News

Dr. Jan Michiels joined Assembleon (assembleon.com) in September 2007, after spending the bulk of his career at Akzo and Unilever, where he gained experience in everything from telephony and cable TV to polyester yarns, construction materials and PCBs. Educated in Eindhoven as an analytical chemist, Michiels describes himself as a “detail-oriented manager with a good memory for structures.” That may be why, then, he believes the future of flexible manufacturing lies not in a single machine, but in integrated solutions. He was interviewed by Circuits Assembly’s Mike Buetow.

CA: Traditionally, companies procuring a new line start with the placement machine. Do you see that changing, and how in your estimation have customer evaluation trends evolved over the past decade?

JM: The placement machine is the heart of the assembly process, and it looks like it will stay that way. The pick-and-place cycle is the basic action of assembly – it primarily determines both the first-pass yield and the cost of placement for an electronics manufacturer. Without accurate placement, the whole process falls down. Other equipment is, of course, as important to board quality – if there’s insufficient solder paste on a board, components could slide off no matter how accurately they’ve been placed.

For the pick-and-place cycle, tolerances have to be down to a maximum of 50 µm for 01005 components and even 20 µm for large ICs. Each individual component has to be placed accurately, and that means checking each of over 120,000 placement cycles per hour. That puts phenomenal demands on, for example, the vision system. With soldering, on the other hand, all components go through the machine in a single pass. Because of this, the placement machine determines the overall process capability. Customers come knowing their specification limits. Process control limits must be well inside these to give zero defects – then even some drift gives process engineers time to act and find the special cause that has led to the drift. If you’re placing just inside the specification limits, any drift will soon mean rejects. Rejects tend to come in bunches – once the process is out of control, you’ll get large numbers of faulty boards spewing from your production line.

To continually improve processes, we are generally more tied into customers’ overall development programs. Customer evaluation, therefore, tends to start much earlier than it used to. With most of our customers, we have built up a long-term relationship based on trust. We now are extending this to our suppliers, and we are developing closer relationships with them.

CA: In times of excruciating margin pressures, how do you prioritize long-term R&D planning? In other words, how do you ensure funding for significant new platforms – versus modest upgrades – and if faced with the choice, what do you cut first?

JM: We are in the luxurious position of having recently released our new line, and will not need a significant new platform for some time. Our A-Series brought a leap in quality and output, and since then we have been making evolutionary improvements. It comes down to the quality cycle – you gather customer experience with your products, and feed that back to make the next improvements. Important will be shortening the time-to-market of the evolutionary improvements.

Our design principle is modular, and we have robust machines. We have been able to increase machine speeds a couple times with software upgrades since the A-Series was introduced, even on machines already installed.

We are very aware our machines are our customers’ bread-and-butter, so cost, quality and reliability improvements drive our long-term R&D planning. Developing a new platform is never an aim in itself, but will be driven by enabling us and our customers to realize functions not possible with the existing platform. A well-balanced R&D portfolio will have both short-term improvements and longer-term breakthroughs.

CA: Assembléon has had a 21-year arrangement with Yamaha to co-develop and sell machines. How would you characterize the relationship today? And under what circumstances would it possibly expire?

JM: It is the longest such arrangement in the industry, and we are proud of it. We are very happy with our development cooperation; we get a good feel for each other’s customer requirements and can use them to improve our machines.

We have very clear lines with Yamaha; we market Yamaha machines as our M-Series and X-Series (Assembléon’s own machines are the A-Series). Our agreement with Yamaha means we don’t sell these into Asia – apart from the MG-8R, which we sell only as an end-of-line machine because in some cases it is useful as an alternative to our AX-201 for line balancing when mopping up large and odd-form components. The cooperation with Yamaha has been very valuable to us, and we don’t see it expiring any time soon.

CA: Several major placement manufacturers are publicly or privately “for sale.” Can this be good for the customer base?

JM: It depends on how the sale is handled. A sudden sale can be disconcerting both to staff and to customers. We have been looking for a suitable buyer for some time, but our parent company Philips has been very understanding. We are looking for the right buyer. We need someone who knows the importance of long-term relationships, which is the key to the industry.

CA: Is there an emerging market (besides China) Assembléon is betting on for the next five years? And are there any such markets you believe to be “over-hyped?”

JM: Eastern European countries like Romania, Russia and Ukraine have a very promising future. We have had a presence in Eastern Europe for around 15 years, and over 12 years in Russia. Electronics production has expanded steadily over that time, and we have built very strong sales and support there. Local sales support is essential in these regions, particularly because there are so many different cultures. We have built an excellent group of local agents/service providers through the region, and now have our own representative in Russia (Assemrus).

It’s not time to write off manufacturing in America and the rest of Europe, though. We are selling pick-and-place machines here, too. If you have high rework costs, you have to locate in a low-wage economy to make manufacturing pay, but even then quality can still suffer; you can’t build quality into products by selection or rework. In our opinion, if the total supply chain is doing its job well, Europe and the Americas can also keep a healthy manufacturing sector. Indeed, we have recently been seeing trends for electronics manufacturing to return close to the European and American end-user markets because of fast-rising logistics/transport costs.

CA: Assembléon has won the Circuits Assembly Service Excellence Award five times in a row. How do recognition and customer feedback figure into your continuous improvement methods?

JM: Well, customers are the start of any improvement process, and the SEA to us represents high-level feedback about what the industry and our customers want. We have worked hard to make the pick-and-place cycle a “closed loop”; our parallel placement design means we can check component location from pick right through to place. We are also closing the wider business “loop” to feed back customer requirements, so we can improve machine design and support.

The most significant contribution to our winning the award, we believe, is our Installed Base Solutions program. This separates the three basic services we provide, and we tailor them to each customer. Performance Services optimize the output, functions and flexibility of the machines. Knowledge Services give training and support tools for running and maintaining production lines effectively and efficiently. Technical Services help reduce running costs, minimize unscheduled downtime and maximize efficiency.

Central to our approach is not just to sell equipment that gives our customers a competitive advantage; we also ensure our customers stay competitive into the future. Where technically possible, we can even help extend the economic lifecycle of our customers’ whole SMT production lines.

Finally, I mentioned the recently increasing logistics/transport costs. The energy costs of manufacturing electronic equipment are increasing more widely, and the equipment’s environmental effects becoming more significant. We have made lifecycle calculations that show the most significant environmental effect of a pick-and-place machine is the energy it consumes over its life. So, we have for some time been working to give our machines the lowest possible energy consumption per output. The low accelerations and decelerations that control our placement process also significantly reduce energy consumption. The motors don’t have to make sudden movements and have lower masses to drive, with more precise vacuum control also reducing energy use. Our machines are compact, which reduces the factory floor space that has to be treated by heating, ventilation and air conditioning systems. And low defects per million rates mean much less scrap, and a much smaller rework department for customers to power.