CCD Packaging: Part 1: Concepts Part 2: Cost and Schedule Jim Fast

1
CCD PackagingPart 1 ConceptsPart 2 Cost and
ScheduleJim Fast
2
Packaging Issues from LBL meeting

• Structure glued to CCD surface must extend to
  edges of active area
• No air bubbles are allowed in the bonded region
• Charge migrates at boundaries due to stress
• May not be possible to flat field this effect out
• Need to match the total shrinkage of silicon over
  cool down
• Aluminum nitride hybrid shrinks a bit more than
  silicon
• Invar 36 foot shrinks a bit less than silicon
• Glue bond role is impossible to model must
  prototype
• This is an ELECTRICAL issue (stress moves charge
  around), so we need real sensors and a readout
  system to really investigate the packaging issues
• There is only a 270micron gap between the active
  area and the guard ring structure for the bond
  pads this is VERY tight
• Need to control glue run-out while ensuring no
  voids
• Bond and wedge interference with hybrid are a big
  concern
• See drawing later in talk

3
CCD package LBNL John Bercovitz
4
CCD package LBNL

• John Bercovitz

5
CCD Package assembled
CRIC, Clocks
Molybdenum frame
Invar frame
Aluminum Nitride
CCD
6
Variations
Thermal And Mechanical connection
Coldplate Segment cutaway
PC board
Leads
Coldplate Segment cutaway
Connector
Detector Assembly
7
Package Components
Cold Plate segment
Glue
CRIC, Clocks
Molybdenum frame
Invar frame
Aluminum Nitride
CCD
8
Exploded View from Back
9
Cutaway view, Back
CCD
Flexure Blades
Pins
Aluminum Nitride Substrate
Circuit Board
Invar Frame
M3 -mounts to Coldplate
M2 -into Moly frame
Molybdenum Frame
10
Exploded view, Back
Pogo Pads
Flexure Blades
Copper
Pogos
CCD
Aluminum Nitride Substrate
Invar Frame
Molybdenum Frame
Pins
Circuit Board
11
CTIO package concept very preliminary and
incomplete

• J. Fast

Invar Frame
AlN Hybrid
Wirebonds
CCD
12
CTIO package concept very preliminary and
incomplete

• J. Fast

Invar Frame
Wirebonds
AlN Hybrid
CCD
Edges of active area and guard ring
13
Cost and Schedule Issues

• CRIC chip and CRIC board development and
  production
• Assumes 2 prototype cycles plus final order at 20
  weeks each
• Must start development of board 2/13/04
• Must order first prototypes by 4/22/04 (21K)
• Delivery of first prototypes September 2004
• Delivery of second prototypes July 2005 (22K)
• Production boards and CRIC chips in 2006 (152K)
• CCDs packaging development and production
• Thinned mechanical silicon order in early
  February, in hand late May
• Obtain 10 low grade but functioning CCDs by
  October 2004 (56K)
• Study packaging electromechanical issues through
  early 2005
• Obtain 10 engineering grade CCDs by July 2005
  (60K)
• Finish packaging studies by December 2005
• Production CCD procurement November 2005 (342K)
• Test stands, tooling and mounting foot parts
• 5 test stands one this year (38K) and four in
  2005 (246K)
• Tooling and foot development in 2004 (31K) and
  2005 (13K)
• Production tooling in 2006 (12K)
• Production foot procurement in 2006 (105K)