EE Times Europe: In the past year, Micronas has ceased its consumer activities and put all its eggs in one basket: the automotive market. This decision was taken at the climax of the automotive crisis. Has this been an act of despair or of smart sense?

Matthias Bopp: Of course, it looks like an unfavourable timing for a shift in direction. However one has to approach the automotive business with a long-term perspective. The short-term decline we saw last year should not affect our long-term strategy. I was not involved in this decision, after all, I started my term as Micronas' CEO on January 1, 2010. However I am fully committed to this automotive focus, otherwise I would not have accepted this task. Now is the time to shape this basic decision.

EE Times Europe: How did Micronas manage to get through the crisis with just one of its former two supporting legs?

Bopp: The restructuring has strongly affected our 2009 figures and we have cut almost 50 percent of our jobs, we also moved away from our consumer business. In the automotive segment, sales declined by around 35 percent in 2009. This is quite on par with what we see among our competitors, hence we have not been hit harder than others.

We did not lose market share in Hall sensors which represents our core business, and we did not lose any important projects. However, the situation has changed since Q1 2010. We see that some customers are already trying to catch up with their roll-out plans. In some cases, they even brought forward the introduction of new platforms or cost-optimized solutions. This creates new opportunities for us, but also new challenges because we need to speed up some developments and bring some products faster to market.

EE Times Europe: So cost optimization is the driving force for your customers?

Bopp: Costs are always an issue. Even in a recovering market you need to keep an eye on system costs. The specifications of our products, in particular Hall sensors, are very much determined by system costs, in particular when it comes to the more advanced solutions, even for reasons that have nothing to do with the Hall sensor component.

In some cases, higher integration solves the cost problem, but sometimes it is possible to reduce our customers’ costs by choosing a different system approach without affecting our sales or having to accept average selling price (ASP) reductions. Another innovation driver for our customers is Hybrid Electric Drives (HEVs) and, to a growing extend, “green” solutions, for example the reduction of fuel consumption and CO2 emissions. This does not only refer to the powertrain but to many other design aspects of cars, for instance cable looms and data networks, where designers try to use less copper and thus reduce weight.

EE Times Europe: How does this strategic decision affect your semiconductor technology?

Bopp: We have defined a product roadmap to align our technology to our target markets and applications. The technology for our latest geometry node, currently undergoing a qualification process, can be used for all of our three core business segments – Hall sensors, embedded microcontrollers and gas sensors.

Thus we see good synergies across our technology development and our different business segments. This allows us to combine these applications if our customers wish us to do so. Take the Hall sensors, for instance: today, many Hall sensors contain an integrated microcontroller. We market this as a sensor solution but on top of the controller these solutions frequently also contain non-volatile memory and are designed to operate in an enhanced temperature range.

EE Times Europe: This would mean that Micronas after all is not a single-technology company but is positioned broader than it seems. NV memory, microcontroller, sensors – isn’t this an integration task in the first place?

Bopp: Yes, but we are not in the mainstream market for non-volatile memory. We do not compete against vendors focusing on, for instance, flash devices. We regard this as an embedded solution. But yes, we have to integrate sensors, microcontrollers and memory within one solution. And of course we focus on segments and applications where we are particularly strong. To give you an example, traditionally we had a rather substantial business with dashboard controllers. To further develop these devices we would need certain technologies and IPs. But we do not see our future strengths in this 16- and 32-bit controller market segment.

For this reason, we will cease the respective activities and focus on embedded 8-bit controllers. Here we intend to activate synergies by combining sensors, microcontrollers and driver circuits for motors, butterfly valves, displays and the like. These designs will deliver embedded 8-bit controllers optimized for the specific requirements of a car environment. For instance they could be supplied directly by the on-board voltage without requiring additional power management. They could be integrated directly into a LIN bus and provide interfaces to sensors and actuators, such as brushless motors used to implement a power window or a seat adjuster.

EE Times Europe: How does Micronas position itself with regard to semiconductor production?

Bopp: Having focused on automotive we manufacture about 90 percent of our production in-house. Our fab in Freiburg (Germany) embraces two modules – one for 6” wafers and one for 8” wafers. A large part of the production today is 6” since the 8” line has been used for consumer products in the first place. In terms of geometries, we mainly produce at the 0.8µm and 1.2µm nodes; currently we are in the process of qualifying the 0.45µm node for automotive on both 6” and 8” lines. We design all our new products on this node. Making the transition to 0.45µm as well as moving to 8” wafers will enable us to achieve a much higher productivity.

EE Times Europe: Is fab lite or fabless a topic worth to consider?

Bopp: We have an eye on this topic since we have to think beyond the current geometry node. There is no dogma to manufacture everything in-house. We have a good set-up for the mid-term, but one needs to look five or six years ahead. However, our customers are usually very sensitive when it comes to the question of production outsourcing, in particular now as some suppliers run into bottlenecks. Because we have full control over all the production parameters, we are in an excellent position. We also have the back-end in-house in our facilities in Freiburg and in Glenrothes (Scotland).

EE Times Europe: In 2009, Micronas and the Freiburg University have jointly integrated a fuel cell on chip level. Do you consider commercializing this technology?

Bopp: This is still a research project. Besides the fuel cell, we also conduct research on bio sensors. We have chosen to move the gas sensor technology from research to product design status.

EE Times Europe: How can a semiconductor company grow exclusively with automotive products? How could business look for Micronas?

Bopp: For the automotive industry we expect an average growth rate of 8 percent annually over the next five years. The market for automotive semiconductors will grow by some 12 percent annually. The Hall sensors business, which is our supporting leg with a worldwide market share of 30 percent, should grow 14 percent per year. I would say this forms a good basis for growth.

EE Times Europe: What future developments do you see in this market? What is the shape of innovation?

Bopp: Today we serve more than 50 different automotive applications with Hall sensors alone. This ranges from wear-free seat belt locks to the gas pedal. Modern gear transmissions, in particular dual-clutch transmissions, call for significantly higher sensor quantities. HEVs and electric cars also require many sensors, in classic chassis applications as well as for additional current sensors.

EE Times Europe: You said dual-clutch transmissions. Given the current e-car hype, how long do you believe there will be vehicles around with dual-clutch transmission or with transmission altogether?