AERO-U REU
2007 Highlights
NSF REU Site:
AERO-U: Aerospace Engineering Research Opportunities for Undergraduates
http://aero.tamu.edu/research/undergraduate/aero-propulsion-fluids/
The following are the faculty mentors who worked with REU students during the program. The list provides the name, position, degree and research interests of each faculty mentor.
Dr. Rodney D. W. Bowersox, Associate Professor. Ph.D., Virginia Polytechnic Institute and State University. Gasdynamics, aerothermochemistry, high-speed and unsteady aerodynamics, aero-propulsion, turbulence modeling, numerical simulations, instrumentation development and wind tunnel design.
Dr. Sharath S. Girimaji, Professor. Ph.D., Cornell. Turbulent and hypersonic flow modeling. Partially-Average Navier-Stokes (PANS) model development. Lattice-Boltzmann methods. Plasma turbulence.
Dr. Othon Rediniotis, Professor. Ph.D. from Virginia Tech. Director, Aero and Fluid Dynamics Laboratory. Dr. Rediniotis specializes in Fluid Mechanics/Aerodynamics and Intelligent Systems in Aerospace Engineering. Research interests include: Experimental Techniques in Fluid Dynamics and Smart Materials and Structures, Flow Control, Intelligent Structures and Vehicles, Biomimetic Vehicle Locomotion and Control, Micro-Electro-Mechanical Systems (MEMS) and Applications to Fluid Dynamics and Intelligent Structures.
Dr. Jacques C. Richard, Senior Lecturer & Research Associate Professor. Ph.D., Rensselaer Polytechnic Institute, Troy, NY. Plasma and gas dynamics computational modeling. Lattice-Boltzmann methods. Spectral element methods. Modeling plasma jets in magnetic fields. Electric propulsion (EP): ion thruster optics plasma flow, Magneto-hydro-dynamics (MHD).
Dr. Adonios Karpetis, Assistant Professor. Ph.D., Princeton. Experimental turbulent combustion and high-speed flow visualization, Microcombustion. Laser Diagnostics.
Dr. Paul Cizmas, Associate Professor. Ph.D., Duke. Unsteady Aerodynamics and Heat Transfer Fluid-solid Interaction. Propulsion. Computational Fluid Dynamics (CFD). Massive Parallel Processing.
Figure 1. Students after touring Ad Astra Rocket Company and the Neutral Buoyancy Lab (NBL) at NASA Johnson Space Center in Houston, TX. From keft: Elizabeth Jesse, David Williams, Alex Bayeh, Chi Mai, David Alcocer, Meghan Patzke, Robert Fiebisohn, Julia Cosse.
Figure 2. Students who did not attend the tour (a) Nicole Mendoza, and (b) Daniel Araya
The following are some of the students that participated in the AERO-U REU during the Summer 2007 program. The personal information is not shown but they are from different institutions of various gender and ethnicity, major, GPA (>3.25), expected graduation dates, etc. Sample research titles and abstracts are shown below.
David Alcocer, University of California, Berkeley. Mechanical Engineering major. Expected graduation. Spring 2009. Research Title: ÒLBM Simulations for MHD Boundary Layer Control over a Flat Plate.Ó [Mentor: Dr. Jacques C. Richard]
Abstract: This paper investigates Magnetohydrodynamic (MHD) boundary layer control simulated with lattice Boltzmann methods (LBM). Plasma flows over a flat plate are replicated. Specifically, this paper focuses on minimizing boundary layer thickness rather than delaying laminar to turbulent transition. In addition, cases are run for several magnetic field strengths and multiple fluid Reynolds numbers.
Figure 3. (a) axial velocity vs. height above plate for Re = 900 (higher Re than last year) at x = 50 with different applied magnetic fields (b) similar effects on velocity profiles due to jet injection [McCormick, 2000, AIAA-00-16398].
Daniel Araya, Texas A&M University. Aerospace Engineering major. Expected graduation Spring 2008. Research Title: ÒSensitivity of Static Pressure Measurement for the NASA T38 Pitot-Static ProbeÓ [Mentor: Dr. Othon Rediniotis]. Paper submitted to the AIAA Journal for 2008. Applying to grad school at TAMU.
Abstract: The ability of an aircraft to safely maneuver through populated airspace relies on consistently accurate instruments. This study aims to investigate one critical instrument, the pitot-static probe. Specifically, this study examines the sensitivity of the static pressure measurement for a NASA T-38 jet trainer pitot-static probe. A Computational Fluid Dynamics (CFD) model of the T38 probe was created and validated against NASA results. The preliminary results show that there is both an effect due to surface corrosion and, even more so, due to axial location on the measurement of static pressure.
Figure 4. (a) Side view of NASA T-38 jet trainerÕs pitot-static probe in wind tunnel used to study effects of probe surface corrosion on measurements. (b) Static pressure contours from computational fluid dynamic (CFD) simulations of flow over probe. (c) Pressure coefficient from CFD results. (d) Pressure coefficient from NASA Langley Research CenterÕs CFD results.
Alexander C. Bayeh, Texas A&M University. Aerospace Engineering major. Expected graduation Spring 2008. Research Title: ÒProperties of Mach Disks from an Underexpanded Supersonic Nozzle for Cold FlowÓ [Mentor: Dr. Adonios Karpetis]. Abstract submitted to the 46th AIAA Aerospace Sciences Meeting and Exhibit, in Reno, NV on Jan. 7 - 10. Applying to grad school at TAMU.
Abstract: The objective of the work is to collect pressure and temperature data on Mach disks using Raman spectroscopy. Using this technique allows us to take measurements for pressure and temperature without introducing probes into the flow, and thus disrupting it. By using a Schlieren system we will find the location of the Mach disks and thus the appropriate position for the laser probe from which data is acquired. The spectroscopy techniques being used to take the measurements are the rotational and vibrational Raman scattering techniques using a Nd:YAG laser operating at a wavelength of 532 nm. This research focuses on measuring the properties of Mach disks in order to help further research in the area of supersonic combustion and improve present day high-speed air breathing engines..
Figure 5. (a) Mach disk formation diagram from Scott, J., ÒShock Diamonds and Mach Disks,Ó AerospaceWeb.org, URL: http://aerospaceweb.org/question/propulsion/q0224.shtml [cited 20 July 2007]. (b) Experimental Schlieren photographs of Mach disks from an underexpanded nozzle. Throat diameter 3 mm, exit diameter 5 mm, and back pressure of 60 psig.
Julia T. CossŽ, University of Rochester. Mechanical Engineering major. Expected graduation Spring 2008. Research Title: ÒRotational and Vibrational Raman line-imagingÓ [Mentor: Dr. Adonios Karpetis]. Applying to grad school at TAMU.
Abstract: The main objective of my work was to set up a combined rotational and vibrational line-imaging spontaneous Raman spectroscopy system using a 532 nm laser. This system will be used to study under-expanded supersonic jets for cold flow that are simultaneously being developed in the laboratory. By capturing the rotational Raman scattering we will be able to directly measure temperature. Through the vibrational Raman scattering we will be able to directly measure the number density of scattering species, which directly corresponds to the density and mass specific gas constant. Thus by knowing temperature and density we will be able to deduce the pressure of the under-expanded supersonic jet in a non-isobaric setting.
Figure 6. Raman spectroscopy system and the Raman spectrometer.
Richard A. Deresz, Working under the TAMU Undergraduate Summer Research Grant (http://essap.tamu.edu/usrg/) program and help integrate the non-TAMU students into TAMU and into the LBM-MHD modeling research team. Texas A&M University. Aerospace Engineering major. Expected graduation Spring 2008. Research Title: ÒAcceleration of a Plasma Flow by Oscillating Magnetic MirrorsÓ. [Mentor: Dr. Jacques C. Richard]. Abstract submitted to the 44th AIAA/ASME/ASEE/SAE Joint Propulsion Conference in Hartford, CT, July-20-24, 2008 (AIAA.org). Applying to grad school at TAMU.
Abstract: This paper explores the phenomenon of Fermi acceleration for possible application to plasma propulsion devices. A magnetohydrodynamic (MHD) model based upon the lattice-Boltzmann method (LBM) simulates plasma response to an externally applied magnetic field. The applied field configured as a magnetic bottle traps plasma injected into it via a rectangular jet. Moving the mirrors toward each other contracts the bottle and accelerates the plasma. Plasma loss from the magnetic bottle is a desired side effect constituting the propulsive mechanism for a device. Different time-varying magnetic fields, injection velocities, and magnetic Reynolds numbers are tested. Cases demonstrating high exit velocity in the axial direction and minimal reverse flow are examined. The best cases exhibit exit flow accelerated to approximately twice its injection velocity in the axial direction and three times the injection velocity in the radial direction. A magnetic field generated by loops with positions varying sinusoidally with time shows the best results. The field produced by loops with positions varying linearly with time and
return to their original positions in one time step is the most likely implementation for a propulsion or flow control device. Proportionally increasing injection velocity and current loop frequency does not increase exit velocity. The paper concludes that Fermi acceleration would be better used as a propulsion supplement and not as a stand-alone device.
Figure 7. (a) Magnetic field configuration of a magnetic bottle. (b) Computational configuration with boundary conditions for magnetic field and current loops at opposite ends of the domain. (c) Axial-direction magnetic field component of points on the centerline when the current loops positions vary linearly with time and approach once. (d) Axial-direction velocity component of points on the centerline when the current loops positions vary sinusoidally with time and approach once
Robert Fievisohn, Clarkson University. Mechanical Engineering major with a Math minor. Expected graduation Spring 2008. Research Title: ÒNew Solution of the Collision Operator for LBM Schemes Using a Spectral Solution for the Homogeneous Boltzmann Equation.Ó [Mentor: Dr. Jacques C. Richard]. Abstract submitted to the 5th International Conference on Mesoscopic Methods in Engineering and Science in Amsterdam, The Netherlands, June 16-20, 2008 (ICMMES.org). Applying to grad school at TAMU.
Abstract: Recent developments in mathematics have allowed for the collision process of the Boltzmann equation to be discretized using spectral methods. This new method can be coupled
with existing LBM schemes. Doing so allows us to stop using the BGK approximation and
get more accurate results for a broader range of problems. This paper discusses how to
incorporate the spectral solution as the collision operator for current LBM schemes along
with results for the new method.
Figure 8. Time evolution of the probability distribution function. The exact solution at time 0,3,6, and 9 is plotted as well as the points evaluated at each time using the spectral method.
Elizabeth M. Jesse, Embry-Riddle Aeronautical University. Aerospace Engineering major. Expected graduation Spring 2009. Research Title: ÒPreparation of the National Aerothermochemistry Lab Shock Tunnel and Ludwig Tunnel.Ó [Mentor: Dr. Rodney D. W. Bowersox]. Chi Mai, working under the TAMU Undergraduate Summer Research Grant (http://essap.tamu.edu/usrg/) program, and with Nicole Mendoza below, helped integrate the non-TAMU students into the wind tunnel and other facilities near the airport.
Abstract: shock tunnels use the shock wave caused by the bursting of a diaphragm to create high enthalpy, pressure and temperature flow. They are primarily used to study hypersonic flow and chemical reactions. During the summer of 2007, the shock tunnel was prepared for the National Aerothermochemistry Lab, which simulates up to Mach 10.7 flows. This included the construction of the chock tunnel as well as a program that allows users to predict the flow based on input characteristics. The program determines the path of the shock, between the driver and driven gases and expansion fans. It also determines the gas properties in each region and can tailor the input properties to give the best-run condition. This paper will focus on the expansion region of the flow as determined by the method of characteristics.
Figure 9. (a) Current Assembly of Ludwieg Tube. (b) Path of expansion waves or Òexpansion fansÓ in Ludwig tube.
Shalom Johnson, Texas A&M University. Aerospace Engineering major. Expected graduation Spring 2008. Research Title: ÒRotating Stall Suppression Using Oscillatory Blowing Actuation on Blades.Ó [Mentor: Dr. Paul Cizmas]. Abstract submitted to the 44th AIAA/ASME/ASEE/SAE Joint Propulsion Conference in Hartford, CT, July-20-24, 2008 (AIAA.org). Applying to grad school at TAMU.
Abstract: In the highly maneuverable aircraft designs of today, aircraft are required to have a
propulsion system (jet engine) that can operate during sudden accelerations and rapid changes in angle-of-attack. Consequently, the compressor of the jet engine occasionally must operate at low-flow rates and rapid changes in inlet conditions. The high-angle-of-attack and low-flow regime of compressor operation is often plagued by rotating stall and surge. Rotating stall and surge can result in loss of engine performance, rapid heating of the blades, and severe mechanical stresses. Traditional methods for suppressing rotating stall and surge require the use of complex variable inlet guide vanes and variable inlet stator vanes. The goal of this research is to design a stall suppression system that will introduce oscillatory blowing to energize the air in the boundary layer of one rotor blade (super blade). The energized boundary layer will increase the stall angle by several degrees; therefore, suppressing rotating stalls and surge. The objective of the research is to design the Òsuper bladeÓ and a new disk for the Texas A&M Propulsion laboratory compressor test rig.
Figure 10. (a) Variable inlet and stator vanes in the compressor at the front of a jet engine. (b) Expanded Assemble of the Disk, Blade, "super blade', and the pulse module. (c) Slice of the Disk, FEA stress results. (d) Blade + Dovetail, FEA Stress results. (e) Blade Internal Flow.
Nicole Mendoza, Texas A&M University. Aerospace Engineering major. Expected graduation Spring 2008. Research Title: ÒThe Tailored-Interface Hypersonic Shock Tunnel: Design and Operation[Mentor: Dr. Rodney D. W. Bowersox]. Applying to grad school at TAMU.
Abstract: Several tasks were completed during this summer session. The first and largest of the tasks was to write a MATLAB program to model the entire flow-field inside a high-enthalpy shock tunnel. The scope was limited to a seniorÕs knowledge of compressible flow. An x-t diagram was produced, as well as other useful output such as a ReynoldÕs number map, flight envelope graph, and capabilities data. A second task was to physically construct the shock tunnel itself. The entire support system was designed to give specific capabilities, then acquire and assemble the pieces. This also required a building to put the shock tunnel in, so a storage shed was converted to a working laboratory. The last task was to construct a Ludwig tunnel in the same manner. A short program was written to model the flow field inside the Ludwig tunnel. The scope of this paper will be primarily limited to discussing the incident and reflected shock portion of the tunnel and program, as well as explaining the tailoring conditions.
(d)
Figure 11. Illustration of path of shock wave (double line) and expansion waves or fans (sequence of lines) in tunnel. (a) Tunnel before diaphragm breaks. (b) Shock wave goes right while expansion waves go left. (c) Shock wave reflected off right end now goes left and expansion waves reflected off left end now go right. (d) Shock Tunnel.
Meghan M. Patzke, Texas A&M University. Aerospace Engineering major. Expected graduation Spring 2008. Research Title: ÒBlade Clocking of a Turbine with Cooling Effect.Ó [Mentor: Dr. Paul Cizmas]. Abstract submitted to the 44th AIAA/ASME/ASEE/SAE Joint Propulsion Conference in Hartford, CT, July-20-24, 2008 (AIAA.org). Applying to grad school at TAMU.
Abstract: Blade indexing, or clocking, has been implemented to maximize efficiency in turbine stages. Relative motion between rotor and stator airfoils leads to unsteadiness caused by potential flow and viscous interactions. An existing flow solver was used to analyze the results of clocking on rotor and stator airfoil effects. The CFD (computational fluid dynamics) flow solver uses an implicit approach to model flow and combustion in a turbine with Reynolds-averaged Navier-Stokes equations coupled with the species conservation equations. Future research includes evaluating clocking and cooling in a combustor and determining whether blade indexing is beneficial in this special case.
Figure 12. Detail of the O- and H-grid of the computational model of the turbomachinery.
David M. Williams, University of Michigan. Aerospace Engineering major. Expected graduation Spring 2008. Research Title: ÒIncorporating Gravity into the Equilibrium Distribution in a Lattice Boltzmann Scheme.Ó [Mentor: Dr. Jacques C. Richard]. Applying to grad school at TAMU.
Abstract: The standard form of the lattice Boltzmann method accounts for changes in velocity due to collisions between particles. Missing from this approach is a methodology for modeling body forces, in particular gravity. In this paper, we examine a technique which models gravity within the incompressible limit in a lattice Boltzmann method. Specifically, we examine the effectiveness of the addition of a gravitational potential term directly to the equilibrium velocity probability density function. A gravitational term modeled by Buick and Greated is adapted from a hexagonal grid to a rectangular grid. This term is incorporated into the equilibrium distribution, and the resulting lattice Boltzmann method is used to simulate a gravity-driven Poiseuille flow. The accuracy of this approach is examined under various Reynolds numbers and gravitational strengths. It is shown that the gravitational modifications to the lattice Boltzmann method are only accurate under extremely low Reynolds numbers, and low gravity conditions.
Figure 13. (a) Analytical Solution Compared to Numerical Solutions for Re = 5, 10, and 20 cases. (b) Analytical Solution and Numerical Comparison for g = 5 x 10-9, 1 x 10-8, and 5 x 10-8 cases.