LH2 Absorber R & D
Illinois Consortium for Accelerator Research (IIT, NIU, UC, UIUC), U Miss
Oxford U
Mary Anne Cummings
NIU
Feb. 9, 2002
Mucool FNAL
Mucool LH2 Absorber Collaboration
E. L. Black, M. Boghosian, K. Cassel D. M. Kaplan, W. Luebke, Y. Torun
Illinois Institute of Technology
S. Ishimoto, K. Yoshimura
KEK
M. A. Cummings, A. Dyshkant, D. Hedin, D. Kubik
Northern Illinois University
D. Errede, M. Haney
University of Illinois, Urbana-Champaign
M. Reep, D. Summers
University of Mississippi
Y. Kuno
Osaka University
G. Barr, W. Lau
Oxford University
C. Darve, C. Johnstone*, A. Klebaner, B. Norris, M. Popovic, S. Geer
FNAL
* also research faculty at IIT
Feb. 9, 2002
Mucool FNAL
Absorber Window Design
Modified Torispherical integrated window
and flange design (tapered detail at left).
Machine parameters shown.
r
r
R
Precision measurement of window
and flange with CMM at FNAL
Industrial Center.
Feb. 9, 2002
Mucool FNAL
Window manufacture
(U of Miss)
Flange/window unit machined
from aluminum piece
Backplane for window
pressure tests
Feb. 9, 2002
Mucool FNAL
Window Test Setup
Test setup at NIU for window , front view:
pressure control system
Backplane with connections,
and with window attached
Strain gages
applied to
window
Feb. 9, 2002
Mucool FNAL
Photogrammetry
FEA calc. for
displacement
• Non-contact measurement of strain
by calculating displacement
• Compare with strain gage readout and
FEA calculations
Feb. 9, 2002
Mucool FNAL
Photogrammetry measurements during earlier pressure test.
Note the projected dots!
Feb. 9, 2002
Mucool FNAL
Latest window burst test
Feb. 9, 2002
Mucool FNAL
1.
Rupture tests
130 m window
The Latest: 340 m window
3.
Leaking appeared at 31 psi
..outright rupture at 44 psi!
2.
350 m window
Burst at ~ 120 psi
Feb. 9, 2002
Mucool FNAL
Burst at > 120 psi
Comparison of results on 130m window 1 at 0.18
Mpa test pressure:
FEA, elastic
region only
Deformation of Window end-cap Vs test pressure
0.6
FEA, non-elastic
region included
Test pressure (MPa)
0.5
0.4
0.3
P=.18MPa
0.2
0.1
0
0
40
60
80
100
120
distance from end-cap centre line (mm)
NIU Photogrammetry results &
non-linear FEA calculation
Feb. 9, 2002
20
FEA results by Oxford
Mucool FNAL
140
160
Comparison of results on 130m window at 0.24 Mpa test pressure
Comparing the Oxford and linear FEA results with the NIU photogrammetry
data
Deformation of Window end-cap Vs test pressure
0.8
Test pressure (MPa)
0.7
0.6
0.5
0.4
P=.24PMa
0.3
0.2
0.1
0
0
20
40
60
80
100
120
distance from end-cap centre line (mm)
FEA, non-elastic
region included
Feb. 9, 2002
Mucool FNAL
140
160
FEA Calculations
• 350 micron window (W. Lau)
• Non-elastic region included
• Three dimensional analysis,
necessary for vibrational analysis
Stress distribution when first
yield developed at 77.7psi
(0.536 MPa) pressure
Feb. 9, 2002
Mucool FNAL
Flow Tests
Density maintenance:
1.
Nozzle design and arrangement (E.Black, S. Dyshkant)
Bates LH2 target heat carrying performance determined by total eN
cross section measurements. We’ll need other confirmation.
2.
3-dim LH2 absorber flow simulations (W. Lau)
Will run 3-d sims for water, air, LH2 – demonstrate for water
with the “analog” simulation at NIU
3.
Absorber instrumentation
Absorber integrated into a cryo system, with extreme temperature and
pressure variations considered for safety.
Heat loss/feedthrough
Feb. 9, 2002
Mucool FNAL
Results of the fluid flow analysis:Oblique nozzle arrangement with air flow at room temperature
Air inlet
Inlet velocity = 100 m/s
Velocity profile in Y-direction
Feb. 9, 2002
Mucool FNAL
Safety
Mu-Cool Safety Panel has been formed. It consists of:
Wes Smart (Chairman)
Jim Kilmer
Jim Priest
Safety guidelines
1. Results of FEA calculations which show that the maximum window stress at 25 psid
2.
3.
4.
5.
internal pressure is less than (0.25)x(ultimate strength).
Results of a cold pressure test per paragraph II.C.3.b.(i) (for metal flasks) of the LH2
Target Guidelines. I would suggest that at least one window be burst under cold
conditions. Burst test at NIU with nitrogen
Results of a room temperature pressure test per paragraph II.C.3.b.(ii) of the LH2 Target
Guidelines. Burst test
Copy of the Material Certification Sheet of the material from which the windows are
fabricated provided
Other; including results of tensile testing samples of the material.
Feb. 9, 2002
Mucool FNAL
Design of LINAC LH2 Absorber Beam Test
Feb. 9, 2002
Mucool FNAL
Next
 Burst test with nitrogen at NIU
 1st safety report to committee
 Flow tests/ 3-d FEA study
 Heat exchange/containment for LH2 cryostat
 Instrumentation
Feb ‘02
Mar ‘02
Apr ’02
May ‘02
Jun ’02 - ??
 LH2 safety
 LH2 monitoring
 LH2 physics
 11 cm radius absorber
Data acquisition
Jun ’02 -
LH2
 Beam
 Cryotests for convection at NIU
Jul ’02 - ??
The program gets redefined after every milestone!
Feb. 9, 2002
Mucool FNAL
Thin Window Design
Minimize window thickness:
Minimum thickness depends on
shape (ASME Standard)
Hemispherical:
t = 0.5PL / (SE - 0.1P)
s = 0.5D
Ellipsoidal:
t = 0.5PD / ( SE - 0.1P)
s = .25D
Torispherical:
t = 0.885PD / ( SE - 0.1P)
s = .169D
t = min. thickness s = sagitta
Feb. 9, 2002
s
Variable thickness near window edges
can further reduce the minimum
window thickness near beam:
r
Title:
Title:
Creator:
Creator:
Previe w :
This EPS picture w as not saved
w ith a preview inclu ded in it.
Comment:
This EPS picture w ill prin t to a
PostScrip t prin ter, but not to
other types of prin ters.
Previe w :
This EPS picture w as not saved
w ith a preview inclu ded in it.
Comment:
This EPS picture w ill prin t to a
PostScrip t prin ter, but not to
other types of prin ters.
D
R
r
displacement
displacement
r
L
R
r
Mucool FNAL
beam axis
beam axis
ANSYS Finite Element Analysis, Zhizing
Tang, FNAL.
Strain Gage Measurement
• Measure the strain on window with strain
gages
• Use Finite Element Analysis program (ANSYS)
to relate vessel pressure to maximum window
strain – confirm this with strain gages applied
to window surface
FEA mesh for calcs. on the
window-flange unit
< two different gages
Instrumented
window
Feb. 9, 2002
Mucool FNAL
Strain gage data 350m window
Strain vs. time – the gage ceases to return to resting position after ~ 80 psi;
demonstrating
point.
Strain vs time -the
can seeyield
that rosette
did not return to 0 ustrain after experiencing ~80psi
2000
1.20E+02
Uniaxial at 132mm
Rosette at 12 mm Left gage.
Pressure
1.00E+02
Rosette Gage at 12 mm yielding..
Rosette at 12mm showing yielding..
1500
5000
8.00E+01
4500
1000
4000
PSI
Microstrain
6.00E+01
3500
4.00E+01
3000
2.00E+01
microstrain
500
2500
2000
0
0
500
1000
1500
2000
2500
0.00E+00
1500
1000
-500
-2.00E+01
500
Time (s)
-20.00
0
0.00
20.00
40.00
60.00
-500
PSI
Feb. 9, 2002
Mucool FNAL
80.00
100.00
120.00
Forced-Flow Absorber Design
External Heat Exchange:
Mucool ~ 100W
(E. Black, IIT)
Large and variable
beam width =>
large scale turbulence
Establish transverse
turbulent flow with
nozzles – complicated, hard
to simulate
Feb. 9, 2002
Mucool FNAL
Convection absorber design
Convection cell is driven by heater and
particle beam.
Heat exchange via helium tubes near
absorber wall.
Internal heat exchange:
Flow is intrinsically transverse.
Output from 2-dim Computational Fluid
Dynamics (CFD) calcs. illustrate the
concept. (K. Cassel, IIT)
<
Lines indicate greatest flow near beam
center.
Studies are encouraging, but there is a poorly
known parameter: hLH2, coefficient of
convective heat transfer. Prototyping of this
design is being done by Shigeru Ishimoto et al
at KEK.
Feb. 9, 2002
Mucool FNAL