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.

Impulse response of a plate_Time domain problem

Please login with a confirmed email address before reporting spam

Hello

I have problem with finding the impulse response of the system.

I have a very simple rectangular geometry and I want to find the impulse response of the system in different points of the plate (both point response and transfer response).

For this reason I have defined a point load in point A and need to have displacement as a function of time in different points on the plate. The problem is that I do not know how to define an impulse force. I thought of using a very narrow gaussian pulse or a step function (and do derivation after solving the problem), but it is not possible to use these functions as a point load.

Can you please help me with this problem.
Thank you.

1 Reply Last Post 04.11.2010, 02:59 GMT-4
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 04.11.2010, 02:59 GMT-4
Hi

normally one should try to keep BC's (boundary conditions) and in particular loads only on the two highest dimensions of a model, that means that points loads are not "good" mainly because they introduce singularities and the model is bad or wrong particularly in the vicinity of the "point". But this does not mean that it's not possible.

one way around is to define a "small" region around a point (edge in 2D or boundary in 3D) and apply the load there.

second is then the impulse. a Dirac is not appreciated by any solver and will give difficulties or not converge at all in time dependent analysis, so you should "smooth it, with a gaussian or a derivable step up combined shortly thereafter with a step down, there are the Heavysides function, and the V4 step or pulse function include the Heaviside functions to ramp up/down.

last you could consider a harmonic response and pass entirely in the frequency domain, there are a few examples in the doc (the doc of 4.1 is better, but for the aaproach the 3.5 doc still remains valid apart from the GUI useage that has changed

--
Good luck
Ivar
Hi normally one should try to keep BC's (boundary conditions) and in particular loads only on the two highest dimensions of a model, that means that points loads are not "good" mainly because they introduce singularities and the model is bad or wrong particularly in the vicinity of the "point". But this does not mean that it's not possible. one way around is to define a "small" region around a point (edge in 2D or boundary in 3D) and apply the load there. second is then the impulse. a Dirac is not appreciated by any solver and will give difficulties or not converge at all in time dependent analysis, so you should "smooth it, with a gaussian or a derivable step up combined shortly thereafter with a step down, there are the Heavysides function, and the V4 step or pulse function include the Heaviside functions to ramp up/down. last you could consider a harmonic response and pass entirely in the frequency domain, there are a few examples in the doc (the doc of 4.1 is better, but for the aaproach the 3.5 doc still remains valid apart from the GUI useage that has changed -- Good luck 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.