A Wavelet-based Approach for Ball Grid Array (BGA) Substrate Conduct Paths Inspection

The aim of this paper is to locate the boundary defects such as open, short, mousebite, and spur on Ball Grid Array (BGA) substrate conduct paths using machine vision. The 2-D boundaries of BGA substrate conduct paths are initially represented by the 1-D tangent angles. The tangent angles are evaluated from the covariance matrix eigenvector over a region of support on a boundary segment. Boundary defective region results in irregular tangent angle variations. Then, the wavelet transform decomposes the 1-D tangent angles and captures the irregular angle variations by indicating larger magnitude of wavelet coefficients on finer scale. Finally, the upper limit for the wavelet coefficients of BGA substrate conduct paths can be established by Quality Control (QC) skills. A boundary defect can be easily located by if its wavelet coefficients exceed upper limit. Real BGA substrates with various boundary defects are used as test samples to evaluate the performance of the proposed method. Experimental results show that the proposed method achieves 100% correct identification for BGA substrate boundary defects by appropriate wavelet basis and decomposition level. The proposed method is invariant with respect to the orientation of the BGA substrates, and it does not require pre-stored templates for matching. This method is suitable for various types of BGA substrates in small batch production because precise positioning of BGA substrates and the prestored templates are not required.