Wirebonding for TOB and TEC Susanne Kyre University

Wirebonding for TOB and TEC Susanne Kyre University of California, Santa Barbara DOE review, January 20, 2004 Susanne Kyre Slide 1

Wirebonding • • • Equipment and Personnel Scale of the project Bonding times Bond quality TEC Outlook DOE review, January 20, 2004 Susanne Kyre Slide 2

Equipment and Personnel Equipment: • K&S 8090 wirebonder • K&S 8060 wirebonder (receive end of January) • Dage 3000 pulltester • Stereo microscope • Precision measuring microscope Personnel: • 1 main wirebond technician (Julie Stoner) • 1 backup wirebond technician (Andrea Allen) • In the process of hiring a second wirebond technician • 1 support technician (David Staszak) DOE review, January 20, 2004 Susanne Kyre Slide 3

Scale • UCSB will build ~4000 Modules with about 10 million wirebonds (that is 3 times the amount of wirebonds done in all of CDF; All of UCSB’s wirebonding for Ba. Bar was 150, 000 bonds, we will do that in a week) • Full production TEC and TOB: • 15 Modules per day • 30 Hybrids per day DOE review, January 20, 2004 Susanne Kyre Slide 4

Bonding times • Starting point: Bond centers had bonding times of up to 40 minutes with about 10 bonds that didn’t stick and needed to be rebonded • Wirebond programs now bond an entire Module or Hybrid (previously had separate programs for each row of bonds), reduced bonding time by 50% • Wirebond programs for all Module and Hybrid types now use pattern recognition (bonding time reduction by about 2 -3 minutes) • Pure bonding time for a module < 5 min, for a hybrid < 4 min (CMS record) • Additional time (set up, database entry, pull testing): ~11 min for a module, ~ 8 min for a hybrid • Total processing time can be reduce by using a support technician • Actual bonding times during mini production run: • 2. 5 hours for 10 modules (one technician) DOE review, January 20, 2004 Susanne Kyre Slide 5

Bond quality • CMS wirebonding spec: pull strength > 6 g • Extensive pull testing was done to determine a set of standard bonding parameters for each bonding surface Hybrids: • We pull test 10 bonds on a test area on each PA • Average pull strength: 10. 0 g, std. Deviation: 0. 6 g DOE review, January 20, 2004 Susanne Kyre Slide 6

Bond quality Modules: • First 20 parts of each type, pull test every 50 th bond • PA to Sensor bonds: • Average pull strength: 9. 9 g, std. Deviation: 0. 7 g • Sensor to Sensor bonds: • Average pull strength: 8. 2 g, std. Deviation: 0. 8 g DOE review, January 20, 2004 Susanne Kyre Slide 7

Bond quality • 79 Modules and 92 Hybrids bonded so far • Faults introduced during bonding: • • 1 pinhole (first TEC module) out of ~84000 bonds Shorts: 1 repaired short on a hybrid Open channels: 13 on modules, 5 on hybrids Bad channels introduced in bonding: 0. 034% for modules, 0. 012% for hybrids (historically 0. 5% – 2%) DOE review, January 20, 2004 Susanne Kyre Slide 8

TEC production • We will do TEC Ring 5 and 6 Module production • TEC Module and Hybrid wirebonding: • • All fixtures for Rings 5 and 6 are designed TEC Ring 6 module carrier and wirebond fixture were built and tested Wrote program and wirebonded 4 Ring 6 modules Wrote program and wirebonded 6 Ring 6 hybrids DOE review, January 20, 2004 Susanne Kyre Slide 9

Future plans • Produce hybrid and module carriers and wirebond fixtures in machine shop • Write remaining wirebond programs for TEC Modules and Hybrids • Installation and set up of new 8060 Wirebonder (end January) • Training of new wirebond technician (mid February) • Training of wirebond technician from Mexico • Mini production run • 2 weeks at full production capacity of 15 Modules per day (start January 26) • Production bonding! DOE review, January 20, 2004 Susanne Kyre Slide 10