Beam Windows

1
Beam Windows for high transverse density beams
By the CERN Groupe Méthodes Presented
by L.Bruno AB/ATB Targets Dumps Section
2
Beam Windows For high transverse density beams
OUTLINE 1. Statement of the problem 2. Standard
window systems 3. New developments
4. Window materials 5. Summary
3
The structural problemPrimary static membrane
stresses
Titanium a 30 mm(Beam Line)sc 264
MPayc 1.4 mma 50 mm(Target Unit)sc 339
MPayc 2.7 mm
q1 bar
0
sc
Thickness t0.1mm
The structural problem implies large
displace-ments (ycgtt/2), which call for iterative
solutions.
Solid circular disc fixed and heldunder
uniformly distributed pressure
4
Comparison Beryllium windowsPrimary static
membrane stresses
Beryllium Ø 2.5 (a32mm) (Beam Line)sc 166
MPayc 0.37 mm Ø 4 (a51mm)(Target
Unit)sc 222 MPayc 0.84 mm
q1 bar
0
sc
Thickness Ø 2.5 t0.020 (0.508mm) Ø
4 t0.025 (0.635mm)
The structural problem implies large
displace-ments (ycgtt/2), which call for iterative
solutions.
Solid circular disc fixed and heldunder
uniformly distributed pressure
5
Temperature distributionCNGS beam parameters
Beam sigma 0.53 mmNominal int. 2.4 1013
pUltimate int. 3.5 1013 p
Ultimateintensity
Temperatures too high
Nominalintensity
M.Sans
r mm
The temperature distribution induced by the CNGS
beam was evaluated by FLUKA. Temperatures higher
than 50C are affected by statistical errors lt1.
The simulation covered 3 107 events with a 0.25mm
x 0.1mm axi-symmetric binning.
6
Comparison Beryllium windows CNGS beam
parameters
Ultimateintensity
Beam sigma 0.530 mmNominal int. 2.4 1013
pUltimate int. 3.5 1013 p
SF0.64 _at_ 200C SF0.73 _at_ 300C
a32mm
Nominalintensity
SF0.85 _at_ 200C SF0.97 _at_ 300C
a51mm
M.Sans
r mm
The temperature distribution induced by the CNGS
beam was evaluated by FLUKA. Temperatures higher
than 50C are affected by statistical errors lt1.
The simulation covered 3 107 events with a 0.4mm
x 0.1mm axi-symmetric binning.
7
Temperature distributionTest case Beam sigma
0.795 mm
Ultimateintensity
Beam sigma 0.795 mmNominal int. 2.4 1013
pUltimate int. 3.5 1013 p
SF0.61 _at_ 300C SF0.69 _at_ 400C
a30mm
Nominalintensity
SF0.78 _at_ 300C SF0.88 _at_ 400C
a50mm
M.Sans
r mm
The temperature distribution induced for a test
case beam was evaluated by FLUKA. Temperatures
higher than 50C are affected by statistical
errors lt1. The simulation covered 3 107 events
with a 0.4mm x 0.1mm axi-symmetric binning.
8
Temperature evolutionat the beam spot Test case
t ms
Titanium has a relatively low thermal diffusivity
of 2.7 mm2/s (as a comparison, graphite has 75
mm2/s). This value implies a build-up of
temperature at the second extraction.
9
Brazed Windows (SPS beam obstacles and monitors)
Brazed joint
knife
knife
Weld
Air
Vacuum
Flange
Flange
Foil
10
Captive Windows (nTOF Fast Induction Chamber)
Vacuum system
11
Reinforced Captive Windows
Vacuum
Reinf.
Air
Helicoflex joint
Foil
12
Window Materials
Captive reinforced windows
Captive windows
(L.Bruno, S.Sgobba)
13
Status and Future Activities

• High specific heat,thermal diffusivity and
  physical/chemical stability (e.g. crystal growth,
  oxidation resistance) are key factors
• De-coupling the pressure-bearing from the
  leak-tighness function eases design and lowers
  costs
• Standard iterative analytical design formulas are
  very conservative for ductile materials.
• Thermal stresses, stress waves, fatigue, axial
  shocks are to be studied.
• Tests are planned at CERN under CNGS nominal beam
  load.