Discussion Forum

One possibility is that the input Search for eigenfrequencies around has a value that is far from the expected values.

Thank you very much Henrik for your kind reply. I inserted the "eigenfrequencies around value" that I expected because i experimentally evaluated. But it seems to doesn t work. does it is possible, according your experience, that itis caused by the bad mesh? Or there is something I miss? Thank you very much for the help.

Usually, eigenvalues and corresponding modes (at least the lower ones) are not that sensitive to the mesh, so I doubt that the mesh quality is the culprit here. Can there be a difference betwen the real world boundary conditions, and those in the FE model?

Have you measured more than one natural frequnecy, and if so, do the others match?

Hi Henrik and thank you very much for your replay. Unfortunatelly, I have experimentally evaluated only the first one because I was interested in the "lower" frequency response of the sensor and, actually I am can not do further measurements. Probably you are right, the èroblem is in the boundary conditions. Actually, the structure is very simple, it is a cantilever with a fixed end and the rest free to move. There is a piezo layer betwen two other layer to constitute the sandwich structure, I did not used particular boundary constrain. Can you have some suggestions? Thank you very much again.

Without getting into details- that means, details about the experiment and looking at the .mph file...it is impossible to do more than make guesses. Probably a long talk with someone more experienced...who will ask lots of questions about what and how you are doing things...will eventually lead to a solution.

But...having done a lot of experiments myself, some on similar structures...I can say the following: When experiment and theory do not agree, one or both are wrong.

Yes, i agree with you Dave and, unfortunatelly I really think that I am wrong in the FE simulation!!!