Upload
donga
View
213
Download
0
Embed Size (px)
Citation preview
IFE Chambers:Modeling and Experiments at UCSD
Farrokh Najmabadi
5th US-Japan Workshop on Laser IFE
March 21-23, 2005General Atomics, San Diego
Electronic copy: http://aries.ucsd.edu/najmabadi/TALKSUCSD IFE Web Site: http://aries.ucsd.edu/IFE
UCSD Research under the High-Average-Power Laser Program
Final opticsTarget engineeringChamber dynamicsChamber armorBlanketSystem integration
I. Simulation of IFE Chamber Dynamics
The focus of our research effort is to model and study the chamber dynamic behavior on the long time scale, including Hydrodynamicsand Energy & particle transfer mechanisms
I. Simulation of IFE Chamber Dynamics
The focus of our research effort is to model and study the chamber dynamic behavior on the long time scale, including Hydrodynamicsand Energy & particle transfer mechanisms
Spartan Chamber Simulation Code
SPARTAN features:Physics: Navier-Stokes equations with state dependent transport properties; Coronal model for radiation.Numerics: Godunov solver; Embedded boundary; and Adaptive Mesh RefinementTwo-Dimensional: Cartesian and cylindrical symmetryUse results from rad-hydro codes (BUCKY) as initial condition.
SPARTAN features:Physics: Navier-Stokes equations with state dependent transport properties; Coronal model for radiation.Numerics: Godunov solver; Embedded boundary; and Adaptive Mesh RefinementTwo-Dimensional: Cartesian and cylindrical symmetryUse results from rad-hydro codes (BUCKY) as initial condition.
All results presented are for a chamber filledwith 50 mTorr of Xe
All results presented are for a chamber filledwith 50 mTorr of Xe
Without Radiation, a high temperature zone would be developed in the chamber
t = 0.5 ms
Tmax = 5.3 104 K Tmax = 3.1 105 K Tmax = 2.2 105 K
t = 3 ms(After one bounce)
t = 8 ms
Tmax = 1.3 105 K Tmax = 5.1 104 K Tmax = 3.2 104 K
t = 20 ms t = 100 mst = 65 ms
Convergence of shocksleads to a hot zone.
Convergence of shocksleads to a hot zone.
Hot zone does notdisappear
Hot zone does notdisappear
Radiation is the Dominant Energy Transfer Mechanism
0.0
0.5
1.0
1.5
2.0
2.5
3.0
0 20 40 60 80 100
time [ms]
Ene
rgy
[105 J
]
Case I: No Radiation
Internal Energy
Kinetic Energy Conducted to wall
After the first “bounce” kinetic energyof gas is converted into internal energy(hot zone). Conduction to the wall is too slow.
After the first “bounce” kinetic energyof gas is converted into internal energy(hot zone). Conduction to the wall is too slow.
Case II: With Radiation
0.0
0.5
1.0
1.5
2.0
2.5
3.0
0 20 40 60 80 100
time [ms]
Ene
rgy
[105 J
]
The energy is dissipated by radiation. The gas temperatureis effectively pinned byradiation (~5,000K for Xe)
The energy is dissipated by radiation. The gas temperatureis effectively pinned byradiation (~5,000K for Xe)
Internal Energy
No Radiation
Radiation
Conducted to wall
A variety of chamber geometries has been considered
Infinite cylinder (Cartesian symmetry)
Infinite cylinder (Cartesian symmetry)
Finite cylinder (Cylindrical symmetry)
Finite cylinder (Cylindrical symmetry)
Sphere (Cylindrical symmetry)
Sphere (Cylindrical symmetry)
Octagon(Cylindrical symmetry)
Octagon(Cylindrical symmetry)
Impact of Chamber Geometry on Chamber Temperature at 100 ms
Tmax = 4,420 KTavg = 1,430 K
Tmax = 4,310 KTavg = 1,440 K
Tmax = 4,580 KTavg = 1,440K
Tmax = 4,610 KTavg = 1,610 K
A large number of cases with different geometries were simulated: 3-D effects can be observed and sufficiently described by using a combination of Cartesian and axisymmetric cylindrical 2-D models.Shape of the chamber makes little difference in peaks and averages of the temperature (< 10%).Peak Xe temperature is set by radiation at about 5,000K.
A large number of cases with different geometries were simulated: 3-D effects can be observed and sufficiently described by using a combination of Cartesian and axisymmetric cylindrical 2-D models.Shape of the chamber makes little difference in peaks and averages of the temperature (< 10%).Peak Xe temperature is set by radiation at about 5,000K.
Infinite Cylinder(Cartesian Symmetry) Cylinder SphereTall Cylinder
II. Armor Simulation Experiments At Dragonfire Facility
II. Armor Simulation Experiments At Dragonfire Facility
Thermo-Mechanical Response of Chamber Wall Can Be Explored in Simulation Facilities
Capability to simulate a variety of wall temperature profiles
Capability to simulate a variety of wall temperature profiles
Requirements:
Capability to isolate ejecta and simulate a variety of chamber environments & constituents
Capability to isolate ejecta and simulate a variety of chamber environments & constituents
Laser pulse simulates temperature evolution
Laser pulse simulates temperature evolution
Vacuum Chamber provides a controlled environment
Vacuum Chamber provides a controlled environment
A suite of diagnostics:Real-time temperature (High-speed Optical Thermometer)Per-shot ejecta mass and constituents (QMS & RGA)Rep-rated experiments to simulate fatigue and material response
Relevant equilibrium temperature (High-temperature sample holder)
A suite of diagnostics:Real-time temperature (High-speed Optical Thermometer)Per-shot ejecta mass and constituents (QMS & RGA)Rep-rated experiments to simulate fatigue and material response
Relevant equilibrium temperature (High-temperature sample holder)
Real-time Temperature Measurements Can Be Made With Fast Optical Thermometry
Temperature deduction by measuring radiance at fixed λOne-color: Use tables/estimates for ε(λ1,Τ) Two colors: Assume ε(λ1,T) = ε(λ2,T) Three colors: Assume d2ε/dλ2 = 0 [usually a linear interpolation of Ln(ε)
is used]
Temperature deduction by measuring radiance at fixed λOne-color: Use tables/estimates for ε(λ1,Τ) Two colors: Assume ε(λ1,T) = ε(λ2,T) Three colors: Assume d2ε/dλ2 = 0 [usually a linear interpolation of Ln(ε)
is used]
Spectral radiance is given by Planck’s Law (Wien’s approximation):
L(λ,T) = C1 ε(λ,T) λ-5 exp(-C2/λT)
Since emittance is a strong function of λ, T, surface roughness, etc., deduction of temperature from total radiated power has large errors.
Spectral radiance is given by Planck’s Law (Wien’s approximation):
L(λ,T) = C1 ε(λ,T) λ-5 exp(-C2/λT)
Since emittance is a strong function of λ, T, surface roughness, etc., deduction of temperature from total radiated power has large errors.
Our observationsTwo-color method achieves sufficient accuracy (~1%). Three-color method is too difficult.
Our observationsTwo-color method achieves sufficient accuracy (~1%). Three-color method is too difficult.
Status of High-Speed Thermometer
Single fiber from head to splitter/detector
Band-pass filter/focuser
PMT
Expander/neutral filter
50-50 splitter
PMT
A Variety of Measure has reduced the noise in thermometer signal considerably
Tem
pera
ture
(K)
Time (10-7s)
Little difference in thermometer signal when averaged over 4, 8, 16, and 32 shots.
Little difference in thermometer signal when averaged over 4, 8, 16, and 32 shots.
Thermometer is calibrated based on the melting point of tungsten
Thermometer is calibrated based on the melting point of tungsten
Time (10-7s)T
empe
ratu
re (K
)
T Melt point
Armor Irradiation Test Matrix
Test matrix: Initial Temp. ∆T No. of ShotsSample 1A: RT 2,000oC 103 (100s)Sample 2A: RT 2,000oC 104 (~16 mins)Sample 1B: RT ~ 2,000oC 105 (~2.8 hr)Sample 3A: RT ~ 2,500oC 103
Sample 2B: RT 2,500oC 104
Sample 3B: ~ RT 2,500oC 105
Sample 4A: 500oC 2,000oC 103
Sample 5A: 500oC 2,000oC 104
Sample 4B: 500oC 2,000oC 105
Sample 6A: 500oC 2,500oC 103
Sample 5B: 500oC 2,500oC 104
Sample 6B: 500oC 2,500oC 105
Samples: Powder metallurgy tungsten samples from Lance Snead.
Test matrix: Initial Temp. ∆T No. of ShotsSample 1A: RT 2,000oC 103 (100s)Sample 2A: RT 2,000oC 104 (~16 mins)Sample 1B: RT ~ 2,000oC 105 (~2.8 hr)Sample 3A: RT ~ 2,500oC 103
Sample 2B: RT 2,500oC 104
Sample 3B: ~ RT 2,500oC 105
Sample 4A: 500oC 2,000oC 103
Sample 5A: 500oC 2,000oC 104
Sample 4B: 500oC 2,000oC 105
Sample 6A: 500oC 2,500oC 103
Sample 5B: 500oC 2,500oC 104
Sample 6B: 500oC 2,500oC 105
Samples: Powder metallurgy tungsten samples from Lance Snead.
Powder Metallurgy Tungsten Samples After Laser Irradiation
Optical microscope at low resolution“Black” areas appear black because of the “dull finish” (they appear as whitish to the naked eye)
Optical microscope at low resolution“Black” areas appear black because of the “dull finish” (they appear as whitish to the naked eye)
10,000 Shots
Samples are polished to a “mirror-like”finish. The “damaged” area has a “dull” finish.A brown background is placed in the photograph to enhance contrast.
Samples are polished to a “mirror-like”finish. The “damaged” area has a “dull” finish.A brown background is placed in the photograph to enhance contrast.
Effects of Shot Rate and Temperature Rise
103 shots 105 shots104 shots
370mJ (~2000oC ∆T), RT
530mJ (~2500oC ∆T), RT
Effects of Shot Rate and Temperature Rise
103 shots 105 shots104 shots
370mJ (~2000oC ∆T), 500oC
530mJ (~2500oC ∆T), 500oC
High Magnification
Effects of Shot Rate and Temperature Rise
103 shots 105 shots104 shots
530mJ (~2500oC ∆T), 500oC
530mJ (~2500oC ∆T), RT
Material Response: At First Glance
It appears that samples evolves at two different time scales:Low shot count: Defect planes appear, High shot count: Individual “nuggets” form (are we seeing the powder constituents breaking apart?)
Higher equilibrium temperature leads to less damageHighly visible in low shot counts, For example, 1,000 shots at ∆T ~ 2,500oC with 500oC sample is “almost” damage free while the corresponding RT sample shows damage.At high shot count, samples with higher equilibrium temperature also show “slightly” less damage.
It appears that samples evolves at two different time scales:Low shot count: Defect planes appear, High shot count: Individual “nuggets” form (are we seeing the powder constituents breaking apart?)
Higher equilibrium temperature leads to less damageHighly visible in low shot counts, For example, 1,000 shots at ∆T ~ 2,500oC with 500oC sample is “almost” damage free while the corresponding RT sample shows damage.At high shot count, samples with higher equilibrium temperature also show “slightly” less damage.