Prediction of Transient Skin Temperature of High Speed Vehicles through CFD

Mukkarum Hussain Fluid Mechanics Department, Institute of Space Technology

Karachi, Pakistan mrmukkarum@yahoo.com

Muhammad Nauman Qureshi Fluid Mechanics Department, Institute of Space Technology

Karachi, Pakistan mnaumanqureshi@gmail.com

Abstract—Severe aerodynamic heating is experienced by vehicles flying at supersonic and hypersonic speeds. This aerodynamic heating can result in very high skin temperatures of the flight vehicle that can even cause mission failure. Hence a detailed knowledge of the aerodynamic heating load is essential for designing an appropriate thermal protection system. Although there is considerable knowledgebase available in high speed aerothermodynamics but uncertainties in the prediction of the transient skin temperatures still exist. This paper describes the procedure which determines the transient skin temperatures of high speeds vehicles using FLUENT. Solid-Fluid Coupling and transient boundary condition capabilities of FLUENT are used to predict temperature transients. The available X-15 flight data for three different flight trajectories (Flight A, B and C) are used for validation of the method. Laminar and turbulent computations are carried out to calculate temperature transients at wing mid-span chord location of X-15 and compared with the available in-flight data. The results obtained for skin temperatures at different locations are found both qualitatively and quantitatively in good agreement with in-flight data. This validates the methodology utilized in modeling the transient aero-thermal analysis of high speed vehicles. This method could be very useful in predicting the aerodynamic heating loads of other high speed vehicles.

Keywords—transient skin temperature; X-15; solid-fluid interaction; hypersonic flow, aerodynammic heating

I. INTRODUCTION Large flow gradients are produced at the surface of vehicle

that affects the characteristics of the boundary layer and heat transfer. Within the boundary layer, kinetic energy of fluid is transformed into the thermal energy which cause large amount of heat generation and high temperatures at near wall regions. Particularly in high speed flows, this phenomenon could be catastrophic if material of the vehicle is not high temperature resistant. The severity of aerodynamic heating depends on the flight trajectory (velocity, altitude and angle of attack histories) and vehicle geometry. In-depth investigation is therefore required to compute structural and thermal loads on high speed vehicles. Designing of thermal protection system requires accurate prediction of thermal loads which is very critical in the case of manned mission.

The objective of the present work is to validate the methodology developed for transient skin temperatures prediction [1] of high speeds vehicles using CFD software FLUENT. CFD helps us analyzing the flow field and heat

transfer around a vehicle if the grid is appropriate and correct boundary conditions are used [2-6]. However, large compute time and storage especially for 3D transient problems is required. CFD methodology must be validated before utilizing the CFD results.

NASA tested experimentally an aircraft named X-15 to generate aero-thermal flight data [7]. This data is used for the benchmarking of different methods developed for the transient skin temperatures prediction of high speeds vehicles [8-11]. Transient temperature data for three different trajectories of X-15 are available. Maximum Mach nos. are 4, 5, and 6 for Flight A, Flight B, and Flight C trajectories respectively. Maximum altitudes is around 30 Km for Flight B & C and 65 Km for Flight A, while angle of attack varying upto 20o.

The temperature data of three flights is acquired through 290 thermocouples mounted at inner surface of skin and 190 thermocouples placed on the internal structure. The location of several thermocouples placed near leading edge is shown in Figure 1. Numerous additional thermocouples are used to measure temperatures in the various systems in the airplane. The overall efficiency of the system for the conditions existing during the three flights covered was claimed to be 12K (±200 F).

In present study, transient skin temperatures at 4%, 20%, and 46% (Figure 2) wing mid-span chord locations at windward and leeward sides are predicted and evaluated with experimental data. Table I summarize thermocouples detail used for validation purpose. X-15 airplane’s wing is a multi-spar composition which used NACA 66005 airfoil as its fundamental cross-section [7]. The wing skin is of Inconel-X and its material properties are given in Table II.

Fig. 1. Different Thermocouple Location near Leading Edge of X-15 Aircraft Wing [7] (dimensions in inches)

723978-1-4673-6396-9/13/$31.00 ©2013 IEEE

FLUENT’s 2D solver is used to compute transient skin temperatures at wing mid-span chord location of X-15. Mesh used in present study is shown in Figures 3&4. The near wall clustering was done in such a manner that the Y+<5 for the whole trajectory at all locations. The boundary conditions applied are described in Figure 5. A solid-Fluid interfacing capability of FLUENT [12] is used to compute accurate results, as shown in Chart 1. Transient boundary conditions are defined by using User Define Function (UDF) and profile features of FLUENT. Spalart-Allmaras (SA) turbulence model with SIMPLE scheme is used. Numerical computations with PISO and Roe-FDS scheme are also carried out. No major effects on results with these schemes are found. In future, effect of different turbulence model would be studied. Approximately it took 100 hours for one complete computation on workstation having 2 x quad core processors operating at 2.4 GHz and 16 GB RAM.

.

Fig. 2. Wing Mid-Span Chord Locations [7]

Table I: Thermocouple Location Detail

Location Thermocouple Number

4% Chord, upper/lower 32/38

20% Chord, upper/lower 33/39

46% Chord, upper/lower 35/41

Table II: Material Properties (Inconel-X)

Density 8.28 g/cc

Specific Heat 0.431 J/g-oC

Thermal Conductivity 12 W/m-k

Emissivity 0.76

Fig. 3. Mesh around Wing Mid-Span Chord

Fig. 4. Mesh near Leading Edge of Wing Mid-Span Chord

Fig. 5. Boundary Condition used in Transient Aero-thermal Analysis

Fig. 6. Flow Chart for Transient Aero-thermal Analysis

724

II. FLIGHT CONDITIONS X-15 flight data of three different flight trajectories are used

for the validation of procedure for predicting transient skin temperature of high speed vehicles through CFD. Freestream conditions are shown in Figures 7-9. Free-stream conditions are such that perfect gas and no slip boundary condition assumptions are valid.

Fig. 7. Free-Stream Conditions, Altitude vs. Time

Fig. 8. Free-Stream Conditions, Angle of Attack vs. Time

Fig. 9. Free-Stream Conditions, Mach vs. Time

III. RESULTS AND DISCUSSION

A. Flight A (4% Chord Location) Figures 13-14 show results at 4% chord locations for Flight

A trajectory. Initially both the laminar and turbulent simulations match with the flight data for around 40 seconds. This can be explained through laminar to turbulent criteria plot

[13] as shown in Figure 10. The flow for first 20 seconds is laminar and then it transits to turbulence and remain turbulent upto 70 seconds. For the rest of trajectory it remains laminar. This can also be observed from simulations that the transient skin temperatures of laminar simulation match well with the flight data from 100 seconds till end.

B. Flight A (20% Chord Location) The transition criteria plot for the 20% chord location of

Flight A trajectory shows that upto around 100 seconds the flow at this location is turbulent. Then it becomes laminar and remain laminar upto 250 seconds. For the subsequent time it remains turbulent. However, from present simulations it is remarked that both for windward and leeward flow, turbulent simulations produce much accurate results, as shown in Figures15-16. A possible reason for such behavior could be that during the initial turbulent phase the surface temperature achieved at 110 seconds remains constant in laminar flow regime upto 250 seconds. As the flow regime becomes turbulent due to its trajectory the surface temperature starts increasing again.

C. Flight A (46% Chord Location) The flow nature is same as that of 20% location. From

simulations it is observed that both for windward and leeward flow, the turbulent simulations produce better results than laminar flow predictions, as shown in Figures17-18. The reason for such behavior could be the same as given in previous case.

D. Flight B The flow remains turbulent for all three locations, as

shown in Figure 11. From Figures 19-24 it is observed that the turbulent predictions for skin temperature matches the flight results very well for all flow locations and for both windward and leeward sides.

E. Flight C (4% Chord Location) At 4% location the turbulent criteria plot for the trajectory

show that the flow remains laminar throughout the trajectory, as shown in Figure 12. From CFD it is also observed that the laminar simulations predict much accurate results than turbulent, as shown in Figures 25-26.

F. Flight C (20% Chord Location) The flow at this location is initially turbulent upto 50

seconds (Figures 27-28). From 50 to 220 seconds it remains laminar and afterwards flow transitions to turbulent again. For windward side, the turbulent simulations predict well the skin temperatures when compared with flight data. For leeward side, the turbulent predictions are much nearer to flight data with slight over predictions.

G. Flight C (46% Chord Location) At this location the flow remains turbulent during most of

the flight time. The CFD simulations with turbulence predict very well the transient skin temperature when compared with flight data, as shown in Figures 29-30.

725

Fig. 10. Laminar to Turbulent Criterion, Flight A

Fig. 11. Laminar to Turbulent Criterion, Flight B

Fig. 12. Laminar to Turbulent Criterion, Flight C

Fig. 13. Temperature at 4% Chord, Leeward, Flight A

Fig. 14. Temperature at 4% Chord, Windward, Flight A

Fig. 15. Temperature at 20% Chord, Leeward, Flight A

Fig. 16. Temperature at 20% Chord, Windward, Flight A

Fig. 17. Temperature at 46% Chord, Leeward, Flight A

Fig. 18. Temperature at 46% Chord, Windward, Flight A

726

Fig. 19. Temperature at 4% Chord, Leeward, Flight B

Fig. 20. Temperature at 4% Chord, Windward, Flight B

Fig. 21. Temperature at 20% Chord, Leeward, Flight B

Fig. 22. Temperature at 20% Chord, Windward, Flight B

Fig. 23. Temperature at 46% Chord, Leeward, Flight B

Fig. 24. Temperature at 46% Chord, Windward, Flight B

Fig. 25. Temperature at 4% Chord, Leeward, Flight C

Fig. 26. Temperature at 4% Chord, Windward, Flight C

727

Fig. 27. Temperature at 20% Chord, Leeward, Flight C

Fig. 28. Temperature at 20% Chord, Windward, Flight C

Fig. 29. Temperature at 46% Chord, Leeward, Flight C

Fig. 30. Temperature at 46% Chord, Windward, Flight C

VI. CONCLUSION A methodology which determines the transient skin

temperatures of high speeds vehicles using FLUENT is developed and validated. The available X-15 flight data for three different flight trajectories (Flight A, B and C) are used for validation of the method. Laminar and turbulent

computations are carried out to calculate temperature transients at wing mid-span chord location of X-15 and compared with the available in-flight data. The results obtained for skin temperatures at different locations are found both qualitatively and quantitatively in good agreement with in-flight data. This validates the methodology utilized in modelling the transient aero-thermal analysis of high speed vehicles. This method could be very useful in predicting the aerodynamic heating loads of other high speed vehicles. The effect of different turbulent models like two-equation, k-kl transition, and 4 equation SST transition models for predicting transient skin temperatures remain to be explored.

V. REFERENCES [1] Mukkarum Husain, Shamoon Jamsheed, M. Nauman

Qureshi, “Transient Aero-thermal Analysis of High Speed Vehicles using CFD.” Proceedings of IBCAST 2012 Islamabad, Pakistan, January 09 – 12, 2012.

[2] J. C. Tannehill, D. A. Anderson, R. H. Pletcher, “Computational Fluid Mechanics and Heat Transfer”, Second Edition, January 1997.

[3] John D. Anderson, Jr., “Computational Fluid Dynamics, the Basics with Application”, McGraw-Hill Series in Mechanical Engineering, 1995.

[4] Toro EF., “Riemann Solvers and Numerical Methods for Fluid Dynamics: A Practical Introduction”, Springer, 1997.

[5] Klauss A. Hoffmann, Steve T. Chiang, “Computational Fluid Dynamics. Vol. I”, Fourth Edition, Engineering Education System, Wichita, Kansas, 2001.

[6] Klauss A. Hoffmann, Steve T. Chiang, “Computational Fluid Dynamics. Vol. II”, Fourth Edition, Engineering Education System, Wichita, Kansas, 2001.

[7] Joe D. Watts, Ronald P. Banas, “X-15 Structural Temperature Measurments and Calculations for Flights to Maximum Mach Numbers of Approximately 4, 5, 6”, NASA Technical Memorandum X-883.

[8] Murray Palitz, “Measured and Calculated Flow Conditions on the Forward Fuselage of the X-15 Airplane and Model at Mach Numbers from 3.0 to 8.0”, NASA Technical Notes, NASA NT D-3447.

[9] Richard D. Banner, Albert E. Kuhl, Robert D. Quinn, “Preliminary Results ofAerodynamic Heating Studies on the X-15 Airplane”, NASA Technical Memorandum X-638.

[10] Robert D. Quinn, Leslie Gong, “A method for calculating transient surface temperatures and surface heating rates for high speed aircraft”, NASA/TP-2000-209034, 2000.

[11] Michael R. Tauber, “A Review of High Speed Convective, Heat Transfer Computation Methods”, NASA Technical Paper 2914, 1989.

[12] “ANSYS FLUENT Theory Guide”, Release 14.0, 2011. [13] Quinn R. D., Gong L., “A Method for Calculating

Transient Surface Temperature and Surface Heating Rates for High-Speed Aircraft”, NASA/TP-2000-209034, 2000.

728