Question on dynamic explicit analysis

[Tinni1]

Hi,

I am performing a dynamic explicit analysis in Abaqus, with contact, to capture the post-buckling behavior of cold-formed steel lipped channel section of 1 m height. I have modeled discrete rigid 3D shell plates at both ends. The boundary condition is top pinned, bottom fixed and applied 250mm displacement with a tabular amplitude for a time period of 2 sec. This time period is more than a 3xtime period of the system(I calculate with a natural frequency analysis first).

I am checking the ratio of kinetic energy to the internal energy of the system, which should be below five percent as per the guide-line when we perform a pseudo-static analysis so that the inertia effect is not dominant.

I am getting the below result for the initial few time steps:

1. For the First time step, Internal energy =0, Kinetic energy=119.6, Can this happen? If this is erroneous, what should I do to mitigate this?
2. For the next 18 time steps the ratio of internal energy to kinetic energy gradually reduces from 22 % to 5%, and for the rest of the time steps, it is less than 5%.-Is this result acceptable, or do I need to increase the time period by more than 2 sec to reduce the inertia effect?

Could anyone help with this?

Thanks in advance!

 

[FEA way]

Do you use mass scaling in this analysis ? Generally speaking, kinetic energy might be high at the initial stage of the analysis when the material is moving without developing significant deformation. The use of a smooth step amplitude can reduce this effect. Also, take a look at the plot of these energies for the whole analysis and try to determine whether it looks fine globally or not.

 

[Tinni1]

Thanks for your response.

I did not use mass scaling, as I have used a reasonably, smaller element size (2mm for the column, 5mm for the discrete rigid plate).

I have plotted the internal energy and kinetic energy of the whole system as attached. Looks like, they are O.K. But could you also advise?

The answer to my first question is not clear to me,
For the First time step, Internal energy =0, Kinetic energy=119.6, Can this happen? If you could clarify, please?

Also, please let me know how do I use a smooth step amplitude?
Presently, I have used amplitude as below:

1. Time-->0, amplitude---->0
2. Time---->2, amplitude---->1

Thanks,

 

[FEA way]

This plot looks fine to me, kinetic energy is visible only at the beginning of the analysis and then behaves as expected. I wouldn’t worry too much about this initial peak but you can take a closer look at other global energies and examine the remaining results and the deformed shape as well as part of a standard verification. This way you may notice what causes the initial peak of ALLKE.

Smooth step amplitude is defined in the same manner as the default tabular one but you have to choose this type in CAE or add proper parameter value to the input file.

 

[Tinni1]

Great! many thanks!