Learning Forum

[chema.pinero.vilela]

Hello,

I am trying to perform a transient analysis using a modal analysis approach.

The problem involves imposing a displacement profile to excite the base of a stair structure.

When checking the displacement profile at the base, it matches the input profile exactly. However, when I compare the velocity and acceleration profiles obtained at the base, they differ significantly from the ones that should analytically result from differentiating the displacement profile with respect to time.

On the other handa, when applying an acceleration profile instead of a displacement profile, all the acceleration, velocity and displacement profiles at the excited base match closely with the analytically calculated ones.

This occurs while using the same number of steps and time steps for both analyses:

Number of Steps: 3400 (each one 0.5ms)

Time step: 0.25 ms

This provides many integration points for better accuracy.

Could someone explain the reason behind this behaviour? The stresses obtained applying displacement or acceleration profiles are quite similar.

Thank you in advance.

[Erik Kostson]

 

Hi 

 

Do not exactly why that is (make sure to use modal super pos. and base excitation – acc. based is most logic or perhaps common I would imagine – set displacement to relative base – stresses are the most important anyhow ).

 

For another point you do not need 3400 steps 🙂 – just use 1 substep with a global time step of 0.25 ms in one single step that lasts for the total simulation time – see here for diff., between steps and substeps.

https://www.padtinc.com/2011/11/08/you-dont-wanna-step-to-this-breaking-down-loadsteps-and-substeps-in-ansys-mechanical/

We also have a course on dynamics which I would suggest (it is free and on our website here)

All the best

Erik

 

 

[chema.pinero.vilela]

The methodology/normative that I am using for this analysis says that a displacement profile has to be imposed. Then, I have to obtain the acceleration profile in different locations of the model. But these acceleration profiles are quite higher as I said before.

If I impose an acceleration profile at the base, I can obtain acceleration profiles much "logical" in those locations.