Query about explicit solver

[Tinni1]

09Hello,

I am using the Abaqus explicit solver to conduct a post-buckling analysis of a cold-formed steel U section.
The support and the loading points are made of 3D discrete rigid solid elements. The supports and the loading is applied through a reference point attached to the rigid body.
When I submitted the data check, an error message appeared that a mass and rotary inertia must be applied to the reference node attached to the rigid body reference point.
I applied a nominal mass and rotary inertia, both of magnitude 1E-09.
The error was removed.

My query is
Are the magnitude of the mass and rotary inertia have to be exact? Or such a nominal value (e.g., 1E-09)application is OK?

Could you please advise?

 

[FEA way]

It's better to set it to some reasonable values, at least of the same order of magnitude as they would be if those parts were modeled as deformable. It's important for contact.

 

[Tinni1]

Thanks for your response. I need some further clarification and advice, please, on this.

In my model, the only deformable part is the cold-formed steel U section. Therefore, I have defined mass for it through the property module.

As I mentioned, the supports and the loading applicators are made of discrete rigid solid parts (Half-round). I have attached the point mass and the rotary inertia to this support and load applicators' reference point.

Should I calculate each half-round mass and rotary inertia to define a reasonable magnitude?


The pic of my Abaqus model is attached here is:

 

[FEA way]

It would be best to calculate the mass properties of those rigid parts when they are still solid (for discrete rigid parts we use surface geometry) and then assign them as point mass/inertia.

 

[Tinni1]

Thank you. What do you mean that for discrete rigid, we use the surface geometry to calculate mass? Is that something different from normal solid elements in Abaqus?
Does that mean that, I need to calculate the mass of those solid half-rounds as per their geometry?

 

[FEA way]

I'm just talking about defining discrete rigid parts - Abaqus will ask you to turn them into surface geometries so that they can be meshed with discrete rigid elements. But to calculate the mass properties you would need a solid geometry so before turning it into a shell for meshing, evaluate the properties of the original solid parts.

 

[Tinni1]

Understood, thanks!