GRAN SASSO - PowerPoint PPT Presentation

About This Presentation
Title:

GRAN SASSO

Description:

GRAN SASSO'S HADRON STOP. Temperature's behaviour under specified beam conditions ... decided to simplify the model making the pessimistic hypothesis that between ... – PowerPoint PPT presentation

Number of Views:35
Avg rating:3.0/5.0
Slides: 17
Provided by: barbarac72
Category:

less

Transcript and Presenter's Notes

Title: GRAN SASSO


1
GRAN SASSOS HADRON STOP Temperatures
behaviour under specified beam conditions Barb
ara Calcagno
2
CNGS Facilities
3
Beam Conditions
2D Energy deposition map FLUKA
FORTRAN interpolation program to convert the
energy deposition map 2D ? 3D
Integration of the data over the whole volume
ET-2D Total energy - 2D model17 kW ET-3D
Total energy - 3D model 16.9 kW Difference lt
0.6? negligible for the purpose of the
calculations performed
4
PURPOSES OF THE BEAM DUMP
  • Contain in a relatively small volume the
    radioactivity introduced by the beam particles
    reaching the end of the decay tunnel
  • Shield the detectors from the beam particles
    reaching the end of the decay tunnel
  • Contain the energy deposited by the beam
    particles and keep the local temperature within
    acceptable limits, during the longest data taking
    period (200 days) at the highest possible beam
    intensity (13.8x1019 pot/year) and at a proton
    beam energy of 400 GeV
  • Contribute to the shield of a possible near
    detector against muons produced by pions decaying
    in the decay tunnel

5
PRELIMINARY CONSIDERATIONS
  • The mass of graphite and cast iron are made of
    blocks the maximum roughness measured and,
    consequently, the maximum air gap between
    graphite and cast iron blocks is of 2 mm and 4 mm
    respectively. The contact between blocks, even if
    not perfect, is at least guaranteed over 50 of
    each surface. It has been decided to simplify the
    model making the pessimistic hypothesis that
    between blocks there is no contact but a layer of
    air, considered a solid with low conductivity,
    neglecting the effects of radiation and
    convection.
  • Preliminary calculations without cooling system
    shown that the temperature reach a value of 700
    oC ? needing of a cooling system

6
COOLING SYSTEM
Preliminary calculations without cooling system
700 0C after 200 days ? Destruction of concrete
structures ? Needing of a cooling system
HYPOTHESES - Cooling power 50 kW - diameter of
the tubes 4 cm - number of tubes 12 -
Turbolent regime U 0.2 m/s - Tbulk
17 0C ? -
Flow rate for each tube 6.5 l/min -Nu0.012
(Re0.87-280) Pr0.4
1.5? Pr? 500 3x103? Re? 106 ? -
h ? 500 W/m2K
?
7
CALCULATIONS HYPOTHESES
COMMON ASSUMPTIONS ? STANDARD CASE Graphite
blocks - horizontal and longitudinal airgap 2
mm Kxx 7.9 W/mK Kzz 9.4 W/mK
- vertical airgap 0.5 mm
Kyy 26.4 W/mK
Iron blocks - horizontal and
longitudinal airgap 4 mm Kxx 6.0 W/mK Kzz
9.8 W/mK -
vertical airgap 2 mm
Kyy 10.9 W/mK ?INITIAL TEMPERATURE
20 0C ? EXTERNAL SURFACES OF THE MODEL
INSULATED (worst case) ?NO CONVECTION IN THE
MUON PIT the air is considered as one block
material A-GEOMETRY
B-GEOMETRY ?VERTICAL SIMMETRY ? NO FLUX THROUGH
THE SURFACES OF SIMMETRY ? Only a quarter of the
dump is modeled ? 1 COOLING SYSTEM located
immediately below the aluminum box, with the
same cooling power of 50 kW.
?VERTICAL SIMMETRY - HORIZONTAL SIMMETRY ? NO
FLUX THROUGH THE SURFACES OF SIMMETRY ? Only a
quarter of the dump is modeled ? 2 COOLING
SYSTEMS located immediately below and above the
aluminum box, with the same cooling power of 50
kW.
8
A-GEOMETRY of the HADRON STOP
GRAPHITE
ALLUMINUM
TRANSVERSAL SECTION
HEAT SINK
2.6 m
4.0 m
2.8 m
6.0 m
CAST IRON
AIR
0.2 m
CONCRETE
LONGITUDINAL SECTION
3.0 m
5.0 m
5.0 m
18.2 m
23.8 m
9
B-GEOMETRY of the HADRON STOP
TRANSVERSAL SECTION
GRAPHITE
ALLUMINUM
1.3 m
1.6 m
2.0 m
1.4 m
HEAT SINK
CAST IRON
AIR
CONCRETE
LONGITUDINAL SECTION
3.0 m
3.2 m
5.0 m
18.2 m
5.0 m
10
A-Geometry.Temperature profile on the transversal
section related to the maximum temperature
reached in graphite after 200 days - 4.5x1019
pot/year
6 HOURS
10 DAYS
5 DAYS
12 HOURS
24 HOURS
TEMPERATURE0C
50 DAYS
25 DAYS
100 DAYS
200 DAYS
11
A-Geometry.Temperature profile on the transversal
section related to the maximum temperature
reached in graphite after 200 days - 4.5x1019
pot/year
TEMPERATURE0C
24 HOURS
5 DAYS
50 DAYS
10 DAYS
100 DAYS
200 DAYS
12
A- Geometry load cases studied
The results of the worst case considered -
13.8x1019 pot/year-without concrete - show the
needing of studying the possibility of using a
second cooling system.
13
B-Geometry.Temperature profile on the transversal
section related to the maximum temperature
reached in the graphite after 200 days -
13.8x1019 pot/year
6 H
24 H
5 Days
3 Days
12 H
TEMPERATURE0C
In 5 days the transient could be considered
concluded the maximum temperature in graphite
(194 0C) is reached
Starting from the 10th day, the temperature
profile doesnt change with the time
10 Days
20 Days
200 Days
14
Temperature Profile after 200 days of running
and comparison between the maximum temperatures
obtained with 1 and 2 cooling systems
LONGITUDINAL Temperature Profile related to the
maximum temperatures in iron and in graphite
TEMPERATURE0C
10 m
15
Temperature Profile after 200 days of running on
the external surfaces of the hadron stop
TEMPERATURE0C
TOP VIEW
LONGITUDINAL VIEW
FRONT VIEW
16
CONCLUSIONS
? The choice of two cooling systems allows to
keep the maximum temperatures under
reasonable limits, always below the values
accepted for analogous structures used in
previous experiments, also in the case of
continuos running with the maximum number
of protons on target 13.8x1019 pot/year ? Each
assumption which has been adopted is referred to
the worst case under the thermal point of
view? the results obtained guarantee that the
structure modelled would work in safety
conditions under the thermal loading
previously specified.
Write a Comment
User Comments (0)
About PowerShow.com