Warpage Measurement of Board Assemblies Using Projection MoirÉ System With Improved Automatic Image Segmentation Algorithm

Abstract

Out-of-plane displacement (warpage) has been a major reliability concern for board-level electronic packaging. Printed wiring board (PWB) and component warpage results from coefficient of thermal expansion (CTE) mismatch among the materials that make up the printed wiring board assembly (PWBA). Warpage occurring during surface-mount reflow process and normal operations may lead to serious reliability problems. PWB/PWBA warpage can be measured by many different optical techniques, among which moire methods have emerged as real-time, noncontact, full-field, and superior techniques with high resolutions. In this paper, a warpage measurement system based on the projection moire technique and improved automatic image segmentation algorithm is presented. In order to use the projection moire system to accurately and separately determine the warpage of a PWB and assembled chip packages in a PWBA, a novel automatic image segmentation algorithm was developed. The algorithm uses mask image models to detect and segment assembled packages from the PWB, and defines their warpage, respectively. This approach resulted in higher resolution and processing rate compared to the original algorithm based on snakes. The warpage of the PWB and chip packages in a PWBA can be successfully determined individually irrespective of the package locations and orientations. Real-time continuous and composite Hermite surface models were generated to evaluate the PWB warpage values underneath packages. The improved segmentation algorithm was integrated with the projection moire system so that the system is able to accurately evaluate the warpage of the PWB and chip packages in a PWBA, respectively.