Welcome to the forums. Please post in English or French.

You are not logged in.

#1 2017-09-07 14:22:01

DeuxPi360
Member
Registered: 2016-04-01
Posts: 8

Contact LAC, quadratic elements, leopard skin on the contact surfaces

Hello

I'm trying to use the new LAC contact algorithm to use quadratic tetrahedrons (generated by Salome-Meca).

The simple calculation (two block pressed against each other) converges but the stresses on the contact surface don't look good: only a few high stress sectors thus the leopard skin (see picture in the attachment).

Could it be because of nodes close to each other from the master and slave meshes?

I also ran into oscillations during the calculation trials. These could be avoided by altering the mesh finesse.

I'm thankful for any advice to tune the contact settings for the quadratic elements.

Merci
Larry


Attachments:
170906_Forum.rar, Size: 1.83 MiB, Downloads: 301

Offline

#2 2017-09-07 14:23:54

DeuxPi360
Member
Registered: 2016-04-01
Posts: 8

Re: Contact LAC, quadratic elements, leopard skin on the contact surfaces

picture


Attachments:
Screenshot at 2017-09-06 13:26:17.png, Size: 766.13 KiB, Downloads: 439

Offline

#3 2017-09-13 16:42:03

DeuxPi360
Member
Registered: 2016-04-01
Posts: 8

Re: Contact LAC, quadratic elements, leopard skin on the contact surfaces

For anyone struggling with the same problem:

I was able to use the CONTINUE contact method (without LAC) with the quadratic tetrahedrons by creating a linear submesh with the slave faces of each contact pair (one needs to merge the double edges of the resulting lin-quad mesh).

Not sure if the results of such a simulation can be trusted... the linear elements are too stiff.

Offline

#4 2017-09-14 10:30:55

RichardS
Member
From: Munich, Germany
Registered: 2010-09-28
Posts: 563
Website

Re: Contact LAC, quadratic elements, leopard skin on the contact surfaces

Hi,
there should not be any stiffness issues as the surface elements don't carry any stiffness anyways.
The only problem is the resolution of the contact curvature which obviously isn't as accurate any more.

Best,
Richard


Richard Szoeke-Schuller
Product Management
www.simscale.com
We are hiring! https://www.simscale.com/jobs/

Offline

#5 2022-11-24 07:53:36

markuss
Member
From: Leipzig
Registered: 2020-01-20
Posts: 7

Re: Contact LAC, quadratic elements, leopard skin on the contact surfaces

DeuxPi360 wrote:

For anyone struggling with the same problem:

I was able to use the CONTINUE contact method (without LAC) with the quadratic tetrahedrons by creating a linear submesh with the slave faces of each contact pair (one needs to merge the double edges of the resulting lin-quad mesh).

Not sure if the results of such a simulation can be trusted... the linear elements are too stiff.

Hi, how do you define the linear submesh in Smesh/Code Aster for contact analysis? Which kind of element do you use?
Kindly regards
Markus

Offline