Monitoring wire bonding via time-frequency analysis of horn vibration

Author

Zhang, Dong ; Ling, Shih-Fu

Author_Institution

Sch. of Mech. & Production Eng., Nanyang Technol. Univ., Singapore

Volume

26

Issue

3

fYear

2003

fDate

7/1/2003 12:00:00 AM

Firstpage

216

Lastpage

220

Abstract

In this paper, a technique is developed to monitor a wire-bonding process. Vibration of the horn of a bonder is picked up by a piezoelectric material patch attaching on the surface of the horn and analyzed by joint time-frequency analysis. With neural network, features extracted from the time-frequency distributions of the horn vibration are then exploited for classifying bonding machine setting and predicting bond quality. The results demonstrate the advantages of the time-frequency analysis of horn vibration in monitoring wire bonding.

Keywords

acoustic signal processing; electronics packaging; feature extraction; feedforward neural nets; lead bonding; process monitoring; quality control; time-frequency analysis; ultrasonic bonding; vibration measurement; automatic bonding; bond quality; bond strength; bonding machine setting; electronic packaging; feature extraction; filtered sensor signal; horn vibration; in-process monitoring; interconnections; joint time-frequency analysis; neural network; piezoelectric patch; three-layer feedforward structure; ultrasonic wedge bonder; wire bonding monitoring; Bonding forces; Bonding processes; Feature extraction; Monitoring; Neural networks; Phase locked loops; Piezoelectric materials; Time frequency analysis; Vibrations; Wire;

fLanguage

English

Journal_Title

Electronics Packaging Manufacturing, IEEE Transactions on

Publisher

ieee

ISSN

1521-334X

Type

jour

DOI

10.1109/TEPM.2003.820800

Filename

1256724