Browse Prior Art Database

Program for 2-Dimensional, Multiple Site Process Control Indices

IP.com Disclosure Number: IPCOM000113181D
Original Publication Date: 1994-Jul-01
Included in the Prior Art Database: 2005-Mar-27
Document File: 2 page(s) / 59K

Publishing Venue

IBM

Related People

Wiley, JP: AUTHOR

Abstract

A program is disclosed that calculates process control indices appropriate for multiple sites and in two dimensions. It is specifically applied to hole locations in high performance circuit boards, and is applicable to any two-dimensional pattern or device having many sites of variable error which must all be within specification simultaneously.

This text was extracted from an ASCII text file.
This is the abbreviated version, containing approximately 64% of the total text.

Program for 2-Dimensional, Multiple Site Process Control Indices

      A program is disclosed that calculates process control indices
appropriate for multiple sites and in two dimensions.  It is
specifically applied to hole locations in high performance circuit
boards, and is applicable to any two-dimensional pattern or device
having many sites of variable error which must all be within
specification simultaneously.

Procedure is as follows:

1.  Find parameters for systematic error as a function of position on
    the part [1,2].

2.  Calculate maximum systematic error based on the parameters and
    the known active area of the part.

3.  Calculate failure probability for a single site assuming the
    random or non-systematic portion of the error follows Rayleigh
    distribution.  The specification is eccentric to the distribution
    due to systematic error.  The Rayleigh distribution must be
    integrated over the volume which lies beyond the spec.; a
    numerical technique (not disclosed) is required for this purpose.

4.  Find the failure probability for many sites.  The number of sites
    is a fraction of the total.  It is assumed that the maximum
    systematic error is constant over one corner and the systematic
    error over the remainder of the surface is zero.  The number of
    sites is assumed to be one-eighth of the total number of sites in
    this analysis.  Other fractions may be appropriate for other
  ...