Pick Performance for SMT Placement Machines

Mr. Black was the President and Co-Founder of Zevatech in 1977 and introduced first Pick and Place System at Nepcon West 1980. Bob is now the President, CEO and Co-Founder of Juki Automation Systems. He is also a Co-Founder of the SMEMA Council of IPC. He serves as a member of SMTA and SEMI..

Georgian Simion is an independent consultant with 20+ years in electronics manufacturing engineering and operations.
Contact me at georgiansimion@yahoo.com.

Mr. Koep is responsible for product planning and technical marketing for the Preform Products at Alpha. He is the co-author of several patents in the areas of soldering applications focusing on reflow and alternative methods.

Scott Wischoffer has been in the SMT industry since 1986 with experience in service, training, sales & applications, and marketing.

How is Pick performance quantified? How do you hold a pick and place company to a 99% pick performance? Is this listed on a spec sheet of all pick and place companies?

There are a lot of factors that come in the place with pick performance: good parts, good feeders, good and clean nozzles. Is there a generic board and parts that can be used to quantify pick performance?

While a lot of performance data is claimed here, let me set you straight. It does you no good whatsoever if you can place a bazillion parts per hour with a 99.9999999999% accuracy rate, if you can't keep the machine running at least 4-6 hours non-stop during each 8-hour production shift, in a low-volume, high-mix product scenario as an example.

Losing 4 hours of "available" machine run time every single day is 2,084 hours per year that entire LINE is not running. If you figure a cost-of-doing business for an "average" SMT production line is anywhere from a minimum of perhaps $5000 per shift, (this cost includes everything a typical company might pay for, including utilities, labor, operating permits, building licenses, indirect charges, materials, etc, etc,) then you can see that even a small mom/pop business will see $2000 of lost "production per hour" per shift.

Well, $2000 times two shifts per day is $4000, times 5 days a week is $20,000, times 52 weeks, is at least $1,000,000! Therefore you need to consider just how easily or quickly any pick and place machine can be turned around between each job.

Even for companies that run the same cellphone boards all day, every day the set-up times or conversely, the machine run times are always critical. If your pick and place machine does not have the capability to notify the operator prior to the end of each feeder fill, indicate which feeders are empty and can be pulled and replaced while the machine is still running and then have the machine come back and finish picking from the replenished feeder(s), or cannot indicate when each feeder is no longer needed as the last board of a production lot is built, the time wasted waiting for the entire build to finish can be very, very costly.

So, when considering a new placer, be sure to understand exactly what has to happen during tear down of one job and set-up of the next. I have seen so many people buy a machine based on pick speed and accuracy, but realize afterwards that they just shot themselves in the foot because they need to re-teach each feeder location after the set-up is done, or they can't use cut tapes, or they can't fit enough matrix trays on, or they can't automatically update the MRP inventory using the machine management software, or they can't automatically identify only which feeders that need to come off for one job, and which ones can stay on for the next, or..well, you get my message.