Applied Statistical Consultants 12 Wildwood Drive Springvale, Maine 04083 Contact Us

Electronics Case Study   |   Neural Networks Case Study

Case Study:  Electronics Packaging

Background:
Pop-corning is a problem known to occur in the manufacture and use of electronic components.  When excessive moisture is trapped inside the component during the encapsulation process, the parts tend to bubble and crack when heated, such as during solder reflow operations.  This can be an obvious problem to users of electronic components, but it is also a serious issue for manufacturers of electronic components.  Since this problem is not seen until the product is completely through the assembly process, any yield issue directly impacts the manufacturer's bottom line profitability.

Problem:
Recently, a manufacturer of electronic components was suffering with a yield issue attributed to pop-corning.  An industry expert was contracted to solve the problem.  After several months of investigation and experimentation, several possible root causes were identified and a corrective action was recommended.  The cost of this project was well over $100,000, including the contractor's fee, internal technician resources, and work-in-process.  After the corrective action was implemented and the contract was up, the pop-corning continued at the same rate as before. 
 

ASC Solution 

Goal:
To eliminate pop-corning as a yield concern in the manufacture of electronic components.  

Objective:
To test the hypothesis that pop-corning is a function of certain materials and their associated curing profiles. 

Results:
A fractional factorial experiment was designed, run, and analyzed at a total estimated cost of under $6,000.  Pop-corning was found to be a function of one of the materials and its curing profile.  A simple adjustment was made to the process and pop-corning was eliminated as a defect mode for this component.

► Neural Networks Case Study