• Registered: 19-05-2014
  • Posts: 26

Topic: LinearEdgeLoad

Hi there,
I'm trying to make a working model that uses LinearEdgeLoad on 3d beams.

LinearEdgeLoad 1 loadTimeFunction 1  csType 0 Components 2 0 -100  DOFs 1 3

Is this line correct if I want 0 load at the first node and -100 at the second just for the 3rd dof?
If this is correct then I can't understand why the reactions are completely wrong in the attached model.
Following the code with a debugger the mid gauss point gets a load value of -125, seems the coordinates aren't handled properly at line 237/238 of beam3d.C

Post's attachments

test.in 1.15 kb, 8 downloads since 2018-08-31 

You don't have the permssions to download the attachments of this post.
  • Registered: 25-06-2013
  • Posts: 241

Re: LinearEdgeLoad

Hi,
I think that the input of load should look like in the attached file. The problem is, that you need to specify all te dofs for loads, it is not like boundary condition where you specify only the selected dofs.

Post's attachments

testCorrected.in 1.2 kb, 12 downloads since 2018-09-10 

You don't have the permssions to download the attachments of this post.

3 Reply by fponta (edited by fponta 13-09-2018 22:37:30)

  • Registered: 19-05-2014
  • Posts: 26

Re: LinearEdgeLoad

Dear Martin,
I still get strange results using all the dofs:

    R E A C T I O N S  O U T P U T:
    _______________________________


    Node        1 iDof  1 reaction  0.0000e+00    [bc-id: 2]
    Node        1 iDof  2 reaction  0.0000e+00    [bc-id: 2]
    Node        1 iDof  3 reaction -1.7764e-15    [bc-id: 2]
    Node        1 iDof  4 reaction  0.0000e+00    [bc-id: 2]
    Node        2 iDof  1 reaction  0.0000e+00    [bc-id: 3]
    Node        2 iDof  2 reaction  0.0000e+00    [bc-id: 3]
    Node        2 iDof  3 reaction -7.5000e+01    [bc-id: 3]
    Node        2 iDof  4 reaction  0.0000e+00    [bc-id: 3]

and the result is wrong (I'd expect +100 and +50)

Using

Components 12 0 0 0 0 0 0 0 0 -100 0 0 0

to make it like my model I get the exact results as my model:

    R E A C T I O N S  O U T P U T:
    _______________________________


    Node        1 iDof  1 reaction  0.0000e+00    [bc-id: 2]
    Node        1 iDof  2 reaction  0.0000e+00    [bc-id: 2]
    Node        1 iDof  3 reaction  1.5000e+02    [bc-id: 2]
    Node        1 iDof  4 reaction  0.0000e+00    [bc-id: 2]
    Node        2 iDof  1 reaction  0.0000e+00    [bc-id: 3]
    Node        2 iDof  2 reaction  0.0000e+00    [bc-id: 3]
    Node        2 iDof  3 reaction  2.2500e+02    [bc-id: 3]
    Node        2 iDof  4 reaction  0.0000e+00    [bc-id: 3]

which is also wrong because I'd expect it to be at least the symmetric case..  (I'd expect +50 and +100)
There's something odd with the conversion between local-global coordinates when oofem computes the integration weights

4 Reply by nitramkaroh (edited by nitramkaroh 14-09-2018 16:02:05)

  • Registered: 25-06-2013
  • Posts: 241

Re: LinearEdgeLoad

You are right, there was a bug and your original input was correct.
I just pushed the fix.