Dismiss
InnovationQ will be updated on Sunday, Oct. 22, from 10am ET - noon. You may experience brief service interruptions during that time.
Browse Prior Art Database

Smart Quality Monitoring of Dicing Process

IP.com Disclosure Number: IPCOM000116114D
Original Publication Date: 1995-Aug-01
Included in the Prior Art Database: 2005-Mar-30
Document File: 2 page(s) / 87K

Publishing Venue

IBM

Related People

Chaussard, J: AUTHOR [+3]

Abstract

Disclosed is a continuous spectrum system for analyzing mechanical vibrations caused during the dicing operation. The wafer dicing operation is improved by connecting a piezo electric accelerometer on the dicing spindle. The accelerometer generates an analog electrical signal that is illustrative of mechanical vibrations and shocks. The electrical signal is applied to a spectrum filter with selected response that is capable to identify the signature of the dicing operation, to detect any malfunction and lastly to take the appropriate action. The functional frequency range is between 600 and 4 000 Hz.

This text was extracted from an ASCII text file.
This is the abbreviated version, containing approximately 100% of the total text.

Smart Quality Monitoring of Dicing Process

      Disclosed is a continuous spectrum system for analyzing
mechanical vibrations caused during the dicing operation.  The wafer
dicing operation is improved by connecting a piezo electric
accelerometer on the dicing spindle.  The accelerometer generates an
analog electrical signal that is illustrative of mechanical
vibrations and shocks.  The electrical signal is applied to a
spectrum filter with selected response that is capable to identify
the signature of the dicing operation, to detect any malfunction and
lastly to take the appropriate action.  The functional frequency
range is between 600 and 4 000 Hz.

      The feasibility has been demonstrated with a piezo electric
accelerometer.  The signal is monitored for instance by using the
multipurpose monitor.

      Fig. 1 shows the signature of a normal dicing process.  Fig. 2
is typical of an edge chipping occurrence, whereas Fig. 3 shows the
signature in case of an untapped wafer.

      A more intelligent and adaptive analysis system may use neural
system expert system software for self-learning and for total
automation of the dicing process.