ATLAS Pixel Detector Pixel Support Tube Design Prototyping

ATLAS Pixel Detector Pixel Support Tube: Design, Prototyping, and Production PST Progress Update September 2002 N. Hartman LBNL

Pixel Detector ATLAS September 17 th Review Schedule • • • September 2002 9: 00 9: 30 9: 45 10: 00 10: 30 10: 50 11: 05 11: 25 12: 15 PST Design Update Shell Prototyping Rail Prototyping Mounts and Interfaces Break Heater Testing Heater Design and Fabrication Production Planning, Costs, Schedules Questions/Comments N. Hartman LBNL

Pixel Detector ATLAS Pixel Support Tube (PST) Overview • Design Updates – Flange design • Reduced from 26 pieces to 2 • Length Shortened – Rail design • • Impact on stiffness of forward shell calculated Reduced from 2 pieces per rail to 1 Shell design augmented Local stiffness analyzed • Prototyping – Material test results received – Shell and rail prototypes fabricated (covered in subsequent presentations) • Shell produced with heaters, and in hybrid form • Short rails produced and measured • Production covered later September 2002 N. Hartman LBNL

ATLAS Pixel Detector PST Overview September 2002 N. Hartman LBNL

Pixel Detector ATLAS Support Condition of Pixel Support Tube in Inner Detector Side C +X +Z +Y Vertical View from top—all Tube Supports are Horizontal and Co-planar Side A ID Flat Rail (float XZ) (constrained Y) TRT SCT Flat Rail (float XZ) (constrained Y) Fixed XYZ SCT Vee Rail (float Z/dogged Z) (constrained XY) Fixed YZ (N/A) Fixed XY Fixed Y Properties TBD Constraint TBD ID Vee Rail (float Z/dogged Z) (constrained XY) Flexure Mounts September 2002 N. Hartman LBNL 5

ATLAS Pixel Detector PST Key Structures Mount Pad Flexure d ar C w or F SCT Flexures and mount pads l e rr Ba Forward End flange and Flexure, installation rail rd A a w or F PST Flanges September 2002 N. Hartman LBNL

ATLAS Pixel Detector Rail Overview DETAIL Flat Rail DETAIL V Rail Vee and Flat rails were chosen to provide pseudo-kinematic support for the detector during delivery to the support points. September 2002 Rails are used only for delivery, not support. N. Hartman LBNL 7

ATLAS Pixel Detector Flange Design September 2002 N. Hartman LBNL

Pixel Detector ATLAS Initial Flange Concept • Flange Face (machined layup) • Stiffeners (layups) Base Piece – – Face Piece – – • Laid up individually ½ mm thick 24 parts Assembly – Flange base (Layup) Laid up as plate Machined to size Reinforcements – – – • ½ mm thick Laid up as hoop, sized to fit shell – 26 pieces bonded simultaneously as one assembly Flange assembly bonded to PST Shell Flange bolts September 2002 N. Hartman LBNL

Pixel Detector ATLAS Revised Flange Concept • Stiffeners eliminated – – • Flange shortened – – – Backing Ring Single piece skirt and face (“L” shape) • September 2002 From 40 mm long to 25 mm Allows thicker “skirt” in order to machine ID, while approximately conserving material amount from old design Still extremely conservative bond stresses Two piece design – – Flange Cross Section Not required for stiffness Reduces part count by 90% – Single piece skirt and flange face, provides good shear coupling to shell ID of skirt machined, but face is not Backing piece provides extra thickness for required stiffness N. Hartman LBNL

Pixel Detector ATLAS Revised Flange Analysis • ANSYS analysis – – • Results – – • – Max Bolt Force Area September 2002 Sub-micron displacement in flange Max bolt load ~100 N (at topmost bolt) Glue Stress Calculations – Glue Shear Stress Calculation Stiffening ribs removed Flange constrained over bolt stress areas only (~2. 5*Bolt Dia. ) Bolts omitted on diameter (planned pin locations) 2 mm forward end offset used (worst case) – – Simple shear stress calculated (Shear = Axial Force/Area) Max bolt load used and area assumed to be ½ of 1 stiffening unit (1/48 th of flange circumference) Max stress assumed of 21 Mpa (Hysol Adhesive) Factor of Safety = 140 for 25 mm long flange N. Hartman LBNL

ATLAS Pixel Detector Rail Design September 2002 N. Hartman LBNL

Pixel Detector ATLAS Rail Design Summary • So far, the PST has been modeled without considering the effect of rails in the bending stiffness of the shell – Provided for faster/easier modeling – Will result in higher displacements in the SCT when rails are added • Rails have conflicting design demands – Rail deflection must be minimal, to assure installation of detector – Rail stiffness must also be minimal, to reduce impact on SCT • Initial analysis showed problems – Rail deflections were perhaps acceptable (~150 -200 microns) – However, impact on stiffness unacceptable (increase of 85%) • Design shifted to one piece rail – Goal to increase local section modulus of rail, but with lowest cross sectional area possible • “Hollow” shape more efficient • However takes up more space in PST – Fiber changed to high strength carbon (rather than high modulus) in order to lower contribution to overall shell stiffness September 2002 N. Hartman LBNL

ATLAS Pixel Detector Evolution of Rail Design Initial rail shape designed to use as little space as possible inside PST, and to allow placement of sliders anywhere along frame V-rail changed to “inverted” v shape. Increases inertia of section, and can be used either as v or inverted v. One piece design chosen to fill maximum volume (and increase bending stiffness of rail itself). This was made possible by decision to place sliders/rollers at end of frame, freeing up space for rail inside. September 2002 N. Hartman LBNL 14

Pixel Detector ATLAS Rail FEA Model simulates prototype of rails and 300 mm long shell (initial two-piece rail shape). tapers on ends for rail misalignment Pixel Mass (1/4 of 35 kg) applied to PEEK slider. Slider impacts rail through contact elements. Shell is constrained along edges (where flanges or stiffeners would be). center bearing Section (R = 10 mm, L = 20) Shell modeled as both quasi-isotropic glass laminate and composite hybrid laminate of carbon and glass. Prototype PEEK slider shell constrained on edges slider rail 300 mm shell slider Cross section of v-rail and slider September 2002 N. Hartman LBNL 15

Pixel Detector ATLAS Rail Analyses Composite Carbon/Glass Shell (Carbon in Hoop Direction) Eaxial = 21 GPa; Ehoop = 147 GPa Quasi-isotropic Glass Shell E = 19 GPa Slider made from PEEK E = 3. 5 GPa Rail Quasi-isotropic CN 60 E = 126 GPa Load Applied = 8. 75 kg Dmax = 185 microns September 2002 Hybrid Shell reduces rail displacement by 20% Rail Quasi-isotropic CN 60 E = 126 GPa Load Applied = 8. 75 kg Dmax = 154 microns N. Hartman LBNL 16

ATLAS Pixel Detector Expected Rail Performance • Rails displace more in beam mode than shell mode (displacements are primarily not in the cross sectional plane) – Deflection scales by stiffness (EI) of rail itself (to first order) – However, adding additional hoop plies of YSH 80 (in the forward) does help by about 20% • Different rail designs were compared for optimization – FEA results used as a starting point and comparison – Different designs compared by calculating EI, and then scaling to find expected stiffness and deflection implications Deflection in final rail shape is anticipated to be on the order of 125 microns (5 mil). September 2002 N. Hartman LBNL

ATLAS Pixel Detector Anticipated Loads/Displacements Induced in SCT With Stiffer Forward PST Shells, Due to Installation Rails Highest Displacements and Forces Still Arise from Gravity and CTE Loading, Which are not affected by an increase in the Forward Shell Stiffness. September 2002 Z constrained flexure is located on side C, negative X (in this coordinate system). N. Hartman LBNL 18

ATLAS Pixel Detector Prototyping September 2002 N. Hartman LBNL

Pixel Detector ATLAS Prototyping Plan • Material Testing – First test completed – Results are fairly consistent, but disagree with calculations • Shells – presented seperately – Completed – Successfully demonstrate ability to reliably make tubes of given size • Rails – presented seperately – Partially Complete – foot long rails have been made – Successful so far, but issues remain – Rail Sliders and/or rollers need to be fabricated and tested • Flanges – To be outsourced, not yet complete • Hoop Stiffeners – May not be prototyped (fabricate during production phase only) • Mount Pads/Flexures – presented seperately – To be fabricated in-house, not yet complete • PST Assembly (bonding) – Not yet complete – To be fabricated in-house and/or outsourced – Design yet to be completed September 2002 N. Hartman LBNL

Pixel Detector ATLAS Material Test Results • Calculations Differ substantially from results attained – Modulus of YSH 80 samples is almost 40% low in cases (this modulus would be expected with CN 60 type fiber) – Modulus of CN 60 sample is approximately 20% low – Fiber volume from one sample is low (other samples not tested) • However, measurements are fairly consistent – – September 2002 YSH 80 samples are both very low E 1 and E 2 directions are similar (quasi-iso layups) Spread in test data (from multiple coupons) is not extreme Void content in one sample is fairly low (other samples not tested) N. Hartman LBNL

Pixel Detector ATLAS Major Outstanding Items • Design – Rail Riders • Conservative choice is a rolling mechanism for detector – Space available at end of frame – Detector is more than half of sliding mass (on four support points) • Sliders will be retained for service structure – Space for rollers is probably not available – Each support sustains lower load – Rails • Is 35% increase in bending stiffness of forward tube acceptable? • Rigorous FEA model of new rail design must be completed, along with tests to validate stiffness – Flange/Mount Pads • Design must be modified for new flange (without ribs) • Prototyping – Material Properties • Discrepancies must be reconciled (Test accuracy or fabrication? ) – Hoop Stiffeners • Layup separately or incorporate in shell layup (this would require prototyping) – Bond Tooling • All design must be completed in order to finish prototype phase September 2002 N. Hartman LBNL