*The text is written in January 2009 for the **http://groups.google.com/group/mor4ansys** group.*

Yongquan Wang, Hualing Chen, Xueming He

System-level modeling and simulation for an electrostaticfeedback microaccelerometer based on equivalent electrical representation

Microsyst Technol (2007) 13:613–624

http://dx.doi.org/10.1007/s00542-006-0361-2

I have browsed the paper above. It is a good application of the nodal approach to develop a compact model for MEMS system level simulation.

My feeling is that Laura’s approach for RF-microresonators should also work in this case. You can compare the method in paper with that described at

http://modelreduction.com/Applications/Electromechanical.html

As usual, each method has advantages and disadvantages. The model in the nodal approach can be parametrized more easily. It should be also easier to introduce nonlinearities. On the other hand, one should develop a compact model by hand without the use of the FEM model.

With model reduction one starts directly with the FEM model and obtains a reduced model automatically. Yet, it is more difficult to preserve parameters and the nonlinear model must be linearized around some operation point.

RSS