Mesh error

[curiousinvite]

I was trying to test a hemisphere of 2 layer each of 50mm thickness made of different material for impact load and pressure. I received the above error after submitting the job. I modelled the shell(hemisphere) as solid 3d deformable. I think the error is due to meshing, can anybody help me with it ?

 

[FEA way]

Those are just warnings. What are the error messages ?

 

[curiousinvite]

this is the error message

 

[curiousinvite]

I have also attached my file.

 

[FEA way]

The mesh is way too coarse (especially in the thickness direction) and parts are thin-walled so you should also consider using a different type of elements (e.g. continuum shell).

 

[curiousinvite]

Thank you for the response. I have tried using shell but again I received an error message. I have attached my file. In the error it says"Excessive rotation of nodes in node set ErrNodeExcessRotation-Step1."

 

[FEA way]

This new shell mesh should also be refined, the elements on both surfaces are too large.

 

[curiousinvite]

I am new to abaqus so I am having a lot of difficulties. It shows element distortion error when i try to make the mesh size small at impact location. I did managed to make my mesh size resonably small for the solution. But, there is variation in my result if i change my time duration why does it affect my result so much. And, what should the distance between the surface of impact and impact object ?

 

[FEA way]

The time period in dynamic analyses corresponds to physical time so it's very important. For example, your projectile will travel a different distance depending on the time period of the step.

The standard approach is to place the impactor right next to the target since there's no need to simulate its whole travel to the target - there's no air resistance anyway.

 

[curiousinvite]

In my case I kept the time period as 0.05 and space between the target surface and the impact object is 100mm and velocity of the object is 22m/s. 0.05 time period is enough for impact. I changed my time period to 0.1 or 0.01 which is also enough for the impact but why do i get different stress and deflection values for different time period?

 

[FEA way]

Can you share that new file (the one that solves without errors but gives different results when time period is changed) ?

 

[curiousinvite]

I have attached my file .

 

[FEA way]

When I submit the analysis for the "shell" model it still doesn’t complete. Is that the right file ?

 

[curiousinvite]

Thank you so much for all the replies. That's the file. I wanted to ask you one more thing; Do we need to create a separate static step for pressure and a dynamic step for impact, or we can create a dynamic step for both pressure and impact?

 

[FEA way]

Abaqus/Standard might be more efficient for preload steps but you can't combine it with Abaqus/Explicit in a single analysis. You would have to use the import procedure for that.