CERN Cooling Meeting, Wim de Boer, Univ. Karlsruhe 1 12.11.2008 A CO2 cooling test system Outline: 1. Reduce material budget (by combining cooling pipe, mechanical support and current leads into single structure?) 2. Cooling with CO2 3. First test results
22
Embed
Outline: 1. Reduce material budget (by combining cooling ...
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
CERN Cooling Meeting, Wim de Boer, Univ. Karlsruhe 112.11.2008
A CO2 cooling test system
Outline:
1. Reduce material budget(by combining cooling pipe, mechanical support andcurrent leads into single structure?)
2. Cooling with CO2
3. First test results
CERN Cooling Meeting, Wim de Boer, Univ. Karlsruhe 212.11.2008
Material budget reduction
Combine functionality of
PowerleadsCooling tubesMechanical support
CERN Cooling Meeting, Wim de Boer, Univ. Karlsruhe 312.11.2008
Ladder design by powering via cooling pipes
Cooling channels
Very important:sensor and hybridcan be at differenttemperatures!Thermal contact ONLY via bondwires
Modules mounted alternately above and under the alu structure
Supply and return line for CO2
CERN Cooling Meeting, Wim de Boer, Univ. Karlsruhe 412.11.2008
2.2 cm strixel design on 9x10 cm sensor
Granularity improved by factor of 8 (before 2 sensors=18 cm strips, now 2.2 )Power for 130 nm electronics factor 5 less/channel, so total power same order.Can use same technology of hybrids, if 512 ch preamps (before 256) and pitch of 130 μm (bond pads 65 μm)
CERN Cooling Meeting, Wim de Boer, Univ. Karlsruhe 512.11.2008
+
-~50A
~1A
Problem:every hybrid see´s a different voltage.
Powersupply: 2.4V, 40A(with sense wire)
Zener diode stabilizeseach ladder at 1.2V(so no change if 1 module fails)
2m COLD pure Al only 2.5 mΩ, sovoltage drop 100 mV,but similar on + and –
Hybrids see SAMEvoltage (to +- 20 mV)
No need for DC/DCconverters on hybrids
Voltage compensation in long ladders
Current back on neighbouring ladder in orderto have connections outsideand use current twice (combinedserial/parallel powering)
CERN Cooling Meeting, Wim de Boer, Univ. Karlsruhe 612.11.2008
Long Barrel with endcap at small radii
All connections for cooling and current outside tracking volume
CERN Cooling Meeting, Wim de Boer, Univ. Karlsruhe 712.11.2008
Strixel disks: bent ladders
Can use same sensors as in barrel, if somemore overlap allowed.Can have SAME hybridsand same “circular” laddersfor each ring
Will have middle ring at back of rohacell, so enoughspace for mountingon cooling rings
Note: disks in two halvesin order to mount withbeam pipe in place
CERN Cooling Meeting, Wim de Boer, Univ. Karlsruhe 812.11.2008
Numbers for Strawman C
All numbers similar to present tracker, except for #ch*7, power and current similar Note 1: only 2 types of sensors and all ladders the same. 32 sensors/ladder const.Note 2: Sensors and hybrids very similar to present ones, except i) strixels
ii)parallel powering iii) CO2 cooling. In total 1448 ladders, so if two in series fail, only 0.2% lost.
CERN Cooling Meeting, Wim de Boer, Univ. Karlsruhe 912.11.2008
Strixel Disks for Strawman C
Suppose 2 pixel disks at z=270 and z=360 mmand 7 strixel disks between z=450 and z=1550 mmStrixels add only 2Mch to 58 Mch for long barrelThe rings in the disk can be build as (bent) ladders with same sensors as barrel!All services from disks and pixels OUTSIDE tracking volume
CERN Cooling Meeting, Wim de Boer, Univ. Karlsruhe 1012.11.2008
Material Budget
Material budget :
•Mechanical structure: 0. 6%X0 / layer
• Si sensors: 0.4% X0/layer
•Hybrid + control lines: 1% X0/layer
Total: 2,0% X0/layer
CERN Cooling Meeting, Wim de Boer, Univ. Karlsruhe 1112.11.2008
Material budget with “endcaps” as inner discs
η
Mat
eria
l/X0
CERN Cooling Meeting, Wim de Boer, Univ. Karlsruhe 1212.11.2008
Material budget
CERN Cooling Meeting, Wim de Boer, Univ. Karlsruhe 1312.11.2008
Summary of cooling liquids at LHC
Notes:Single phase cooling simplest, but large pumps needed(Pumps: CMS: 1 floor, ATLAS: 1 room, LHC-b: 1 rack)Two-phase evaporation in principle much better, because
heat of evaporation much larger than specific heat, butany pressure changes means a temperature change, so be careful about tube bending, tube sizes etc.
CO2 has largest heat of evaporation, is non-toxic, non-flammable, industrial standard, liquid at room temperature, but high pressure (73 bar at 31 C)
300
CERN Cooling Meeting, Wim de Boer, Univ. Karlsruhe 1412.11.2008
CERN Cooling Meeting, Wim de Boer, Univ. Karlsruhe 1912.11.2008
Flowmeters
bybrid withheater andT-sensor
Some pictures
Cold liquid sent through ladder.Blue temperature curve shows position of liquid.
CERN Cooling Meeting, Wim de Boer, Univ. Karlsruhe 2012.11.2008
-70
-60
-50
-40
-30
-20
-10
0
10
20
30
0 10 20 30 40 50 60
time [min]
tem
pera
ture
[°C
]
sensor1sensor2sensor4
Regulating temperature with pressure
11,5bar 8bar 6,5 5,5
very easy to set and hold temperature: just keep pressure constant
CO2 pressurein [bar]
CERN Cooling Meeting, Wim de Boer, Univ. Karlsruhe 2112.11.2008
beginning dry-out
050
100150200250300
0 0,5 1 1,5 2 2,5 3
flow in kg/hour (both tubes)
tota
l hea
ting
pow
er [W
]
Test results
easy to cool large powers with little flow of CO2,
flow was tested up to 3,7 kg/hour (max. of flowmeters) with negligible pressure drop
Even much bigger flow seems possible with tolerable pressure drop
CERN Cooling Meeting, Wim de Boer, Univ. Karlsruhe 2212.11.2008
Summary
• sLHC requires cooling of sensors below -25 0C to get leakage current noise down
• Requires strixels of 2.2 cm to get S/N similar as for LHC(signal down by ¼, so capacitance down by ¼)
• Reduction of material budget possible by combining cooling, support and current leads into single structure
• Low temperatures and all connections outside volume possibleby CO2 cooling, which allows 6m long cooling pipes
• Using CO2 allows to build outer barrel tracker from long ladderswith only one type of sensor shape and endcaps at small radii withALL service connections outside tracking volume (reduces material budget INSIDE tracker, not so much total!!!)
Simple blow system with CO2 precooled bottle working very well as test system