Re: SMT Thru-hole Data Collection

I'm trying to get a feel for what the industry is doing with regards to collecting assembly defect data in the SMT/PWA process. Is it sampling or 100%? What are the points within the process where critical assembly defect data is collected. We run a medium board volume of mixed technology and high product mix. Our current sampling of 3-4 boards per shift is done after Reflow and Wave Solder. Yields are between 80-85%, but that is a representation of the boards sampled and not the process. Any help that would enable us to implement a more effective quality monitoring system would be greatly appreciated.

Regards, Mark Milward

reply »

| I'm trying to get a feel for what the industry is doing with regards to collecting assembly defect data in the SMT/PWA process. Is it sampling or 100%? What are the points within the process where critical assembly defect data is collected. We run a medium board volume of mixed technology and high product mix. Our current sampling of 3-4 boards per shift is done after Reflow and Wave Solder. Yields are between 80-85%, but that is a representation of the boards sampled and not the process. Any help that would enable us to implement a more effective quality monitoring system would be greatly appreciated.

How many of you folks use the "dpmo" metric for defect data collection? | | Regards, | Mark Milward | |

reply »

| I'm trying to get a feel for what the industry is doing with regards to collecting assembly defect data in the SMT/PWA process. Is it sampling or 100%? What are the points within the process where critical assembly defect data is collected. We run a medium board volume of mixed technology and high product mix. Our current sampling of 3-4 boards per shift is done after Reflow and Wave Solder. Yields are between 80-85%, but that is a representation of the boards sampled and not the process. Any help that would enable us to implement a more effective quality monitoring system would be greatly appreciated. | | Regards, | Mark Milward | | Mark, My company has a much more extensive quality data collection system than checking 3 or 4 boards per shift. We feel that quality is our top concern in production so we do a 100% inspection of SMT boards and about 10% on boards that have only axial and radial mounted parts. (no solder, yet) We collect data everywhere! After reflow, axial and radial mounted parts, wave, DC and Function. We keep up with what we call leakage (any errors found after the product leaves the machine mount area) and record it within the machine centers quality data. Our goal is to have less than 200ppm defect ratio against parts mounted. I feel that if you really want to know what is happening within your process you have to collect and document all the data that you can. Hope this helps! Bill

reply »

| I'm trying to get a feel for what the industry is doing with regards to collecting assembly defect data in the SMT/PWA process. Is it sampling or 100%? What are the points within the process where critical assembly defect data is collected. We run a medium board volume of mixed technology and high product mix. Our current sampling of 3-4 boards per shift is done after Reflow and Wave Solder. Yields are between 80-85%, but that is a representation of the boards sampled and not the process. Any help that would enable us to implement a more effective quality monitoring system would be greatly appreciated.

Mark: You sound like work at a "poster child" company for some quality control guru to make call and come-off looking like a genius. You have lousy yields, do virtually no quality control, and couldn't possibly know if your processes are capable or in control. Get help!!! A guru named Wheeler writes a column on statistical process control in "Quality Magazine" each month. Good luck. Dave F | | Regards, | Mark Milward | |

reply »