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.

transient analysis with harmonic point load

Please login with a confirmed email address before reporting spam

Hi,

I am trying to model a ractangular plate driven by sinusoidal load with very high frequency (approximately

15000 Hz). What I want to get is the time dependent displacement of a certain point on my plate.

After defining a point load (F_z = sin(15000*2*pi*t)) I pick my solver parameters as follows:

Solver Parameters 0:0.0000001:0.001

(The reason why I pick such a small delta_t (0.0000001) is because the period of the applied load is 1/15000 seconds)

I leave the absolute and relative tolerance values as they are, i.e., 0.01 and 0.010 respectively.

After doing the meshing/solving I postprocess to get my time vs z_velocity plot at a certain point. The problem

is both the displacement and velocity plots are choppy and no matter how small I pick the delta_t (time

iteration) I get the same plots. How can that happen? I am expecting to get higher resolution in time as I

decrease the delta_t but that does not happen.

What am I missing? Can it be because of the tolerance values? Or maybe e-7 is too small for an accurate

transient analysis?

Any help is greatly appreciated.

Thanks

1 Reply Last Post Apr 15, 2010, 1:27 a.m. EDT
Ivar KJELBERG COMSOL Multiphysics(r) fan, retired, former "Senior Expert" at CSEM SA (CH)

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Apr 15, 2010, 1:27 a.m. EDT
Hi

Take a look at the "Solve - Solver parametres - Time stepping tab", and check the corresponding doc, I suppose you need to tewak the "Time steps taken by the solver".

On the other side expect long calculations as you are demanding very many points N=max time/resolution*M where M is the number of sub-steps COMSOL needs to follow your time step and to converge (situation dependent) and also "point" constrints are not very clean, they represent a simplification that does not make the analysis of any derivative veryx easy, hence not that good convergence.

Have fun Comsoling
Ivar
Hi Take a look at the "Solve - Solver parametres - Time stepping tab", and check the corresponding doc, I suppose you need to tewak the "Time steps taken by the solver". On the other side expect long calculations as you are demanding very many points N=max time/resolution*M where M is the number of sub-steps COMSOL needs to follow your time step and to converge (situation dependent) and also "point" constrints are not very clean, they represent a simplification that does not make the analysis of any derivative veryx easy, hence not that good convergence. Have fun Comsoling Ivar

Note that while COMSOL employees may participate in the discussion forum, COMSOL® software users who are on-subscription should submit their questions via the Support Center for a more comprehensive response from the Technical Support team.