Several studies extolled the benefits of monitoring wrist straps and work surfaces to control ESD. But it wasn’t until Motorola installed a new system that tracked and reported the ESD failure occurrences of wrist straps and work surfaces that we appreciated just how far-reaching the benefits could be.
The system was installed by placing ground monitors at workstations and wiring them to a computer through a satellite box. The monitors continually checked the grounded work surface and the operator’s wrist strap. Each satellite box could handle up to 16 monitors and several satellites could be daisy-chained to form a system.
The software displayed a visual image of the area and what was taking place at each station. By the change in the color of the workstation picture, we could see if a high, low or good condition existed. The data for that workstation was then stored to an ASCII file for later tabulation.
We began tabulating the data after the first seven days. Operator connect time was analyzed for total connect time, frequency of connect, maximum duration of connect, average duration of connect, high fault count (>10 MW series resistance from ground to operator to ground) and low fault count (<2.8 MW series resistance from ground to operator to ground). Fault duration was too short to be significant.
A total of 1,055 operator connects were recorded. There were 42 high faults or open connections and 97 low faults or grounds recorded. The high faults were randomly distributed across the time period. They indicated that, with our temperature and humidity, we could expect about a 4% contact problem that usually can be reduced by applying ESD-control lotion. We also expected the faults to decrease as the operators became accustomed to the system.
The low faults were more difficult to resolve. Instead of a random distribution, we found that 61 of 97 events occurred between 10 p.m. Saturday night and 2 a.m. Sunday morning. The area was not staffed during those hours so we had an interesting puzzle on our hands.
Our investigation led to an engineer who was working on a development project during this time. His favorite wrist strap, old and extremely battered, was the culprit. A new wrist strap brought the next week’s low faults to below 1% randomly distributed across the time period.
An unforeseen benefit of the monitor system was insight into station utilization across all shifts—with minimal effort for the engineering staff. Analysis of test systems showed a surprising delta in the average duration of operator connect time. The difference, about 4 minutes, was attributed to poorly planned changes in station layout to accommodate new handling mechanisms. The layout caused the operator to unplug several times each shift to perform peripheral tasks, which decreased productivity.
Other utilization information showed that one packaging station was used only 10 times during the week, and two QA stations were used 71 times but for a total of 9.2 hours. A strong argument was made to eliminate or combine the three with no significant loss in efficiency and at a net savings of 36 ft2 of floor space.
Historically, ESD has been a hidden cost of factory operation. Having a system that can supply the engineering department with information to solve ESD issues and give the production manager information to reduce cost is very appealing.
In today’s manufacturing, lost revenue due to ESD failures, low workstation utilization or loss of floor space impacts delivery and customer satisfaction. A production manager is always under pressure to reduce costs, be more productive and reduce floor space. Since this system provides data for both worlds, it was easy to get support for continuing the project.
The system is new and we still have a lot of work to do, but tools like this will make the task more successful. Everyone needs tools, but multifunctional tools like this do the job better.
About the Authors
Michael R. Hoogstra is a Senior Electronic Technician for the Advanced Power Group in the Analog Integrated Circuits Division at Motorola. He has a degree in electrical and electronic technology. Motorola, Advanced Power and Computer Peripherals Operation, Analog IC Division, 2100 E. Elliot Rd., Tempe, AZ 85284, (602) 413-3957.
Steve Nosek is a Test Engineer with Motorola ASIC. He has been in electronics for 35 years, and has been addressing ESD problems since 1972. Motorola, ASIC Final Manufacturing Test, Gate Array Products, Application Specific Integrated Circuits Division, 1300 N. Alma School Rd., Chandler, AZ 85224, (602) 814-5439.
Copyright 1996 Nelson Publishing Inc.
September 1996
|
|