Learning Forum

[Amin1372]

Hello ,

I am doing a project that I want to see the reaction force of a part of my model after applying sinusoidal force with different frequencies and constant magnitude. I was wondering if you could help me out and tell me which method ( harmonic response or transient structural) I should use.

 

Thanks,

Amin

[peteroznewman]

Sinusoidal loads over a range of frequencies is called Harmonic Response.

There are two ways of solving it. Direct (stand alone), where all the nodes are computed, or Modal Superposition (MSUP), where a Modal Analysis solution feeds into the Setup cell of Harmonic Response and the modes are used to compute the response. Modal Superposition usually takes less time to solve, but you have to make sure to include a sufficient number of modes to have an accurate solution.

Harmonic Response is a linear analysis, so you are limited to linear materials, linear contact, small displacement assumption. If you need anything to be nonlinear, then you have to go with Transient Structural, but that is a Time-History simulation not a Frequency-based analysis, so you have to input the periodic load as time-history and post-process the results to extract frequencies.

 

[Amin1372]

Hello Peter,

 Thanks for your help.  So it seems I need to use Transient Structural because of friction between some parts which makes the model nonlinear. I would appreciate if could help me out with my question:

[Amin1372]

I wanted to find the reaction force of some parts when I apply a 0.3N force with two frequencies of 30 and 120 Hz.

 

[peteroznewman]

I recommend you build two (or three) linear Harmonic Response models that will bound the range of potential outputs before you go with a nonlinear Transient Structural model.

You say you have frictional sliding motion between some parts.  Build a first model where the parts are bonded and there is no sliding.  Build a second model where there is No Separation Contact so the sliding is frictionless.  The results from each model will be slightly different, and the frictional sliding solution will be somewhere in-between.  If the difference between the two results is small enough to be unimportant, you don't need to go to the trouble of building, solving and post-processing the Transient Structural model.

A third optional model, if it makes sense in your particular case, is to have no contact where the friction occurs.

[Amin1372]

Hello Peter,

Yes that's a good idea. Is it true to define 0.3 N as real , zero as imaginary force and  0 to 30 Hz (or 0 to 120 Hz) for frequency range ? and do I need to feed the Harmonic Response with Modal Analysis? if so, How many modes should I choose to be calculated?

Thanks,

Amin

 

[peteroznewman]

Where did you get 0.3 N of force from?

What gives the force a periodic frequency and what determines the frequency?

If you use a Modal Analysis to precede the Harmonic Analysis, this post has information on how many modes to include.

[Amin1372]

[Amin1372]

Hello Peter,

This is my model which includes human teeth and a device which vibrates and applies force with magnitude of 0.3N and two different frequencies of 30 and 120 Hz. 

[peteroznewman]

Fair enough.