MEMS Continue to Fulfill Their Promise

Freescale pioneered the integration of circuitry with MEMS sensors for high-volume applications in the early 1980s with micro-machined pressure sensors and you are celebrating your 25th anniversary this year. Automotive has been a traditional driver for MEMS technology. Is this still true?

The number of sensors continues to increase in the vehicle and that number is still expected to grow. The airbag application was one of the first drivers for MEMS technology. Occupant safety applications are still the largest volume application for acceleration sensors in the automotive industry today. More recently, there has been the addition of the tire pressure monitoring system, which has a promising demand in volume. Another recent and rapidly growing use for sensors is in the rollover sensing application.

Are other markets providing sufficient demand for new products or technology changes?

The medical market continues to be a driver for MEMS technology for specialized medical applications such as noninvasive and invasive blood pressure monitoring, hospital and critical care beds, dialysis systems, drug delivery, and more. The miniature pressure sensor packaging is fit for sub-module components or disposable units. We are starting to see the consumer market as the next driver for MEMS technology. Products ranging from cell phones and anything handheld to hard disk drive protection can benefit from MEMS technology. One example is in free-fall detection. This monitoring can determine when an object is falling and allows implementation of protection mechanisms for a variety of electronic products.

What issues/demands are you hearing from your customers?

For accelerometers in the consumer market, Freescale is still hearing the device needs to be extremely small with multiple features, including three-axis detection, low current consumption, and a power save mode as well as other features. For pressure sensors in the automotive market for tire pressure monitoring, Freescale is working toward satisfying the demand for a more integrated device that uses low power for the battery life requirement.

For a MEMS manufacturer, integration can be either at the silicon or at the package level. How do you determine which approach is best?

When we start looking at what governed this integration we established that there were three aspects that were very important to that decision. One, is the phase lag between different technologies and the ability to integrate them. If we were to bring forward technologies that integrate both, which is the case in some choices we have made, we clearly have to have significant technology investment up front and must make sure we have the return. On the other hand if we were to exercise system in package integration it would allow faster time to money, better financial returns in some cases. For example, as you integrate more and more functionality, you end up with larger silicon that is governed by the defect density calculations that end up decreasing the overall yield. The third one is the total investment to bring forward an integrated technology. Right now, tire pressure monitors integrate on a single piece of silicon, some analog functionality, early stages of signal conditioning with the transducer itself. We don't always draw the line segregating the transducer from the ASIC.

Is MEMS fulfilling its technology promise?

MEMS devices are generally much smaller than more conventional solutions. The power consumed is typically miniscule. The devices are much more reliable and thus have longer service lives. As more MEMS solutions are placed in consumer applications, I believe the technology will also fulfill its promise just as it has done for decades now in the automotive market.

Demetre Kondylis General Manager, Sensor and Analog Products Division, Transportation & Standard Products Group Freescale Semiconductor, Inc. http://rbi.ims.ca/4408-531