Detector Cooling

1
Detector Cooling

• Jaak Lippmaa
• HIP, University of Helsinki

2
State of play OCT 2006

• Detectors are in Secondary Vacuum
• To reduce Conductive and Convection heat load
• To minimize DetectorltgtPocket Wall distance
• Detectors are in Neutral Gas Atmosphere
• To avoid vacuum equipment in the tunnel
• Simpler mechanics

3
State of play OCT 2006

• Should we cool the Beam Pipe to reduce Radiative
  heat transfer from Pocket Wall to Detectors
• Should we cool part of Pockets facing Detectors
  (Stainless Steel is bad heat conductor)
• Should we discard Radiative heat load altogether
  since the effective area of Detectors is small
  (40 mm2 for 10 planes)

4
Pocket PlayCourtesy of Berend Winter
Benoit reported reduction of inward bend down to
80 micron, but we still want to study possible
bending due to local cooling of the Pocket Wall
300 micro-meter wall thickness (purple)
Pressure vectors (Pointing inwards 100,000 Pa)
3 mm wall thickness (green)
5
Pocket PlayCourtesy of Berend Winter
This is how mere vacuum deforms the pocket. Even
the 100 micron range deformations should be
accounted for in the Alignment Procedures
6
Pocket PlayCourtesy of Berend Winter
-30ºC Cold Spot in the Pocket Wall ?T50ºC
(ambient temperature is 20ºC)
7
Pocket PlayCourtesy of Berend Winter
Pressure
Pressure and thermal
Undeformed reference (deformation is amplified
in the graphics)
Deformation pulled back due to thermal
contraction as much as 30 microns
8
Pocket Play

• Local cooling introduces inevitable mechanical
  strain on the Thin Window
• Thermal gradients influence total deformation
  caused by atmospheric pressure on the Pocket
• Conclusion Cooling Pockets partially may not be
  such a good idea after all!!!

Convective Heat Transfer increases Cool Area to
a section when Detectors are not in Secondary
Vacuum
9
Pocket Play

• Should we cool the Beam Pipe to reduce Radiative
  heat transfer from Pocket Wall to Detectors
• Probably NOT as it needs loads of cooling power
• Should we cool part of Pockets facing Detectors
  (Stainless Steel is bad heat conductor)
• Probably not as it introduces additional
  deformations to the Thin Window AND would be
  mechanically complicated AND increase the cool
  mass anyway

Present cooling through Copper Heat Sinks is
constrained certain temperature gradient over
the detector area in X direction is unavoidable.
WE NEED TO REDUCE CONVECTIVE HEAT TRANSFER !!!
10
Heat Loads
BOX L100 mm W134 mm H100 mm
H
L
W
I2R Load Convection Conduction Insulation
Loss Radiative Transfer
11
Heat Loads

• I2R Load assume 25 W
• Radiative
• SiCu -gt0.01 mW
• Box -gt0.002 W
• Convective
• (N2) 0.4 W
• (vacuum) 0 W
• Conductive
• Only contact is Heat sink
• Load Essentially I2R

Tamb 25 C Tc -30 C TDET -20 C
Copper Heatsink
Detector
Beam
BOX
12
Cooling Options

• Since the heat load inside the Box is essentially
  electrical heat generated by the Detector
• Especially when the Detectors are in Secondary
  Vacuum
• Need to integrate CF cooling with TOTEM and ATLAS
• We should consider Thermoelectric Cooling (TEC)

13
TEC - Intro

• No moving parts
• MTBF 100000 h
• No gases involved
• Temperature control is easy down to fraction of
  degree
• Tubes not needed
• Vacuum compatible devices available

14
TEC Radiation Hardness

• Bi-Te and Pb-Te compounds are used in TEC units
• Measurements of doped TEC compounds (Seebeck
  Coefficient and Conductivity) have been carried
  out in neutron fluxes up to 1.6 x 1019 (fast
  neutrons) by General Electric Company (Knoll
  Atomic Power Laboratory) in 1960
• Compounds are radiation tolerant (Bi-Te more so)
• Increase in resistivity and SC is due to decrease
  of carrier concentration
• Radiation damage can be annealed at temperatures
  below 300 C
• Changes can be compensated by adjusting current
  and voltage
• TEC elements are in use in ATLAS laser cooling

15
TEC Selection

• One TEC could be sufficient
• Two (Mounted Top and Bottom) is better due to
  twice smaller heat load

16
SUMMARY

• Cooled radiation shield around Tracker (Benoits
  Green Box as proposed by Mimmo)
• Vacuum is absolutely necessary if we want GASTOF
  and Silicon Trackers in the same box
• Thermoelectric cooling is an option
• Pocket wall stays warm