datasheets.com EBN.com EDN.com EETimes.com Embedded.com PlanetAnalog.com TechOnline.com  
Events
UBM Tech
UBM Tech

Design Article

Sensor fusion and MEMS for 10-DoF solutions

Lj. Ristic, Managing Director, Petrov Group

9/3/2012 6:07 AM EDT

Barometric sensing
Let us turn now to barometric sensing. The key part of a barometer is a pressure sensor. The crucial part of every pressure sensor is a diaphragm. There are several variations of MEMS technology on how to make a silicon diaphragm. In part, this is related to underlying principle of operations of pressure sensor. Generally, two basic principles are exploit: piezoresistive effect and capacitive method. In the case of the piezoresistive effect, there is a piezoresitor embedded into a diaphragm. When the diaphragm moves it creates a change in the piezoresistor, which in turn is correlated to pressure. In the case of capacitive sensing, the diaphragm itself represents one of the electrodes of the capacitor, and the change in capacitance is correlated to the change in pressure. Here, we will discuss several examples of pressure sensors that represent the main stream in barometric sensing.


Figure 5: Piezoresitive pressure sensor made by using wet anisotropic etching

Figure 5 shows one of the pressure sensors and its diaphragm. The diaphragm is made of silicon crystal by using wet anisotropic etching of the silicon wafer with the (100) orientation. This approach is known as a bulk-micromachining. The etch rate of the silicon crystallographic planes with (111) orientation is much slower than the surface with (100) orientation which leads to the unique sloped truncated, and pyramidal shape beneath the diaphragm. The piezoresistive resistors are placed at the edge of diaphragm. The MEMS die is completed by bonding the MEMS wafer to the holder wafer whereby a sealed cavity (reference chamber) for absolute pressure sensor is created. This approach is used by Honeywell and Freescale.

Figure 6 shows another pressure sensor with a different diaphragm. In this case, a combination of dry and wet etching with monocrystaline silicon growth, CMP process, and sacrificial etching is used. It leads to the formation of a silicon diaphragm with a sealed cavity at the top of the wafer without the need for wafer-to-wafer bonding. This technique leads to a smaller MEMS die compared to the die made by using wet anisotropic etching. The piezoresitive elements are again placed at the edge of the diaphragm where the stress is the largest. This approach is used by STM – they call the process VENSENS.

It should be pointed out that the final barometer product includes two dies in a single package. One is the pressure sensor MEMS die and the other is the CMOS IC die. The two dies are typically packaged side-by-side. For example, STM and Freescale use the LGA package for their barometers. Naturally, there is a hole in the LGA package to allow ambient pressure to reach the pressure sensor. The CMOS IC is an ASIC that performs signal processing and also provides a digital output that is compatible with I2C and PSI protocols.


Figure 6: Piezoresistive pressure sensor made with a grown-monocrystalline Si-layer

Finally, an example of a pressure sensor based on capacitive sensing is shown in Figure 7. In this case, a surface micromachining is used in combination with a thin polysilicon layer (about 2 um) and sacrificial etching. The sealed cavity and the thin diaphragm that also serves as one of the plates of the capacitor are seated at the top of the MEMS die. This technology is essential for high pressure sensors. The capacitive pressure sensor is at the core of sensing solutions for TPMS (tire pressure monitoring system). An excellent example of this type of product is a Freescale’s TPMS solution that includes a capacitive pressure sensor and CMOS ASIC on a single chip, an RF-transmitter chip for wireless data transmission, and an MCU chip, all in a single package (all together three chips). This product is the darling of the automotive industry and is used in both cars and trucks for monitoring tire pressures.


Figure 7: Capacitive MEMS pressure sensor

Conclusion
The basic aspects of sensor fusion have been described in this paper including the 10-DoF solution. Sensor fusion is a powerful concept that demonstrates how the whole is greater than the sum of its parts. Then, the focus turned to motion, inertial, and barometric sensing since they are the key components of 10-DoF solutions. The select MEMS sensor examples shown here are representatives of real products. They illustrate well that there is no lack of inventiveness when it comes to MEMS technology. Some MEMS solutions are cautious, some are more radical. Either way, they deserve our full attention because the products made by MEMS technology are reliable, and cheap, and they fuel new applications that were unthinkable only a few years ago. One can state with certainty that MEMS technology and MEMS sensors are essential to sensor fusion and the 10-DoF solutions as well as to the next generation of m-DoF solutions.

About the author:

Lj. Ristic is Managing Director at Petrov Group LLC, (Palo Alto, Calif.)

Visit Petrov Group website.




docdivakar

9/7/2012 5:49 PM EDT

The author has to be congratulated for an excellent article on sensor fusion and its future directions.

I also would like to call his attention (since he is local to Silicon Valley) to the event on a closely related topic being held on a Saturday this month (29th):

Next Generation Circuit & Systems, Communication and Sensor Technologies in Mobile Devices

http://ewh.ieee.org/r6/scv/comsoc/

MP Divakar

Sign in to Reply



Hrishik

10/15/2012 12:32 PM EDT

Hello Mr. Ristic,

I would like to appreciate the succinct description of sensor fusion provided in the aforesaid article. But, while working on sensor fusion for the the estimation of velocity of a body frame, I came across some real world challenges, for which I seek advice.
*When the mobile device is mounted on a moving body frame, there are 3 coordinate systems, the geomagnetic earth frame, the frame of device mount and the frame of motion of the body itself. In this case, how can we project, the acceleration vector on to the body frame only?
*Furthermore, from inertial navigation, we can use gyroscopes to lock in inertial vectors in the reference frame. But, how do we account for centripetal forces in this case?

It would be really helpful if the doubts enumerated above could be clarified.

Thank you,
Hrishik Mishra

Sign in to Reply



Please sign in to post comment

Navigate to related information

Datasheets.com Parts Search

185 million searchable parts
(please enter a part number or hit search to begin)