| "This seemed like a great opportunity to assist a local manufacturer while giving a student experience..." Professor Don Ermer |
Artesyn Technologies had a good thing going but wanted to make it even better. As a leading manufacturer of high-tech communication products -- such as printed circuit boards (PCBs) for high-quality, modular electronic hardware and software for teledata networks -- the Madison company called on the College of Engineering to help improve its production process for real-time, single-board computer systems.
To build its PCBs, Artesyn uses a wave soldering operation to provide an electrical connection between the through-hole components and the PCBs. With a new development in the industry, the company saw an opportunity to eliminate a time-consuming step in its wave soldering operation and produce a more robust, environmentally friendly and consistent product.
Helping streamline a key production process at Artesyn Technologies were (from left) Fay Susanto, Todd Treichel, quality engineering manager, and Professor Donald S. Ermer. |
The development, a change in the flux, or agent, used to deoxidize the components and the PCBs before the wave soldering operation, meant getting rid of the stage that washed off the toxic flux residue. This no-clean or no-wash flux would lessen the cost of running the expensive wash process by eliminating the need for a large wash machine and ionized water, decreasing electricity demands, reducing "solderability" problems and eliminating toxic residue.
"The change sounded like a good one, but in order to institute it, the complete production process had to be studied and altered," says Todd Treichel, Artesyn's quality engineering manager. For this, the company turned to experts in planned experimentation, or design of experiments, at the College of Engineering. Donald S. Ermer, a Procter & Gamble/Bascom Professor in Total Quality for the Departments of Industrial and Mechanical Engineering, and Fay Susanto, a graduate student in industrial engineering, volunteered to take on the project.
"This seemed like a great opportunity to assist a local manufacturer while giving a student experience that cannot be learned from a textbook," says Ermer. "This kind of a situation puts into practice the students' classroom and textbook knowledge, and the experience of professors, while solving a real industry problem. We find this type of experience invaluable to graduating students. They learn to work in a professional setting, and use problem solving and communication skills." Each semester, one or two of Ermer's students take part in such projects.
In this case, Ermer and Susanto's goal was to find wave solder settings that would minimize the amount of flux residue and the number of solder balls and solder-bridging defects during the wave soldering process, while using the no-clean flux.
Ermer, Susanto and Treichel utilized a design-of-experiment plan to study the conditions on Artesyn's production line. This multi-factor approach involves testing several variables at the same time. Each combination of variables is run during the experiment to determine the main and interaction effects. "Interaction terms always need to be determined in these physical processes in order to fully understand them," explains Ermer.
In this instance, the design-of-experiment method allowed the group to adjust several processing conditions at the same time -- including speed, temperature and wave height -- in order to identify the significant interactions and main effects. Susanto says that their results with the design-of-experiment method illustrated the tremendous value of running proper, planned experimentation and not trial and error.
Along with Treichel, Susanto planned and conducted experiments at Artesyn during her semester of independent study. Running a total of 160 dummy boards (10 per trial) in a variety of machine setting combinations, she was able to identify the wave solder machine settings that would minimize the amount of flux residue and the number of solder balls and shorts, using the new no-wash flux. Susanto also performed a confirmation trial with 30 circuit boards and found no solder defects.
The continued operation is showing an 80-percent improvement in product quality with better solder joints, in addition to a significant cost savings and a reduction of cycle time, says Treichel. The typically two-day operation was reduced by approximately eight hours by eliminating the wash and dry stages. "We also saw cost savings up to 30 percent on certain products.
"This was a real win-win situation," concludes Treichel. "Fay could get real-life experience and we could reap the rewards of what Ermer has studied and is teaching."
--By Amy Krueger--
| For further information, please contact: |
Donald S. Ermer, 608/262-2557
ermer@engr.wisc.edu
Copyright 1999 The Board of Regents of the University of Wisconsin System
Content by perspective@engr.wisc.edu
Markup by webmaster@engr.wisc.edu
Date last modified: Wednesday, 03-Mar-1999 12:00:00 CST
Date created: 03-Mar-1999
