B2.1 - Simulation Based Design of Sensors

SENSOR+TEST Conferences 2011
2011-06-07 - 2011-06-09
Proceedings SENSOR 2011
B2 - Sensor Design and Modeling
M. Kaltenbacher - Alpen-Adria University Klagenfurt (Austria)
228 - 233


In most cases, the fabrication of prototypes within the design process of modern sensors is a lengthy and costly task. Therefore, an increasing need for reliable and usable computer modeling tools capable of precisely simulating the multi-field interactions arises. Such appropriate computeraided engineering (CAE) tools offer many possibilities to the design engineer. Arbitrary modification of sensor geometry and selective variation of material parameters are easily performed and the influence on the sensor behavior can be studied immediately. In addition, the simulation provides access to physical quantities that cannot be measured, and simulations strongly support the insight into physical phenomena. Thus, a CAE-based design can tremendously reduce the number of necessary prototypes. However, we want to emphasize that a direct physical control of the sensor design is possible only with the help of experiments, whereas the computer simulation is always based on a model of reality. Therefore, the
quality of the results depends on the suitability of the physical model as well as the material parameters.

In an early state of the design process it is often advantageously to base the modeling for the numerical simulation on a system of ordinary differential equations (ODEs), which uses finite network elements. Therewith, the engineer can check in a quite short time, if the design goals can be approximately achieved. In the second step, one needs mostly a very accurate model, which will result in a system of coupled partial differential equations (PDEs). For the efficient solution of these models we have developed the simulation tool CFS++ (Coupled Field Simulation in C++) based on the Finite Element (FE) method. The program is applicable for simulating capacitive micro-machined transducers, piezoelectric transducers, electromagnetic and magneto-mechanical sensors and actuators. A typical example is the investigation in the total harmonic distortion (THD) of silicon microphones. The nonlinear electrostatic force, acting in such a capacitive, micro-machined device, results in higher harmonics for any monofrequent excitation. Such numerical computations are the key for an advanced design process.