Discussion Closed This discussion was created more than 6 months ago and has been closed. To start a new discussion with a link back to this one, click here.
Eigen-mode filter
Posted Feb 28, 2013, 9:42 p.m. EST MEMS & Nanotechnology, MEMS & Piezoelectric Devices, Structural Mechanics 1 Reply
Please login with a confirmed email address before reporting spam
Hi All,
I am using Comsol to simulate a mechanical structure. I am trying to find some specific eigen-modes that have ONLY x,y direction strain and no z direction strain. The COMSOL eignmode solution gives me a lot of modes that I am not looking for. As the frequency goes higher, the junk modes become more and more. I am not sure how to perform a filter to filter out the undesired modes.
I am wondering if I can use a x,y direction harmonic excitation to excite the desired mode, but in the frequency domain I would need many frequencies to analysis which wound take a lot of simulation time. Also, the step between two frequencies would ignore my eigenfrequency.
Could any one help me on this? Thank you
I am using Comsol to simulate a mechanical structure. I am trying to find some specific eigen-modes that have ONLY x,y direction strain and no z direction strain. The COMSOL eignmode solution gives me a lot of modes that I am not looking for. As the frequency goes higher, the junk modes become more and more. I am not sure how to perform a filter to filter out the undesired modes.
I am wondering if I can use a x,y direction harmonic excitation to excite the desired mode, but in the frequency domain I would need many frequencies to analysis which wound take a lot of simulation time. Also, the step between two frequencies would ignore my eigenfrequency.
Could any one help me on this? Thank you
1 Reply Last Post Mar 1, 2013, 3:33 a.m. EST
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
Hi
the best is to hit the eigenfrequency solver node (far down in the tree) and change from RMS normalisation to participative mass normalisation. Then solve again.
You will get now 3 new Global variables, their squares are the "relative mass participating in the mode, then you plot the square value of all three in a table and you pick the ones with highest "z" component
Normally if you add up all square of the given values, foe all modes, you arrive to the total mass per direction.
This is a measure of the relative energy in each mode, you should not strictly consider it as a classical "mass" even if given the name
Unfortunately, in the present version of COMSOL the rotations (participative inertias) are not (yet?) considered
--
Good luck
Ivar
the best is to hit the eigenfrequency solver node (far down in the tree) and change from RMS normalisation to participative mass normalisation. Then solve again.
You will get now 3 new Global variables, their squares are the "relative mass participating in the mode, then you plot the square value of all three in a table and you pick the ones with highest "z" component
Normally if you add up all square of the given values, foe all modes, you arrive to the total mass per direction.
This is a measure of the relative energy in each mode, you should not strictly consider it as a classical "mass" even if given the name
Unfortunately, in the present version of COMSOL the rotations (participative inertias) are not (yet?) considered
--
Good luck
Ivar