Highlights from a panel interview with Trevor Galbraith, Global SMT & Packaging and Dr. Subodh Kulkarni, President & CEO, CyberOptics Corporation

TG: Welcome to our debate this morning on Metrology in the Inspection Process. AOI needs to become data-driven to work alongside SPI and X-ray to deliver measurable feedback in real-time to the placement machine. A panel of global leaders will debate this challenge and chart the course for inspection equipment and its critical role in the smart factory environment. How important is metrology in the inspection process?

SK: Our view is somewhat consistent. We are all seeing similar issues as circuits are getting more complex, the parts are getting smaller. Metrology was always an important factor in inspection, always was there, we always collected data, but customers are demanding more and more of the data, particularly the high-end demanding customers. My view is that in the semiconductor industry, which is where my background comes from, metrology always played a big role in inspection in the semiconductor industry and to some extent as substrates like PCBs - if you open up an iPhone X right now, it's hard to see a PCB in iPhone X and when you are making those kinds of very advanced parts, metrology plays are very important role in inspection.

TG: Given the nature of solder paste, is it possible to make every potential defect a data-driven decision?

SK: The question is not just can every defect be a data-driven decision, because you are dealing with a statistical sampling here. You are looking at trends so it is a collective decision of where the process is, what may be happening with the defects, and what the statistical impact of that is. So I do think AI is going to play a bigger and bigger role and that we definitely see, where the demanding customers want to see trend analysis and more of a macro decision-making process on what do the defects mean and how it impacts me in overall in my yield and other areas, so it isn't a single defect decision.

TG: For the doubters who are thinking Industry 4.0 is marketing hype, a lot of rubbish, which of course it isn't, they are wanting evidence of what potential yield improvement they are going to get.

SK: For the advanced products the yields are not at 95% and in many cases, we find that for the latest memory chips there is a serious shortage in the industry - an acute shortage, new technologies are coming in the memory industry and the yields are not even close to 95% and those cases inspection equipment becomes a critical enabling tool to improve the yields to the 90+%. So these guys are starting at yields much lower than 50% and this is a serious issue at the high-end of memory manufacturing for instance. Inspection is a critical enabler in those cases to drive the yields up, so it's not just a question of 95% going to 96%. In advanced products you are helping processes stabilize themselves to get from 30-40% to 80+%.

TG: Do you think that post-placement AOI is going to become a critical element of the smart factory line?

SK: It already is I would say. Post-placement AOI is a critical component of a smart factory line today.

TG: But many people tend to put AOI post-reflow rather than post-placement.

SK: It depends on the customer and the configuration, but many of the really smart factory lines are using AOI in the post-placement, not necessarily post-reflow. They certainly have post-reflow, that is how AOI started, we all know that historically, but many customers who are wandering the tradeshow floor here are thinking of AOI bigger than just post-reflow.

TG: Thank you for joining us.

CyberOptics Corporation
Technology Leadership. Global Solutions.
Industrial Metrology Solutions »

CyberOptics Corporation published this content on 19 January 2018 and is solely responsible for the information contained herein.
Distributed by Public, unedited and unaltered, on 19 January 2018 18:39:07 UTC.

Original documenthttp://cyberoptics.com/conews/metrology-in-the-inspection-process/

Public permalinkhttp://www.publicnow.com/view/4DA1E7179212E8859D96289EE2512ABF36449FDF