Thermal Cycling Tests of Dummy Modules

1
Thermal Cycling Tests of Dummy Modules

• Setup
• Pre-cycling resistance measurements
• IZM module results
• Alenia module underway

2
Setup

• Modules built with CERN Flex V.2 and bare dummy
  module.
• Used Ecobond glue under MCC, pigtail bonding
  field, and 4 corners.
• Modules used for FE wirebonding tests prior to
  thermal cycling
• First cycled modules unconstrained, then attached
  to sector using CGL and UV-cure tacks on 4
  corners.
• Cycled 21oC to 35oC in environmental chamber.
• Monitored module temperature and chain resistance
  in real time

3
Setup (cont.)
Only 4 columns monitored in real time during
cycling
W
W
All columns measured on probe station before and
after
4
Column Resistances (before cycling)
Bad chip
Alenia
IZM
5
IZM Module results

• No change after 2 cycles of unconstrained module
• No change DURING 19 cycles of module on sector
• Open columns detected AFTER overnight storage at
  21oC

On bad chip, all columns were open prior to tests
chip fell off after removing module from
sector. Solder bumps left ½ on chip ½ on sensor.
RED location of open column 32/270 columns
failed
6
Possible Explanation
UV Tack
UV Tack
Carbon-carbon on sector
CGL

• Results are consistent with differential movement
  of dummy chips and dummy sensor.
• For example, assume sensor doesnt move under
  cooldown(probably not correct, just example),
  then contraction of each chip is about (2.5x10-6)
  x(either 3.7 or 5.5x103 microns)x560.5-0.8
  microns.
• Worst case is corners of chip, but we only
  measure columns.
• Approximate strain is (0.5-0.8)/(20) or
  2.5-3.8.. Solder stress vs strain becomes
  non-linear for strains above about 1.
• Creep rupture is possible explanation for failure
  after sitting.

7
Alenia Module Tests Underway

• 3 Cycles to 35o unconstrained
• 3 Cycles to 35o mounted on sector
• Time at 35o 30-40 minutes/cycle
• Let sit for a day
• No failures yet
• Tests continuing