Comments

• disla

  March 6

  All geometries have initial penetration.
  Some contact definitions are solved as an interference fit. take a look at the adjust parameter.
  
• Fatmac

  March 6

  This issue has confused me a lot in the past - now my work flow for any contact analysis includes an initial run with no loads applied to see if there are any odd interferences. In this case I see the interference here. The odd part of it is that when I use view cracks I see this I have seen this before where the "open cracks" view suggest there is no interference but the mesh actually has some overlaps. I have also seen scenarios where the .step files have no interference but after meshing they do

  Interfernece.jpg 162.9K

  InterferneceB.jpg 93K
• kuhl

  March 7

  Thanks for the advice.
  From the information in the manual I thought that interferences or gaps are dealt with without causing stress, because it says:
  "Elastic, CCX solver:
  This creates elastic springs between the surfaces. As with Tie, the nodes of the slave surface are
  moved in a strain-free way so they lie on the master surface."
• Victor

  March 13

  I've seen this before. It seems to be common for elastic contact on curved surfaces with incompatible meshes, even when there's no interference. I wonder if it might be a bug in CCX.
• kuhl

  March 17

  I had another look at it. It seems as though the block is pulled inwards while the rod is pulled outwards, independet of the initial gap (1: slave pojected onto master, 2: 0,1mm gap, 2: 0,1mm interference)
  
  
  For the rectangluar rod, the slave node in the corner does not seem to know where to move if there is an initial gap.
  
  
  
  
  
  it moves to the top, but not to the left...
  
  I cant really make sense of it. Any ideas?