×

Notice

The forum is in read only mode.
  1. Forum
  3. Public Forums
  5. Finite Element Modelling - Code_Aster
  7. Flexure problem with ASTER

Flexure problem with ASTER

  • Boloni
  • Topic Author
  • Offline
  • Junior Member
  • Junior Member
More
16 years 3 months ago #2589 by Boloni
Flexure problem with ASTER was created by Boloni
Hello,
I'm trying to learn how to use ASTER and seem to be blocked at a simple problem. Therefore I seek help here :)

The model involves a beam of dimensions: 20x1x1m, of steel E=200 GPa clamped at both ends. The load of 5 10^5 Pa is applied uniformly on the upper surface of the beam.

I applied PRESS_REP and FORCE_FACE on a group of faces and the results are identical, but when I compare them to results issued from a commercial software (not sure if I can mention other software name) and a simple analytical model, the results (displacements) are 25% smaller.

Could you please guide me to find where am I using wrong ASTER ?

Thanks in advance,
Francisc
More
16 years 3 months ago #2590 by kwou
Replied by kwou on topic Re:Flexure problem with ASTER
Hi Francisc
do you use the same boundary conditions in the commerial software and Aster? Do you use the same elements? From your post it is not clear what kind of elements you use but I guess you applied 3D elements instead of simple beam elements (because you use surface loads).
Furthermore, the mesh itself and specially the number of elements across the thickness has influence on the results, see e.g:
caelinux.com/CMS/index.php?option=com_jo...;id=2529&catid=4
where a similar problem has been described.
Can you be a bit more specific on element type, mesh and boundary conditions?

Kind regards - kees<br /><br />Post edited by: Kees Wouters, at: 2009/02/11 17:59
Interest: structural mechanics, solar energy (picture at 'my location' shows too little pv panels)

--
kind regards - kees
More
16 years 3 months ago #2591 by kwou
Replied by kwou on topic Re:Flexure problem with ASTER
Sorry - this message should not be here.
Kind regards - kees<br /><br />Post edited by: Kees Wouters, at: 2009/02/11 17:52
Interest: structural mechanics, solar energy (picture at 'my location' shows too little pv panels)

--
kind regards - kees
  • Boloni
  • Topic Author
  • Offline
  • Junior Member
  • Junior Member
More
16 years 3 months ago #2592 by Boloni
Replied by Boloni on topic Re:Flexure problem with ASTER
About the mesh, in the commercial software I use hexahedrons (mapped) elements. In SALOME I defined it using Hexahedron(i,j,k) in 3D, then Qudrangle (mapping) in 2D and Wire discretisaton with Automatic length hypothesis in 1D.
To refine the mesh I adjust the Automatic length bar in 1D.

The BCs in the commercial software, were defined selecting the 2 extreme surfaces (clamped) and imposing 0 displacement for all the components.
In SALOME, I created 2 groups of facets (saw it in a tutorial), for the clampings, then defined in EFICAS the displacement constraints DDL_IMPO for these 2 groups, putting the components to 0.

Thanks for your answer, hope we can find the solution :)
More
16 years 2 months ago #2601 by kwou
Replied by kwou on topic Re:Flexure problem with ASTER
Hi Francisc
Your boundary conditions are a bit overwhelming. Specially because the axial displacement of the beam is restrained I can imagine that the your construction behaves stiffer than expected. The contraction of the elements due to the poisson ratio v may cause this. Do you use v around 0.3? So for testing pls put v to zero and compare the results.
Of course, still better is that you remove the boundary condition in axial direction at one end of the beam. At least that is the way I look at it now.

Pls let us know the results.
Though it remains strange that your commercial software behaves differently.

Kind regard - kees<br /><br />Post edited by: Kees Wouters, at: 2009/02/14 20:24
Interest: structural mechanics, solar energy (picture at 'my location' shows too little pv panels)

--
kind regards - kees
  • Boloni
  • Topic Author
  • Offline
  • Junior Member
  • Junior Member
More
16 years 2 months ago #2615 by Boloni
Replied by Boloni on topic Re:Flexure problem with ASTER
I already use the Possion ratio equal to 0.
As for removing the BC in axial direction, it changes the problem, the deformations shape, theferefore it doesn't solve the problem.
I'll try putting BC on a group of nodes and not a group of surfaces in ASTER. I'll let you know if anything changes.

Edit: I used groups of nodes instead of groups of surfaces to impose the BC for the clamping. It doesn't change the results.<br /><br />Post edited by: Boloni, at: 2009/02/16 16:13
Moderators: catux
  1. Forum
  3. Public Forums
  5. Finite Element Modelling - Code_Aster
  7. Flexure problem with ASTER
Time to create page: 0.172 seconds
Powered by Kunena Forum