• Journal of the Korean Society for Precision Engineering
• /
• v.29 no.8
• /
• pp.879-886
• /
• 2012

The ultrasonic horn used for bonding of flip chip has been designed to vibrate at a natural frequency. The ultrasonic horn must be manufactured accurately in physical terms, because the small change of mechanical properties may result in the significant change of natural frequency. Therefore, tight tolerance is inevitable to keep the natural frequency in acceptable range. However, since tightening of the tolerance increases the manufacturing cost significantly, trade-off between the cost and accuracy is necessary. In this research, an attempt was made to design the ultra sonic horn within acceptable natural frequency while the manufacturing cost was kept as low as possible. For this purpose, among the 18 tolerances of physical terms of the ultrasonic horn, the most important 4 factors were selected using Taguchi method. The equation to relate those main factors and the natural frequency was made using response surface method. Finally, optimal design scheme for minimum manufacturing cost without a loss of performance was determined using SQP method.

• Proceedings of the Korean Society for Noise and Vibration Engineering Conference
• /
• 2010.10a
• /
• pp.279-280
• /
• 2010

• Proceedings of the Korean Society for Noise and Vibration Engineering Conference
• /
• 2010.05a
• /
• pp.592-593
• /
• 2010

• Journal of the Microelectronics and Packaging Society
• /
• v.14 no.4
• /
• pp.79-85
• /
• 2007

This study was focused on the feasibility of ultrasonic bonding of Au stud flip chip bumps on the flexible printed circuit board (FPCB) with three different surface finishes: organic solderability preservative (OSP), electroplated Au and electroless Ni/immersion Au (ENIG). The Au stud flip chip bumps were successfully bonded to the bonding pads of the FPCBs, irrespective of surface finish. The bonding time strongly affected the joint integrity. The shear force increased with increasing bonding time, but the 'bridge' problem between bumps occurred at a bonding time over 2 s. The optimum condition was the ultrasonic bonding on the OSP-finished FPCB for 0.5 s.

• Proceedings of the Korean Society for Noise and Vibration Engineering Conference
• /
• 2011.10a
• /
• pp.289-290
• /
• 2011

• Journal of Welding and Joining
• /
• v.26 no.1
• /
• pp.31-36
• /
• 2008

• Proceedings of the KSME Conference
• /
• 2008.11b
• /
• pp.2750-2753
• /
• 2008

This paper introduce the CFD analysis for predicting the heat transfer at the Ultrasonic horn. Approximately Ultrasonic horn separates two part. One is preheating part and the other is cooling part. Temperature of preheating part rise up by $260^{\circ}C$ that make it possible to attach a chip to a semiconductor. Also there is a piezo material in the cooling part. When piezo work, it generates heat of $100^{\circ}C$. It can stand by $150^{\circ}C$. But the high temperature conducted from the preheating part has a bad affect on the piezo. These situation make it necessary cooling at piezo. Previously except of the piezo, all of them are composed of the SUS440c that has good thermal conductivity. This study shows way that not only cooling the piezo but also cutting off the conduction between preheating part and cooling part by using the Ti and Duralumin that have low thermal conductivity compare with the SUS440c. Conclusion of CFD analysis that the heat coming from the piezo can't be transferred the horn cause of the Ti and Duralumin.

• Journal of the Microelectronics and Packaging Society
• /
• v.13 no.1 s.38
• /
• pp.23-29
• /
• 2006

Ultrasonic soldering of Si-wafer to FR-4 PCB at ambient temperature was investigated. The UBM of Si-substrate was Cu/ Ni/ Al from top to bottom with thickness of $0.4{\mu}m,\;0.4{\mu}m$, and $0.3{\mu}m$ respectively. The pad on FR-4 PCB comprised of Au/ Ni/ Cu from top to bottom with thickness of $0.05{\mu}m,\;5{\mu}m$, and $18{\mu}m$ respectively. Sn-3.5wt%Ag foil rolled to $100{\mu}m$ was used for solder. The ultrasonic soldering time was varied from 0.5 s to 3.0 s and the ultrasonic power was 1,400 W. The experimental results show that a reliable bond by ultrasonic soldering at ambient temperature was obtained. The shear strength increased with soldering time up to a maximum of 65 N at 2.5 s. The strength decreased to 34 N at 3.0 s because cracks were generated along the intermetallic compound between Si-wafer and Sn-3.5wt%Ag solder. The Intermetallic compound produced by ultrasonic soldering between the Si-wafer and the solder was $(Cu,Ni)_{6}Sn_{5}$.