Introducing CFD and Wind Tunnel Testing in an Undergraduate Fluid Mechanics Course

Journal of Energy and Power Engineering 9 (215) 55-516 doi: 1.17265/1934-8975/215.6.1 D DAVID PUBLISHING Introducing CFD and Wind Tunnel Testing in an Undergraduate Fluid Mechanics Course Wael Mokhtar and Shirley Fleischmann School of Engineering, Grand Valley State University, Grand Rapids, MI 4954, USA Received: February 27, 215 / Accepted: April 16, 215 / Published: June 3, 215. Abstract: CFD (computational fluid dynamics) is following the trend of CAD and FEA (finite element analysis) to undergraduate education especially with recent advances in commercial codes. It will soon take its place as an expected skill for new engineering graduates. CFD was added as a component to an experiment in a junior level fluid mechanics course. The objectives were to introduce CFD, as an analysis tool, to the students and to support the theoretical concepts of the course. The students were asked to complete an experimental two-dimensional study for a wing in a wind tunnel, to use CFD to simulate the flow, and to predict the aerodynamic lift using CFD as well as the experimentally obtained pressure distribution. In addition, they had to compare their results to published data for the studied wing. Details of the course, the wind tunnel test and the CFD simulations are presented. Samples from the students work are used in the discussion. The lab activities were successfully completed by the students and the learning objectives were well addressed. One of the valuable outcomes from this lab was the opportunity for the students to integrate multiple fluid mechanics analysis tools and learn about the limits for each tool. CFD also enhanced the learning in the lab activities and increased students interest in the subject. Key words: Undergraduate teaching, computational fluid dynamics, experimental fluid mechanics. 1. Introduction In the last two decades, engineering hands-on skills have expanded to include numerical methods. CAD was the first skill to make its way to industry. Later, FEA (finite element analysis) was added as a powerful design tool. It is now expected that, a mechanical engineering graduate will know the basics of CAD and FEA. Most engineering schools offer courses to support these design skills. Software developing companies are competing to develop more user-friendly packages with wider varieties of built-in tools to attract more users. From an educational point of view, it is very important to teach the students the limits of these numerical tools and stress the fact that physical testing is not being totally replaced by numerical simulation. The integration of experimental and numerical methods should be the goal of the users. Corresponding author: Wael Mokhtar, Ph.D., assistant director, research fields: teach methods, CFD and aerodynamics. E-mail: mokhtarw@gvsu.edu. CFD (computational fluid dynamics) is currently following the same track to be added to hands-on skills. Starting as a research tool taught at the graduate level, it has recently started to appear in undergraduate level courses. Mokhtar [1] introduced CFD to an undergraduate fluid mechanics course. In his course, he introduced basic concepts of CFD through a set of projects in the lab where the students built their skills by a practice-learning approach. He indicated that, the use of CFD supported the introduction of critical thinking and creativity. Later, Mokhtar [2] expanded his project-based learning to teach an undergraduate CFD course. Recently, Mokhtar [3] discussed the introduction of a wind energy project in his CFD undergraduate course. In his examples, Mokhtar presented assessment data to show the success of introducing CFD in the undergraduate level. He presented a balanced approach between software training and theoretical foundation. He used a commercial CFD package (Star CCM+) in these

56 Introducing CFD and Wind Tunnel Testing in an Undergraduate Fluid Mechanics Course courses. For more than a decade, instructors have been exploring the introduction of CFD to undergraduate courses, Hailey, et al. [4] introduced CFD concepts in junior level fluid mechanics course. Their goals were: to support the fluid mechanics topics of the course, to motivate the students to take the advanced CFD course, and to introduce the students to CFD tools used in summer internships. In their courses, they used an in-house CFD solver. The authors indicated the success of introducing the CFD at undergraduate level with some minor concerns on the students responses to the software. Sert, et al. [5] used CFD as a teaching tool for a basic fluid mechanics course. In their approach, the students were given virtual flow lab software (FlowLab) where they used a simplified version of CFD to solve basic problems. Details of the numerical methods were not the focus, instead, CFD general concepts such as meshing, boundary condition and flow properties were the main objectives. Blekhman [6] presented further evaluation of using FlowLab to support experiments in a fluid mechanics course. He also indicated the success of using CFD as a teaching tool for undergraduate students with a limited introduction for its concepts. Cummings, et al. [7] discussed the development of an undergraduate CFD aerodynamics course in the US Air Force Academy. The focus of the course was on the use of CFD as a practical tool. Topics such as accuracy, stability, meshing and turbulence modeling were covered in the course. Commercial software packages were used for meshing and post-processing such as GridGen, FieldView, Tecplot and MatLab. In-house panel method (vortex-lattice) software was used as solvers. The authors indicated that, advanced topics such as the details numerical algorithms and turbulence modeling were not deeply covered. They also reported that, the course helped the students to better understand fluid mechanics concepts and introduced them well to CFD as a design and analysis tool. In another example, Guessous, et al. [8] developed a hands-on undergraduate course in CFD. Through a set of experiments and the use of a commercial code (Fluent), senior students were introduced to CFD as a testing tool without getting into the details of it. The authors reported that, the course was successful and the learning objectives were met. The objectives of this approach were to introduce CFD as a black box tool and support the fluid mechanics concepts taught in the course. Navaz, et al. [9] presented two courses for undergraduate students where CFD was introduced. The first was an applied CFD course where basics of CFD including some solver details were covered in the lecture while in the lab the students used CFD commercial codes (Fluent) and a couple of pre- and post-processing tools. In the second course, CFD was utilized to teach compressible flow without focusing on the details for CFD methods. Both courses had experimental lab activities where CFD results were verified. They reported that, the two courses were successful to introduce the students to CFD and teach them through practice the limits of both computational and experimental tools. Bullough, et al. [1] developed modules in fluid mechanics lab where the students studied several designs of diffusers using CFD, experimental and analytical methods. Fluent was used for the CFD part and it was used as an analysis tool without many details about its theory. The objective was to teach the students the importance of using these three methods and to get them to observe the limits for each approach. The authors indicated that, their students learned through this hands-on experience some of the challenges and limits to generate accurate results using CFD. As a design tool, Stubley, et al. [11] introduced CFD to undergraduate students. In their course, a brief overview about CFD concepts was presented in the beginning of the course. Then, a set of pre-computed simulations was given to the students where the focus was to use this CFD tools to analyze complex flow. The simulations were made by a

Introducing CFD and Wind Tunnel Testing in an Undergraduate Fluid Mechanics Course 57 commercial package (CFX) and the students used the post-processing tool that came with the package in their study. The authors indicated that using CFD as teaching tool was helpful to add the realization and visualization aspect of fluid mechanics. Stern, et al. [12] presented the details of a virtual fluid mechanics lab that was developed by CFD. In this tool, details CFD concepts were not the main focus. The main objective was to provide fluid mechanics instructors with an easy to use teaching tool to support the concepts covered in their courses. Burban [13] used CFD as a design tool for a team of undergraduate students in the SAE (Society of Automotive Engineering) Super-Mileage competition. As he reported, over a year of mentoring, his team was successful to utilize CFD in the design of their vehicle. The focus on this case was the correct use of CFD with limited information about the theoretical concepts. Mazumder [14] mentored undergraduate students in a CFD research study. He reported that the students were successful to use a commercial package (Fluent) to complete a two phase flow study. He indicated that the students utilized CFD with limited knowledge in its theoretical details. Their results compared well with the published experimental data. 2. Present Approach It can be noticed that, over the last two decades, CFD has been introduced to undergraduate students. The advances in commercial software could be one of the reasons that helped this transition. Four forms of CFD introduction can be summarized as follows: an applied CFD course with limited background in CFD methods; project-based learning for CFD packages; introducing CFD as a teaching tool in undergraduate courses without focusing on its details; CFD as a design and research tool for extracurricular students activities. It is clear that CFD as an applied tool is ready for undergraduate students. In the present approach, CFD was introduced to support a lab activity in junior level undergraduate fluid mechanics course. Basics of CFD were introduced and the students completed a wind tunnel test for a wing. They then compared results from both the experimental and CFD study to the published data for the studied wing. Details of the lab philosophy are presented follow by the wing wind tunnel test details and the CFD component that was added. Results from the students work are also presented. 3. Design in the Fluid Mechanics Lab In the mechanical engineering program at Grand Valley State University, fluid mechanics course is taught in the end of the junior year as the second required course in the thermo-fluid track. The first course is thermodynamics and the last course is heat transfer. Throughout the thermo-fluid track, design skills are introduced in the form of design projects, lab activities and research topics. Fleischmann, et al. [15] presented more information about the design philosophy in the thermo-fluid track. Fluid mechanics is the first course to include a lab in the track. The lab is a three-hour session every week and the lecture part is also 3 h. The focus in this paper is on the lab portion. The lab activities support the theoretical part of the course and introduce design skills in the form of experimental design and a final design and build project. Each week, the lab starts with a brief introduction of the experiment objectives and the tools that can be used in the study. Then the students form teams and develop their experimental plan. After the instructor approval, each team preforms the experimental study. Then, each student is individually responsible to report this data and interpret the results. 4. Preparation for the Wing Experiment The wing experiment occurred at the middle of the semester. In preparation to understand the concept of determining force by integrating a surface pressure

58 Introducing CFD and Wind Tunnel Testing in an Undergraduate Fluid Mechanics Course distribution on flat as well as curved surfaces, the students completed homework problems and an experiment using a hydrostatic pressure distribution. This experiment was completed by the third week of the semester. Just before the wing experiment they completed a measurement of drag on a circular cylinder using an experimentally obtained surface pressure distribution. In this case, the model is very similar to the wing model in that it vertically spans the 1 foot by 1 foot test section using the top and bottom of the test section as end plates to assure 2-D flow. Pressure taps are located around the mid-section of the cylinder at 15 increments just as the surface pressure taps are located around the mid-section of the wing at various percent of chord on the top and bottom of the wing. And also with a 4 inch diameter, the cylinder shows blockage effects just as the wing does at high angles of attack. The models are physically similar. In our wind tunnel, the maximum Re D is 2.2 1 5 which, for a smooth cylinder assures a laminar attached boundary layer with a turbulent wake and an early separation point. The speed is close enough that a slight trip (a sand strip on the front) will transition the boundary layer and on half of the cylinder students will observe very low pressures associated with high acceleration in the flow outside the boundary layer affecting the attached flow. Flow separates much later on this half and follows the ideal flow (C P = 1 4sin 2 θ) much longer. This allows students to see what separated flow looks like so that they will recognize stall from the surface pressure distribution. They will recognize regions of favorable and unfavorable pressure gradient and the effects on the surface pressure distribution when flow stays attached longer as it will on an un-stalled wing. The blockage effect also clearly shows up as C P values even less than -3 as predicted by the ideal flow with perfect no-slip wall but only for the attached flow (the turbulent boundary layer side). The regular geometry of the cylinder enables the students to evaluate drag coefficient as an integral of the pressure coefficient times the cosine of the angle relative to forward stagnation. For the wing, with a more complex geometry especially at various angles of attack it is clear that, the lift can be evaluated by this method of integrating the pressure distribution but the drag would be difficult. This sets the stage for recognizing the benefit of CFD modeling. Because it is also easy to compare the predicted pressure coefficient for ideal flow to the experimental pressure coefficient, students are prepared to expect differences between viscous and inviscid solutions. This is easily obtained with the software used. 5. Wing Experiment At this point, students were ready to benefit from an experimental study of a two-dimensional wing (airfoil). The goal of the experiment was to calculate the aerodynamic lift by integrating the measured surface pressure. The study showed the effect of wind speed and angle of attack. The experiment was conducted in the open loop closed test section wind tunnel, as shown in Fig. 1. The model was mounted to a base plate in the floor of the test section that could be turned for a range of angles of attack. Fig. 2 shows a Clark-Y wing model that was used in the experiment. As was the case for the circular cylinder, the model vertically spanned the test section using the upper and lower walls as end plates to allow for the two dimensional study. The wing was equipped with 18 pressure openings located at %, 7.5%, 1%, 2%, 3%, 4%, 5%, 6% and 7% chord on both the upper and lower surfaces. Fig. 3 shows the wing inside the test section. The aerodynamic lift was covered in the lecture part of the course before this lab activity. Also the lab manual summarized the basics and the method to calculate the lift from the surface pressure coefficient. Fig. 4 shows part of the theoretical concepts that were given to the students in the beginning of the lab. Each team was asked to develop the details of their experiment including the wind speeds, angles of attack and wind tunnel settings.

Introducing CFD and Wind Tunnel Testing in an Undergraduate Fluid Mechanics Course 59 Lift FL dfl df cos surface top bottom Fig. 4 Basics presented to the students. 6. CFD Component df cos Fig. 1 A photo of the wind tunnel used in the study. Fig. 2 Clark-Y wing model. The wing experiment had been a part of the lab in fluid mechanics for many years however without a substantial comparison to inviscid flow solutions or other predictive tools such as CFD. The availability of CFD tools as well as free software that could be easily downloaded made this a natural improvement to the lab. An overview of basics of CFD was presented to the students using a commercial code (Star CCM+) that is used in the technical elective taught in the senior year [2]. The demonstration included the general features of an effective mesh and some of the capabilities and limits of CFD as an analysis tool. At this level, no further details about CFD algorithms and solver schemes could be presented. A case study similar to the wing experiment was used in the demonstration to make the discussion more relevant to the students. Figs. 5 and 6 show a zoom in for the two-dimensional mesh and part of the post-processing for the studied airfoil. The demonstration also included some the flow structures such as boundary layer separation, wake formation, stagnation regions and lift force calculation. Having a commercial code allows the use of flow animation and tools such as vector field and pressure and velocity contours to enhance the discussion. Fig. 3 The wing model inside the test section. Fig. 5 CFD class demonstration, zoom in mesh.

51 Introducing CFD and Wind Tunnel Testing in an Undergraduate Fluid Mechanics Course Fig. 6 CFD class demonstration, velocity contours. A commercial code (Star CCM+) was successfully used before to introduce CFD to undergraduate students in fluid mechanics course through a series of demonstrations similar to the one outlined here [1]. This training needed more time in the class and lab which was not case for this course. To shorten the needed training time and achieve the learning objectives, simplified software XFOIL was introduced. It is a panel method code that was developed by Drela [16] to study airfoil sections. XFOIL includes several features such as: boundary layer formation [17]; trailing edge treatment [18]; boundary layer tripping; meshing control tools; aerodynamic loads calculation; inviscid and viscous analysis; airfoil design tools (inverse method); and more features [19-21]. In the lab, basics of XFOIL use were presented to the students with a focus on the importance of meshing, inviscid and viscous modeling. The profile of the Clark-Y airfoil section was given to the students as well. Fig. 7 shows the airfoil section and examples of meshing (paneling) that could be used in XFOIL to add clustering near to the leading and trailing edges. 7. Published Data Another important part of the lab activities was to get the students to compare their experimental and CFD results to the published data. Clark-Y airfoil section has been used in low speed applications since the 193 s. The students were asked to research and report their references for the polar curves of this Fig. 7 Part of the XFOIL demonstration. airfoil. Among references, NASA (National Aeronautics and Space Administration) [22] reported wind tunnel data in 193 for a long list of airfoil section including Clark-Y. A more recent source was a book by Lyon, et al. [23] where they also reported data for many airfoil sections. XFOIL was one of the tools used besides wind tunnel testing in this book. 8. Students Work The students were divided into teams where each team developed the experiment plan and complete the wind tunnel test for the wing. Each team was from four to five students. Fig. 8 shows one of the teams during the wind tunnel test. Each student was asked to complete the CFD study, research the published data and interpret all the data in a written report. In this section, samples from the students results will be discussed to observe the level students responses to the lab activities. Parameters that each team decided were: range of wind speeds; range of angle of attack; wind tunnel accuracy; similarity requirements between CFD and wind tunnel test; sources for published data. Figs. 9-14 show the pressure coefficient distribution reported by three students. Student A presented the upper and lower surface in separate graphs for three

Introducing CFD and Wind Tunnel Testing in an Undergraduate Fluid Mechanics Course 511 C P Chord position (ft) Fig. 11 Pressure coefficient as function of position for α of, student B. Fig. 8 One of the students during the wind tunnel testing. 1. Pressure coefficient -1. 1. α = α = 1 α = -1 C P 15 15-2. Fig. 9 Pressure distribution on upper surface, student A. Chord position (ft) Fig. 12 Pressure coefficient as function of position for α of 15, student B. Percent of chord Pressure coefficient α = α = 1 α = -1 C P Fig. 1 Pressure distribution on lower surface, student A. Fig. 13 Pressure distribution for α of, student C.

512 Introducing CFD and Wind Tunnel Testing in an Undergraduate Fluid Mechanics Course Percent of chord C P Fig. 14 Pressure distribution for α of 14º, student C. angles of attack:, 1 and -1. Student B reported separate graphs angles of attack:, 7 and 15. The third student reported angles of attack:, 4 and 14. Student A studied positive and negative values for the angle of attack which is suitable for cambered airfoil section such as Clark-Y. Students B and C ran a high angle of cases to explore the stall conditions. Student C ran two sets of testing through these angles with wind speeds: 8 mph and 1 mph to report the effect of the wind speed. The first two students did not report that part of the study. Students A and B plotted the pressure coefficient C p in regular scale while the usual practice in aeronautical application is to have C p presented in a reverse scale as the way student C reported his data. To run a CFD simulation using XFOIL, the students had to meet the similarity conditions between the two cases. Reynolds (Re) and Mach (M) numbers were calculated for the tested wing and the same values were used in the simulations. Figs. 15-2 show samples of the CFD results reported by the students using XFOIL. The software allows both inviscid and viscous modeling. Students A and B used viscous modeling and student C used inviscid modeling, all of them used the right modeling condition (Re and M). It can be seen that, the measured pressure distribution compared well with the CFD results. The next step in the study was to calculate the lift coefficient C L. The CFD software, XFOIL, reports the Fig. 15 Simulated pressure coefficient for an angle of attack α =, XFOIL results, student A. Fig. 16 Simulated pressure coefficient for an angle of attack α of 1, XFOIL results, student A. Fig. 17 Simulated pressure coefficient for an angle of attack α of 7, XFOIL results, student B.

Introducing CFD and Wind Tunnel Testing in an Undergraduate Fluid Mechanics Course 513 Fig. 18 Simulated pressure coefficient for an angle of attack α of 15, XFOIL results, student B. values directly. For the wind tunnel test, the students had to calculate the lift coefficient by integrating the measure pressure distribution. Fig. 21 shows the set of equations reported by one of the students for calculating C L. Tables 1-4 show samples of the lift coefficients reports by four students. The first student compared their results to the published data while the rest compared the wind tunnel and the CFD results. All of them commented on the reasons of differences between the data. One of the lessons they learn at this part was that, each method had its own sources uncertainty. Below are some the error sources: accuracy of pressure measurements, wind tunnel; blockage interference at high angle of attack, wind tunnel; mesh refinements near to leading and trailing edge, CFD; large separation regions, CFD. 9. Further Discussion and Assessment Fig. 19 Simulated pressure coefficient for an angle of attack α of, XFOIL results, student C. Fig. 2 Simulated pressure coefficient for an angle of attack α of 14, XFOIL results, student C. Performing a fluid mechanics study using a wind tunnel and CFD and verifying the results against published data were the main goal of this experiment. In the process, the students got to learn about the limits and accuracy of each method. In this section, some of their concluding statements will be discussed. Surface d d cos top d cos bottom d cos d cos d cos cos d shown in Fig. 4 d d d d d d d cos shown in Fig. 4 d cos.5 2 d cos d cos.5 2 cos d Fig. 21 Calculation of the drag coefficient, student sample.

514 Introducing CFD and Wind Tunnel Testing in an Undergraduate Fluid Mechanics Course Table 1 Lift coeficient, student A. Angle of attack 1-1 Experimental.81 1.34 -.16 Published.42 1.25 -.5 CFD analysis.435 1.6285 -.7721 Table 2 Lift coeficient, student B. Angle of attack Experimental C L Theoretical C L.4987.435 4 1.129.916 14 1.8916 2.941 Table 3 Lift coeficient, student C. 95 ft/s 15 ft/s -.77 -.71 6 -.8 -.73 1 -.8 -.77 Table 4 Lift coeficient, student D. α EXP XFOIL Error Uncertainty.882.47 87.7 16. 1 1.49 1.36 14.1 18.4-1.834-7.52-98.9 13.8 One of the students reported in his conclusion the following statement: Simulations provide an effective tool for evaluating the validity of experimental measurements and vice versa, but regions of separation and unpredictable flow are very difficult to model correctly. It is clear that, he learned through the practice, the limits of a CFD study and that there are some flow conditions where wind tunnel testing will be more accurate than CFD. Another student commented about the wind tunnel blockage: The cross-sectional area of the wind tunnel in which the airfoil was placed was 1 foot by 1, and when the angle of attack for the airfoil was adjusted to ±1, it took up a considerable amount of cross-sectional area. This set-up could cause interference between the affected boundary layer around the wing and the sides of the tunnel and in return could affect the pressure at the wing The limits of wind tunnel testing and the size of the tested model is one of the standard constraints for accurate testing. It is clear that, this student got this part when comparing his results to CFD and published data. In another student s comment about the accuracy in calculating the lift coefficient by integrating the pressure distribution, he reported: From the previously presented data, the accuracy of the coefficient of lift increased as the angle of attack increased. This was due to the difficulty of numerically integrating C p versus chord length plots that have outliers, which are present at the stagnation point in situations with a smaller angle of attack Some of the students focused on one source of error while others commented on more than one source of uncertainty. They all realized the value of the integration of CFD and wind tunnel. It is also important to mention that, some of the students did not get that deep in the discussion and got busy with reporting the results without focusing on the big picture. However, this group also completed both

Introducing CFD and Wind Tunnel Testing in an Undergraduate Fluid Mechanics Course 515 sides of the lab activities successfully. The authors ran this lab activities several times in the last couple of years. From their point of view, this integrated lab activities (CFD and wind tunnel) enriched the lab experience for the students and expanded their view for value of both tools. Nearly all the students learned the CFD software after one demonstration with minimal assistance after that. The lab also encouraged many students to take the CFD technical elective in their senior year. One of the students who completed this lab, student A in the previous sections, took the CFD technical elective and another independent study in CFD before his graduation. Another student got even more interested in CFD and he is currently working on his master degree in CFD applications. 1. Conclusions A lab activity was designed to integrate CFD and wind tunnel testing in an undergraduate fluid mechanics course. Details of the lab were discussed through samples from the students work. This lab activity was repeated several times in the last couple of years. The majority of the students were attracted to the CFD tool. Many of them learned the limits of using CFD and the accuracy of running a wind tunnel test to complete a fluid mechanics study. Positive feedback from both the students and the instructors about this lab activity was observed. As a final remark, CFD can serve as a good tool to support junior level fluid mechanics classes. This work is an example of that use. Usually, students learn this type of software very fast and that saves time for the instructors to focus on theoretical parts of the course. In addition, CFD provides the students with a good visualization tool to enhance the learning environment. Acknowledgments The authors would like to thank the students who participated in the lab activity and provided the samples presented in this paper. References [1] Mokhtar, W. 21. Using Computational Fluid Dynamics to Introduce Critical Thinking and Creativity in an Undergraduate Engineering Course. The International Journal of Learning 17 (9): 441-58. [2] Mokhtar, W. 211. PBL (Project-Based Learning) An Effective Tool to Teach an Undergraduate CFD Course. In Proceedings of the ASEE (American Society for Engineering Education) Annual Conference, 973. [3] Mokhtar, W. 214. Introducing Wind Energy to an Undergraduate CFD Course. In Proceedings of the ASEE North Central Section Conference, 11. [4] Hailey, C., and Spall, R. 2. An Introduction of CFD into the Undergraduate Engineering Program. In Proceedings of the ASEE Annual Conference, 1556. [5] Sert, C., and Nakiboglu, G. 27. Use of CFD (Computational Fluid Dynamics) in Teaching Fluid Mechanics. In Proceedings of the ASEE Annual Conference, 156. [6] Blekhman, D. 27. Lessons Learned in Adopting a CFD Package. In Proceedings of the ASEE Annual Conference, 83. [7] Cummings, R., and Morton, S. 25. Computational Aerodynamics Goes to School: A Course in CFD for Undergraduate Students. In Proceedings of the 43rd AIAA (American Institute of Aeronautics and Astronautics) Aerospace Sciences Meeting and Exhibit, 172. [8] Guessous, L., Bozinoski, R., Kouba, R., and Woodward, D. 23. Combining Experiments with Numerical Simulations in the Teaching of Computational Fluid Dynamics. In Proceedings of the ASEE Annual Conference, 222. [9] Navaz, H., Henderson, B., Berg, R., and Nekcoei, S. 2. A New Approach to Teaching Undergraduate Thermal/Fluid Sciences Courses in Applied Computational Fluid Dynamics and Compressible Flow. International Journal of Mechanical Engineering Education 3 (1): 35-49. [1] Bullough, W., Hart, J., and Chin, S. 23. Comparative Studies: CFD, Experimental and Analytical Techniques in the Fluids Laboratory. International Journal of Mechanical Engineering Education 31 (2): 15-9. [11] Stubley, G., and Hutchinson, B. 22. The Role of Case Studies in CFD Education. ASME (American Society of Mechanical Engineers). [12] Stern, F., Xing, T., Yarbrough, D., Rothmayer, A., Rajagopalan, G., Otta, S., Caughey, D., Bhaskaran, R., Smith, S., Hutchings, B., and Moeykens, S. 24. Development of Hands-on CFD Educational Inter Face

516 Introducing CFD and Wind Tunnel Testing in an Undergraduate Fluid Mechanics Course for Undergraduate Engineering Courses and Laboratories. In Proceedings of the ASEE AC (Annual Conference) Conference, 1526. [13] Burban, P., Hegna, H., and Zavodney, L. 28. Role of CFD in Undergraduate Research in the Design of a Supermileage Competition Vehicle. Presented at the ASEE NC (North Central) Conference, Arizona, USA. [14] Mazumder, Q. 29. Integration of Computational Fluid Dynamics Analysis in Undergraduate Research Program. In Proceedings of the ASEE North Central Regional Conference, 73. [15] Fleischmann, S., Sozen, M., and Mokhtar, W. 21. A Green Heat Transfer Design Project to Introduce Globalization and Society Awareness. In Proceedings of the ASME 21 International Mechanical Engineering Congress & Exposition, IMECE21-38285. [16] Drela, M. 1989. XFOIL: An Analysis and Design System for Low Reynolds Number Airfoils. In Proceedings of the Conference on Low Reynolds Number Airfoil Aerodynamics, 1-12. [17] Drela, M. and Giles, M. B. 1987. Viscous-Inviscid Analysis of Transonic and Low Reynolds Number Airfoils. AIAA Journal 25 (1): 1347-55. [18] Drela, M. 1989. Integral Boundary Layer Formulation for Blunt Trailing Edges. Presented at the 7th AIAA Applied Aerodynamics Conference, Seattle, USA. [19] Drela, M. 199. Elements of Airfoil Design Methodology. In Applied Computational Aerodynamics, edited by Henne, P. Vol. 125. Washington, DC: AIAA Inc., 167-89. [2] Drela, M. 1988. Low-Reynolds Number Airfoil Design for the MIT Daedalus Prototype: A Case Study. Journal of Aircraft 25 (8):724-32. [21] Drela, M. 1998. Pros and Cons of Airfoil Optimization. In Frontiers of Computational Fluid Dynamics, edited by Caughey, D. A., and Hafez, M. M. Edge, NJ: World Scientific Publishing Co. Pte. Ltd. [22] Louden, F. 193. Collection of Wind-Tunnel Data on Commonly Used Wind Sections. NASA report. [23] Lyon, C., Broeren, A., Giguere, P., Gopalarathnam, A., and Selig, M. 1997. Summary of Low-Speed Airfoil Data. Vol. 3. Virginia: SoarTech Publications.