Target Tracking, Tracking/Beam Steering
Interface, and Target Fabrication Progress
presented byRon Petzoldt
HAPL Project ReviewAtlanta, Georgia
February 5-6, 2004
Outline
I. Target Tracking- Experimental system progress- Effect of chamber gas pressure on target placement and tracking
II. Tracking/Beam Steering Interface- A new concept for Tracking/Steering alignment
III. Target Fabrication - Insulating foam
Flange
Timing (20 mm)
Position Measurement
window40 mm
20 mm22 mm
18 mm
LasersLens Flange apertures Lens
Photodiode
Line scancamera
Timing
Position
45 mm
40 mm
~18 mm
~50 mm
A 4 mm diameter target will decrease the photodiode current more than 10% (trigger) over a range of 13 to 14 mm.
The line scan camera sees both sides of the target over similar range.
The functional span is about ±6 mm
How to get micron-leveltracking accuracy (1)
Detector calibration is accomplished with target on translation stage
- Flat field correction corrects for variability in laser intensity and camera response - Also accomplished automatically before each shot sequence
01000200030004000500060007000
10 15 20 25
Target Height (mm)..
Shadow Center
(pixels)
How to get micron-leveltracking accuracy (2)
Flat field corrected data
We achieved tracking with high reliability on all three detector stations
Air rifles were used to fire “surrogate targets” through the tracking stations
- Previously achieved ~ 3 m tracking repeatability in stationary tests- Now working on in-flight
- Modular development - using 4.5 mm BB’s (and pellets)
- Up to 300 m/s with surrogate targets- Multiple shot capability- Installed back on main line
Target “shadow” (raw data)
We have calculated the displacement of the target due to chamber gas velocity*
• Output from SPARTAN code was used to estimate the deflection of a target injected 100 ms after a target explosion
• These preliminary calculations are conservative and results should be considered mostly as illustrative.- Stokes law assumed
F = 3ud- Chamber gas velocity based on initial SPARTAN code results which did not include radiation effects
• Target displacement calculated as a function of injection velocity and gas imparted acceleration
• Three target trajectory paths considered
(*from Z. Dragojlovic, UCSD)
Three Injection Paths Considered:Path I
Example results for target at 600 m/s• 50 mTorr Xe• 0.1 s after microexplosion• Target within 5 mm of center• Other paths had much less deflection
5 mm circle around chamber center
Target trajectory
Injection Path I
Peak gas velocity approximately 300 m/s
Path II
Path III
Preliminary Results are Encouraging
• Depending on the injection path, injection velocity of ~100-600 m/s required tomaintain displacement within ~ 5 mm
• These results are preliminary and need to be confirmed based on updated chamber gas velocity profiles
• The force imparted on the target by gas flow also needs to be better assessed.
B
B
B
J
J
J
H
H
0.0
1.0
2.0
3.0
4.0
5.0
6.0
0 200 400 600 800 1000
Injection Velocity (m/s)
Target Injection Path:
III
III
Preliminary results indicate proper chamber design can minimize target displacement
Data point from previous example
Chamber gas also affects tracking accuracyThe time between position measurements must be guided by
measurement accuracy and acceleration uncertainty.Example:Position measurement uncertainty X0 = 10 m Acceleration uncertainty a = 300 m/s2
Initial velocity uncertainty V0 = 2 X0/t +at/2Time measurement uncertainty t = 0 (very small)Position measurement interval t = 250 s
= 4 kHz or 0.1 m at 400 m/s
€
ΔX =ΔX0 +ΔV0t+0.5Δat2 =49μm 0.1 m 0.1 m
Chamber center
Next to lastmeasurement
Finalmeasurement
at 0.1 m interval; in-chamber tracking may be inadequate
“Continuous” tracking with extremely fast beam steering response could improve accuracy
• Mirrors would be “continuously” aligned with target tracking measurements.• Alignment actuators would have to be closely spaced because material
- sound speed (5100 m/s for aluminum)and damping will limit steering response time.
• Continuous tracking requires ~10 m accuracy at ~ 10 kHz at ~10 m distance- 100 m accuracy (10 m resolution) achieved commercially at 480 Hzfrom 2 m with 2 m (56 degree) field of view* (accuracy scales with FOV)
• Discrete trackers could acquire the target, reduce the required field of view, and hand off to continuous tracking system.
*Tracy McSheery Private Communications, 23 Jan 2004 - PhaseSpace Inc.
InjectorDiscretetrackers
Continuoustracker
A new concept for target tracking and driver alignment*
*Mark Tillack and Ron Petzoldt
Moving target
PSD?
Quad Quad
Driver beamFull or reduced power
Removable mirrors?
Tracking beam
Beam combiner
Alignment mirrors
Beam and tracking final pointing mirror
X, Y translation
, angle
Optical filter
A method must be developed to ensure a common reference for driver beams and target tracking One concept is to somehow measure where beams actually hit the targetsThis new concept uses common optics for driver and trackingRequires separate tracking for each driver beam
Current PSD accuracy is not adequate for in-chamber tracking
Hamamatsu S1881 PSD
22 mm
22 mm
Active area
Zone A
Zone B
Resolution = 2.8 micronZone A Position error = ±150 micronZone B Position error = ±400 micron
Voltage rise time is 3 s which is also too long for high-speed tracking(the target moves 0.4 mm/s)
Conclusion: An alternative sensor would probably be needed to perform the PSD function
We measured Young’s modulus for RF foam
Simple measuring system
Force (grams) vs Compression
Linear part of measurementsgives E = 0.25 MPa
105 mg/cm3 RF foam DimensionsH = 6.1 mmW = 6.3 mmL = 7.4 mm
1-D estimates for compression of foam by accelerated target are given by:
050
100150200250300350400450
0 0.5 1 1.5 2 2.5Compression (mm)
Forc
e (
gra
m)
Similar measurements:14 mg/cc TPX foam E= 0.11 MPa100 mg/cc DVB foam E= 0.76 MPa
RF foam sample
Ansys calculations show 1.3 m compression RF foamand 0.4 m for DVB
Acceleration
Comp
Comp.
Low modulus for DVB and RF foam will decrease buckle pressure (and increase fill time)
Foam Thickness( )m
Density( /kg m3)
Y ’oungsmodulus
MPa
0.365E/(AR)2
(a )*tm
BucklepressureANSYS(a )tm
PS 200 100 31.5 1.21 2.67200 100 0.76 0.029 0.057DVB260 100 0.76 0.047 0.095200 105 0.255 0.010 0.029RF260 105 0.255 0.016 0.034
Target Radius = 1950 mExternal PS membrane 1 m (2 m for PS foam calc)
Buckle Pressure Calculations
*Roark and Young, Formulas for Stress and Strain (1982) -Real Buckle Pressure
ANSYS buckling model
Liquid surface tension during insulating foam drying could damage foam cells
Options for drying insulating foam
Freeze drying1. Fill targets with DT gas.2. Freeze DT.3. Draw vacuum to sublimate DT gas from insulating foam.
Alternate method1. Fill targets with DT gas.2. Supercritical evaporation from insulation to critical point
(39.4 K, 1.77 MPa).3. Continued drying with reduced pressure in insulation to prevent
condensation.4. Slightly higher pressure inside seal coat to allow condensation.
Requires care to avoid shell rupture.
Fill chamber
Insulating foam
Seal Coat
Inner DT
€
ΔP =2σΔrr
≈2×108Pa2μm2mm
=0.2MPaBurst pressure
Insulating foam must be very concentric with target to achieve uniform DT layer thickness
He gas greatly increases thermal conductivity in insulating foamDT Fuel
Permeation boundary
Open cell foam
DT gas radius r1
€
Q=4 mgcm3
0. 25g
⎛
⎝ ⎜ ⎜
⎞
⎠ ⎟ ⎟
0.0487 Wcm3
⎛
⎝ ⎜
⎞
⎠ ⎟ =0.78 mW
€
ks+g ≈kg +fs23
−fs3
⎛
⎝ ⎜
⎞
⎠ ⎟ ρf
ρsks
≈0.024W /m⋅K +0.923
−0.93
⎛
⎝ ⎜
⎞
⎠ ⎟ 0.1( ) 0.07W /m⋅K( )
=24mW/m⋅K +2.3mW/m⋅K =26mW/m⋅K
Beta decay heat causes temperature drop across insulating foam during layering
Small non-concentricities cause changes in outer DT temperature
A
B
A< BOffset = (B-A)/2Nominal foam thickness = 150 mNominal DT thickness = 450 m
0 1 2 3 4 50
500
1000
1500
2000
2500
3000
3500
4000
T (K)
Offset (m)
DT outer surface temperature difference vs foam offset
(assumes uniform outer foam temperature)
A small non-concentricity of the foam layer results in a much greater DT layer non-concentricity
0 0.4 0.8 1.2 1.6 20
2
4
6
8
10
Offset of the foam (m)=(B-A)/2
DT Offset (m)=(C-D)/2
The DT is repositioned during Beta layering to maintain a uniform inner surface temperature
k(DT) = 330 mW/m•K k(foam) = 26 mW/m•K
A
B
C
D
Conclusion: DT offset requirement is unknown but expect foam non-concentricity must be < 1%
Summary and Conclusions
I. Target Tracking- Tracking station reliable operation was developed offline- Detectors are ready of online operation- Preliminary calculations indicate that proper chamber design and injection path selection can minimize gas induced target displacement
II. Tracking/Beam Steering Interface- A new concept for Tracking/Steering alignment is under consideration
III. Target Fabrication - Low measured DVB and RF foam strength would increase fill time- Insulating foam thickness must be very uniform
Next step: Focus on online target tracking