Engineering topics

Mehmed, Oral

Publications and source records attributed to Mehmed, Oral.

At least 19 records

NASA NTRS2006

Quiet High Speed Fan (QHSF) Flutter Calculations Using the TURBO Code

A scale model of the NASA/Honeywell Engines Quiet High Speed Fan (QHSF) encountered flutter wind tunnel testing. This report documents aeroelastic calculations done for the QHSF scale model using the blade vibration capability of the TURBO code. Calculations at design speed were used to quantify the effect of numerical parameters on the aerodynamic damping predictions. This numerical study allowed the selection of appropriate values of these parameters, and also allowed an assessment of the variability in the calculated aerodynamic damping. Calculations were also done at 90 percent of design speed. The predicted trends in aerodynamic damping corresponded to those observed during testing.

Bakhle, Milind A.↗

NASA NTRS2005

Five-Axis, Three-Magnetic-Bearing Dynamic Spin Rig

The Five-Axis, Three-Magnetic-Bearing Dynamic Spin Rig is an apparatus for vibration testing of turbomachine blades in a vacuum at rotational speeds from 0 to 40,000 rpm. This rig includes (1) a vertically oriented shaft on which is mounted an assembly comprising a rotor holding the blades to be tested, (2) two actively controlled heteropolar radial magnetic bearings at opposite ends of the shaft, and (3) an actively controlled magnetic thrust bearing at the upper end of the shaft. This rig is a more capable successor to a prior apparatus, denoted the Dynamic Spin Rig (DSR), that included a vertically oriented shaft with a mechanical thrust bearing at the upper end and a single actively controlled heteropolar radial magnetic bearing at the lower end.

Morrison, Carlos R.↗

NASA NTRS2004

Fully Suspended, Five-Axis, Three-Magnetic-Bearing Dynamic Spin Rig With Forced Excitation

The Five-Axis, Three-Magnetic-Bearing Dynamic Spin Rig, a significant advancement in the Dynamic Spin Rig (DSR), is used to perform vibration tests of turbomachinery blades and components under rotating and nonrotating conditions in a vacuum. The rig has as its critical components three magnetic bearings: two heteropolar radial active magnetic bearings and a magnetic thrust bearing. The bearing configuration allows full vertical rotor magnetic suspension along with a feed-forward control feature, which will enable the excitation of various natural blade modes in bladed disk test articles. The theoretical, mechanical, electrical, and electronic aspects of the rig are discussed. Also presented are the forced-excitation results of a fully levitated, rotating and nonrotating, unbladed rotor and a fully levitated, rotating and nonrotating, bladed rotor in which a pair of blades was arranged 180 degrees apart from each other. These tests include the bounce mode excitation of the rotor in which the rotor was excited at the blade natural frequency of 144 Hz. The rotor natural mode frequency of 355 Hz was discerned from the plot of acceleration versus frequency. For nonrotating blades, a blade-tip excitation amplitude of approximately 100 g/A was achieved at the first-bending critical (approximately 144 Hz) and at the first-torsional and second-bending blade modes. A blade-tip displacement of 70 mils was achieved at the first-bending critical by exciting the blades at a forced-excitation phase angle of 908 relative to the vertical plane containing the blades while simultaneously rotating the shaft at 3000 rpm.

Morrison, Carlos R.↗

NASA NTRS2003

Unsteady Flowfield in a High-Pressure Turbine Modeled by TURBO

Forced response, or resonant vibrations, in turbomachinery components can cause blades to crack or fail because of the large vibratory blade stresses and subsequent high-cycle fatigue. Forced-response vibrations occur when turbomachinery blades are subjected to periodic excitation at a frequency close to their natural frequency. Rotor blades in a turbine are constantly subjected to periodic excitations when they pass through the spatially nonuniform flowfield created by upstream vanes. Accurate numerical prediction of the unsteady aerodynamics phenomena that cause forced-response vibrations can lead to an improved understanding of the problem and offer potential approaches to reduce or eliminate specific forced-response problems. The objective of the current work was to validate an unsteady aerodynamics code (named TURBO) for the modeling of the unsteady blade row interactions that can cause forced response vibrations. The three-dimensional, unsteady, multi-blade-row, Reynolds-averaged Navier-Stokes turbomachinery code named TURBO was used to model a high-pressure turbine stage for which benchmark data were recently acquired under a NASA contract by researchers at the Ohio State University. The test article was an initial design for a high-pressure turbine stage that experienced forced-response vibrations which were eliminated by increasing the axial gap. The data, acquired in a short duration or shock tunnel test facility, included unsteady blade surface pressures and vibratory strains.

Bakhle, Milind A.↗

NASA NTRS2003

Spin Testing for Durability Began on a Self-Tuning Impact Damper for Turbomachinery Blades

NASA and Pratt & Whitney will collaborate under a Space Act Agreement to perform spin testing of the impact damper to verify damping effectiveness and durability. Pratt & Whitney will provide the turbine blade and damper hardware for the tests. NASA will provide the facility and perform the tests. Effectiveness and durability will be investigated during and after sustained sweeps of rotor speed through resonance. Tests of a platform wedge damper are also planned to compare its effectiveness with that of the impact damper. Results from baseline tests without dampers will be used to measure damping effectiveness. The self-tuning impact damper combines two damping methods-the tuned mass damper and the impact damper. It consists of a ball located within a cavity in the blade. This ball rolls back and forth on a spherical trough under centrifugal load (tuned mass damper) and can strike the walls of the cavity (impact damper). The ball s rolling natural frequency is proportional to the rotor speed and can be designed to follow an engine-order line (integer multiple of rotor speed). Aerodynamic forcing frequencies typically follow these engineorder lines, and a damper tuned to the engine order will most effectively reduce blade vibrations when the resonant frequency equals the engine-order forcing frequency. This damper has been tested in flat plates and turbine blades in the Dynamic Spin Facility. During testing, a pair of plates or blades rotates in vacuum. Excitation is provided by one of three methods--eddy-current engine-order excitation (ECE), electromechanical shakers, and magnetic bearing excitation. The eddy-current system consists of magnets located circumferentially around the rotor. As a blade passes a magnet, a force is imparted on the blade. The number of magnets used can be varied to change the desired engine order of the excitation. The magnets are remotely raised or lowered to change the magnitude of the force on the blades. The other two methods apply force to the rotating shaft itself at frequencies independent of the rotor speed. During testing, blade vibration is monitored with strain gauges and laser displacement probes.

Duffy, Kirsten↗

NASA NTRS2003

Dynamic Spin Rig Upgraded With a Five- Axis-Controlled Three-Magnetic-Bearing Support System With Forward Excitation

The NASA Glenn Research Center Dynamic Spin Rig is used for experimental evaluation of vibration analysis methods and dynamic characteristics for rotating systems. Measurements are made while rotors are spun and vibrated in a vacuum chamber. The rig has been upgraded with a new active magnetic bearing rotor support and excitation system. This design is expected to provide operational improvements over the existing rig. The rig will be able to be operated in either the old or new configuration. In the old configuration, two ball bearings support the vertical shaft of the rig, with the test article located between the bearings. Because the bearings operate in a vacuum, lubrication is limited to grease. This limits bearing life and speed. In addition, the old configuration employs two voice-coil electromagnetic shakers to apply oscillatory axial forces or transverse moments to the rotor shaft through a thrust bearing. The excitation amplitudes that can be imparted to the test article with this system are not adequate for components that are highly damped. It is expected that the new design will overcome these limitations.

Morrison, Carlos R.↗

NASA NTRS2002

A Comparison Study of Magnetic Bearing Controllers for a Fully Suspended Dynamic Spin Rig

NASA Glenn Research Center (GRC) has developed a fully suspended magnetic bearing system for the Dynamic Spin Rig (DSR) that is used to perform vibration tests of turbomachinery blades and components under spinning conditions in a vacuum. Two heteropolar radial magnetic bearings and a thrust bearing and the associated control system were integrated into the DSR to provide noncontact magnetic suspension and mechanical excitation of the 35 lb vertical rotor with blades to induce turbomachinery blade vibration. A simple proportional-integral-derivative (PID) controller with a special feature for multidirectional radial excitation worked very well to both support and shake the shaft with blades. However, more advanced controllers were developed and successfully tested to determine the optimal controller in terms of sensor and processing noise reduction, smaller rotor orbits, and energy savings for the system. The test results of a variety of controllers we demonstrated up to the rig's maximum allowable speed of 10,000 rpm are shown.

Choi, Benjamin↗

NASA NTRS2002

ASTROP2-LE: A Mistuned Aeroelastic Analysis System Based on a Two Dimensional Linearized Euler Solver

An aeroelastic analysis system for flutter and forced response analysis of turbomachines based on a two-dimensional linearized unsteady Euler solver has been developed. The ASTROP2 code, an aeroelastic stability analysis program for turbomachinery, was used as a basis for this development. The ASTROP2 code uses strip theory to couple a two dimensional aerodynamic model with a three dimensional structural model. The code was modified to include forced response capability. The formulation was also modified to include aeroelastic analysis with mistuning. A linearized unsteady Euler solver, LINFLX2D is added to model the unsteady aerodynamics in ASTROP2. By calculating the unsteady aerodynamic loads using LINFLX2D, it is possible to include the effects of transonic flow on flutter and forced response in the analysis. The stability is inferred from an eigenvalue analysis. The revised code, ASTROP2-LE for ASTROP2 code using Linearized Euler aerodynamics, is validated by comparing the predictions with those obtained using linear unsteady aerodynamic solutions.

Reddy, T. S. R.↗

NASA NTRS2002

Calculation and Correlation of the Unsteady Flowfield in a High Pressure Turbine

Forced vibrations in turbomachinery components can cause blades to crack or fail due to high-cycle fatigue. Such forced response problems will become more pronounced in newer engines with higher pressure ratios and smaller axial gap between blade rows. An accurate numerical prediction of the unsteady aerodynamics phenomena that cause resonant forced vibrations is increasingly important to designers. Validation of the computational fluid dynamics (CFD) codes used to model the unsteady aerodynamic excitations is necessary before these codes can be used with confidence. Recently published benchmark data, including unsteady pressures and vibratory strains, for a high-pressure turbine stage makes such code validation possible. In the present work, a three dimensional, unsteady, multi blade-row, Reynolds-Averaged Navier Stokes code is applied to a turbine stage that was recently tested in a short duration test facility. Two configurations with three operating conditions corresponding to modes 2, 3, and 4 crossings on the Campbell diagram are analyzed. Unsteady pressures on the rotor surface are compared with data.

Bakhle, Milind A.↗

NASA NTRS2002

Self-Tuning Impact Dampers Designed for Turbomachinery Blade Vibration Suppression

Turbomachinery blades are subject to aerodynamic forces that can lead to high-cycle-fatigue (HCF) failures. These failures will only increase as engineers begin to design blades without shrouds or as integrally bladed disks (blisks). These new designs will decrease blade damping significantly because the mechanical damping from shroud and blade joints will be eliminated. Also, it is difficult to design dampers for the engine environment with its extremely high centrifugal loads and high temperatures. The self-tuning impact damper has been designed to provide the additional damping required to avoid HCF while withstanding the harsh engine environment. In addition, the damper is placed within the engine blade itself rather than external to it.

Duffy, Kirsten P.↗

NASA NTRS2002

Aeroelastic Calculations of Quiet High- Speed Fan Performed

An advanced high-speed fan was recently designed under a cooperative effort between the NASA Glenn Research Center and Honeywell Engines & Systems. The principal design goals were to improve performance and to reduce fan noise at takeoff. Scale models of the Quiet High-Speed Fan were tested for operability, performance, and acoustics. During testing, the fan showed significantly improved noise characteristics, but a self-excited aeroelastic vibration known as flutter was encountered in the operating range. Flutter calculations were carried out for the Quiet High-Speed Fan using a three-dimensional, unsteady aerodynamic, Reynolds-averaged Navier-Stokes turbomachinery code named "TURBO." The TURBO code can accurately model the viscous flow effects that can play an important role in various aeroelastic problems such as flutter with flow separation, flutter at high loading conditions near the stall line (stall flutter), and flutter in the presence of shock and boundary-layer interaction. Initially, calculations were performed with no blade vibrations. These calculations were at a constant rotational speed and a varying mass flow rate. The mass flow rate was varied by changing the backpressure at the exit boundary of the computational domain. These initial steady calculations were followed by aeroelastic calculations in which the blades were prescribed to vibrate harmonically in a natural mode, at a natural frequency, and with a fixed interblade phase angle between adjacent blades. The AE-prep preprocessor was used to interpolate the in-vacuum mode shapes from the structural dynamics mesh onto the computational fluid dynamics mesh and to smoothly propagate the grid deformations from the blade surface to the interior points of the grid. The aeroelastic calculations provided the unsteady aerodynamic forces on the blade surface due to blade vibrations. These forces were vector multiplied with the structural dynamic mode shape to calculate the work done on the blade during one vibration period, then this result was converted to an aerodynamic damping. Flutter occurs when the aerodynamic damping becomes negative, if structural damping is ignored. The results of these aeroelastic calculations are summarized in this plot of aerodynamic damping versus mass flow rate at a constant rotational speed. As the backpressure is increased, the mass flow rate through the fan decreases and the fan operating point moves towards the stall line. The aeroelastic calculations showed that the aerodynamic damping decreases as the stall line is approached, as observed during testing. In addition, the aeroelastic calculations with the TURBO code correctly predicted the aeroelastic parameters: the most unstable vibration mode and interblade phase angle, as observed during testing. The Quiet High-Speed Fan demonstrated significant noise reductions during testing, but flutter imposed limits on its operating range. The accurate calculation of the aeroelastic characteristics using the TURBO code is a significant step toward eliminating flutter from the operating range and toward realizing the benefits of reduced fan noise. The aeroelastic calculations described here were performed under a grant by University of Toledo researchers in collaboration with Glenn's researchers.

Bakhle, Milind A.↗

NASA NTRS2000

On a Self-Tuning Impact Vibration Damper for Rotating Turbomachinery

A self-tuning impact damper is investigated analytically and experimentally as a device to inhibit vibration and increase the fatigue life of rotating components in turbomachinery. High centrifugal loads in rotors can inactivate traditional impact dampers because of friction or misalignment of the damper in the g-field. Giving an impact damper characteristics of an acceleration tuned-mass damper enables the resulting device to maintain damper mass motion and effectiveness during high-g loading. Experimental results presented here verify that this self-tuning impact damper can be designed to follow an engine order line. damping rotor component resonance crossings.

Duffy, Kirsten P.↗

NASA NTRS1999

Neural Networks Used to Compare Designed and Measured Time-Average Patterns

Electronic time-average holograms are convenient for comparing the measured vibration modes of fan blades with those calculated by finite-element models. At the NASA Lewis Research Center, neural networks recently were trained to perform what had been a simple visual comparison of the predictions of the design models with the measurements. Finite-element models were used to train neural networks to recognize damage and strain information encoded in subtle changes in the time-average patterns of cantilevers. But the design-grade finite element models were unable to train the neural networks to detect damage in complex blade shapes. The design-model-generated patterns simply did not agree well enough with the measured patterns. Instead, hybrid-training records, with measured time-average patterns as the input and model-generated strain information as the output, were used to effect successful training.

Decker, Arthur J.↗

NASA NTRS1998

Vibrational Analysis of Engine Components Using Neural-Net Processing and Electronic Holography

The use of computational-model trained artificial neural networks to acquire damage specific information from electronic holograms is discussed. A neural network is trained to transform two time-average holograms into a pattern related to the bending-induced-strain distribution of the vibrating component. The bending distribution is very sensitive to component damage unlike the characteristic fringe pattern or the displacement amplitude distribution. The neural network processor is fast for real-time visualization of damage. The two-hologram limit makes the processor more robust to speckle pattern decorrelation. Undamaged and cracked cantilever plates serve as effective objects for testing the combination of electronic holography and neural-net processing. The requirements are discussed for using finite-element-model trained neural networks for field inspections of engine components. The paper specifically discusses neural-network fringe pattern analysis in the presence of the laser speckle effect and the performances of two limiting cases of the neural-net architecture.

Decker, Arthur J.↗

NASA NTRS1998

A Magnetic Suspension and Excitation System for Spin Vibration Testing of Turbomachinery Blades

The Dynamic Spin Rig (DSR) is used to perform vibration tests of turbomachinery blades and components under spinning conditions in a vacuum. A heteropolar radial active magnetic bearing was integrated into the DSR to provide non-contact magnetic suspension and mechanical excitation of the rotor to induce turbomachinery blade vibrations. The magnetic bearing replaces one of the two existing conventional radial ball bearings. Prior operation of the DSR used two voice-coil type linear electromagnetic shakers which provided axial excitation of the rotor. The new magnetic suspension and excitation system has provided enhanced testing capabilities. Tests were performed at high rotational speeds for longer duration and higher vibration amplitudes. Some characteristics of the system include magnetic bearing stiffness values up to 60,000 lb./in., closed loop control bandwidth around 500 Hz, and multi-directional radial excitation of the rotor. This paper reports on the implementation and operation of this system and presents some test results using this system.

Johnson, Dexter↗

NASA NTRS1998

Model-Trained Neural Networks and Electronic Holography Demonstrated to Detect Damage in Blades

Detect Damage in Blades Electronic holography can show damaged regions in fan blades at 30 frames/sec. The electronic holograms are transformed by finite-element-model-trained artificial neural networks to visualize the damage. The trained neural networks are linked with video and graphics to visualize the bending-induced strain distribution, which is very sensitive to damage. By contrast, it is very difficult to detect damage by viewing the raw, speckled, characteristic fringe patterns. For neural-network visualization of damage, 2 frames or 2 fields are used, rather than the 12 frames normally used to compute the displacement distribution from electronic holograms. At the NASA Lewis Research Center, finite element models are used to compute displacement and strain distributions for the vibration modes of undamaged and cracked blades. A model of electronic time-averaged holography is used to transform the displacement distributions into finite-element-resolution characteristic fringe patterns. Then, a feedforward neural network is trained with the fringe-pattern/strain-pattern pairs, and the neural network, electronic holography, and video are implemented on a workstation. Now that the neural networks have been tested successfully at 30 frames/sec on undamaged and cracked cantilevers, the electronic holography and neural-network processing are being adapted for onsite damage inspection of twisted fan blades and rotormounted blades. Our conclusion is that model-trained neural nets are effective when they are trained with good models whose application is well understood. This work supports the aeromechanical testing portion of the Advanced Subsonic Technology Project.

Decker, Arthur J.↗

NASA NTRS1998

Damping Experiment of Spinning Composite Plates With Embedded Viscoelastic Material

One way to increase gas turbine engine blade reliability and durability is to reduce blade vibration. It is well known that vibration can be reduced by adding damping to metal and composite blade-disk systems. As part of a joint research effort of the NASA Lewis Research Center and the University of California, San Diego, the use of integral viscoelastic damping treatment to reduce the vibration of rotating composite fan blades was investigated. The objectives of this experiment were to verify the structural integrity of composite plates with viscoelastic material patches embedded between composite layers while under large steady forces from spinning, and to measure the damping and natural frequency variation with rotational speed.

Mehmed, Oral↗

NASA NTRS1997

Vibrational Analysis of Engine Components Using Neural-Net Processing and Electronic Holography

Decker, Arthur J.↗