Twin Disc Assembly in Abaqus 6.9 1

[OptiEng]

Hello All,

I am looking to run a contact model between two discs in abaqus 6.9. I have modelled the disc in Solidworks 2009.

It is better to create the assembly in Solidworks or abaqus? Is there any way of importing the assembly into abaqus from solidworks other than parasolid (don't have the licence to do this).

Also when I define one of the disc's to be rigid when I import it, I can not create an assembly instance as I get an error message explaining that "a solid rigid part cannot be instanced in an assembly".

Thanks

OptiEng

 

[rstupplebeen]

If your geometry is easy enough to create is CAE then I would do it there because it's easier to modify a diameter or width in your case. Even semi complex geometry I usually go to SW. I usually start by trying to export as a SAT which is CAE's native format. After that I try STEP and then IGS if geometric errors pop up. There is also an add-on SW translator for CAE that might be of help.

Try importing it as the other type of rigid (discrete or analytical). I can never remember which is which. For your geometry I might consider not making it rigid unless you have steel in contact with plastic or something similar.

I hope this helps.


Rob Stupplebeen

 

[OptiEng]

Hello All,

I am trying to model a shaft which stays fixed in position and resists a disc when I a applied a force in the x direction. However, when I run the analysis, my disc just goes straight through the shaft likes it no even there. Can anyone help?

Thanks

OptiEng

 

[OptiEng]

P.S I have created a surface to surface contact through the interactions menu between the outside diameter of the shaft and inner diameter of the disc, but still have the same problem of the disc floating through the shaft.

 

[rstupplebeen]

Try adaptive stabilization and bringing your initial step size down.

Rob Stupplebeen

 

[OptiEng]

Ok, I will try this, thanks. I am used to modelling individual parts, now I am modelling an assembly of parts... do you have any tips on modelling assemblies, i.e. I might be missing something really major required for interactions in FE. For exmaple, how do they model a crash structure impacting a wall in FE?

Thanks

OptiEng

 

[rstupplebeen]

General rules are difficult but here goes.
1. Stick with the defaults as much as possible.
2. Make parts that will contact in contact at the beginning of analysis if possible (RBM)
3. Switch up master and slave even if it violates 'best practices'
4. Try both node to surface and surface to surface because they have different default settings. N2S has the legacy settings.
5. A trivial amount of friction sometimes helps even in frictionless (?) situations.
6. Adaptive stabilization fixes most initial RBM issues just make sure the energy associated with it does not get too high.
7. Slight wiggling of a model such as mesh size or initial step size sometimes gets a solution to converge.
8. Use tech support they are great and will save you a lot of headache.
I hope this helps.

Rob Stupplebeen

 

[OptiEng]

Great tips, thank-you very much, I will give them a try today.

 

[OptiEng]

I am trying to open up some exmaple problems in the Abaqus example manual 6.9. For the example I am looking at there are input files. E.g. hertzcontact_2d_relem.inp, how do I open these up in abaqus? I have tried a few different ways without any success.. can anyone help?

Thanks

OptiEng

 

[rstupplebeen]

File>Import>Model

Rob Stupplebeen

 

[OptiEng]

Hi,

I am still having issues with the disc moving through the other disc like its invisible. I am not sure how to aply and use adaptive stablization as you suggested above. Any other tips?

Thanks

 

[OptiEng]

Ok, I believe I found this under the ALE Adaptive Mesh Constraints Option. However I get an error message when I run the analysis which says:

"The *adaptive mesh constraint option may not be used in analysis with linear geometry"

Can anyone help with what this means?

Thanks

 

[rstupplebeen]

Adaptive stabilization is in the step definition. You should also have the nonlinear flag checked there too.


I don't think you need to use adaptive remeshing. That is meant for very large deformations.

Can you post your CAE file?

Rob Stupplebeen

 

[OptiEng]

Hi,

I have posted the CAE file at

http://my.engineering.com/pg/file/OptiEng

Any advice would be much appreciated.

OptiEng

 

[OptiEng]

By reducing the deformation scale factor I seemed to solve that problem described on 14 Nov 7.01....

 

[OptiEng]

Plotting and saving images of viewports

I am trying to save images (e.g. JPEG's) of my deformed plots and XY data plots. Anyone know how this is done in ABAQUS 6.9 as I am new to the interface of ABAQUS.

Thanks

Opti Eng

 

[rstupplebeen]

Why not use symmetry instead of BC "Disc_1_BC_2"? That constraint basically makes it so the disk can not get fatter.
Changes
1. Stabilization
2. Nonlinear geometry
3. Default interaction property
Attached is a picture.
I hope this helps.

Rob Stupplebeen

 

[rstupplebeen]

CTRL-C copies the viewport to the clipboard.

Rob Stupplebeen

 

[OptiEng]

Thanks again. Really helpful tips, when you say "symmetry" do you mean just removing the "x" constraint?

 

[rstupplebeen]

Your welcome.
See attached.

Rob Stupplebeen