Applying a Time-Varying Voltage

[Anonymous]

I am simulating an RF MEMS resonator using Autodesk Simulation Multiphysics 2012. The resonator consists of a polysilicon disk suspended above a silicon substrate by a short stem. The disk is surrounded by two electrodes. I would like to apply a time varying voltage to the input electrode in order to determine the electrostatic force acting on the disk (which is biased at 10V). This will then be used to calculate the disk displacement, specifically the expansion and contraction of the disk as a function of time. Different frequency sinusoids will be applied to determine the frequency response of the device which will be used to determine resonance, input and output impedance, etc. Using "Electrostatic Field Strength and Voltage" I am able to apply only a DC voltage. Is there a way to apply a time-varying voltage, a sinusoid for instance? If not, will one of the other analysis types help me move forward? I have used "Natural Frequency (Modal)" analysis to determine the resonant frequency and type of modes available and "Frequency Response" analysis to determine how the disk will respond to a specific force applied at one node/frequency but need a more complete picture. Thanks in advance for any advice! Steve

[S.LI]

You want to see the structure response under an electrical load, don't you?

I will go MES analysis type if I understand you correctly.

1.) apply voltage to all nodes you want;

2.) Analysis parameters setup dialog -> Thermal/Electrical tab, select "model file" for "source of nodal voltages"; and assign it with an appropriate load curve index.

3.) Go to Load curves tab, and edit the load curve shape as you wish (such as sinusoid).

4.) then your model is good to run.

One thing worth mentioning here is that all electrical loads have to follow the same load curve.

[John_Holtz]

Sorry, but the calculation of the electrostatic forces is based on a constant voltage.

If I understand your explanation, you want to apply Voltage = A * sin(frequency * time). I have no experience in this area (MEMS devices,  electrostatics, electricity), so I do not know if it is appropriate to assume that the electrostatic force at the maximum voltage is equal to the electrostatic force when a constant voltage = A is applied. If that were acceptable, then you could calculate the electrostatic forces and use Mechanical Event Simulation (MES) to vary the forces sinusoidally.

The other factor to consider is whether the displacements are large enough that they affect the electrostatic force distribution. If so, then the electrostatic forces vary based on time (known) and displacement (unknown), so the entire process becomes iterative between calculating the electrostatic force and the displacement.

Good luck.

[Joey.X]

Generic electric field analysis can be simplified according to assumptions

(a) Transient electrical conduction

This is a fully transient analysis with time dependent term in governing equations.

(b) AC conduction

This assumes sinusoidal steady state, with converting current continuity equation into frequency domain; the governing equation is simplified to complex quantity, where two Poisson equations are to be solved.

(c) DC conduction

Steady state (or DC current) is assumed, and time dependent term in governing equation is dropped and the governing equation is simplified to Poisson equation.  Refer to ASIM document here

Note that ASIM's electrostatic analysis can do (c) only.  smceee's question appears located in first, in my opinon, it has no workaround because of different governing equations except user don't care the effect of time dependent item. 

[Anonymous]

All,

Thank you for your responses.  Is there any chance that an upcoming software update will address Transient Electrical Conduction?  If not, any thoughts on a product out there that will allow me to import a model from Autodesk and run the analysis?

Also, I am trying to follow the other approach mentioned to see what the results look like.  I completed  Electrostatic simulations for several applied voltages and then used the results to calculate the expected resonator displacement given the applied DC voltage.  I would like to apply these results to my model in a sinusoidal fashion to a specific surface over a certain period of time and determine the resulting displacement, again as a function of time and applied voltage.  As this period of time more closely matches device resonance, the displacement should increase and the mode should closely approximate the mode I am trying to excite from Modal Analysis.  I am curious to see if this is the case approximating the situation as mentioned.

Thanks again for any help you can offer getting this set up using MES or any other analysis mode.

Steve