Modelling Bolt Connection with Contact SLIDE in Optistruct?

Philipp Link

Hello,

 

I have modelled a bolt connection using a Sliding Contact between bolt and part. Because an errror accoured during the Calculation I have changed the contacts to Freeze Contact and it worked. Which contact is best suited for a bolt connection or do you recommend another modelling approach?

 

Thank you for your help

 

Philipp

 

Thomas Quenaon_21507

Hello Philipp Link,

This depends of what you want to simulate here ?

BR-

Thomas

 

Philipp Link

Hello Thomas,

 

the picture represents my upper mount for a Linear Buckling Analysis of a strut.

I guess the Freeze Contact does not model the physical behaviour...

 

Adriano Koga_20259

Philipp Link said:

Hello Thomas,

 

the picture represents my upper mount for a Linear Buckling Analysis of a strut.

I guess the Freeze Contact does not model the physical behaviour...

 

the problem with sliding contact in this case is that it is frictionless, so it will have all freedom to rotate and might cause some instabilities/ill-coditioning.

 

What kkind of error did you get? Try sharing the .out file or any error messages.

Thomas Quenaon_21507

Philipp Link said:

Hello Thomas,

 

the picture represents my upper mount for a Linear Buckling Analysis of a strut.

I guess the Freeze Contact does not model the physical behaviour...

 

A freeze contact has a linear behavior: nodes in contact are completely bonded.

If you use Sliding Contact, that means that you might want to take into account non-linear beahaviour (contacts).

Which case you want to simulate ?

BR-

Thomas

Philipp Link

I am just interested in the linear buckling of the strut, so am I right to use the Freeze Contact?

 

Thank you!

 

Philipp

Thomas Quenaon_21507

Philipp Link said:

I am just interested in the linear buckling of the strut, so am I right to use the Freeze Contact?

 

Thank you!

 

Philipp

Freeze contact can bring additional virtual stiffness in your model (locally around the contact area) and it will prevent some mouvement that could occur in real application (bolt sliping, gap...)

BR-

Thomas

pstang

It depends on what boundary condition you want to simulate in your buckling analysis.  Assuming your bolt axis is in x-axis and your vertical direction is z-direction.  If your force was applied in negative z on the blue component, then freeze contact on the bolt and blue component would give a fixed end condition.  If sliding contact was used on the bolt and blue component, it would give a pin end condition.  To prevent the blue component to rotate around the bolt, you need a y constraint, may be a y-SPC on the center of blue component top surface.

Philipp Link

Adriano Koga_20259 said:

the problem with sliding contact in this case is that it is frictionless, so it will have all freedom to rotate and might cause some instabilities/ill-coditioning.

 

What kkind of error did you get? Try sharing the .out file or any error messages.

I have received the following Error:

ERROR # 312 FROM SUBROUTINE UKUERR
For static load case 1 the compliance is large or negative.
Buckling load case 2 cannot be solved.

If I change the Contact to Freeze, the Simulation works.