent air and water, and also at several shaft speeds and operating conditions. Two methods were used to provide active excitation to the blades. The first was to impart a mechanical impulse to the inducer shaft and observe the strain gauge response. The second was to install a stator plate upstream of the inducer to create a non-uniform but regular pressure field at the inducer inlet. There are challenges with using either approach, but damping information was extracted using both methods. Results showed that the critical damping ratio for a given blade mode increased from air to quiescent water, and also increased when the inducer was spinning.
Mulder, Andrew↗
Effects of Alternate Leading Edge Cutback on the Space Shuttle Main Engine Low Pressure Fuel Pump
A higher order cavitation oscillation observed in the SSME low pressure fuel pump has been eliminated in water flow testing of a modified subscale replica of the inducer. The low pressure pump was modified by removing the outboard sections of two opposing blades of the four-bladed inducer, blending the "cutback" regions into the blades at the leading edge and tip, and removing material on the suction sides to decrease the exposed leading edge thickness. The leading edge tips of the cutback blades were moved approximately 25 degrees from their previous locations, thereby increasing one blade to blade spacing, decreasing the second, while simultaneously moving the cutback tips downstream. The test was conducted in MSFC's inducer test loop at scaled operating conditions in degassed and filtered water. In addition to eliminating HOC across the entire scaled operating regime, rotating cavitation was suppressed while the range of both alternate blade and asymmetric cavitation were increased. These latter phenomena, and more significantly, the shifts between these cavitation modes also resulted in significant changes to the head coefficient at low cavitation numbers. Reverse flow was detected at a slightly larger flow coefficient with the cutback inducer and suction capability was reduced by approximately 1 velocity head at and above approximately 90% of the reference flow coefficient. These performance changes along with more intense reverse flow are consistent with poor flow area management and increased incidence in the cutback region. Although the test demonstrated that the inducer modification was successful at eliminating the higher order cavitation across the entire scaled operating regime, different, previously unobserved, cavitation oscillations were introduced and significant performance penalties were imposed.
Mulder, Andrew↗
Rotating Balances Used for Fluid Pump Testing
Marshall Space Flight Center has developed and demonstrated two direct read force and moment balances for sensing and resolving the hydrodynamic loads on rotating fluid machinery. These rotating balances consist of a series of stainless steel flexures instrumented with semiconductor type, unidirectional strain gauges arranged into six bridges, then sealed and waterproofed, for use fully submerged in degassed water at rotational speeds up to six thousand revolutions per minute. The balances are used to measure the forces and moments due to the onset and presence of cavitation or other hydrodynamic phenomena on subscale replicas of rocket engine turbomachinery, principally axial pumps (inducers) designed specifically to operate in a cavitating environment. The balances are inserted into the drive assembly with power to and signal from the sensors routed through the drive shaft and out through an air-cooled twenty-channel slip ring. High frequency data - balance forces and moments as well as extensive, flush-mounted pressures around the rotating component periphery - are acquired via a high-speed analog to digital data acquisition system while the test rig conditions are varied continuously. The data acquisition and correction process is described, including the in-situ verifications that are performed to quantify and correct for known system effects such as mechanical imbalance, "added mass," buoyancy, mechanical resonance, and electrical bias. Examples of four types of cavitation oscillations for two typical inducers are described in the laboratory (pressure) and rotating (force) frames: 1) attached, symmetric cavitation, 2) rotating cavitation, 3) attached, asymmetric cavitation, and 4) cavitation surge. Rotating and asymmetric cavitation generate a corresponding unbalanced radial force on the rotating assembly while cavitation surge generates an axial force. Attached, symmetric cavitation induces no measurable force. The frequency of the forces can be determined a priori from the pressure environment while the magnitude of the hydrodynamic force is proportional to the pressure unsteadiness.
Skelley, Stephen↗
Fluctuating Pressure Environments and Hydrodynamic Radial Force Mitigation for a Two Blade Unshrouded Inducer
Fluctuating pressure data from water flow testing of an unshrouded two blade inducer revealed a cavitation induced oscillation with the potential to induce a radial load on the turbopump shaft in addition to other more traditionally analyzed radial loads. Subsequent water flow testing of the inducer with a rotating force measurement system confirmed that the cavitation induced oscillation did impart a radial load to the inducer. After quantifying the load in a baseline configuration, two inducer shroud treatments were selected and tested to reduce the cavitation induced load. The first treatment was to increase the tip clearance, and the second was to introduce a circumferential groove near the inducer leading edge. Increasing the clearance resulted in a small decrease in radial load along with some steady performance degradation. The groove greatly reduced the hydrodynamic load with little to no steady performance loss. The groove did however generate some new, relatively high frequency, spatially complex oscillations to the flow environment.
Mulder, Andrew↗
Fluctuating Pressure Environments and Hydrodynamic Radial Force Mitigation for a Two Blade Unshrouded Inducer
Fluctuating pressure data from water flow testing of an unshrouded two blade inducer revealed a cavitation induced oscillation with the potential to induce a radial load on the turbopump shaft in addition to other more traditionally analyzed radial loads. Subsequent water flow testing of the inducer with a rotating force measurement system confirmed that the cavitation induced oscillation did impart a radial load to the inducer. After quantifying the load in a baseline configuration, two inducer shroud treatments were selected and tested to reduce the cavitation induced load. The first treatment was to increase the tip clearance, and the second was to introduce a circumferential groove near the inducer leading edge. Increasing the clearance resulted in a small load decrease along with some steady performance degradation. The groove greatly reduced the hydrodynamic load with little to no steady performance loss. The groove did however generate some new, relatively high frequency, spatially complex oscillations to the environment.
Mulder, Andrew↗
Summary of Recent Inducer Testing at MSFC and Future Plans
This viewgraph presentation covers water flow tests on the RS-83 Main LOX Inducer for the Space Shuttle Main Engine (SSME). The presentation lists recent water tests on the SSME liquid oxygen (LOX) pump inducer, includes images and diagrams of the water test facility at Marshall Space Flight Center (MSFC), profiles inducer hydrodynamic forces, and diagrams the performance of the RS-83 inducer.
Skelley, Stephen↗
Water Flow Performance of a Superscale Model of the Fastrac Liquid Oxygen Pump
As part of the National Aeronautics and Space Administration's ongoing effort to lower the cost of access to space, the Marshall Space Flight Center has developed a rocket engine with 60,000 pounds of thrust for use on the Reusable Launch Vehicle technology demonstrator slated for launch in 2000. This gas generator cycle engine, known as the Fastrac engine, uses liquid oxygen and RP-1 for propellants and includes single stage liquid oxygen and RP-1 pumps and a single stage supersonic turbine on a common shaft. The turbopump design effort included the first use and application of new suction capability prediction codes and three-dimensional blade generation codes in an attempt to reduce the turbomachinery design and certification costs typically associated with rocket engine development. To verify the pump's predicted cavitation performance, a water flow test of a superscale model of the Fastrac liquid oxygen pump was conducted to experimentally evaluate the liquid oxygen pump's performance at and around the design point. The water flow test article replicated the flow path of the Fastrac liquid oxygen pump in a 1.582x scale model, including scaled seal clearances for correct leakage flow at a model operating speed of 5000 revolutions per minute. Flow entered the 3-blade axial-flow inducer, transitioned to a shrouded, 6- blade radial impeller, and discharged into a vaneless radial diffuser and collection volute. The test article included approximately 50 total and static pressure measurement locations as well as flush-mounted, high frequency pressure transducers for complete mapping of the pressure environment. The primary objectives of the water flow test were to measure the steady-state and dynamic pressure environment of the liquid oxygen pump versus flow coefficient, suction specific speed, and back face leakage flow rate. Initial results showed acceptable correlation between the predicted and experimentally measured pump head rise at low suction specific speeds. Likewise, only small circumferential variations in steady-state were observed from 80% to 120% of the design flow coefficient, matching the computational predictions and confirming that the integrated design approach has minimized any exit volute-induced distortions. The test article exhibited suction performance trends typically observed in inducer designs with virtually constant head rise with decreasing inlet pressure until complete pump head breakdown. Unfortunately, the net positive suction head at 3% head fall-off occurred far below that predicted at all tested flow coefficients, resulting in a negative net positive suction head margin at the design point in water. Additional testing to map the unsteady pressure environment was conducted and cavitation-induced flow disturbances at the inducer inlet were observed. Two distinct disturbances were identified, one rotating and one stationary relative to the fixed frame of reference, while the transition from one regime to the next produced significant effects on the steady state pump performance. The impact of the unsteady phenomena and the corresponding energy losses on the unexpectedly poor pump performance is also discussed.
Skelley, Stephen↗
Water Flow Performance of a Superscale Model of the Fastrac Liquid Oxygen Pump
As part of the National Aeronautics and Space Administration's ongoing effort to lower the cost of access to space, the Marshall Space Flight Center has developed a rocket engine with 60,000 pounds of thrust for use on the Reusable Launch Vehicle technology demonstrator slated for launch in 2000. This gas generator cycle engine, known as the Fastrac engine, uses liquid oxygen and RP-1 for propellants and includes single stage liquid oxygen and RP-1 pumps and a single stage supersonic turbine on a common shaft. The turbopump design effort included the first use and application of new suction capability prediction codes and three-dimensional blade generation codes in an attempt to reduce the turbomachinery design and certification costs typically associated with rocket engine development. To verify the pump's predicted cavitation performance, a water flow test of a superscale model of the Fastrac liquid oxygen pump was conducted to experimentally evaluate the liquid oxygen pump's performance at and around the design point. The water flow test article replicated the flow path of the Fastrac liquid oxygen pump in a 1.582x scale model, including scaled seal clearances for correct leakage flow at a model operating speed of 5000 revolutions per minute. Flow entered the 3-blade axial-flow inducer, transitioned to a shrouded, 6-blade radial impeller, and discharged into a vaneless radial diffuser and collection volute. The test article included approximately 50 total and static pressure measurement locations as well as flush-mounted, high frequency pressure transducers for complete mapping of the pressure environment. The primary objectives of the water flow test were to measure the steady-state and dynamic pressure environment of the liquid oxygen pump versus flow coefficient, suction specific speed, and back face leakage flow rate. Results showed excellent correlation between the predicted and experimentally measured pump head rise at low suction specific speeds. Likewise, only small circumferential variations in steady-state impeller exit and radial diffuser pressure distributions were observed from 80% to 120% of the design flow coefficient, matching the computational predictions and confirming that the integrated design approach has minimized any exit volute-induced distortions. The test article exhibited suction performance trends typically observed in inducer designs with virtually constant head rise with decreasing inlet pressure until complete pump head breakdown. Unfortunately, the net positive suction head at 3% head fall-off occurred far below that predicted at all tested flow coefficients, resulting in a negative net positive suction head margin at the design point in water. Additional testing to map the unsteady pressure environment was conducted and interesting rotating phenomena at the inducer inlet were observed. These rotating phenomena's cell numbers, direction, and speed were correlated with pump operating parameters. The impact of the unsteady phenomena and their corresponding energy losses on the unexpectedly poor pump performance is also discussed.
Skelley, Stephen↗