10kW級Φ75mm大口径 ICP加熱器の気流特性 に関する研究

8th European Symposium on Aerothermodynamics for Space Vehicles
Spectroscopic Measurement of Plasma Flow Produced
by 10 kW and Φ 75 mm Large Diameter ICP Heater
Satoshi Miyatani (University of Tokyo)
Kazuhiko Yamada (ISAS/JAXA)
Takashi Abe (ISAS/JAXA)
Background
Aerodynamic Heating during atmospheric entry
Conventional System
RESIST
Thermal tile
Flare-type Membrane Aeroshell
REDUCE
Flexible aeroshell is packed during
launch and operation in orbit
After deploying aeroshell at highaltitude, the reentry vehicle is
separated from the mother ship.
Inflatable Torus
Heat Shield
Thin-membrane
flare
NASA Space Shuttle
Vehicle can decelerate at highaltitude, and aerodynamic heating
is reduced drastically.
Vehicle can also soft land and
float on ocean.
Ablator
JAXA MUSES-C
Capsule
(Payload)
〇 Aerodynamic heating reduction during reentry.
〇Terminal velocity reduction for the soft landing.
〇 Easy integration in launching phase because of packing efficiency.
2
Background
Aerodynamic Heating during atmospheric entry
Conventional System
RESIST
Thermal tile
Flare-type Membrane Aeroshell
REDUCE
Flexible aeroshell is packed during
launch and operation in orbit
After deploying aeroshell at highaltitude, the reentry vehicle is
separated from the mother ship.
Inflatable Torus
Heat Shield
Thin-membrane
flare
NASA Space Shuttle
Vehicle can decelerate at highaltitude, and aerodynamic heating
is reduced drastically.
Vehicle can also soft land and
float on ocean.
Ablator
JAXA MUSES-C
Capsule
(Payload)
〇 Aerodynamic heating reduction during reentry.
Details of research and development of our inflatable
〇Terminal velocity reduction for the soft landing.
〇aeroshell
Easy integration in launching phase because of packing efficiency.
=> 11:50-12:10, Mar 5th, Prof. Yamada (JAXA)
3
Re-entry from Low Earth Orbit
Numerical estimation of heat flux
Initial condition of orbit
• Altitude: 400 km (Circular orbit)
• Orbital velocity: 7.7km/sec
• Deceleration: 90 m/sec
Specification of flight model
• Diameter of Aeroshell: 2.5m
• The total mass of flight model: 15kg
> Maximum heat flux
≈ 130 kW/m2 (with aeroshell)
≈ 600 kW/m2 (without aeroshell)
0.5 m
2.5 m
Comparison of heat flux history at the
stagnation point during atmospheric entry
with or without aeroshell.
Total Mass:
15kg
5
Experimental Facility for Heating
Arc Heat Tunnel
CO2 Laser
×Heat Flux: ~
MW/m2
×Contamination
due to melting
electrode
ICP Heater
×Heating area: small
Hypersonic Wind Tunnel
×Enthalpy: low.
×Duration: short
•
•
•
•
•
No contamination
High enthalpy
Low heat flux (100 kW/m2 ~ )
Large heating area
Long duration time
6
Experimental Facility for Heating
Arc Heat Tunnel
CO2 Laser
×Heat Flux: ~
MW/m2
×Contamination
due to melting
electrode
ICP Heater
×Heating area: small
Hypersonic Wind Tunnel
×Enthalpy: low.
×Duration: short
•
•
•
•
•
No contamination
High enthalpy
Low heat flux (100 kW/m2 ~ )
Large heating area
Long duration time
7
Inductively Coupled Plasma Heater
Plasma Torch
Vacuum Pump
Gas
(air, nitrogen)
Matching Box
真空チャンバー
High-frequency
Power Source
Mass Flow Controller
Plasma Flow
真空チャンバー
Vacuum Chamber
Vacuum Valve
Cooling Water
8
Air plasma and nitrogen plasma
Air
• Violet plasma in the torch.
• Whitish flow in the test section.
Nitrogen
• Violet plasma in the torch.
• Yellowish flow in the test section.
9
Objective of Research
Investigate characteristics of plasma flow
Up until now…
Dynamic Pressure
• 30 ~ 50 Pa at the center of the flow
Heat Flux
• 100 ~ 400kW/m2 at the center
• Diameter of flow core ≈ 40mm
Pitot tube
Gardon gage
Numerical Modeling (Dr. Minghao)
• Rotational and translational
temperature,
• 6500 K in the torch
• 3500K in the test section
Ref: Yu Minghao et al, “Thermochemical Nonequilibrium 2D Modeling of Nitrogen Inductively
Coupled Plasma Flow,” Kyusyu University, 2015.
10
Objective of Research
Investigate characteristics of plasma flow
Up until now…
Dynamic Pressure
• 30 ~ 50 Pa at the center of the flow
Heat Flux
• 100 ~ 400kW/m2 at the center
• Diameter of flow core ≈ 40mm
Pitot tube
Gardon gage
Numerical Modeling (Dr. Minghao)
• Rotational and translational
temperature,
• 6500 K in the torch
• 3500K in the test section
Ref: Yu Minghao et al, “Thermochemical Nonequilibrium 2D Modeling of Nitrogen Inductively
Coupled Plasma Flow,” Kyusyu University, 2015.
11
Objective of Research
Investigate characteristics of plasma flow
Up until now…
Dynamic Pressure
• 30 ~ 50 Pa at the center of the flow
Heat Flux
• 100 ~ 400kW/m2 at the center
• Diameter of flow core ≈ 40mm
Numerical Modeling
Current Investigation
Spectroscopic measurement
• Investigate chemical species
in the plasma flow.
• Experimentally estimate
molecular temperatures by
spectrum fitting.
• Rotational and translational
temperature,
• 6500 K in the torch
• 3500K in the test section
Emission spectroscopic measurement
13
Spectroscopic Measurement
Operating Conditions
Mass
Input
Pressure
Gas Flow Rate
Power
(kPa)
(g/sec)
(kW)
N2
0.65
10
19
Specification of Spectrometer
Manufacturer
StellarNet Corp.
Model
BLUE-Wave UVNb
Wavelength
200-1050 nm
Grating
600 g/mm
Resolution
0.50nm
CCD
2048 pixel
Observed Positions (mm)
14
Spectrum（A, B, C）
N
A
B
C
N2(2+)
N2+(1-)
Unresolved
contributions
N2(1+)
Spectra at C1, D1 and E1 (N2, 0.65g/sec, 10kW, 19kPa)
15
Comparison of C, D, E
N2(1+)
N2+(1-)
Unresolved
contributions
Normalized Intensity
N2(2+)
C
D
E
Spectra at E1, F1 and T1 (N2, 0.65g/sec, 10kW, 19kPa)
Observed Positions
16
Temperature Estimation by SPRADIAN
SPRADIAN ver1.5
• Multipurpose software package for radiation analysis
• Developed for JAXA’s sample return mission (MUSES-C) to evaluate the
radiation heating around its re-entry capsule
Conditions of Calculation
• Chemical species：N2 and N2+
• Temperature (Shape of emission spectrum is determined by rotational and
vibrational temperatures)
– Rotational Temp: 1,000K to 15,000K by 500K or 250K
– Vibrational Temp: 1,000K to 15,000K by 500K or 250K
JAXA MUSES-C
17
Spectrum Fitting Process
𝑶𝑶𝑶𝑶𝑶𝑶𝑶𝑶𝑶𝑶𝑶𝑶𝑶𝑶𝑶𝑶 𝑺𝑺𝑺𝑺𝑺𝑺𝑺𝑺𝑺𝑺𝑺𝑺𝑺𝑺𝑺𝑺 = 𝒂𝒂 × 𝑵𝑵𝟐𝟐 𝑺𝑺𝑺𝑺𝑺𝑺𝑺𝑺𝑺𝑺𝑺𝑺𝑺𝑺𝑺𝑺 + 𝒃𝒃 × 𝑵𝑵+
𝟐𝟐 𝑺𝑺𝑺𝑺𝑺𝑺𝑺𝑺𝑺𝑺𝑺𝑺𝑺𝑺𝑺𝑺 + 𝒄𝒄
•
Integrate
•
a, b, c: adjusted to reduce
residual error by rootmean square method.
Temperatures are
determined from the most
fitted combination of
spectra.
18
Fitting Range (A, B, C, D)
•
Fitting Range: 280nm-480nm
–
–
N
strong emission intensity
band spectra from molecules
A
B
C
D
N2(2+)
N2+(1-)
Unresolved
contributions
N2(1+)
Observed Positions
19
Result (A, B, C, D)
•
•
Estimated temperatures of N2+ is consistent with the numerical modeling.
Nitrogen plasma is close to local thermal equilibrium (Rotational Temp ≈
Vibrational Temp).
Estimated rotational and vibrational temperature of N2+
N2+
N2
N2+
N2+
Pos
A
B
C
D
Rotational, K
6,500 ± 1,000
8,500 ± 2,000
6,000 ± 1,000
5,000 ± 1,000
Vibrational, K
7,500 ± 2,000
7,500 ± 2,000
7,000 ± 1,500
5,500 ± 2,000
Estimated rotational and vibrational temperature of N2
N2
Comparison of numerical and observed spectra at C
Pos
A
B
C
D
Rotational, K
6,500 ± 5,000
8,500 ± 6,500
7,000 ± 5,000
6,500 ± 5,000
Vibrational, K
15,000 ± 7,000
15,000 ± 9,000
15,000 ± 7,000
15,000 ± 6,000
20
Fitting Range (E)
•
Fitting Range: 690nm-900nm
–
–
strong emission intensity
band spectra from molecules
E
Unresolved contributions
N2(1+)
N2+(1-)
Observed Positions
Spectrum at E (N2, 0.65g/sec, 10kW, 19kPa)
21
Estimated Temperature (E)
• Nitrogen plasma is close to local thermal equilibrium.
• The rotational and vibrational temperatures are nearly equal to the
results of numerical modeling.
N2
Pos
Rotational, K
Vibrational, K
E
3,250 ± 1,750
4,250 ± 1,750
Comparison of numerical and observed spectra at T1
22
Spectrum Fitting (LTE)
N2+
•
C
Spectrum Fitting is conducted with the assumption
of LTE.
N2
Estimated Temperatures
(1000K to 15,000K by 200K)
Pos
A
B
C
D
E
E
Temperature, K
6,700 ± 1,000
8,300 ± 2,200
6,100 ± 1,000
5,500 ± 1,000
4,100 ± 1,000
N2
Flow enthalpy at T1:
4.9±1.2 MJ/kg
Observed Positions
23
Radial Temperature Distribution
•
•
•
•
Estimate the radial distribution of temperatures from E to E’ with the assumption of LTE.
Radial distribution of spectra is modified by Abel inversion.
Emission intensity significantly decreases from 40 mm to 60 mm.
Diameter of the flow core: 40 mm. similar result with the heat flux measurement
Radial Temperature Distribution from E to E’
24
Temperature and Pressure
• Pressure: High => local thermal equilibrium (Default: 19kPa)
Low => local thermal equilibrium at C
non-equilibrium at E
Pressure and temperatures of N2+ at C
(N2, 10kW, 0.65g/sec)
Pressure and Temperatures of N2 at E
(N2, 10kW, 0.65g/sec)
25
Summary and Future work
Summary
Spectrum
• There are emission from N2(1+), N2(2+), N2+(1-) and N.
• There is unresolved emission from 500 nm to 700 nm.
Temperature Estimation
• Nitrogen plasma is close to local thermal equilibrium.
• Plasma torch: 6,500 K ~ 8,500 K
• Test section: 4,000 K
• Flow enthalpy: 4.9 MJ/kg
• Pressure and temperature
Plasma Torch
– High pressure: local thermal equilibrium
– Low pressure: non-equilibrium
Future work
• Investigate the unresolved emission.
Test Section
• Investigate the relationship between pressure and temperature.
• Estimate temperatures of the air plasma.
26
Pressure and Spectrum
• Emission intensity is different at different pressure
• Chemical species are the same at different pressure.
Pressure and spectra at C (N2, 10kW, 0.65g/sec)
27
Pressure and spectrum
•
•
Chemical species are different at different pressure in shorter wavelength.
Chemical reaction at low pressure is different from that at high pressure.
Pressure and spectra at E (N2, 10kW, 0.65g/sec)
28
Error range of temperature estimation
• Assume spectrum shape can be calculated within 5% error by SPRADIAN.
⇒ computed spectrum intensity (Ii) has 5% uncertainty.
• The sum of the minimized performance function and the uncertainty gives the
tolerable maximum value of the performance function. => tolerable temperature
Performance function
𝑁𝑁
2
1
𝐼𝐼𝑖𝑖
𝑆𝑆 = � �
− 1�
𝑁𝑁
𝐼𝐼(𝜆𝜆𝑖𝑖 )
𝑖𝑖
i : Number of pixel
Ii: Numerical Spectrum
Ii(λi): Observed
Spectrum
Uncertainty in performance function
𝑁𝑁
2
𝐼𝐼𝑖𝑖
𝐼𝐼𝑖𝑖
𝛿𝛿𝛿𝛿 = � �
− 1� × 0.05
𝑁𝑁
𝐼𝐼(𝜆𝜆𝑖𝑖 )
𝐼𝐼(𝜆𝜆𝑖𝑖 )
𝑖𝑖
Ref: Fujita et al, Spectroscopic Flow Evaluation in Inductively Coupled Plasma Wind Tunnel
,“Journal of thermo physics and heat transfer, 2008.
29
Radial Temperature Distribution
Radial Temperature Distribution at T1
(N2, 10kW, 0.65g/sec, 19kPa)
Radial Temperature Distribution at T1
(N2, 10kW, 0.65g/sec, 3.3kPa) 30
Previous Research Project
31
Reentry Trajectory
Velocity vs. Altitude
Comparison between flight
data and prediction by the
trajectory simulation
Flight data obtained by
secondary surveillance radar
Prediction of the experimental
vehicle with aeroshell
(CB=15kg/m2)
Prediction of only capsule
without aeroshell
(CB=500kg/m2)
Flight data agree with prediction and flexible aeroshell demonstrated decelerating
performance as predicted. Flexible aeroshell can decelerated at higher altitude
than a conventional reentry capsule.
TITANS
初期条件：高度400km円軌道（軌道速度7668m/s）から減速度：90m/sで軌道離脱
突入機諸元：機体重量15kg，エアロシェル直径2.5m，頭部直径0.5m，頭部曲率半径0.5m
（従来型のカプセル（機体重量15kg，エアロシェルなし，機体直径0.5m，頭部曲率半径0.5m）と比較する．）
＜軌道離脱から再突入までの軌道の概略＞
＜大気圏突入時の速度と高度の関係＞
TITANSの終端速度は7m/sで
そのまま安全に緩降下できる
減速度90m/sで軌道離脱した場合，
地球を約半周した後に大気圏に突入する．
TITANSは柔軟エアロシェルにより，
従来システムより高高度（高度100km
以上）から減速が開始される．
TITANS
＜大気圏突入時の加速度履歴＞
TITANSは最大８Gの減速度を受ける（G環境は
従来システムと同等）．突入機重量は15kgなので，
エアロシェルにかかる総荷重は120kgf程度であ
る．１ｍクラスのエアロシェルでは十分に耐えうる
荷重であるが，エアロシェルを大型化した場合に
関しては，今後実証が必要である．
＜大気圏突入時の淀み点熱流束履歴＞
淀み点は最大熱流束130kW/m2 （輻射平衡温
度で1000℃相当）であり，従来システムに比べ，
1/5程度に抑えられている．ただ，エアロシェル
は，700℃程度の温度を経験すると推定される．
そのため現在の技術からのさらなる膜面材料
の高性能化が実利用時の有用性を高める．
System Block Diagram
A-gas (Center, Axial)
B-gas (Around, Axial)
Hand Valve
×３
Argon
Gas
Mass Flow Controller
×３
C-gas (Around, Helical)
Manometer
Manometer
Threeway
valve
Air
Gas
Hand Valve
×３
Cooling Water
System
(Upper Manifold)
Control Panel
HighFrequency
Power Source
4MHz,10kW
Mass Flow Controller
×３
Auto-Matching
Box
Vacuum Pump
Gate Opening
Rate Controller
Filter
Coil (3 turns)
Plasma Torch
Auto-Ignition
System
Cooling Water
System
(Lower Manifold)
Vacuum Valve
(Controllable
Gate Opening
Rate)
Manometer
Hand
Valve
Hand
Valve
Leak Hand
Valve
Atmosphere
Vacuum Chamber
35
Heat Flux Measurement
36
Heat Flux Measurement
37
38
Spectrum at F1
N2(1+)
N2(2+)
NI
N2+(1-)
Unresolved
contributions
Fig4. Spectra at F1(N2, 0.65g/sec, 10kW, 19kPa)
39
Spectrum（T1）
Unresolved contributions
Unresolved
contributions
N2(1+)
N2+(1-)
Fig5. Spectra at C1, D1 and E1 (N2, 0.65g/sec, 10kW, 19kPa)
40
Result (C1, D1, E1)
• N2+: Thermal equilibrium (Rotational
= Vibrational
Pos
• N2: Emission intensity is low =>
C1
large error range
D1
N2+
E1
N2
N2+
F1
Table 6. Estimated Temperature
N2+
Rotational, K
6,500±1,000
9,000
(+3,500/-2,500)
6,000
(+500/-1,000)
5,000
(±1,000)
Vibrational, K
7,500
(+2000/-1,500)
7,500
(+3,500/-1,500)
7,000
(±1,500)
5,500
(+2,000/-1,000)
N2
Rotational, K
6,500
(+8,500/-2,500)
8,500
(+6,500/-5,500)
Vibrational, K
15,000
(+0/-7,000)
15,000
(+0/-9,000)
E1
7,000
(+8,000/-3,000)
15,000
(+0/-6,500)
F1
6,500
(+8,500/-2,500)
15,000
(+0/-6,000)
Pos
C1
D1
Comparison of numerical and observed spectra at E1
41
Fitting at LTE assumption
• There is no large difference between numerical and experimental
spectrum.
Table. Temperature
• The assumption of LTE is reasonable.
N2 and N2+
Spectrum Fitting (at E1)
Pos
N2+
N2
C1
N2+
D1
E1
F1
T1
Rotational, K
Vibrational, K
6,700
(±1,000)
8,300
(+2,200/-1,600)
6,100
(+800/-1,000)
5,500
(±1,000)
3,900
(±600)
cp=1.2 (kJ/K・kg)
Enthalpy
7.2~10MJ/kg
42
A
B
C