Mechanical and Thermal Management for ATLAS Upgrade Silicon Tracking System

Mechanical and Thermal Management Mechanical and Thermal Management for for

ATLAS Upgrade Silicon Tracking SystemATLAS Upgrade Silicon Tracking System

W. O. Miller (iTi)W. O. Miller (iTi)

Carl Haber (LBNL), Gil Gilchriese (LBNL)Carl Haber (LBNL), Gil Gilchriese (LBNL)

VG VG 22W.O. Miller

i T ii T i

ATLAS Silicon Tracker UpgradeATLAS Silicon Tracker Upgrade

• Stave-Type Tracking Detector---- ConsiderationsStave-Type Tracking Detector---- Considerations– Dimensional (Static) stabilityDimensional (Static) stability

• Out-of-plane distortion from gravity to be <50 to 60micronsOut-of-plane distortion from gravity to be <50 to 60microns• Thermal distortions from sub-cooling detector to be <25micronsThermal distortions from sub-cooling detector to be <25microns

– CoolingCooling• Possibly based on ATLAS two-phase flow, with -20 to -25ºC at Possibly based on ATLAS two-phase flow, with -20 to -25ºC at

detector surfacedetector surface• Or possibly high pressure COOr possibly high pressure CO22 at -35 at -35ºC inlet (1000psi)ºC inlet (1000psi)• An approach that assures that thermal runaway is not a problem An approach that assures that thermal runaway is not a problem

– Places cooling directly beneath chip heat loadPlaces cooling directly beneath chip heat load– Radiation length-minimized, consistent with achieving mechanical Radiation length-minimized, consistent with achieving mechanical

objectivesobjectives– Ease of Assembly and maintenance (when required)Ease of Assembly and maintenance (when required)

• Less time consumingLess time consuming• Precision module placement and bonding at earliest construction Precision module placement and bonding at earliest construction

levellevel– Facilitates testing and placement calibrationFacilitates testing and placement calibration– Added benefit in that provides potential for wider participation Added benefit in that provides potential for wider participation

from institutions in building staves and detector assemblyfrom institutions in building staves and detector assembly

VG VG 33W.O. Miller

i T ii T i

ATLAS Tracker: Stave (Upgrade)ATLAS Tracker: Stave (Upgrade)

• Analysis scopeAnalysis scope– Stave StiffnessStave Stiffness, evaluate ability to , evaluate ability to

resist gravity sag, critical at upper resist gravity sag, critical at upper and lower and lower ΦΦ

– Evaluate stave Evaluate stave thermal/mechanicalthermal/mechanical aspects aspects

• Direct heat transfer to Direct heat transfer to evaporative cooling tubesevaporative cooling tubes

• Thermal strains from thermal Thermal strains from thermal gradient from chips to tubesgradient from chips to tubes

• Thermal strains from cool-Thermal strains from cool-down, room temperature to down, room temperature to initial state (minus 25ºC or initial state (minus 25ºC or minus 35ºC)minus 35ºC)

– Coolant tube sizingCoolant tube sizing spaces the spaces the facing separationfacing separation

• Analysis guided by realistic Analysis guided by realistic constraints on radiation constraints on radiation lengthlength

• Radiation length driven Radiation length driven design as always, issues focus design as always, issues focus on:on:– Composite material propertiesComposite material properties

• High modulus, optimized fiber High modulus, optimized fiber orientation for maximum orientation for maximum axial stiffnessaxial stiffness

• High conductivity fiber High conductivity fiber system for heat transportsystem for heat transport

• Issue to be addressed is CTE Issue to be addressed is CTE mismatch in the assembly mismatch in the assembly

– Al cooling tube Al cooling tube

• Stave length, gravity sag, Stave length, gravity sag, bounded between fixed end bounded between fixed end and simple supportsand simple supports

• ~60~60μμm for fixedm for fixed to ~320 to ~320 μμm m for simplefor simple

• Desire limit of <60micronsDesire limit of <60microns

Stave StiffnessStave Stiffness

( beware discussion covers different geometries)( beware discussion covers different geometries)

VG VG 55W.O. Miller

i T ii T i

Stave Static Stiffness Stave Static Stiffness DiscussionDiscussion

• Results of FEA solutionsResults of FEA solutions– Stiffness---solution for purely horizontal and purely verticalStiffness---solution for purely horizontal and purely vertical

• Model comprisesModel comprises– Composite sandwich concept----embedded cooling tubeComposite sandwich concept----embedded cooling tube– Composite Facings and side strips----primary stiffness elementComposite Facings and side strips----primary stiffness element

Fiber choice---K13D2U, very high modulus (120Msi) with Fiber choice---K13D2U, very high modulus (120Msi) with thermal conductivity of 800W/mKthermal conductivity of 800W/mK

Fiber orientation---- 4 to 1 lay up, preferentially aligned Fiber orientation---- 4 to 1 lay up, preferentially aligned along the stave axisalong the stave axis

– Honeycomb----- graphite fiber honeycomb with 3/16in cell sizeHoneycomb----- graphite fiber honeycomb with 3/16in cell size– Al End caps and stainless steel pinsAl End caps and stainless steel pins

VG VG 66W.O. Miller

i T ii T i

Stave Module Layout-Staggered Stave Module Layout-Staggered ModulesModules

Physical length of stave ~99.4cm

6.4cm

width

Modules staggered front to back ~100W Thermal Load

Port card

Original 4.6mm flat tube

Original model

VG VG 77W.O. Miller

i T ii T i

Stave Module Layout-Double-Sided Stave Module Layout-Double-Sided ModulesModules

• Silicon Modules-12.4cm long with 4 sets of chips per strip Silicon Modules-12.4cm long with 4 sets of chips per strip detectordetector

• Chip heat load – 192WChip heat load – 192W

• Module heat load -4W/module or 64W/staveModule heat load -4W/module or 64W/stave

71.5mm

1064mm

More Recent Model

VG VG 88W.O. Miller

i T ii T i

Prototype Stave- ToolingPrototype Stave- Tooling

• Assembly tooling set to longer and wider stave dimensionsAssembly tooling set to longer and wider stave dimensions• Prototype coolant tube initially considered was 7.3mm flat tube Prototype coolant tube initially considered was 7.3mm flat tube

(availability), (availability), but now proceeding with 4.7mm height and 14mil but now proceeding with 4.7mm height and 14mil wallwall (purchase new tube extrusion) (purchase new tube extrusion)

71.5mm

1064mm

VG VG 99W.O. Miller

i T ii T i

Consequences of Staggered vs Double-Consequences of Staggered vs Double-SidedSided

• Silicon Module Lay-outSilicon Module Lay-out– Interest shown in double-sided geometryInterest shown in double-sided geometry– Analysis and lay-outs adapted to provide information for bothAnalysis and lay-outs adapted to provide information for both– Most of the FEA work done on original layout with fewer modulesMost of the FEA work done on original layout with fewer modules

• Adapted the results by adding analytical solutionsAdapted the results by adding analytical solutions– Compared analytical to FEA to form basis for making Compared analytical to FEA to form basis for making

extrapolationsextrapolations

• Separation between facingsSeparation between facings– 4.6mm, 5.88mm, and 7.3mm4.6mm, 5.88mm, and 7.3mm– Now focusing on 4.7mmNow focusing on 4.7mm

• Other factorsOther factors– Thermal heat loadThermal heat load– Coolant pressureCoolant pressure

• CC33FF88 and CO and CO2, 2, substantial difference in pressuresubstantial difference in pressure

– Necessitated adapting design to round tube for stress Necessitated adapting design to round tube for stress considerationsconsiderations

VG VG 1010W.O. Miller

i T ii T i

Stave Support-via Alignment Stave Support-via Alignment PinsPins

• Stave End CapStave End Cap– Two pins at each end for stave support and alignmentTwo pins at each end for stave support and alignment– Pin engagement in stave and end support disk are important to stave Pin engagement in stave and end support disk are important to stave

stiffnessstiffness

• Sag will affected significantly by the end plate designSag will affected significantly by the end plate design

Port cardElectronics chips

Strip detector

~1m length

pins

Strip detector

~1m length


i T ii T i

Stave ConfigurationStave Configuration

• Composite SandwichComposite Sandwich– Facings high conductivity, high Facings high conductivity, high

modulus K13D2U in 4:1 orientation modulus K13D2U in 4:1 orientation ((modulus 1.77 times steelmodulus 1.77 times steel))

– Honeycomb core fills void not Honeycomb core fills void not occupied by cooling tubeoccupied by cooling tube

• Cooling: EvaporativeCooling: Evaporative– Cooling tube: U-shaped, allowing for Cooling tube: U-shaped, allowing for

entrance and exit at one endentrance and exit at one end– Tube flatten slightly to improve Tube flatten slightly to improve

thermal contactthermal contact

• Module Mounting on both Module Mounting on both composite facings: composite facings: balancing balancing CTE mismatch to reasonable CTE mismatch to reasonable levellevel– Detector modules mounted on both Detector modules mounted on both

sides, in either alternating pattern sides, in either alternating pattern or or entirely symmetricalentirely symmetrical

– Hybrids mounted between detector Hybrids mounted between detector modulesmodules

What key factors describe stiffness for this sandwich?

More recently modules on both sidesMore recently modules on both sides


i T ii T i

Stave Parameters Affecting Stave Parameters Affecting StiffnessStiffness

• Stave gravity sag (bending) results from uniform length-wise Stave gravity sag (bending) results from uniform length-wise loadingloading– W=uniform loading, L=unsupported length, C varies from 1 to 5 based on end W=uniform loading, L=unsupported length, C varies from 1 to 5 based on end

fixity, b=width of sandwich (w is generally a function of b so this essentially fixity, b=width of sandwich (w is generally a function of b so this essentially drops out)drops out)

– D=flexural rigidity, E=facing Young’s modulus, t=facing thickness, h=overall D=flexural rigidity, E=facing Young’s modulus, t=facing thickness, h=overall sandwich height, c=core height, sandwich height, c=core height, νν=facing Poisson's ratio=facing Poisson's ratio

• Where do we gain most?Where do we gain most?– Minimizing unsupported stave length, Minimizing unsupported stave length, nextnext– Maximizing end support rigidity, nextMaximizing end support rigidity, next– Separation between facings, Separation between facings, finallyfinally– Facing modulus and facing thicknessFacing modulus and facing thickness

• What difficulties do we encounter?What difficulties do we encounter?– Achieving stave lengths >1m by limiting Achieving stave lengths >1m by limiting δδ < 60microns < 60microns– Cooling both composite faces with single cooling tube, limits cCooling both composite faces with single cooling tube, limits c– Radiation length limits, tRadiation length limits, t– Quasi-isotropic facings, would place a lower limit on E Quasi-isotropic facings, would place a lower limit on E

Db

CwLbending

384)(

4

)1(12

)(2

2

chEt

D


i T ii T i

Stave Sag FEA Models (Three)Stave Sag FEA Models (Three)

• Coolant MassCoolant Mass– Used density of two phase fluid. Used density of two phase fluid.

Mean density is 60kg/mMean density is 60kg/m33, whereas , whereas liquid density is 1660kg/mliquid density is 1660kg/m33

– Round circular tube in half length Round circular tube in half length model (5.88mm)model (5.88mm)

• Sandwich Core (shear Sandwich Core (shear deflection)deflection)– Varied core shear modulus, Varied core shear modulus,

reflected in density change to reflected in density change to materialmaterial

• 66 to 210kg/m66 to 210kg/m33, CVD carbon , CVD carbon foamfoam

• 56 and 110kg/m56 and 110kg/m33, honeycomb, honeycomb

• Sandwich core heightSandwich core height– 4.6, 5.88, 7.33, and now 4.7mm4.6, 5.88, 7.33, and now 4.7mm– Size for prototype stave Size for prototype stave initially initially

drivendriven by tube availability---8mm by tube availability---8mm extrusion, but now we are extrusion, but now we are proceeding with a new extrusionproceeding with a new extrusion

• Full Length (96cm-4.6mm flat Full Length (96cm-4.6mm flat tube) model with wafers, tube) model with wafers, hybrids and cable as dead hybrids and cable as dead weightweight– 0.173in diameter support pins0.173in diameter support pins– Clamped support pins, vertical DOF Clamped support pins, vertical DOF

onlyonly– Core thickness 4.6mmCore thickness 4.6mm

• Half length model with wafers, Half length model with wafers, hybrids and cable as dead hybrids and cable as dead weightweight– 0.173in support pins0.173in support pins– ½ length model used to add ½ length model used to add

structural coupling of wafers, and structural coupling of wafers, and hybridshybrids

– Core thickness 5.88mmCore thickness 5.88mm

• Model of pins and end cap alone Model of pins and end cap alone with stave weight imposedwith stave weight imposed– Two pin diameters (0.173in and Two pin diameters (0.173in and

0.25in)0.25in)


i T ii T i

FEA Sandwich Core Summary-StaggeredFEA Sandwich Core Summary-Staggered


i T ii T i

Stave FEA Results-Core ShearStave FEA Results-Core Shear

• Varied core shear modulusVaried core shear modulus– Looked at carbon foam versus honeycombLooked at carbon foam versus honeycomb– Results did not come out entirely as expectedResults did not come out entirely as expected– Example:Example:

• Increased shear modulus led to increased sagIncreased shear modulus led to increased sag

• Increased shear modulus with carbon foam comes at expense of Increased shear modulus with carbon foam comes at expense of increased core density, this seemed to be over-riding factorincreased core density, this seemed to be over-riding factor

• Suspicion: coolant tube with direct coupling to facing was Suspicion: coolant tube with direct coupling to facing was a major contributor to shear load transfera major contributor to shear load transfer

• ApproachApproach– Estimate degree of sag from core shearEstimate degree of sag from core shear– Uncouple tube by reducing the modulusUncouple tube by reducing the modulus


i T ii T i

Idealized Sandwich StiffnessIdealized Sandwich Stiffness

C=1 for fixed and C=5 for simple

bB

wLshear

8)(

2

c

chhGB

2

• Deflection from Uniform Loading

• Bending Deflection• Core Shear Deflection• End Cap Strain (mounting pins): somewhere between simple and fixed

Db

CwLbending

384)(

4

)1(12

)(2

2

chEt

D

End cap strain

pinsSandwichside

hc

Sandwich dimensions

t

b

L-unsupported length w-load per unit length

G-core shear modulus

E-facing modulus

1.6microns


i T ii T i

Establish Cooling Tube Establish Cooling Tube Interaction Interaction

• 96cm Model of Stave96cm Model of Stave– Use simple edge supports, Use simple edge supports,

K13D2U 4/1K13D2U 4/1– Apply force at quarter points, Apply force at quarter points,

– Extract deflection at Extract deflection at ΔΔ44 and and ΔΔ22, , quarter point and mid-spanquarter point and mid-span

• Use relationship to solve for Use relationship to solve for core shear moduluscore shear modulus

• Result for 4.6mm core with Al Result for 4.6mm core with Al tubestubes– ~128 MPa calculated versus 26.9 ~128 MPa calculated versus 26.9

MPa input for virgin foamMPa input for virgin foam– Tubes contribute most of the Tubes contribute most of the

shear stiffness, except at very shear stiffness, except at very high foam densitieshigh foam densities

G.c11.5 P L1 c1

h1 c1( )2

b 11 .4 8 .2

P/2 P/2

Division between bending and shear, basedon estimate of core shear of 128MPa

Δbending est=35μmΔcore shear est=8.3 μm

Combined Δ=43.3microns (FEA 53.7 μm)

Using sandwich relationships for fixed ends


i T ii T i

Removed Cooling Tube Removed Cooling Tube InteractionInteraction

• Cooling tube as a shear component is affected by Cooling tube as a shear component is affected by compliance of thermally enhanced adhesive bond compliance of thermally enhanced adhesive bond (CGL7018)(CGL7018)– FEA model modification placed reliance for shear transfer on the FEA model modification placed reliance for shear transfer on the

composite fiber honeycomb, 626MPacomposite fiber honeycomb, 626MPa– Increase in sag by small amount (~4microns for HC only)Increase in sag by small amount (~4microns for HC only)

• Experimental tests with stave prototype hopefully will Experimental tests with stave prototype hopefully will reveal extent of contribution from coolant tube on shearreveal extent of contribution from coolant tube on shear– If tube contributes significantly to shear load transfer, what will be the If tube contributes significantly to shear load transfer, what will be the

benefit?benefit?

• Slight increase in stiffness predicted, but if higher than expected, Slight increase in stiffness predicted, but if higher than expected, may be possible to reduce the number of uni-tape layers in the may be possible to reduce the number of uni-tape layers in the composite facingcomposite facing

• Slight benefit in radiation length in using different core material Slight benefit in radiation length in using different core material


i T ii T i

Modification to End Cap Modification to End Cap ConfigurationConfiguration

• PurposePurpose– Lengthen pin engagement with support plateLengthen pin engagement with support plate– Increase internal bonding contact with the composite facingsIncrease internal bonding contact with the composite facings– Add tooling balls for assembly inspectionAdd tooling balls for assembly inspection

Steel pins 0.173in dia Alignment ball

1/4in bonding surface

Shell-like to minimize mass

Cooling tube penetrations


i T ii T i

Stave Radiation Length Stave Radiation Length EstimateEstimate

~ prototype

Sag Est.

54 µms(FEA)

30 µms(extrapolated)


i T ii T i

Vertical Orientation- Version A Vertical Orientation- Version A StaveStave

• Maximum sag Maximum sag 3.5microns3.5microns

• Al end capsAl end caps

• SS pins, SS pins, 0.173in dia.0.173in dia.

• 4.6mm 4.6mm separationseparation


i T ii T i

Horizontal Orientation- Version A Horizontal Orientation- Version A StaveStave

• Maximum sag Maximum sag 55.5 microns55.5 microns

• Al end capsAl end caps• SS pins, SS pins,

0.173in dia.0.173in dia.• 4.6mm 4.6mm

separationseparation


i T ii T i

2m Long Stave2m Long Stave

• Sag Sag – For fixed geometry previously discussed, sag from bending alone For fixed geometry previously discussed, sag from bending alone

would increase by L to the 4would increase by L to the 4thth – 50microns would increase to 800microns, without factoring in 50microns would increase to 800microns, without factoring in

potential increase in core shearpotential increase in core shear

• Increased Sag can be mitigated by increasing Increased Sag can be mitigated by increasing structure moment of inertiastructure moment of inertia– Larger separation between facings for sandwich structureLarger separation between facings for sandwich structure

• From 4.7mm to 20mmFrom 4.7mm to 20mm

• This complicates cooling and requires internal structure to This complicates cooling and requires internal structure to provide high core shear modulusprovide high core shear modulus

– Entirely different structural approach?Entirely different structural approach?

• Any approach where support is provided at two opposing Any approach where support is provided at two opposing ends becomes a beam problemends becomes a beam problem

• Preferred solution is to provide a mid-span support at Preferred solution is to provide a mid-span support at 1m1m– Allows retention of basic construction concept Allows retention of basic construction concept


i T ii T i

Closing Remarks on Stave Closing Remarks on Stave DistortionDistortion

• Sag from lack of stave sandwich stiffness is not the problem!Sag from lack of stave sandwich stiffness is not the problem!– Possible devise a pin-end support that controls sag---defined by near Possible devise a pin-end support that controls sag---defined by near

fixed end supportfixed end support

• First sag problem only for upper and lower stavesFirst sag problem only for upper and lower staves

• Secondly, 1m long stave mass fully loaded is of the order of 5N (~1lb)Secondly, 1m long stave mass fully loaded is of the order of 5N (~1lb)– Reactions at pins very small, distortion of end caps not the issueReactions at pins very small, distortion of end caps not the issue

• What are the problems?What are the problems?– Distortion or lack of dimensional accuracy in the end support plate Distortion or lack of dimensional accuracy in the end support plate

surfacesurface

• Any slight imposed “moment” on the stave end will produce large Any slight imposed “moment” on the stave end will produce large out-of-plane distortionout-of-plane distortion

– Angle at the stave end of just 0.04Angle at the stave end of just 0.04º will cause 250 microns at the º will cause 250 microns at the centercenter

This angle is equivalent to what occurs with a simply This angle is equivalent to what occurs with a simply supported beamsupported beam

• A necessary step is an adjustable mount for the end capA necessary step is an adjustable mount for the end cap– Stability of the end plate support---will become an issueStability of the end plate support---will become an issue


i T ii T i

Closing Stave Remarks Closing Stave Remarks (continued)(continued)

• End Plate SupportEnd Plate Support– Factors to be considered in designing stave end supportFactors to be considered in designing stave end support

• Structure CTEStructure CTE– If positive or negative, how much? If positive or negative, how much?

For 60For 60º temperature change, 1ppm/ºC will cause º temperature change, 1ppm/ºC will cause 60microns in a 1m length60microns in a 1m length

Will a change in support structure length distort the stave Will a change in support structure length distort the stave mounting surface?mounting surface?

• Structure CMEStructure CME– A quasi-isotropic composite (75Msi fiber modulus) with a A quasi-isotropic composite (75Msi fiber modulus) with a

cyanate ester resin will have a CME of 20ppm at 55% RH, or cyanate ester resin will have a CME of 20ppm at 55% RH, or 20microns in 1m20microns in 1m

• Conceivable combined strain will be ~80 microns or moreConceivable combined strain will be ~80 microns or more

• Temperature differential across the staveTemperature differential across the stave– Need to determine out-of-plane movement for temperature gradient Need to determine out-of-plane movement for temperature gradient

normal to stave axisnormal to stave axis

• May not be easy to identify magnitude of the gradientMay not be easy to identify magnitude of the gradient

Thermal SolutionsThermal Solutions


i T ii T i

Thermal Property ParametersThermal Property Parameters

Suspect the cable bus thermal conductivity is somewhat low; thermal property will be determined in order to improve thermal quality of results

FEA Model Properties


i T ii T i

Revised 4.7mm Flat Tube Model (No Wafer Revised 4.7mm Flat Tube Model (No Wafer Heat)Heat)

• Separation between facings Separation between facings 4.7mm- 4.7mm- present prototypepresent prototype

• Modified hybrid stack, now----Modified hybrid stack, now----– Chip, followed by silver adhesive, Chip, followed by silver adhesive,

sitting on dielectric (5W/mK), sitting on dielectric (5W/mK), followed by silver adhesive, sitting followed by silver adhesive, sitting on BeO (240W/mK)on BeO (240W/mK)

– Before the dielectric and BeO were Before the dielectric and BeO were merged into one solid (8W/mK)merged into one solid (8W/mK)

• Cable K=0.4W/mK to simulate Cable K=0.4W/mK to simulate 7575µm of Kaptonµm of Kapton, whereas , whereas model thickness is 250 model thickness is 250 micronsmicrons

• Coolant (-25Coolant (-25ºC, 1800W/mºC, 1800W/m22K)K)– Film coefficient to produce about Film coefficient to produce about

3ºC temperature drop3ºC temperature drop

• Results, average chips -Results, average chips -15.5ºC; module center area -15.5ºC; module center area -17ºC to -17.8ºC17ºC to -17.8ºC

Chips (average)

Detector center area

Cooling tube wall -22ºC

Solution with CFDesign- CFD code


i T ii T i

Revised 4.7mm Flat Tube Model (Add 4W per Revised 4.7mm Flat Tube Model (Add 4W per Wafer)Wafer)

• Separation between facings Separation between facings 4.7mm4.7mm

• Modified hybrid stack, now----Modified hybrid stack, now----– Chip, followed by silver adhesive, Chip, followed by silver adhesive,

sitting on dielectric (5W/mK), followed sitting on dielectric (5W/mK), followed by silver adhesive, sitting on BeO by silver adhesive, sitting on BeO (240W/mK)(240W/mK)

– Before the dielectric and BeO were Before the dielectric and BeO were merged into one solid (8W/mK)merged into one solid (8W/mK)

• Cable K=0.4W/mK to simulate Cable K=0.4W/mK to simulate 7575µm of Kaptonµm of Kapton, whereas model , whereas model thickness is 250 micronsthickness is 250 microns

• Coolant (-25Coolant (-25ºC, 1800W/mºC, 1800W/m22K)K)– Film coefficient to produce about 3ºC Film coefficient to produce about 3ºC

temperature droptemperature drop

• Results, average chips -13ºC, Results, average chips -13ºC, module center area -14.5ºC to -module center area -14.5ºC to -15.3ºC15.3ºC

Chips average

Detector center area

Solution with CFDesign- CFD code


i T ii T i

Thermal Solution with NASTRAN -25Thermal Solution with NASTRAN -25ºC ºC CoolantCoolant

• Latest stave model- section one silicon detector in lengthLatest stave model- section one silicon detector in length– Double-sided detectorDouble-sided detector– 0.5W per chip0.5W per chip– No interface resistancesNo interface resistances Ends affected by cut-off

Chip ~ -16ºC

Silicon module ~ -17.8ºC

H=1800W/m2KSame parameters as in previous slide

Simple solution to recover strains


i T ii T i

Gradient Through Stave MaterialsGradient Through Stave Materials

• Could improve contact with tube, but resistance is in the cableCould improve contact with tube, but resistance is in the cable

• Solution with CFDesign includes the contact interface resistancesSolution with CFDesign includes the contact interface resistances

chip

dielectrichybrid

BeO

cable

facing

Extracted from NASTRAN FEA-chip heat only

module


i T ii T i

Carbon-Carbon LaminateCarbon-Carbon Laminate

• For those that like C-C laminates as opposed to resin matrix laminateFor those that like C-C laminates as opposed to resin matrix laminate• Changing the through the thickness K from 1.44W/mK to 10W/mK we Changing the through the thickness K from 1.44W/mK to 10W/mK we

get approximately a 2get approximately a 2ºC change in chip peak temperature (-15.9ºC ºC change in chip peak temperature (-15.9ºC versus -17.9ºCversus -17.9ºC

Thermal block is in the cable!

Chip heat only


i T ii T i

NASTRAN Thermal Strains From -25NASTRAN Thermal Strains From -25ºC ºC CoolantCoolant

• Latest stave model- section one silicon module in lengthLatest stave model- section one silicon module in length– Double-sided moduleDouble-sided module– 0.5W per chip (temp solution from slide #30)0.5W per chip (temp solution from slide #30)

Peak out-of-plane distortion ~5microns


i T ii T i

• Latest stave model- section one silicon module in lengthLatest stave model- section one silicon module in length– Double-sided moduleDouble-sided module– Estimation of tube strain for 60Estimation of tube strain for 60ºC temperature change---- 6.2micronsºC temperature change---- 6.2microns

NASTRAN Thermal Strains From -35NASTRAN Thermal Strains From -35ºC ºC CoolantCoolant


i T ii T i

Stave Thermal Strains- Stave Thermal Strains- Old Old ModelModel

• FEA solution- FEA solution- alternating alternating modulesmodules– Used 48cm long stave, symmetry Used 48cm long stave, symmetry

BC at mid-lengthBC at mid-length– Limit to stave length permitted Limit to stave length permitted

addition of wafers and hybrids, still addition of wafers and hybrids, still model required large number of model required large number of elementselements

• Thermal strain for change 50ºC Thermal strain for change 50ºC in temperature (RT to -25ºCin temperature (RT to -25ºC– Results for stave with 5.88mm core Results for stave with 5.88mm core

height, round tubeheight, round tube– K13D2U 4:1 fiber orientationK13D2U 4:1 fiber orientation– 0.75mm thick facings0.75mm thick facings– Detector modules and hybrids Detector modules and hybrids

coupled to facing without interface coupled to facing without interface compliancecompliance

• Distortion patternDistortion pattern– ~11micron peak to peak, ~11micron peak to peak,

alternating in z-directionalternating in z-direction Balanced double-sided module stave strain is about half


i T ii T i

Stave Thermal Solution – Stave Thermal Solution – Old Old ModelModel

• FEA solution for both Al and Carbon-Fiber Filled PEEK tubesFEA solution for both Al and Carbon-Fiber Filled PEEK tubes– 6 chips for each hybrid, 0.5W each: heat zone shifts top to bottom6 chips for each hybrid, 0.5W each: heat zone shifts top to bottom

– Film coefficient 3000W/mFilm coefficient 3000W/m22K, typical of Pixel stave for CK, typical of Pixel stave for C33FF8, 8, two parallel tubestwo parallel tubes

– Lower peak wafer temperature favors Al cooling tubeLower peak wafer temperature favors Al cooling tube

-17.5C Chip/Silicon -23.5C -13.4C Chip/Silicon -20.2C

(Unfilled PEEK) Silicon -11C Chips -7.1C


i T ii T i

Evaporative Film CoefficientEvaporative Film Coefficient

• Thermal FEA based on film coefficient from ATLAS Thermal FEA based on film coefficient from ATLAS (3000W/m(3000W/m22K)K)– 3000W/m3000W/m22K chosen to expedite the earlier thermal solutionsK chosen to expedite the earlier thermal solutions– Question: what might the film coefficient be for various diameter tubes?Question: what might the film coefficient be for various diameter tubes?

• Using Kreith-BohnUsing Kreith-Bohn– As a check, calculated for 3mm (8mil wall) round tube (ATLAS) As a check, calculated for 3mm (8mil wall) round tube (ATLAS)

Dh=2.59mmDh=2.59mm• 3040W/m3040W/m22K at entrance (x=0.3 after throttling)K at entrance (x=0.3 after throttling)• 4451W/m4451W/m22K at exit (x=0.8, mostly vapor)K at exit (x=0.8, mostly vapor)

– Based on 4.6mm flatten tube (FEA thermal solution) Dh=4.956mmBased on 4.6mm flatten tube (FEA thermal solution) Dh=4.956mm• 1228W/m1228W/m22K at entrance and 1645W/mK at entrance and 1645W/m22K at exit, K at exit, noticeably lowernoticeably lower

– Now for the 7.33mm flatten tube --Dh=7.791mmNow for the 7.33mm flatten tube --Dh=7.791mm• 811W/m811W/m22K entrance and 1000W/mK entrance and 1000W/m22K at exitK at exit

• Important to note that since Important to note that since ΔΔTTfilmfilm is proportional to inner wall is proportional to inner wall surface area, the lower heat transfer film coefficients have no surface area, the lower heat transfer film coefficients have no real impact for same heat load real impact for same heat load

• 2.42ºC (3mm), 2.96ºC (4.6mm), and 2.94ºC (7.33mm)2.42ºC (3mm), 2.96ºC (4.6mm), and 2.94ºC (7.33mm)– The pressure drop is different, lower for the bigger tubes, since the mass The pressure drop is different, lower for the bigger tubes, since the mass

flow rate is the same for each caseflow rate is the same for each case

Coolant Pressure Drop and Convective Heat Coolant Pressure Drop and Convective Heat Transport Transport

•Objective: 1Objective: 1stst order calculations for the semi- order calculations for the semi-flatten tube and circular tube for both Cflatten tube and circular tube for both C33FF88 and and COCO22


i T ii T i

Tube Hydraulic and Thermal Tube Hydraulic and Thermal EvaluationEvaluation

• Stave with CStave with C33FF88 Coolant: Coolant: Semi-Flattened tubeSemi-Flattened tube– Assumptions: entering quality 5%, exiting 75%Assumptions: entering quality 5%, exiting 75%– Resulting mass flow for 192W, 2.7g/sResulting mass flow for 192W, 2.7g/s

– 4.6mm semi-flattened tube, D4.6mm semi-flattened tube, Dhh=5.0mm=5.0mm

– Entering Fluid -25Entering Fluid -25ºC and 1.67barºC and 1.67bar– Average frictional pressure drop 0.137bar, resulting saturation exiting Average frictional pressure drop 0.137bar, resulting saturation exiting

fluid temperature becomes -27ºCfluid temperature becomes -27ºC– Average film coefficient is 1800W/mAverage film coefficient is 1800W/m22KK

• ConclusionConclusion– From hydraulic and heat transport standpoint the tube is From hydraulic and heat transport standpoint the tube is slightly too slightly too

largelarge– Had to update the thermal solution for lower than estimated film Had to update the thermal solution for lower than estimated film

coefficientcoefficient

• Original estimate based on higher coolant inlet quality and higher Original estimate based on higher coolant inlet quality and higher mass flow rate (injection for ATLAS seemed to be based on 30% mass flow rate (injection for ATLAS seemed to be based on 30% quality)quality)


i T ii T i


• Stave with CStave with C33FF88 Coolant: Coolant: Round tubeRound tube– Assumptions: entering quality 5%, exiting 75%Assumptions: entering quality 5%, exiting 75%– Resulting mass flow for 192W, 2.7g/sResulting mass flow for 192W, 2.7g/s

– 4.6mm round tube, D4.6mm round tube, Dhh=4mm=4mm

– Entering Fluid -25Entering Fluid -25ºC and 1.67barºC and 1.67bar– Average frictional pressure drop 0.41bar, resulting saturation exiting Average frictional pressure drop 0.41bar, resulting saturation exiting

fluid temperature becomes -32ºCfluid temperature becomes -32ºC– Average film coefficient is 2500W/mAverage film coefficient is 2500W/m22KK

• ConclusionConclusion– Coolant temperature varying between -25ºC and -32ºC helps to hold Coolant temperature varying between -25ºC and -32ºC helps to hold

silicon temperaturesilicon temperature

• Similar to having average coolant temperature of approximately -Similar to having average coolant temperature of approximately -28.5ºC28.5ºC

– From hydraulic standpoint the tube is From hydraulic standpoint the tube is slightly smallslightly small

• Average predicted frictional pressure drop encroaches on requisite Average predicted frictional pressure drop encroaches on requisite vapor return pressure vapor return pressure


i T ii T i


• Stave with Stave with COCO22 CoolantCoolant: : Round tubeRound tube– Assumptions: entering quality 5%, exiting 75%Assumptions: entering quality 5%, exiting 75%– Resulting mass flow for 192W, 0.82g/sResulting mass flow for 192W, 0.82g/s

– 4.6mm round tube, D4.6mm round tube, Dhh=4mm (satisfies our core height issue in a =4mm (satisfies our core height issue in a sandwich to minimize gravity sag)sandwich to minimize gravity sag)

– Entering Fluid -35Entering Fluid -35ºC and 12barºC and 12bar– Average frictional pressure drop is negligible, 0.034bar, essentially no Average frictional pressure drop is negligible, 0.034bar, essentially no

change in saturation temperature over the 2m tube lengthchange in saturation temperature over the 2m tube length– Average film coefficient is 4650W/mAverage film coefficient is 4650W/m22KK

• ConclusionConclusion– From hydraulic and heat transport standpoint the tube is From hydraulic and heat transport standpoint the tube is more than more than

adequateadequate– Stress, for a 1000 psi max pressure is 6545 psi, which would be Stress, for a 1000 psi max pressure is 6545 psi, which would be

considered allowable for 3003 work-hardened Aluminum tubing (or considered allowable for 3003 work-hardened Aluminum tubing (or age-harden Al alloy)age-harden Al alloy)


i T ii T i

Hydraulic and Thermal Hydraulic and Thermal Summary Summary

• Silicon Module Temperature for 2m (round trip) stave with Silicon Module Temperature for 2m (round trip) stave with 192W heat load192W heat load– If module requirement is -25If module requirement is -25ºC, the coolant most likely must enter at -ºC, the coolant most likely must enter at -

35ºC or at least minus 30ºC.35ºC or at least minus 30ºC.– CC33FF88 saturation pressure at -35ºC is 1.04bar, which means a 0.67bar saturation pressure at -35ºC is 1.04bar, which means a 0.67bar

reduction in available return pressure to the compressor from ATLAS reduction in available return pressure to the compressor from ATLAS conditions, may be a system problemconditions, may be a system problem

• Is -30ºC acceptable from system standpoint?Is -30ºC acceptable from system standpoint?– Frictional pressure loss in a 4.6mm semi-flatten tube with a hydraulic Frictional pressure loss in a 4.6mm semi-flatten tube with a hydraulic

diameter of 5mm at -35ºC coolant inlet is estimated to be 0.18bar, diameter of 5mm at -35ºC coolant inlet is estimated to be 0.18bar, further contributing to return pressure problemfurther contributing to return pressure problem

– For a “hard” silicon temperature of -25ºC, COFor a “hard” silicon temperature of -25ºC, CO2 entering at -35ºC is entering at -35ºC is preferablepreferable

• Prospects for CProspects for C33FF88 at -25ºC---most of problem comes from at -25ºC---most of problem comes from the 192W heat loadthe 192W heat load– Need to improve thermal modeling by quantifying the thermal Need to improve thermal modeling by quantifying the thermal

conductivity of the componentsconductivity of the components– Consider strategy of augmenting heat spreading beneath the chipsConsider strategy of augmenting heat spreading beneath the chips

Coolant Tube Stress Analysis

Compare Semi-Flatten Tube versus Round Tubes


i T ii T i

High Pressure Coolant TubeHigh Pressure Coolant Tube

• Geometry of coolant circuit modified to accommodate a round tubeGeometry of coolant circuit modified to accommodate a round tube

• Aluminum tube: 4.6mm OD (4mm ID), hoop stress for 60bar Aluminum tube: 4.6mm OD (4mm ID), hoop stress for 60bar (1000psi) would be 6546psi (3003 Al has adequate strength)(1000psi) would be 6546psi (3003 Al has adequate strength)

• Saturation of COSaturation of CO2 2 at 25at 25ºC is 933psi.ºC is 933psi.

Foam bonded in segments and feathered to tube

Round tube is supplemented with graphite foam transition to enhance thermal contact

Foam pieces


i T ii T i

FEA-Section of TubeFEA-Section of Tube

• Tube ShapeTube Shape– Both prototype stave tube and tube for the “Both prototype stave tube and tube for the “proposed staveproposed stave” geometry ” geometry

have a semi-flatten shapehave a semi-flatten shape– Flat portion is ~2mm across, to enhance thermal contact.Flat portion is ~2mm across, to enhance thermal contact.– Both tubes are Al tubes with 12mil wallBoth tubes are Al tubes with 12mil wall– For simplicity the shape of the sides are assumed to be circular with For simplicity the shape of the sides are assumed to be circular with

radius defined by core height (distance between composite facings) radius defined by core height (distance between composite facings)

4.6mm 7.3mm


i T ii T i

4.6mm Semi-Flattened Tube4.6mm Semi-Flattened Tube

• NASTRAN solution NASTRAN solution – Internal pressure of 100 psiInternal pressure of 100 psi– Stress bar is Pascal; average Stress bar is Pascal; average

von Mises stress in middle between von Mises stress in middle between two truncated faces is two truncated faces is 6062 psi6062 psi. .

– Deflection of corresponding node in Deflection of corresponding node in this region is 15.5 microns. this region is 15.5 microns.

• 100 psi pressure is equivalent to 100 psi pressure is equivalent to 6.9bar6.9bar– Pressure of CPressure of C33FF88 at -25 at -25ºC is 1.67 bar ºC is 1.67 bar

and at 25ºC the pressure is 8.7barand at 25ºC the pressure is 8.7bar– Allowable yield stress for work-Allowable yield stress for work-

hardened 3003 Series Al is ~27,000 hardened 3003 Series Al is ~27,000 psi psi

• But what happens in the stave?But what happens in the stave?

Middle is region of interest

Stress at ends affected by end conditions


i T ii T i

Prototype StavePrototype Stave -4.67mm -4.67mm HeightHeight

• Present Design----Double-sided modulePresent Design----Double-sided module– Partially flattened tube, 4.67mm heightPartially flattened tube, 4.67mm height– 100psi internal pressure load100psi internal pressure load– Peak deflection of stave surface ~1.5microns, versus nominally Peak deflection of stave surface ~1.5microns, versus nominally

15.5microns for unsupported tube (previous slide)15.5microns for unsupported tube (previous slide)– Results in much lower tube stress Results in much lower tube stress

NASTRAN FEA

Stave PrototypesStave Prototypes


i T ii T i

Stave Prototype – Slightly Flatten Stave Prototype – Slightly Flatten TubeTube

• Flat portionFlat portion– 2mm flat to improve composite facing thermal contact with tube2mm flat to improve composite facing thermal contact with tube

• Thermal adhesiveThermal adhesive– CGL7018, to reduce axial thermal strain from CTE mismatchCGL7018, to reduce axial thermal strain from CTE mismatch– Option: Use EG7658 semi-rigid “filled” adhesive with grafoil to Option: Use EG7658 semi-rigid “filled” adhesive with grafoil to

accommodate thermal mismatch accommodate thermal mismatch


i T ii T i

Stave Prototypes-Round TubeStave Prototypes-Round Tube

• Stave with round tube for high pressure COStave with round tube for high pressure CO22

– Heat transfer from facing to tube enhanced with graphitic foam

• Typical values quoted: 150W/mK through thickness, 50W/mK in-plane

• Free space occupied by graphite fiber honeycombFree space occupied by graphite fiber honeycomb

• Stave length and separation between facings in both prototypesStave length and separation between facings in both prototypes

Separation between facings same in both staves

Foam pieces


i T ii T i

Summary: Prototype Stave Summary: Prototype Stave InvestigationInvestigation

• Construction DetailConstruction Detail– Gain experience on technicalities of building a meter long staveGain experience on technicalities of building a meter long stave

• Controlling tolerances- actually fairly precision requirement to Controlling tolerances- actually fairly precision requirement to produce a flat controllable surface for the modules produce a flat controllable surface for the modules

• Thermal Thermal – Heat transport evaluation from chip to cooling tubeHeat transport evaluation from chip to cooling tube

• Number of thermal resistances that impede heat transportNumber of thermal resistances that impede heat transport– Most notably the cable, which is mostly KaptonMost notably the cable, which is mostly Kapton

• StructuralStructural– Gravity sagGravity sag

• Influence of end cap design and end plate supportInfluence of end cap design and end plate support– Survivability of bond joint between cooling tube and facing after thermal Survivability of bond joint between cooling tube and facing after thermal

cyclingcycling

– Pressure containment of COPressure containment of CO22, and resulting deformation of module , and resulting deformation of module stressesstresses

• Weakness in simulationsWeakness in simulations: all adhesive interfaces are modeled as rigid : all adhesive interfaces are modeled as rigid connectionsconnections

Mechanical and Thermal Management for ATLAS Upgrade Silicon Tracking System

Documents

Transcript of Mechanical and Thermal Management for ATLAS Upgrade Silicon Tracking System