|
|
PROCESS 'OPTIMIZATION' USING DESIGNED EXPERIMENTS IN AN EMS PROVIDER'S FACILITY
|
|
Authors:
|
Sandeep Tonapi
|
|
Company:
State University of NY
|
|
Date:
|
01/25/2000
Conference:
Pan Pacific Symposium
|
| |
|
Abstract: The trend in the electronics packaging industry towards product miniaturization has resulted in the decrease in component density and the number of leads (bumps) per package along with a concurrent decrease in component lead (or bump) pitch. In addition, manufacturing process engineers have to cope with the need to maximize process yields and throughput while concurrently ensuring more than adequate solder joint reliability. The move towards the increased use of Electronics Manufacturing Service (EMS) providers is a result of the need to increase assembly efficiency and decrease processing costs. However, the increase in outsourcing has enhanced the need for systematic process 'optimization'.
This paper discusses the use of designed experiments (both classical and Taguchi) to identify the process window and the 'optimum' print parameters for the stencil printing process. The use of classical as well as Taguchi methods for stencil printing of fine (20 mils) and ultrafine pitch (12 mils) SMCs is discussed. Stencil printing is also used for solder paste deposition for mixed technology assemblies where the Through Hole Components (THCs) are reflow soldered in conjunction with the Surface Mount Components (SMCs) in a single reflow pass. For these assemblies, depositing the required volume of paste at the through hole locations and achieving the desired holefill is as critical as depositing adequate paste at the surface mount locations. The use of designed experiments for these applications is also presented. Statistical analysis and response surface methodology is used to determine the optimum conditions.
Keywords: design of experiments, surface mount technology, printed circuit boards.
|
| |
Cost to download:
Members: Free! (Log on to receive the member rate)
Non-Members: $10
|
Back
|
_160x300.jpg)