Learning Forum

[kangjs1420]

Hello.

I am a student studying welding engineering.

I want to develop a basic thermo-mechanical analysis model for FSW.

I tried to develop an analysis model with Transient Structual, but some problems arise.

The model includes two substrates and one tool, and these two substrates are in the form of lap joint.

 

1. First, the tool was rotated without the tool and the substrate being in contact (the tool is located above the top of the base metal 0.1mm), but the radius of the tool continues to increase during the rotation step.

Left: Start step, Right: End step of Total Deformation

The rotation of the tool was performed using remote displacement.

I would appreciate it if you could let me know why this is happening.

 

2. In the time step of the tool moving into the substrate, the analysis does not converge.

Can't we implement the phenomenon of tool digging into the substrate on Transient Structual?

If I can't implement the phenomenon with Transient Structual, I'd appreciate it if you could let me know which method is appropriate to use.

 

I'm not familiar with Ansys because I mainly used software other than Ansys workbench.

I look forward to hearing from you. Thank you.

 

[dlooman]

The apparent increase in the radius of the tool can occur if large deflection is not turned on (nlgeom,on).  Without large deflection the circumferential motion continues tangent to the original node location and develops a radial component.  This effect can also occur with large deflection if the displacements are scaled graphically.  Frictional sliding tends to be hard to converge due to the stick-slip phenomenon.  Have you reviewed Example 28 in the APDL Technology Showcase Example Problems?  It's an FSW simulation.

[kangjs1420]

Thank you for your response.

I've reviewed the example, but it's a pinless FSW form of the butt joint.

The FSW process type I want is a tool with pin, which is a thermo-mechanical analysis of welding for lap joint.

When I activated the large deflection option, there was a problem that the substract element of the location where the tool dug in was over-deformed and did not converge.

[dlooman]

It's probably not possible to model the tool digging into the plates with an implicit solver.  If that's required considered using an explicit code like LS-DYNA.

[kangjs1420]

 

Thank you. I will try it.