on the use of computers for teaching fluid mechanics · pdf filecomputers can assist the...

NASA Technical Memorandum 106587 ft

On the Use of Computers for TeachingFluid Mechanics

Thomas J. BensonLewis Research CenterCleveland, Ohio

Prepared for theASME Fluids Engineering Division Summer Meetingsponsored by the American Society for Mechanical EngineersLake Tahoe, Nevada, June 19-23, 1994

(NASA-TM-106587) ON THE USE OF N94-32773COMPUTERS FOR TEACHING FLUIDMECHANICS (NASA. Lewis Research

National Aeronautics and Center) 18 p UnclasSpace Administration

63/61 0005514

https://ntrs.nasa.gov/search.jsp?R=19940028267 2018-04-28T16:33:38+00:00Z

On the Use of Computers for Teaching Fluid Mechanics

Thomas J. Benson*Internal Fluid Mechanics Division

NASA Lewis Research CenterCleveland, Ohio 44135

April 8, 1994

ABSTRACT

Several approaches for improving the teaching ofbasic fluid mechanics using computers are presented.There are two objectives to these approaches; to in-crease the involvement of the student in the learningprocess, and to present information to the studentin a variety of forms. Items discussed include: thepreparation of educational videos using the resultsof computational fluid dynamics (CFD) calculations,the analysis of CFD flow solutions using worksta-tion based post-processing graphics packages, andthe development of workstation or personal com-puter based simulators which behave like desk topwind tunnels. Examples of these approaches are pre-sented along with observations from working withundergraduate co-ops. Possible problems in the im-plementation of these approaches as well as solutionsto these problems are also discussed.

INTRODUCTION

Recent advances in computer related technologiescould revolutionize the teaching of undergraduatefluid mechanics in the twenty first century. Theworld wide networking of computers is ushering in anew age of information and communication. The in-troduction of CD-ROM storage devices have enabledentire encyclopedias to be stored and accessed froma small disc. The availability, computing power, andmemory capacity of personal computers exceeds themainframes of the past. Advances in mainframe sizeand speed, computing algorithms and software, andparallel processing have enabled the computational

'Senior Research Engineer°This paper is declared a work of the U.S. Government and

is not subject to copyright protection in the United States.

analysis of ever more complex problems. The cou-pling of this hardware and software growth with newuser-friendly operating systems, windowing capabil-ity, and the development of graphical user interfaces(GUI) has created an entirely new computing envi-ronment. Interestingly, it is an environment in whichundergraduate students are quite comfortable, hav-ing grown up with video games and personal com-puters at school or in the home. The challenge foreducators is to learn, develop, and apply the newtechnology to the traditional problems of fundamen-tal understanding in fluid mechanics.

The teaching of undergraduate fluid mechanics hasalways been a two phase process; the developmentof intellectual or mental knowledge, and the devel-opment of intuitive or gut knowledge. Intellectualknowledge would include the recognition of termi-nology, concepts, basic assumptions, and the deriva-tion of the appropriate flow equations. Intuitiveknowledge would include the recognition of the re-lationships between variables and the understand-ing of how things worked. Lecture classes most of-ten provides intellectual knowledge while laboratoryclasses provide intuitive knowledge. When the au-thor was an undergraduate, one developed an intel-lectual understanding of boundary layer theory us-ing Schlichting (1968), and developed an intuitiveunderstanding using the movies by Shapiro (1961).Computational tools to assist in the development ofintellectual knowledge would include symbol manip-ulation packages, such as MACSYMA (Rand(1984)),which could aid in the derivation of equations. Onecould use the storage, retrieval, display, and network-ing of computers to make basic information morequickly and readily available to students. There are,for example, software packages now available in the

medical profession which allow the student to viewanatomical pictures quickly and interactively. Simi-lar types of software could be developed for fluid me-chanics which would store and display existing texts.

It is, however, in the development of intuitiveknowledge that computers could play a major role,particularly at a time when laboratories are becom-ing even more expensive to operate and maintain.Personal computers and workstations are particu-larly well suited for the development of intuitiveknowledge. They are readily available, relatively in-expensive, and becoming much easier to use. Theoutput from a modern personal computer is princi-pally visual; the most perceptive sense used in educa-tion. Some more advanced machines include audio,which can reinforce the lesson. The proper softwarecan make computers highly interactive, causing thestudent to respond in some fashion to the lesson andnot merely observe it. With motion available on thecomputer screen, the medium is truly three dimen-sional; two space dimensions plus time, or the illu-sion of three space dimensions. Traditional books,paper and chalkboards are only a two dimensionalmedium. The added dimension available in the com-puter screen can dramatically change the student'sperception and understanding of the fluids problem.

Computers can assist the teaching of fluid mechan-ics in several ways. The output from large CFDcodes are often displayed at conferences throughvideo tapes. Videos which highlight a fundamentalfluids problem can now be used in the normal class-room environment. For fundamental flow problems,dedicated computations and videos can be producedsolely for educational purposes. Videos are producedusing the output from a CFD code as input to agraphics package running on a workstation. It is pos-sible for the CFD output data sets and the graphicspackages to be made available to universities for stu-dent use. With the appropriate scripts and some ex-perience, the student can use the graphics package toexplore the flow solution and develop the student'sown understanding of the flow phenomenon. Con-sidering the computing power of some personal com-puters and workstations it is now possible to producesoftware packages which can both solve and graph-ically display basic fluids problems on the work-station. Several personal computer based softwarepackages have been developed recently (Koening andHodge, 1993; Fox and McDonald, 1992; and Mat-

tingly, et.al., 1987) to assist in the teaching of fluidmechanics and propulsion. This paper will presentsome ideas and examples of two packages by Benson(1994a and 1994b) which are under development atthe NASA Lewis Research Center for the teaching offluid mechanics to undergraduate and graduate stu-dents. Eventually the large CFD codes can also bemade available for students to formulate solutions toflow problems.

A particularly intriguing approach to using com-puters in education is the development of flow sim-ulators which act like desk top wind tunnels. On aworkstation, the student is presented with a physicsproblem in one view window. In another window isa control panel which the student uses to vary theconditions of the physics problem. In a third win-dow raw output data from the problem is presentedto the student. The student can then select whichdata to analyze, and in another window record thedata and plot the data on a view screen. If the prob-lem is properly chosen, the equations governing theflow can be almost instantly solved on the worksta-tion and the entire package behaves like a wind tun-nel simulator. A preliminary flow simulator has beendeveloped by Benson (1994a) for the basic problemof supersonic flow past wedges. It could be rathereasily modified to consider alternate physics prob-lems, such as a supersonic nozzle, a sting-mountedwedge at angle of attack, or a diamond airfoil, Thebasics of the code could be retained and other classesof fluids problems coded, including potential flow,conformal mapping, shock tubes, stability and con-trol, and basic airfoil theory. Expanding the idea toother areas of science, basic thermodynamics, mag-netics, electrostatics and optics problems also easilylend themselves to presentation in this way.

EDUCATIONAL VIDEOS

Educational videos can be used to supplement reg-ular classroom lessons on various flow phenomenaif several problems are overcome. Since the ad-vent of the supercomputers in the early 1980's manyhighly detailed three dimensional flow calculationshave been performed and the results presented attechnical meetings by means of video tape. Thereare, however, several problems associated with thesevideos. The first problem is identifying what videosexist for particular flow solutions since a video pre-sentation is seldom noted in conference proceedings

or schedules. This problem could be overcome by us-ing the Internet system to build a catalog of existingvideos. Inquiries could be posted in various news-groups and the results catalogued for future refer-ence. The second problem is that flow solutions pre-sented are often too complicated for undergraduates.For instance a video describing flow around the fore-body of a fighter aircraft contains too many phenom-ena (vortex shedding, boundary layer buildup, shockwaves, separated flows, ...) for detailed understand-ing by undergraduates. Single phenomenon calcula-tions are preferred and this may require that new cal-culations of selected problems be performed. The se-lection of the appropriate problems for study must beprovided by the educational community. The thirdproblem is the variation in style and content of exist-ing videos. For instance, some videos may be work-ing level tapes; without captions, problem descrip-tions, voice over, or any other kind of explanation,while others might be described as public relationstapes with lots of music and visual impact but littletechnical value. A possible solution to this problemwould be the preparation of a series of videos bysome centralized group with the videos prepared un-der guidelines provided by the educational commu-nity. Old videos can be reworked in a standardizedstyle.

To demonstrate some of the possibilities of usingCFD to prepare videos for educational purposes, avideo has been prepared which shows in great detailthe periodic vortex shedding from a circular cylin-der, the Karman vortex street. The calculation ofthe flow field was performed by Kim and Benson(1992) as part of a code development and validationstudy. The flow is incompressible, at Reynolds num-ber of 100 based on free stream velocity and cylinderdiameter and the calculated Strouhal number is .16.The video was prepared at the NASA Lewis Graphicsand Visualization Laboratory using FAST, the FlowAnalysis Software Toolkit, (Walatka et.al. (1994)).The video runs about 30 minutes, without music orvoice-over, but with several still shots separating theaction and describing the flow variables and sceneswhich follow. The video presents the same flow fieldin several different ways; velocity vectors, stream-lines, particle traces, color-tagged particles, vorticitycontours, and injected fluids. This flow field is timedependent and the video allows the student to visu-alize the flow field changing with time; the figures

in this report cannot even begin to present this in-formation. Color is used in the videos to highlightvarious features and this aspect is also not availablein this paper. Each of the different representationspresent the viewer with information not available inthe other representations. For instance, the particletraces shown in Figure 1 demonstrate the formationof the top and bottom alternately shed vortices. Inthe video, the mixing of the flow at the rear of thecylinder is quite apparent. Figure 2 shows the sameflow field at the same time but now contour plotsof vorticity are shown. The student can see the highshear regions on the front of the cylinder and the factthat the vorticity is quickly dissipated downstream,facts not available in the particle traces of Figure 1.In Figure 3, fluid is injected from inside the cylin-der and convected downstream. This technique canbe compared to experiments, but again presents adifferent view of what occurs in the flow field com-pared to Figure 1 or Figure 2. Combining all of thedifferent representations, the student can come to abetter intuitive understanding of this physical pro-cess than is available from any one presentation, orfrom a textbook picture.

COMPUTER GRAPHICS

Computer graphics have been used for many yearsto display the output from large CFD codes. Thetrend in recent years has been towards workstationbased highly interactive packages. These packagespost-process data sets which are created by the flowsolver and contain the results of the flow calculation.Because the two processes of analyzing the resultsand producing the results have been split, a studentcan learn the basic physics of certain flow problemswithout first learning all of the intricacies of CFD.The flow results can be computed by someone else.All that is required for this type of educational ex-perience is a workstation, a graphics package, thecomputed data sets, and some experience with thesoftware and hardware. As previously noted, work-stations are becoming much more available and inex-pensive, and many universities now have computinglabs equipped with workstations. The new graphicspackages, such as FAST, come equipped with on-linedemos, scripts, and some flow data sets. Experiencewith co-op students at NASA Lewis has shown thatundergraduates can quickly learn these new packages

and usually show great interest and initiative in us-ing the packages.

As with the educational videos, there are someproblems to be overcome before computer graph-ics can be widely used in undergraduate education.Most computer graphics packages are licensed soft-ware which must be purchased from vendors and areusually expensive. Perhaps the use of these pack-ages for educational, not commercial, purposes andthe promise that a whole generation of fluid dynam-icists would be schooled in the use of some packagecould lower the expense. Finding the data sets toanalyze is a problem which, like the videos, couldbe solved either through an intense literature searchor through an Internet broadcast. The use of threedimensional flow fields is particularly encouraged inthis effort although the storage requirements can bequite high. Like the videos, it may be desirable torequest certain fundamental flow calculations to becompleted solely for the purpose of producing datasets for analysis. Once the appropriate data sets arelocated or generated, it may be desirable to storeand maintain them at some central location, eithera university or a government workstation accessiblevia the Internet. It may also be desirable to createsome beginner's scripts for the graphics package tomaximize the efficiency of the user's time. However,the central focus of using computer graphics shouldalways be to maximize the student involvement inthe analysis. The interaction of the student and theflow problem will produce the desired result.

As an example of the types of data sets and flowproblems which have been analyzed by undergrad-uate co-ops at NASA Lewis, Figures 4 and 5 showMach 3.0 viscous flow past a sharp fin on a flat plateas computed by Anderson and Benson (1983). Fig-ures 6 and 7 show subsonic viscous flow past a jet incross-flow as computed by Kim and Benson (1993).The graphics package used in both analyses is FAST,although undergraduate co-ops have also studied thesharp fin problem using the older PLOT3D package.

The sharp fin case is interesting because it involvesa three dimensional shock wave boundary layer inter-action. Figure 4 shows a black and white represen-tation of a colored screen dump from a workstation.The figure shows the fin geometry at the top, the flatplate at the bottom and polygon-filled contour plotsof Mach number at three computational planes. Theco-ops learned to recognize shock waves in the free

stream from the abrupt change in color, boundarylayers along the walls from the dark contours nearthe surface. The co-ops could see the interactionbetween the shock and boundary layer in the thick-ening of the boundary layer in the vicinity of theshock, the thinning back towards the wedge, and thecurving of the shock in the vicinity of the wall. Fig-ure 5 shows the result of particle traces near the wallwhich produces several well known features includingthe line of coalescence, and the downstream conicalarea. The students create these patterns by invok-ing the streamline generator and placing a "probe"at some points in the flow field interactively as in-dicated by the white cross hairs in Figure 5. Thestudent can then view the results from different ori-entations by using a mouse to translate or rotate thepicture.

The jet in cross-flow problem is interesting becausethe geometry is quite simple and the flow field con-tains secondary vortices and boundary layer separa-tion and re-attachment. Figure 6 shows two views ofthe secondary vortices generated by the interactionof the injected flow with the approaching boundarylayer. The top view shows a half symmetry viewfrom the side while the bottom view is from upstreamlooking across the injection jet. The co-ops often re-late this flow pattern to smoke issuing forth from achimney on a windy day. Figure 7 shows a topolog-ical analysis of the flow separation around the jet.The view is down onto the flat plate and the sepa-ration and reattachment lines are noted. The co-opwho produced this figure was learning topology whilestudying fluid mechanics and this flow problem andgraphics package satisfied both pursuits.

FLOW SIMULATORS

The newest and most promising effort to use com-puters in the teaching of fluid mechanics involvesthe use of flow simulators for specific fluids prob-lems. Because of the increased computing power ofmodern personal computers and workstations it ispossible to build packages which solve the flow equa-tions for simplified problems nearly instantaneously.Coupling the solver to a GUI produces a flow simu-lator. The student can then develop the understand-ing for the flow problem through interaction with thesimulator, much as students formerly developed un-derstanding through laboratory courses. The GUIalso allows the results of the flow calculations to be

presented to the student in several different formswhich assists in the understanding of basic flow phe-nomenon.

In the past eight months two simulators have beendeveloped: the first solves for supersonic flow pasta single wedge, two opposed wedges, or two wedgesin series; the second solves for the flow in two di-mensional external or mixed compression inlets. Thefirst package is intended for undergraduate educa-tion and contains some unique features to encour-age the students interaction with the package. Thesecond simulator is intended for graduate educationor for preliminary design in industry. In this dis-cussion, we shall concentrate on the first simulatorand the reader can find more details of preliminaryversions of both simulators in Benson (1994a and1994b). Some examples of the results of the simu-lators are given in Figures 8 through 14 which arescreen dumps from the simulators.

Figure 8 shows the basic layout of the simulatorwhich is divided into four main sections: the mainview window is in the upper left, the plotter viewwindow is in the upper right, the output box is inthe lower right, and the input box is in the lowerleft. The main view window shows the geometry,the shock (or expansion), and labels for the hingeand the wedge. On the workstation, these featuresare color-coded, but are presented here in black andwhite. The wedge appears as the lower nearly hor-izontal line with the small semi-circle denoting thehinge location. The upstream flat portion is tagged"0" while the movable wedge is tagged "1". Theshock appears as the upper nearly diagonal line orig-inating at the hinge location and flow is from leftto right above the wedge surface. The plotter viewwindow is located to the right of the main view win-dow. The student selects which sets of variables toplot using the input box. The computed output flowconditions are displayed in the output box below theplotter view window. This data is displayed in twoways; numerically in the row of boxes at the left, andas bar charts to the right. Each bar is a different colorcorresponding to a different flow variable. As theflow conditions are changed in the input box, the re-calculated numbers are displayed and the bar chartsmove much like a thermometer. This type of visualoutput allows the student to immediately sense inwhat direction the flow variables change and by howmuch for a given input. The input box is located to

the left of the output box. It includes some buttonsto select a problem for study and four sub-panels tovary conditions in the problem. The contents of thesub-panels depend on the problem chosen for study.In Figure 8 a single wedge problem is indicated by thedarkened "light" on the button and only three of thefour sub-panels are required for input. The first sub-panel controls geometry and free stream flow condi-tions. The lower two sub-panels are used to controlthe plotting of data; two sub-panels are used to allowflexibility in the types of plots one can generate fordifferent physics problems.

When generating plots, the simulator behaves likea desk top wind tunnel. The student chooses theindependent and dependent variables for plotting inthe sub-panel marked "Variables". As the choices aremade, the axes in the plotter view window are auto-matically labeled and scaled. In Figure 9, the studenthas chosen to plot static pressure ratio versus wedgeangle as indicated by the lights on the buttons. TheMach number then becomes a parameter for gener-ating curves. The student sets a value of Mach num-ber then varies the wedge angle to any desired valueand presses the "Take Data" button. At this point a"*" appears on the graph corresponding to the cho-sen value of independent variable (wedge angle) andcalculated value of dependent variable (pressure ra-tio). The student then uses the input sub-panel toselect a new wedge angle and again "Take Data". Anew point appears on the plot and the procedure isrepeated to a maximum of twenty five data points.The data can be taken in any order, so the studentcan fill in interesting portions of the curve. In Figure9, six data points have been taken. When the stu-dent has completed a trace, the "End Trace" buttonis pushed, a solid color coded line is drawn throughthe data, and a colored label with the value of theparameter is affixed as shown in Figure 10. The stu-dent can then choose a new value of the parameterand begin a new trace as before. Figure 10 showsthe screen dump after the student has begun a newtrace at Mach number equal to 2.0. In this exam-ple, ten data points have been taken along the newtrace, and the old trace has been labeled in the upperright corner of the plotter window. The student canput up to five traces on the plotter. To begin a newplot, the student pushes the "New Plot" button, theold plots are erased, the count boxes for traces anddata are reset to zero, and the "Variables" buttons

can be used to pick new independent and dependentvariables.

This simulator can present different results to thestudent depending on the flow problem. In Figure 11,supersonic flow past opposed wedges has been cho-sen. The sub-panels are different than before sincethe student can now vary the angle of the secondwedge and the distance between the wedges. Thedisplay in the main view window now contains thesecond wedge labeled near the top of the figure andthe multiple flow zones are also tagged. This flowproblem includes a slip line between zones 3 and 4which is noted by the dashed line, and the zone selec-tion button in the output window can now be usedto display the values of all the flow variables in anygiven zone; in this example zone 3 is displayed. Be-cause of the added complexity of this problem, theplotter is used in a different way than it was for thesingle wedge problem. When the plotter is invoked,as shown in Figure 12, a set of cross-hairs appear inthe view window. These cross-hairs can be moved byusing the sliders in the window labeled "Probe". Thestudent can then produce plots of flow variable versusposition rather than flow variable versus Mach num-ber or wedge angle as in the single wedge problem.For this example, static pressure versus X-distancealong the ramp has been selected and data pointshave been recorded in the plotter. The process ofrecording the data is the same as in the previousproblem.

No educational tool would be complete withoutan examination; a question and answer box has beenadded at the lower right corner as shown in Figure 13.The questions and answers are stored in a data setwhich the simulator accesses. Teachers (or for thatmatter students!) can edit this data set and add,modify, or delete questions and answers as required.The questions can appear in the window either se-quentially or randomly as chosen by the student withthe appropriate buttons. The current sequential setof questions and answers have been chosen to pro-mote an interaction of the student with the simulatorand to bring out several important topics regardingcompressible aerodynamics. To use this feature thestudent pushes the "Question" button for a ques-tion, uses the simulator to obtain an answer, thenpresses the "Answer" button to check the answer. Ascurrently configured, the simulator runs through thesame questions every time the simulator is invoked.

This portion of tool could be modified to presentquestions from different data sets or even questionswith a timer but without the provided answers - atrue examination tool.

As an additional example, Figure 14 shows the dis-play employed in the inlet simulator package. Thissimulator does not provide for the plotter used in theundergraduate package but is more concerned withperformance aspects of a high speed inlet, such asits drag and total pressure recovery. The simulatorincludes a model for the subsonic portion of the inletduct based on wind tunnel testing. Again this pack-age was chiefly intended for design and graduate edu-cation because of the multiple problems solved here.

SUMMARY

High speed computer mainframes, personal com-puters, and workstations coupled with interactive op-erating systems and computer graphics provide manyopportunities for new techniques to teach basic fluidmechanics. This paper has presented several waysin which the new technologies can be utilized includ-ing the preparation of educational videos using theresults of CFD calculations, the analysis of flow so-lutions by students using the results of CFD calcu-lations and post-processing graphics packages, andthe development of workstation or personal com-puter based simulators which behave like desk topwind tunnels. Examples of these approaches havebeen presented along with discussion of some possi-ble problems in their implementation.

ACKNOWLEDGEMENTS

The simulators reported here use the FORMSlibrary for graphical user interfaces developed byMark H. Overmars, Department of Computer Sci-ence, Utrecht University, The Netherlands for SiliconGraphics workstations. Some modifications to thislibrary have been made to allow the packages to alsobe used on IBM Rise 6000 machines. All of this soft-ware is public domain and may be copied and usedfor non-commercial products. Copies of the sourceare available from the author at the NASA Lewis Re-search Center as are copies of the video of flow pasta cylinder reported herein.

REFERENCES

Anderson, B.H., and Benson, T.J., 1983, "Nu-merical Solution to the Glancing Sidewall ObliqueShock Wave/Turbulent Boundary Layer Interactionin Three Dimensions," AIAA Paper No 83-0136.

Benson, T.J., 1994a "An Interactive EducationalTool for Compressible Aerodynamics", AIAA Paper94-3117.

Benson, T.J., 1994b "An Interactive, Designand Educational Tool for Supersonic External-Compression Inlets", AIAA Paper 94-2707.

Fox, R.W., and McDonald, A.T., 1992, Introduc-tion to Fluid Mechanics, John Wiley, New York, 4thed.

Kim, S.-W. and Benson, T.J., 1992 "Comparisonof the SMAC, PISO, and Iterative Time-AdvancingSchemes for Unsteady Flows", Journal of Computersand Fluids, Vol. 21, No.3, pp. 435-454.

Kim, S.-W. and Benson, T.J., 1993 "Fluid Flowof a Row of Jets in Crossflow - A Numerical Study",AIAA Journal, Vol. 31, No.5, pp. 806-811.

Koening, K. and Hodge, B.K., 1993, "A Suite ofPersonal Computer Programs to Support PropulsionEducation", AIAA Paper 93-2053.

Mattingly, J.D., Reiser, W.H. and Daley, D.H.,1987, Aircraft Engine Design, AIAA Education Se-ries, Washington, D.C.

Rand, R.H., 1987, "Computer Algebra in AppliedMathematics: An Introduction to MACSYMA", Pit-man Publishing Inc.

Schlichting, H., 1968, Boundary Layer Theory,McGraw-Hill, New York, 6th ed.

Shapiro, A. H., 1961, Shape and Flow, Doubleday,New York.

Walatka, P.P, Clucas, J., McCabe, R.K., Ples-sel, T. and Potter, R., 1994, "FAST User's Guide",NASA Ames Research Center, NAS Division, RNPBranch.

-~-t§.&s-• ̂Ossfls?„ s-x*--"*1

Figure 1 Particle traces from vortex street video.

Figure 2 Vorticity contours from vortex street video.

Figure 3 Injected fluid visualization from video.

Shock Have

Figure 4 Mach number contours. b. Upstream view

Figure 6 Particle traces of jet in cross-flow.

Figure 5 Surface oil traces. Figure 7 Surface topology of jet interaction.

NASA-Lewis VU-SHOCKS

Up

Wedges: [i'single I ID DoubleI |n'opposed| -y | 1.40

•H3Q- Turning DSD-CB-

Plotter

Begin Trace 1

5 Max

End Trace |

Take Data |

25 Max | J

Variables

I

Iff Maeh'l ........... In'Tl/fc .............. I

rn"p'vpo IBfTi/rcT" "1

vs H Sequential 1 [Q Random I [H (3uestion | |jf Answer

I ft"Madh ""1 i" We'dge'Ang

Questions and Answers

Version £0

Figure 8 Screen dump of compressible flow simulator.

10

NASA-Lewis VU-SHOCKS I Redraw] ['Output"! ["HeTpl I"1x0

Wedges: 11 Single I |Q Double I

Sequential I |Q Random"'| rij~Questloril 11' Answer' |

Version 2.0

Figure 9 Acquiring data using the plotter.

11

NASA-Lewis VU-SHOCKS I Redraw! | 'bupul | "Help "| |""Bdt""|

Mnch =3.001

Up

Wedges: [W'siifijle j | IT Double' | |BJ ojaposedj v |J/ig

Mach [ 2.00J | < Enter |

Angle-1 1-301 I |7Trite7|

Plotter

I Begin Trace I

SMax T"

aceDaa

25 Max I 10

Shock } 30.00

pt/ptO 11.000

Variables

[0 Shock Anglein

lO'Maeh'T ........... In"ti/T0 .............. 'I

I i"pl7pO:"' 'I Q' flTrO |

Macti

T/TO

VS Sequentiall |Q Raridorn | 0 Question ] IB Answer I

Mach 11 Wedge Anslel


Version 2.0

Figure 10 Completed plot.

12

NASA-Lewis VU-SHOCKS IWedraw'1 ["outpuTI ["'Help"! r""&dt'"|

Up

Wedges: |l"single""l ITTDoiibte"! [jff'Opp'osedl v FT**"

Plotter

_pn fiToFFlrjewpisn

Begin Trace 1

SMax I 1 I

25 Max

Variables

[R'ShockAnglell ptJ/pJiii I

IB'MacMI'lfTi/tiQ ™1

vs

I'i'Mach 1D Wedge'Angie]

Shock

pt/ptO

I (I Question "I

Q 3: For Mach nunber - 2.0 and wedge angle - 24.0what is the shock angle ? Why ?

Version 2.0

Figure 13 Question and answer box.

15

r

NASA-Lewis VU-INLET

Down Zoom

Compression: Ml External | |j"| M i xed I

Pressire | 4.37 ̂ 1 | 34.22 I Ibs/in

Temperatire .static total"

Engine:MaxAirflow ^ai75j

Cor Airflow I 0.134 |

Special:

Bypass I 0.000 I

72/71

p/pO 124721 T/TO I 1.303 I

Inlet Performance

Drag Sifiersonic Siteonic Engine Face

A0*0 J0:6)9{ Mach \ OJ5i J Macnth ( 0.46 I Mach [ 0.30

D/qOAc I 0.332 I pt/ptO I 0.943 I pt2/ptth I 0.994 I pt2/ptO I 0337

Version 2.0

Figure 14 Supersonic inlet simulator.

16

REPORT DOCUMENTATION PAGEForm Approved

OMB No. 0704-018BPublic reporting burden for this collection of information is estimated to average 1 hour per response, including the time for reviewing instructions, searching existing data sources,gathering and maintaining the data needed, and completing and reviewing the collection of information. Send comments regarding this burden estimate or any other aspect of thiscollection of information, including suggestions for reducing this burden, to Washington Headquarters Services, Directorate for Information Operations and Reports, 1215 JeffersonDavis Highway. Suite 1204, Arlington, VA 22202-4302, and to the Office of Management and Budget. Paperwork Reduction Project (0704-0188), Washington, DC 20503.

1. AGENCY USE ONLY (Leave Wan*) 2. REPORT DATE

June 19943. REPORT TYPE AND DATES COVERED

Technical Memorandum4. TITLE AND SUBTITLE

On the Use of Computers for Teaching Fluid Mechanics

6. AUTHOR(S)

Thomas J. Benson

5. FUNDING NUMBERS

WU-505-62-52

7. PERFORMING ORGANIZATION NAME(S) AND ADDRESS(ES)

National Aeronautics and Space AdministrationLewis Research CenterCleveland, Ohio 44135-3191

8. PERFORMING ORGANIZATIONREPORT NUMBER

E-8851

9. SPONSORING/MONITORING AGENCY NAME(S) AND ADDRESS(ES)

National Aeronautics and Space AdministrationWashington, D.C. 20546-0001

10. SPONSORING/MONITORINGAGENCY REPORT NUMBER

NASATM-106587

11. SUPPLEMENTARY NOTES

Prepared for the ASME Fluids Engineering Division Summer Meeting sponsored by the American Society for Mechani-cal Engineers, Lake Tahoe, Nevada, June 19-23, 1994. Responsible person, Thomas J. Benson, organization code 2670,

. (216)433-5920.12a. DISTRIBUTION/AVAILABILITY STATEMENT

Unclassified - UnlimitedSubject Category 61

12b. DISTRIBUTION CODE

13. ABSTRACT (Maximum 200 words)

Several approaches for improving the teaching of basic fluid mechanics using computers are presented. There are twoobjectives to these approaches; to increase the involvement of the student in the learning process, and to presentinformation to the student in a variety of forms. Items discussed include: the preparation of educational videos usingthe results of computational fluid dynamics (CFD) calculations, the analysis of CFD flow solutions using workstationbased post-processing graphics packages, and the development of workstation or personal computer based simulatorswhich behave like desk top wind tunnels. Examples of these approaches are presented along with observations fromworking with undergraduate co-ops. Possible problems in the implementation of these approaches as well as solutionsto these problems are also discussed.

14. SUBJECT TERMS

Educational; Software

17. SECURITY CLASSIFICATIONOF REPORT

Unclassified

18. SECURITY CLASSIFICATIONOF THIS PAGE

Unclassified

19. SECURITY CLASSIFICATIONOF ABSTRACT

Unclassified

15. NUMBER OF PAGES

1816. PRICE CODE

A0320. LIMITATION OF ABSTRACT

NSN 7540-01-280-5500 Standard Form 298 (Rev. 2-89)Prescribed by ANSI Std. Z39-18298-102

on the use of computers for teaching fluid mechanics · pdf filecomputers can assist the...

Documents