ADVANCED APERTURE RESEARCH Task 22

Many defects created during the semi-conductor assembly and test process are best detected by using measurement techniques. These defects include warpage (co-planarity), component displacement and solder ball deformities. Detecting these defects during the technology development process is critical to the development of a qualified manufacturing process. Current automated inspection machines have difficulty in accurately detecting these defects consistently over many different types of products with many different features and materials. In addition, inspection tools are expensive and the lack of ability to inspect a wide variety of products contributes to the purchase of many different machines to inspect different products and different product features. Manual inspection techniques are very time consuming. Only a small amount of product can be inspected manually for any lot of material which can result in defects being missed and potential escapees. A means to automatically, accurately, and reliably measure specific areas of the assembled unit and determine if any discrepancies exist is needed to ensure a quality process and product. The algorithms developed to measure z-height also have the potential to provide accurate measurements of heights of die, epoxy and components along with component width. These steps are currently performed by operators using time consuming processes that require manual handling of each unit. This limits the number of units that can be inspected and also puts the units inspected at risk from the manual handling. An automated algorithm to detect and measure the height of the various features of the unit can mitigate the manual handling issues, provide accurate measurements and permit a larger sample of units from each lot to be examined safely. In addition, an automated algorithm will have the capability to be tuned for different product sizes, configurations and features to provide a flexible, extensible solution. The algorithms developed under this project will offer a cost effective, flexible, extensible alternative to costly inspection tools and a solution that is optimized for Intel specific needs and defects.