# Discussion Forum

## Simulation of Ram bending of a pipe

I want to study effect of ram bending of a pipe (as shown in attched image) using FEA on comsol. This will be a static structural plastic deformation using elastoplastic material properties. If anyone could guide me how to model and approach this problem in comsol, it would be really helpful.

6 Replies Last Post 2019年10月14日 GMT-4 上午2:26

Posted: 4 months ago
Updated: 4 months ago

Check out this model (although it is in 2D):

https://www.comsol.com/model/compression-of-an-elastoplastic-pipe-12691

Regards,
Henrik

Posted: 4 months ago

Check out this model (although it is in 2D):

Regards,
Henrik

Thank you for your reply and feedback. I had actually consulted this model, but as you said its a 2D model. I want study the stresses on extrados and intrados of pipe at bend and wall deformation at bend. I have defined the model as shown in picture attached. Assigned elastoplastic steel properties to the pipe and set large plasticity. Input defined displacement of the ram. Constrained the bobbins as fixed. The simulation terminates with an error. I think the model is not constrained properly and more over its still complex.

Posted: 4 months ago

Hi

As your problem is symmetric I would propose to "cut" it in two with a symmetry plane to restrict any motion along the initial tube axis, and then to be sure you allow a sliding of the contact pair. In this way you have only one contact pair and less convergence issues.

-------------------
Have fun COMSOLing
Sincerely, Ivar

Posted: 4 months ago
Updated: 4 months ago

Actually, it seems to me like there are two symmetry planes, so that the model can be cut in one quarter. In addition to making your model run a factor of 20 or so faster, the symmetry conditions will provide a number of useful constraints as Ivar points out. You will however probably still need to attach a weak spring in the y-direction to the pipe to keep it stable before the contact solution has settled.

Regards,
Henrik

Posted: 4 months ago

Actually, it seems to me like there are two symmetry planes, so that the model can be cut in one quarter. In addition to making your model run a factor of 20 or so faster, the symmetry conditions will provide a number of useful constraints as Ivar points out. You will however probably still need to attach a weak spring in the y-direction to the pipe to keep it stable before the contact solution has settled.

Regards,
Henrik

Thank you for guidance. One symmetry plane is obvious i.e. YZ plane. By second symmetry plane do you mean XY plane (referring to the screenshot of my comsol model).

Posted: 4 months ago
Updated: 4 months ago

Yes, I was referring to the XY plane.

Regards,
Henrik