Reentry Motion and Aerodynamics of the MUSES-C Sample Return Capsule

Trans. Japan Soc. Aero. Space Sci. Vol. 51, No. 172, pp. 65–70, 2008

By Nobuaki ISHII,1Þ Tetsuya YAMADA,1Þ Koju HIRAKI2Þ and Yoshifumi INATANI1Þ

1ÞThe Institute of Space and Astronautical Science, JAXA, Sagamihara, Japan 2ÞKyushu Institute of Technology, Kita-Kyushu, Japan

(Received June 21st, 2006)

The Hayabusa spacecraft (MUSES-C) carries a small capsule for bringing asteroid samples back to the earth. The initial spin rate of the reentry capsule together with the flight path angle of the reentry trajectory is a key parameter for the aerodynamic motion during the reentry flight. The initial spin rate is given by the spin-release mechanism attached between the capsule and the mother spacecraft, and the flight path angle can be modified by adjusting the earth approach orbit. To determine the desired values of both parameters, the attitude motion during atmospheric flight must be clarified, and angles of attack at the maximum dynamic pressure and the parachute deployment must be assessed. In previous studies, to characterize the aerodynamic effects of the reentry capsule, several wind-tunnel tests were conducted using the ISAS high-speed flow test facilities. In addition to the ground test data, the aerodynamic properties in hypersonic flows were analyzed numerically. Moreover, these data were made more accurate using the results of balloon drop tests. This paper summarized the aerodynamic properties of the reentry capsule and simulates the attitude motion of the full- configuration capsule during atmospheric flight in three dimensions with six degrees of freedom. The results show the best conditions for the initial spin rates and flight path angles of the reentry trajectory.

Key Words: Reentry Aerodynamics, Attitude Motion of Blunt-Nose Capsule

Nomenclature dominant design parameters, governing the attitude motion during reentry. CA: axial force coefficient In previous studies, to determine acceptable reentry con- CLp: roll rate damping coefficient ditions, aerodynamic properties were modeled as a function 7,8) CM: pitching moment coefficient of Mach number and angle of attack. These data were CMq: pitch rate damping coefficient obtained from wind-tunnel tests in the subsonic through CN: normal force coefficient supersonic region, and analyzed numerically in the hyper- Dp: dynamic pressure sonic region. On the other hand, the convective and radiative M: Mach number heating at the stagnation point was also modeled as a func- p0: initial pitch rate at atmospheric entry tion of air density and flight velocity, whose parameters r0: initial spin rate at atmospheric entry were evaluated by a series of aerothermodynamic analyses : angle of attack taking into account the radiant in the high temperature shock 0: initial angle of attack at atmospheric entry layer and succeeding flow and interaction between outer 9–12) n: nutation angle (half coning angle) of capsule defined flow and ablation products. The results of balloon-drop as angle between angular momentum and spin axis of tests enabled data modification in the transonic region and capsule also revealed that the angle of attack must be less than 15 at parachute deployment and less than 5.5 at maximum 1. Introduction dynamic pressure.13–15) This paper uses the modified aerodynamic model to numerically simulate attitude motion to find The Hayabusa spacecraft (MUSES-C) was launched in the set of design parameters, such as position of center of May 2003 and arrived at the small ITOKAWA asteroid in gravity, initial flight path angle, and initial spin rate at atmo- September 2005.1–3) Under the current schedule, the space- spheric entry, to satisfy the above requirements for angles craft will return to the earth in 2010 with a small reentry of attack. capsule4,5) containing asteroid surface soil. The capsule will Section 2 outlines the configuration of the reentry capsule enter the earth’s atmosphere directly at the hypervelocity of and Section 3 summarizes the aerodynamic properties. the interplanetary orbit.6) After severe aerodynamic deceler- Using the more accurate aerodynamic model, the attitude ation, the capsule will deploy its parachute and carry the motion of the capsule is simulated numerically and asteroid sample to the ground. At the reentry phase, the the effects of aerodynamic coefficients are analyzed in initial spin rate of the capsule and the flight path angle are Section 4.

Ó 2008 The Japan Society for Aeronautical and Space Sciences 66 Trans. Japan Soc. Aero. Space Sci. Vol. 51, No. 172

Fig. 2. Axial force coeﬃcient vs. Mach number. Fig. 1. Outline conﬁguration and dimensions of reentry capsule.

2. Conﬁguration of Reentry Capsule

The outline configuration of the reentry capsule is shown in Fig. 1. The shape was determined within restricted weight and dimensions to maximize the internal volume for instal- ling the sample canister, parachute, pyrotechnic devices, and electronic devices.4,5) The capsule has a hemispherical nose with 202-mm diameter and conical side body with 45 half cone angle. The maximum diameter is 404 mm, and the mass is 17 kg. The position of the center of gravity is about 120 mm from the nose. The moments of inertia around the three axes x, y, and z are 0.289, 0.147, and 0.136 kg m2, respectively. The nutation angle (half coning angle) n of the capsule mainly due to the attitude disturbance at capsule separation from the mother spacecraft is Fig. 3. Normal force coefficient vs. Mach number. defined as the angle between the angular momentum and spin axis of the capsule, i.e., the angle n is equivalent to the inverse tangent of ðIyy p0Þ=ðIxxr0Þ. 10% to 20% smaller than the zero angle of attack values. The normal force generated by is quite a lot smaller than 3. Aerodynamic Properties the axial force. 3.2. Static stability In the subsonic, transonic and supersonic regions The pitching moment coefficients CM as a function of (M < 4), sub-scale model wind tunnel tests were conducted Mach number are shown in Fig. 4. It is clear that the capsule using the ISAS high-speed flow test facilities. These wind- is statically stable throughout the flight Mach number rang- tunnel data were modified using the results of the balloon- ing from 30 to zero, although the stability is somewhat lower drop tests.6–8) In the hypersonic region (M > 4), the aerody- in the transonic region. As shown in Fig. 5, the effect of namic properties were analyzed numerically on the basis moving the center of gravity on CM is about 10% for of the Newtonian approximation. In addition, the diffuse 10 mm of movement. reflection model was applied in the rarefied flow regime 3.3. Dynamic damping for Knudsen numbers greater than 1/1000. The wind-tunnel The pitch rate damping coefficients CMq are shown in data were obtained for angles of attack smaller than Figs. 6 and 7. In the subsonic and transonic regimes, CMq 35, while aerodynamic coefficients are assumed constant is positive for angles of attack smaller than 13 and negative for 35. for angles of attack larger than 13. In this situation, at 3.1. Axial and normal forces angles of attack smaller than 13, the attitude is dynamically The axial force coefficients CA and normal force coeffi- unstable due to the positive value of CMq; at angles of attack cients CN are shown in Figs. 2 and 3, respectively. CA has larger than 13 , the attitude is dynamically stable due to the its peak values at M ¼ 1 and Mach-number-independent negative value of CMq. On the other hand, in the hypersonic values in the hypersonic regime. Due to the effect, CA is regime, CMq has small negative values independent of the Aug. 2008 N. ISHII et al.: Reentry Motion of MUSES-C Capsule 67

Fig. 4. Pitching moment coeﬃcient vs. Mach number. Fig. 6. Dynamic damping vs. Mach number.

Fig. 5. Effect of movement of center of gravity on static stability. Fig. 7. Effect of angle of attack on dynamic damping. angle of attack. In this situation, the attitude remains dynam- parameters and can be modified by adjusting the perigee ically stable. altitude of the earth approach orbit. The nominal angle of Accurate estimation of the roll rate damping coefficient attack of the capsule should be zero to reduce aerodynamic CLp is difficult. Although CLp is zero for an ideally axisym- heating, while the maximum error in attitude pointing is metric body in inviscid flow, it has a small negative value in estimated to be 5.4) viscous flows due to surface friction. In the following atti- The initial spin rate is given by the spin-release mecha- tude simulation, a constant value CLp ¼0:01 is assumed nism between the capsule and mother spacecraft. A spring in all flight conditions independent of Mach number and in the mechanism gives the spin rate together with the trans- angle of attack. lational speed for capsule separation. Due to the mechanical fittings between the mother spacecraft and capsule, attitude 4. Analysis of Attitude Motion disturbance is generated in the pitch and yaw rates at separation. From separation tests using a microgravity free-fall Using the aerodynamic coefficients mentioned above, the tower, the maximum disturbance was about 1/3 of the spin attitude motion is simulated numerically in three dimensions rate, and the nutation angle (half coning angle) n, i.e., with six degrees of freedom. equivalent to the inverse tangent of ðIyy p0Þ=ðIxxr0Þ, was 4.1. Initial conditions 9.6.16) The nominal conditions for position, velocity and When considering the initial conditions of reentry trajec- attitude of the reentry trajectory are listed in Table 1. tories, the entry velocity is given from the orbital energy of The following sections analyze reasonable values for the the interplanetary approach orbit to earth and the altitude of flight path angle and the initial spin rate together with the the interface point at atmospheric entry.6) On the other hand, effects of aerothermodynamic environments and aerody- the flight path angle of the reentry trajectory is a changeable namic stability. 68 Trans. Japan Soc. Aero. Space Sci. Vol. 51, No. 172

Table 1. Nominal initial conditions of reentry trajectory.

Altitude 200 km (preset as interface point) Velocity 12.0 km/s (given from interplanetary orbit)

Flight path angle 12 (designed, changeable parameter) Spin rate 60/s (designed, changeable parameter)

(including uncertainties Angle of attack 5 in attitude orientation) (disturbance at capsule separation Pitch rate 20/s is 1/3 of spin rate) Fig. 8. Altitude and velocity of reentry trajectory.

4.2. Reentry environments At shallow entry, the peaks in dynamic pressure and heat flux are small but the total heat input is large due to long flight time. On the other hand, at deep entry, the maximum heat flux is critical to the thermal protection design. To as- sess the aerodynamic heating, the convective and radiative heating at the stagnation point were modeled as a function of air density and flight velocity, whose parameters were evaluated by a series of aerothermodynamic analyses taking into account the radiant in the high-temperature shock layer and succeeding flow, and interaction between outer flow and Fig. 9. Dynamic pressure and flight path angle of reentry trajectory. ablation products.9–12) From the results of ground heating tests of the thermal protection materials,11,12) the acceptable maximum heat flux is 15.5 MW/m2 and the acceptable total heat input is less than 360 MJ/m2. Table 2 summarizes the maximum heat flux and total heat input of several reentry trajectories where the flight path angles are changed para- metrically from 10 through 14. According to the above requirements for heat flux and total heat input, a flight path angle of 12 is best for the nominal reentry conditions. Reentry trajectories with attitude motion were simulated using the above initial conditions. Time histories of altitude and velocity are shown in Fig. 8; dynamic pressure and Fig. 10. Angle of attack and dynamic pressure of reentry trajectory. flight path angle are shown in Fig. 9. At a peak dynamic pressure of 514 hPa, the altitude and velocity are 44.8 km and 7.12 km/s, respectively. 4.3. Aerodynamic stability The time history of the average angle of attack is shown in Fig. 10, where initial angle of attack 0, initial spin rate r0, and initial pitch rate p0 are 5 ,60/s, and 20 /s, respectively. From 0 through 40 sec, aerodynamic effects do not appear. From 40 through 80 sec, as dynamic pressure Dp increases, the average value and oscillation amplitude of the angle of attack decrease due to aerodynamic static and dynamic stability. The minimum value of is about 5 . After 80 sec, increases as Dp decreases. The oscillation Fig. 11. Static stability for initial angle of attack of 30. does not vanish due to shortage of dynamic damping. The effects of static stability are shown in Fig. 11 where

Table 2. Aerothermodynamic environment with respect to entry ﬂight path angle.

Flight path angle at entry [] 10:0 11:0 12:0 13:0 14:0 Requirement Maximum heat ﬂux [MW/m2] 10.1 13.0 15.2 17.0 18.6 < 15:5 Total heat input [MJ/m2] 435 362 327 305 288 < 360 Aug. 2008 N. ISHII et al.: Reentry Motion of MUSES-C Capsule 69

Table 3. Comparison of attitude motion with respect to initial spin rate.

Initial spin rate r0 [ /s] 30 45 60 75 90 Initial pitch rate p0 [ /s] 10 15 20 25 30 Initial angle of attack 0 [ ] 55555 Initial nutation angle n [ ] 9.6 9.6 9.6 9.6 9.6 Requirement Angle of attack Dp [] 3.9 4.8 5.3 5.8 6.4 < 5:5 at maximum dynamic pressure Angle of attack para [] 16.9 15.5 14.5 13.7 17.2 < 15:0 at parachute deployment

Fig. 12. Eﬀect of initial spin rate. Fig. 13. Limit cycle attitude motion in low speed region.

0, r0, and p0 are 30 ,60/s, and 20 /s, respectively. In this case, the center of gravity is 130 mm from the capsule nose. In other words, 10-mm offset is applied to the center of gravity,5,6) and smaller aerodynamic stability is assumed. From 0 through 40 sec, the average angle of attack increases slightly. This tendency does not mean an attitude change, but is due to the change in the flight path angle. As shown in Fig. 11, when the initial angle of attack is 30, the attitude remains stable. However, the minimum value of exceeds 5 at the point of maximum dynamic pressure. Since the Fig. 14. Effect of dynamic damping reduction. maximum expected uncertainty in the position of the center of gravity is less than 5 mm, it does not affect attitude stability or parachute deployment. respect to various initial spin rates. Balloon-drop tests 4.4. Effects of initial spin rate showed that the angle of attack should be less than 15 at The capsule spin is needed to stabilize attitude in the parachute deployment, and less than 5.5 at maximum pre-entry phase before atmospheric entry. However, during dynamic pressure.13–15) As shown in Table 3, an initial spin atmospheric flight, the spin stabilization spoils the aerody- rate of 60/s satisfies both requirements. namic attitude control and parachute deployment. 4.5. Effect of dynamic damping The effects of the initial spin rate are shown in Fig. 12 The effect of dynamic damping is shown in Fig. 13 where where 0, r0, and p0 are 5 ,6/s, and 2 /s, respectively. 0, r0, and p0 are 0 ,3 /s, and 0 /s, respectively. Even if the The initial angle of attack 0 and the initial nutation angle initial angle of attack 0 is zero, the oscillation is enlarged (half coning angle) n, i.e., equivalent to the inverse tangent finally due to inadequate dynamic damping. of ðIyy p0Þ=ðIxxr0Þ, are same as in Fig. 10. Since the initial The results of another simulation are shown in Fig. 14 spin rate is smaller (smaller spin stability), the minimum where 0, r0, and p0 are 5 ,60/s, and 20 /s, respectively. value of at the instant of maximum dynamic pressure is The dynamic damping is intentionally reduced to 1/10 smaller than in Fig. 10. However, the final oscillation after (0:1 CMq is applied in this case), although the initial condi- aerodynamic deceleration is bigger because of smaller spin tions are the same as in Fig. 10. The initial attitude from 0 stability. These results show that a reasonable but not exces- through 80 sec behaves in almost the same way as in sive spin rate is required from the viewpoint of attitude Fig. 10. After aerodynamic deceleration, the final angle of stability at parachute deployment. attack is three times larger than in Fig. 10. It should be noted Table 3 lists the angle of attack at the point of maximum that the attitude oscillation does not diverge, even if a 1/10 dynamic pressure and at the parachute deployment with effect of CMq is assumed. 70 Trans. Japan Soc. Aero. Space Sci. Vol. 51, No. 172

5. Conclusions 50th International Astronautical Congress, Amsterdam, The Nether- lands, October 4 to 8, 1999. 5) Inatani, Y., ed.: Aerodynamics, Thermophysics, Thermal Protection, Using aerodynamic properties obtained from wind-tunnel Flight System Analysis and Design of Asteroid Sample Return tests and numerical analyses, the attitude motion of the Capsule, ISAS Report SP-17, Institute of Space and Astronautical reentry capsule during atmospheric flight was simulated to Science, March 2003. clarify aerodynamic effects, such as static stability, and 6) Ishii, N., Hiraki, K., Yamada, T. and Inatani, Y.: Reentry Dynamics of MUSES-C, J. Space Technology and Science, 14 (1998), pp. 33–39. dynamic damping. 7) Hiraki, K., Inatani, Y., Ishii, N., Nakajima, T. and Hinada, M.: To reduce aerodynamic heating, the best flight path angle Dynamic Stability of MUSES-C Capsule, 21st ISTS, Japan, May for the reentry trajectory is about 12. We also found that 1998. the capsule remains stable for average angles of attack 8) Fujita, K., Inatani, Y. and Hiraki, K.: Attitude Stability of Blunt-Body Capsules in Hypersonic Rarefied Regime, J. Spacecraft Rockets, 41 smaller than 30 , even considering expected uncertainty in (2004), pp. 925–931. the position of the center of gravity. In addition, the initial 9) Suzuki, K., Kubota, H., Fujita, K. and Abe, T.: Chemical Nonequi- spin rate of 60/s is satisfactory for the angle of attack at librium Ablation Analysis of MUSES-C Super-Orbital Reentry maximum dynamic pressure and parachute deployment. Capsule, AIAA Paper 97-2481, June 1997. 10) Hiraki, K., Niikura, K., Nogi, S. and Shimada, T.: Numerical Simula- Although, accurate estimation of the dynamic damping tions of Charring Ablation for the Heat Shields on Super-Orbital coefficient CMq is difficult, the capsule is dynamically Reentry Capsule, 21st ISTS, Japan, May 1998. 11) Yamada, T. and Inatani, Y.: ISAS High Enthalpy Flow Facility for stable, even assuming 1/10 of a moderate value of CMq. The results of the above full model simulation with modi- Thermal Protection Material Tests, Proceedings of 26th IEPC, Oct. 17–21, 1999, pp. 1362–1369. fied aerodynamic properties were applied to the design of 12) Yamada, T., Inatani, Y. and Hiraki, K.: Development of Thermal the MUSES-C sample return capsule, and will be verified Protection System of the MUSES-C/DASH Reentry Capsule, Acta by the actual capsule reentry flight in 2010. Astronautica, 51, 1–9 (2002), pp. 63–72. 13) Hinada, M., Ishii, N., Nakajima, T., Yajima, N., Inatani, Y. and Hiraki, K.: Deployment Test of Parachute for Small Capsule Using Balloon, References 13th ESA Symposium on European Rocket and Balloon Programmes and Related Research, Oland, Sweden, May 26–29, 1997. 1) Kawaguchi, J., Uesugi, K., Fujiwara, A. and Saito, H.: The MUSES-C 14) Hinada, M., Ishii, N., Hiraki, K. and Inatani, Y.: Recovery System of Mission Description and Its Status, 3rd IAA International Conference MUSES-C Reentry Capsule, 14th ESA Symposium on European on Low-Cost Planetary Missions, IAA-L-98-0505, 1998. Rocket and Balloon Programmes and Related Research, Potsdam, 2) Kawaguchi, J.: MUSES-C Mission, 4th IAA International Conference Germany, May 31 to June 3, 1999. on Low-Cost Planetary Missions, IAA-L-0306, 2000. 15) Hinada, M., Hiraki, K., Ishii, N., Inatani, Y., Nakajima, T. and Honda, 3) Kawaguchi, J., Kuninaka, H., Fujiwara, A., Uesugi, K. and Ohnishi, M.: Parachute System of MUSES-C Reentry Capsule, AIAA Paper T.: MUSES-C Launch and Early Operations Report, AAS/AIAA 99-1744, June 1999. Astrodynamics Specialists Conference, AAS Paper 03-662, August 16) Ishii, N., Higuchi, K., Okuizumi, N., Abe, K. and Tatsuta, S.: Spin 2003. Release Mechanism for Small Space Probe Separation, AAS/AIAA 4) Ishii, N., Inatani, Y., Yamada, T., Hiraki, K., Ogawa, H. and Honda, Space Flight Mechanics Meeting 2001, AAS Paper 01-128, Vol. 108, M.: Development Study of an Asteroid Sample Return Capsule, pp. 425–432.