DOE PROGRAM REVIEW June 20071

1
Development of the LSST Camera Cryostat at SLAC
Engineering and RD June 5, 2007

• Rafe H. Schindler (SLAC)

2
CRYOSTAT ASSEMBLY Reference Design as of Aug
2006
Front flange for mounting for L3 Flange
Bosses on GRID for Raft Kinematic Mounts
Copper feedthrus
Fiber-optic feedthrus
3X GRID Z flexures
Rear Access Plate
Vacuum-jacket penetration for cryogen lines
Double O-rings at Plate mount
FP Fast Actuators
GRID
Cryo Plate for FEE cooling
Back flange supports Feedthru Flange and cyl.
mount
Cold Plate for BEE cooling
Cryostat and Valve Box Section view from front
side
3
OUTLINE

• ENGINEERING and RD GOALS FOR 2007
• Cryostat Mechanical Design
• Develop/Formalize Cryostat Interface Definitions
• Advance Baseline Design verify critical
  features with detailed analysis
• Material Trade Studies For Long-Lead Time
  Elements (GRID)
• Contamination Control
• Construct Material Qualification Test Facility
  (MQTF) To Evaluate Preparation and Impact of
  Materials Deployed in The Cryostat
• Develop Contamination Control Plan
• Metrology
• Evolve The Focal Plane Alignment Concept
• Prototype Critical Features of Raft-GRID
  Kinematic Interface
• Prototype Performance of Non-Contact Metrological
  System
• for Camera Assembly
• Study of In-Situ Metrology Wont Discuss Today

4
CRYOSTAT SUBSYSTEM ITS MAJOR PHYSICAL
INTERFACES

CRYOSTAT SUBSYSTEM
INTEGRATION TEST
CONTAMINATION CONTROL
FRONT VAC FLANGE
CRYOSTAT BODY
INSTRUMENTN RING
GRID FLEXURES KIN MOUNT
GRID
REAR VAC. , SIGNAL , FLUID FLANGE
Camera Body Utilities Trunk Thermal System Vac.
System Purge System Controls
L3 Raft Tower Vac System Thermal Sys.
Raft Tower FE 55 Source Diff. Pat. Gen. Controls
BEE Thermal Sys. Vac. System Controls
Raft Tower Guider/WFS Corner Raft Metrology
Sys. Thermal Sys. Controls
CRYO PLATE
COLD PLATE
SHROUD
COOLANT XFER LINES (2X)
Legend
ELEMENTS OF CRYOSTAT
Vac Sys. Thermal Sys. Controls
BEE Thermal Sys. Controls
Raft Tower Guider/WFS Thermal Sys. Controls
Raft Tower Vac Sys. Thermal Sys. Controls
MECHANICAL, ELECTRONIC, THERMAL OR OTHER
INTERFACES
5
DEFINING / FORMALIZING CRYOSTAT INTERFACES

• 2 DAY FACE TO FACE MTING AT SLAC IN MID-MAY TO
  DISCUSS MAJOR CRYOSTAT INTERFACE ISSUES WITH
  SENSOR RAFT-TOWER GROUPS
• RESULTED IN START OF RAFT ?? CRYOSTAT INTERFACE
  DOCUMENT INCLUDING

• RAFT GRID ENVELOPES
• THERMAL LOADS (FEE BEE)
• FOCAL PLANE METROLOGY (METHODOLOGY XFER OF
  REFS)
• ELECTRONICS FEETHRUs

EXAMPLE RAFT??GRID STAY CLEAR DOCUMENT
6
CRYOSTAT ENGINEERING ADVANCING THE BASELINE
DESIGN

• L3 lens and flange structural analysis
• Re-analyzed L3 to verify its ability to carry a
  vacuum load
• Detailed analysis of temperature
• gradients across L3 and
• accompanying distortion
• Cryo Plate structural analysis
• Verify plate can carry spring
  
  inertial loads of
  RAFT
  
  ? with little deflection
• Cryostat Body design changes
• Cylinder ? flared frustrum with in-line flange
• Came out of lens motion sensitivity analysis
  showing back flange support ring not stiff
  enough ? reduced local rotations of focal plane
  by 2X
• GRID Flexures Design Changes
• Elimination of Fast Actuators created GRID shear
  stiffness (bookshelfing) problem
• Change GRID ( Cryoplate) shape from octagons to
  round add stiffening elments
• Change GRID Flexures to A Frame reduces
  tip/tilt sensitivity of Focal Plane
• Complete stress and buckling analyses (see
  example)

L3 X-Section
7
Cryostat Housing Structural Design Change
Flared housing provides more radial space for
feedthrough plate
Cryostat housing tapers back to an in-line flange
One-piece support housing and Utility Trunk
forward section increases mounting stiffness
8
EVOLUTION OF GRID, FLEXTURES, COLD PLATE DESIGN
Flexures attach at 2 points on the Grid to
distribute loads Pre-mounted inspected off
three points prior to insertion mounting to
support flange
Round GRID With Features for A-Flexures and
Closed Plates Around Perimeter for Shear
Stiffening
Original Octagonal GRID With Features for 8 Fast
Actuators and 3 Z-Flexures
Thermal Shroud
Round Cryo Plate
9
Material Manufacturing Trade Studies For GRID

• Silicon Carbide remains material of choice for
  GRID because of its stiffness, high thermal
  conductivity, low thermal expansion coef. vac.
  friendliness
• GRID material and manufacturing trade study being
  conducted in conjunction with detailed FEA for
  the design of the GRID
• Identified numerous prospective SiC manufacturers
• SSG
• POCO Graphite
• Coorstek
• Cercom
• Ceradyne
• ECM
• IBCOL (Germany)
• M-Cubed
• Xinetics
• McCarter, Morgan Advanced Ceramics, Saint-Gobain,
  Trex, Kyocera,
• Identified other viable candidate materials
• High-nickel alloys Invar-36, Nilo, AL-36
• Beryllium alloys O-30H

GRID Fabrication is a Long Lead Time Procurement
- Likely to Require Vendor RD Ongoing
Discussions With Three Vendors Each Have
Different Processes, Finishes, Max-Size
Dimensional Tolerance Capabilities --- LINKED TO
STRUCTURAL ANALYSIS---
10
EXAMPLE DRUMHEAD SHEAR DISTORTION OF SiC GRID
Flexure Location
Flexure Location
Flexure Location
Flexure Location
Flexure Location
Gravity
Flexure Location
Gravity
Maximum z displacement -2mm
Maximum shear displacement -2.9mm
Camera Coordinate System
Camera Coordinate System

• Finite Element Analysis of GRID Displacements
  under gravity
• Includes Load of Raft-Towers
• Example Assume SiC CESIC with E 249 GPa,
  n0.17 r 2700 kg/m3

11
CONTAMINATION TEST CHAMBER (MQTF)

• FEE cables (at the minimum) need to located
  within the Cryostat to achieve desired camera
  performance. They represent a significant surface
  area of non-vacuum friendly materials in close
  proximity to cold CCDs of the FPA.
• In addition to pumping, baffling (reducing
  molecular flow), temp. control of gettering
  surfaces
• developing protocol for material selection,
  processing assembly to mitigate the long term
  contamination risk to the FPA.
• also reviewing feasibility of removing BEE from
  cryostat
• Early Development of Three Contamination Test
  Chambers (MQTF) allow
• study of material preparation (vacuum bakeout) in
  the 1st chamber
• measuring outgassing (rates species) and
  deposition of condensable films onto cold
  surfaces (thickness species) in the 2nd chamber
• measuring transmission vrs l thru condensable
  films (cold) in 3rd chamber
• INFORMATION NEEDED EARLY TO ALLOW
  DESIGN TO PROCEED
• Once a material preparation procedure is
  qualified
• can define minimal tests assuring components
  prepared in this fashion, meet final
  requirements before IT

12
CONTAMINATION TEST CHAMBER IN Bld-33 (Clean
Room)
Sample Preparation Chamber
Outgassing Analysis Chamber
Optical Transmission Chamber
Main Chamber
MAIN
ANTE
FORE
Sample Entry
Straight-Thru Valve
Straight-Thru Valve
Fore or Preparation Chamber
cold finger
Linear Manipulators WS, RGA, QCM, Gas sys., Ion
pump not shown
Scroll Turbopump
Funded Thu RD Program SLAC ( beg, borrow,
(steal)) 3
mo. for bakeout commissioning in Bld-33 Then
Move to Campus Lab
13
3rd OPTICAL TRANSMISSION CHAMBER and OPTICAL TEST
SETUP
Optical Setup Being Tested in Lab
Squat To Reduce Light Path
30cm
cold finger
Contaminated Reference Samples Moved (Cold)
From 2nd Chamber White Light ? Window ?Filter
Wheel (LSST) ? Cold Sample Slide/or
Uncontaminated Slide ? Window ? Precision
Photodiode/PM. Piston Moves Samples Back Forth
In Vacuum Through Same Optical Path to Chop
Measurements (Average Out Instabilities in Light
Output) Expect lt0.1 sensitivity comparable to
photometric target
14
FOCAL PLANE FLATNESS / METROLOGY

• REQUIRE 10 mm (P-V) FLATNESS ACROSS FPA
• FOCAL PLANE BUILT UP OF THREE STRUCTURES
• SENSORS (5.0mm P-V)
• RAFTS OF 9 CCD (6.5mm P-V)
• GRID OF 21 RAFTS (10.mm P-V)
• STRATEGY BUILD PRECISION INTO GRID RAFTS IN
  ADVANCE TO ALLOW A SNAP TOGETHER ASSEMBLY
  ENGINEER IN RATHER THAN CALIBRATE OUT
• Make GRID a Passive Structure Both Mechanically
  Thermally
• Choice of SiC For Stiffness Conductivity
  Thermal Exp.
• Thermally Isolate To Minimize/Control Heat Flow
  Distortion
• Make Rafts Interchangeable Without Further
  Adjustment
• Use a three-vee kinematic couplings with sub-mm
  repeatability
• Setup all rafts relative to their kinematic
  mounts on same fixture

Actual Errors More Complex eg See Accompanying
Slide
15
PRELIMINARY ANALYSIS OF FOCAL PLANE FLATNESS
ERRORS
3s Errors
SENSOR RAFT (7.2mm)
GRID (2.6mm)
DYNAMIC (6.2mm)
16
RAFT TO GRID KINEMATIC MOUNT CONCEPT
RAFT WITH SENSOR PKG, T-STRAPS, CABLES
EMPTY Al Nitride RAFT PLATE
INTEGRAL VEE BLOCKS
SENSOR MOUNT ALIGNMENT STUDS (3x)
SiN BALLS, SiC BOSSES SS CLIPS
GRID
GRID BAY
17
Raft Metrology Fixture Embodies Desired Geometry
Needed at Focal Plane Array
Raft GRID Metrology Strategy To Simplify IT
Raft Metrology Fixture
FEE Cage
AT BNL
Raft
Flat and parallel to 3 balls 100 nm
Set up detectors coincident to the flat reference
surface of the fixture
18
A Master Raft Fixture transfers the Raft
Metrology Fixtures geometry (at BNL) to The
GRID (at SLAC)Raft Kinematic Mounting Points On
GRID Are Measured/Adjusted to Mimic Raft
Metrology Fixture In Advance of LoadingMotion
Stage Used to Lift Master Raft Fixture Onto GRID
Controlling Mating Points on Individual Kinematic
Mounts During Focal Plane Assembly, Same
Stage Lifts Raft Onto the Same Kinematic Mating
Points On GRID From Behind Metrological
Verification Follows Immediately By Differential
Non-Contact System
Raft GRID Metrology Strategy To Simplify IT
AT SLAC.
19
KINEMATIC MOUNT PROTOTYPE FOR RAFT-GRID

• METROLOGY STRATEGY PREDICATED ON USE OF 3-Vee
  KINEMATIC COUPLINGS TO PROVIDE SUB-mm
  REPRODUCABILITY OF THE RAFT-GRID INTERFACE
• LITERATURE SHOWS THAT COUPLINGS ARE SUITABLE BUT
  REQUIRE A BIT OF ENGINEERING ART ON A CASE BY
  CASE BASIS SENSITIVITY TO
• Bulk material properties (stiffness and bearing
  strength)
• Surface condition (finish and friction)
• Shape of vee (wall angles, slotting, arching)
• Mating (controlling mate-up)
• Axial Stiffness (limited)
• RD FOR RAFT-GRID KINEMATIC COUPLING TO VERIFY
  PERFORMANCE
• Developing Prototype Program In conjunction with
  GRID material trade study
• Find Best Geometry, Materials and Surface
  Preparations
• Measure
• Variability from GRID Fabrication (one Raft on
  many Ball sets)
• Repeatability of individual mounts (one Raft on
  one Ball set)
• Interchangeability (many Rafts on one Ball set)
• Temporal Stability of individual mounts

20
KINEMATIC COUPLING PROTOTYPE RD
10 Dewar Window (not shown)
Raft Plate
Raft Plate Assembly with Vee-Blocks
Interchangeable Vee-blocks A286 ss, 6Al4V
Ti, SiC Inconel 718.  Misc. coatings (Nedox,
DLC) for friction hardness. 16 min roughness
Raft Plate
Kinematic Coupling Prototype
Parking Balls
Reference Surfaces
2ND Ball Set
Interchangeable Vee-blocks angle and/or shape
Silicon-nitride CL 5 Ball (round to 5min, smooth
to 0.8 min rms)
Base Mount with Flexure supports off the flange
Epoxy Insertion
Cold straps
Test dewar
Kinematic coupling interface
Ball-and-Cup Assembly (2X3)
Section View of Raft Plate Base Mount
21
RD To Verify Metrology System Concept for FPA
Assembly
Strategy Measure Flatness of Focal Plane Using
a Pair of Commercial Laser Displacement Sensors,
Optical Stage, and a Smaller Reference
Flat Stitch The Entire Surface Together By
Measuring Multiple Overlapping Regions (each
RAFT Size)
Bottom view shows partial buildup of focal plane
Camera focal plane surface
22
DEMONSTRATION OF DIFFRENTIAL NON-CONTACT
METROLOGY SYSTEM IN AIR
XY STAGE TO MOVE LASER HEADS
l/10 OPTICAL FLATS
UP DOWN LOOKING LASER DISPLACEMENT HEADS
XY
23
DEMONSTRATION OF NON-CONTACT SYSTEM IN AIR

• Measure Differential figure between two ?/10
  flats
• Scan over 200mm diameter (300 points) takes 600
  seconds
• Sensor Non-Linearity correction applied
• Computed figure is lt 0.4um P-V over
  200mm diam. surface with better than 0.1um
  repeatability

SCALE OF RAFT
Next Step To Repeat Through Vacuum Window
Cold
24
CONCLUSIONS

• ENGINEERING PROCEEDING ON CRYOSTAT DESIGN
• UPDATING/DETAILING DESIGN WITH FINITE ELEMENT
  ANAL IN CRITICAL AREAS
• DESIGN AND ANALYSIS COUPLED TO MATERIALS
  SELECTION STUDIES
• CRITICAL INTERFACES BEING DEFINED (esp.
  RAFT-TOWER to GRID)
• THERMAL ANALYSIS ON HOLD THIS YEAR FOR
  BUDGETARY REASONS
• TOOLS FOR CONTAMINATION STUDIES UNDER
  CONSTRUCTION
• DEVELOPING MATERIALS TEST FACILITY --
  CONSTRUCTION ADVANCING WELL
• OTHER IMPROVEMENTS TO DESIGN BEING CONSIDERED TO
  FURTHER MITIGATE RISK
• METROLOGY CONCEPTS HAVE MATURED
• OVERALL CONCEPT FOR FOCAL PLANE ASSEMBLY HAS BEEN
  DEFINED
• PROTOTYPE KINEMATIC MOUNT FOR RAFT ALIGNMENT ON
  GRID BEING TESTED AT SLAC
• TESTS OF CONCEPT TO ALIGN SENSORS ON RAFTS TO BE
  CONDUCTED AT BNL
• PROTOTYPE METROLOGY SYSTEM FOR FULL FPA ASSEMBLY
  TESTED SUCCESSFULLY IN AIR SETTING UP FOR
  VACUUM/COLD TEST

25
EXTRA SLIDES

• EXTRA SLIDES

26
Back View Of Contamination Test Chamber
27
Demonstrated fidelity is adequate for FP
metrology feedback
Leitz PMM 12106 (CMM) advertised standard high
(volumetric) uncertainties U3 for 120mm
translation
calibration limit (flatness of reference surfaces)
28
Stitching Metrology Checkout

• Detector surface figure measurement fidelity
  depends upon the tradeoff imposed by measurement
  rapidity vs. completeness.
• We have shown that simulated, bootstrap surface
  figure measurements provide quatifiable error
  propagation for specific sets of measurement
  grid, reference grid and reference shape
  definitions.

29
Parasitic derivation of nonlinearities in each
displacement sensor

• LK-G32 50 µm period, 1.5 µm P-V
• LK-G15 10 µm period, 0.35 µm P-V