Learning Forum

[AmbarNaik13]

Hello,

In my academic project I have to replace weld seam joints by clinching. However I don't have any idea, how to model a clinched joint for static structural simulation. I have read different literature papers but it's difficult to understand and implement them as they are based on other simulation Softwares. The only idea I have at the moment is to design an actual clinch point in CAD, however it will require a lot of computational time. At the moment I want to make a sub-model of a clinched joint and compare it with test results. I have attached a picture of the sub-model. Can anybody give an advice on how to model a clinched joint?

 

 

Regards

Ambar

[peteroznewman]

It will take less time to model in CAD the shape of a clinch than it would be to form the clinch shape from flat metal in a simulation, though that is possible.

https://www.youtube.com/watch?v=JkvXCL6cd7M&vl=en

[AmbarNaik13]

Can you please explain how the joint can be formed in ANSYS? 

[peteroznewman]

Do you mean formed through simulation?  You need a punch and a die much like this discussion recently used.

[AmbarNaik13]

No. Not the forming simulation. I want to actually create a model of a clinched joint, I mean defining the joint between these two sheets in Workbench and then simulate it for tensile and shear load conditions.

[AmbarNaik13]

Hi! I have found a solution which could be close to modelling the joint.

You need a center point or origin of a new co-ordinate system approximately at a distance at which the centroid of the formed geometry lies. 

On each main axis, two point, each in positive and negative direction and having equal distance from the center needs to be created.

These points are used to model springs; one spring in each direction. The stiffness value needs to be obtained from experimental results.

I got a roughly good simulation result till the elastic limit of the joint, however I don't know how one can determine the fracture in this case because as I increase the load, the stress increases without showing the breaking point. I used non linear Structural steel. How can one set up the material model to determine the point at which it will fail?

[peteroznewman]

Another approach is to create the solid geometry in the shape of the finished clench joint.

Then you will get local stress concentration that will act on a nonlinear material model that includes plasticity. If you know the material property: elongation at break, you can use this failure criterion: Total Equivalent Strain > Elongation at Break.

 

[AmbarNaik13]

I will try that and update on further development.

As far as modelling is concerned, I have to build a sub-model in ANSYS which is one of my primary tasks. Modelling the joint in CAD will of course be easy in terms of designing and to parameterize it. 

[AmbarNaik13]

Hi,

I have an update on this topic. To create a submodel of clinched joint which is to be used in a bigger structure, the best recommended method is by using spring connection. One spring required in each direction having stiffness values derived from experimental results of single point clinched joint for shear and tensile tests. There should be two remote points along each axis, 0,05 mm away from the center (co-ordinate origin) in positive and negative direction. These will be the connecting point for the springs. The pinball diameter can be set equal to the diameter of clinched joint.

The calculation of stiffness values obtained from experimental results needs to be as accurate as possible i.e. the selection of force and displacement values. 

The major challenge here is to define the failure criteria.