Thermal Modeling Issues Concerning the Mechanically Pumped Two-Phase CO 2 Cooling for the AMS-2 2 Tracker A.A.Woering, A.Pauw, A.W.G. de Vries, A.A.M. Delil, The Netherlands B. Verlaat National Institute for Nuclear and High-Energy Physics NIKHEF, The Netherlands Contents AMS Tracker Tracker design Analysis Experiments Numerical modeling Numerical modeling and design issues
Alpha Magnetic Spectrometer AMS Silicon Tracker (x,y,z) particle trajectories determined by momentum and charge sign Curvature radius : momentum Curvature direction : charge sign Energy loss in each silicon plane yields charge magnitude
AMS Silicon tracker thermal requirements Silicon wafer thermal requirements: Operating temperature: -10 ºC / +25 ºC Survival temperature: -20 ºC / +40 ºC Temperature stability: 3 ºC per orbit Maximum accepted gradient between any silicon: 10.0 ºC Dissipated heat: 2.0 Watt EOL Hybrid circuit thermal requirements: Operating temperature: -10 ºC / +40 ºC Survival temperature: -20 ºC / +60 ºC Dissipated heat: 192 Watt total, 1 Watt per hybrid pair General requirements Limited mass ( 70-80 kgs) Limited power (< 80 W) Heat dissipation in the AMS-Tracker
TTCS summary 2 Identical completely separated loops (1 for redundancy) 2 serial evaporators in parallel per loop 2 parallel condensers controlled per loop controlled by a 3-way valve. Pressure controlled with a thermal control reservoir Thermal control using standard AMS control module Critical components in redundant configuration (pump, valves) Most fluid components in 2 dedicated TTCS boxes on the support structure at wake side RAM and WAKE heat pipe radiator All hardware is placed in debris-safe areas; a specific debris shield is added when needed
Tracker (component) design 2 parallel evaporator branches 2 parallel condenser branches, 2 radiators pump assembly Evaporator Complete thermal system in side the tracker (Thermal bars+ evaporators) TTCS evaporator connection to inner thermal bars Inner planes evaporator
Radiators Plumbing schematic
Experiments Feasibility test set-up @ NIKHEF Control BBM @ NLR Real size BBM @ NLR operational since early this year First tests are promising Full test campaign expected to start in fall 200 Breadboard wake condenser accumulator LN 2 cooling pump ram condenser evaporator feed line (SS) evaporator line (black)
Numerical modeling Modeling information flows Component modeling fixed with loop temperature Simulation cases for design optimization Radiator size, mass and shape Heat pipe puncture Preheating Detailed modeling of critical components (I-DEAS, NIKHEF) AMS-Tracker Thermal Model Network Cooling system modeling with fluid properties (SINDA/Fluint, NLR/Noordoostpolder) Model exchange Coarse thermal model of detector with conductors represented as links (I-DEAS, NIKHEF) Cooling system set point Model exchange Orbital fluxes boundary conditions Cooling system set point and power dissipation Complete AMS thermal model on International Space Station model to calculate orbital fluxes. (SINDA, CGS/Milan)
$ + $ + # + # + % + ' " + A = I K H A A J 5 E K = J E 6A FAH=JKHA CH=@EA JI B JD A E AH JD AH = > =H 6? + KJAH F = A JD AH = > =H JD AH = IE K =JE
Loop model temperatures, links S/S & environment temperatures, links environment TTCS loop, typical result 293 273 253 Liquid line Vapour line Cond. RAM Cond. Wake Wake RAM temperature [K] 233 213 193 173 153 β=0 MPA flat radiator 133 27000 28000 29000 30000 31000 32000 33000 time [s]
Radiator size, mass and shape Early result, 07/2001 vapour line condenser ram condenser wake liquid line evaporator Radiator mass 293 288 283 Liquid line temperatures beta0,mpa,13kg beta50,ypr152015,13kg beta0,mpa,18kg beta,ypr50152015,18kg Temperature [K] 278 273 268 z 263 258 253 21000 22000 23000 24000 25000 26000 27000 28000 time [s]
Radiator size: Heat pipe failure temperature [K] 288 283 278 273 T1, hpf T1 Reduction of effective radiator area 268 β=0 MPA 263 27000 28000 29000 30000 31000 32000 time [s] Preheating 120 100 Q Q, HX preheating power [W] 80 60 40 Q average = 60 W x 2 branches Q max = 114 W x 2 branches Q average = 14 W x 2 branches 20 11 W, due to setpoint Q ma x = 32 W x 2 branches 0 27000 28000 29000 30000 31000 32000 33000 time [s]
Conclusions Feasibility of TTCS supported by numerical model Several design choices based on and supported by thermal modeling Proven to be an important tool in engineering discussions with AMS experimenters