• Tii
  • Member
  • Offline
  • Registered: 2013-04-15
  • Posts: 14

Topic: Interface Failure Test

Hi,

I modelled a interface failure test very similar to the example NGAP 3D. The model contains just one concrete block and one steel block. In a first load case there is a vertical displacement on the steel plate to have a normal pressure at the interface between the two blocks. In a next step the steel plate is loaded by a horizontal displacement till interface failur will occure. The boundary conditions are choosen equal to your example NGAP so they should be all right. After failure I choosed one node in the middle of the interface and tried to check the failure stress with the Mohr-Coulumb-failure criteria (with the normal pressure sig n and the shear stress tau of the interface). But the results (the shear strenght) are about 1 MPa smaller as they should be after Mohr-Coulomb.
The stiffness-values of the interface are calculated from the stiffness of the concrete Elements (E_c/Element length*10), the friction coefficient is 0.3 and the cohesion 1 MPa. It is very strange that there are the same results if the cohesion is 0 MPa.

Which mistakes could I had make concerning the modelling or the material model? And why is there no influence (or just a very small one) of the cohesion?

Thanks a lot.

Quote Post 1

  • dpryl
  • CC software developer
  • Offline
  • Registered: 2007-07-13
  • Posts: 1,107

Re: Interface Failure Test

Hello Tii, please see the pictures in the Interface Material Dialog and/or read about it in the ATENA Theory Manual. Shortly said, after the interface fails, the cohesion disappears and only dry friction remains.

If you like us to check your model, send it along with your ATENA User ID (WR), as instructed in ATENA Troubleshooting, 2.1.1 I have a problem not listed here.

Regards,

dpryl's Website

Quote Post 2

  • Tii
  • Member
  • Offline
  • Registered: 2013-04-15
  • Posts: 14

Re: Interface Failure Test

Hi,

but in the documentation of your example NGAP you are saying that at the end of the experiment (the interface failed so that the graph in the horizontal load-displacement diagram is constant) the maximal interface shear strength should be tau=c+sig_n*friction-coefficient.
I think your answer above is right. At the end of the experiment it should be just tau=sig_n*friction-coefficient because there is no cohesion any more.
I checked the results. The interface shear stress is going up with increasing horizontal displacement. At one point it suddenly decrease (about 0,6 MPa) and then it stays at this value. So I think this decrease could be the cohesion but it should be 1 MPa (=c).

Quote Post 3

  • dpryl
  • CC software developer
  • Offline
  • Registered: 2007-07-13
  • Posts: 1,107

Re: Interface Failure Test

Are you sure you are looking at the point with max. shear stress on the interface? Once the interface breaks at that point and the contact strats to slip, cohesion is typically lost over a larger area in a single load step.

If you like us to check your model, please send it along with your ATENA User ID (WR).

dpryl's Website

Quote Post 4

  • Tii
  • Member
  • Offline
  • Registered: 2013-04-15
  • Posts: 14

Re: Interface Failure Test

Hi,

thanks for your help. I have another question.
I want to create a diagram with the maximum interface stress (of the interface volume) and the maximum interface normal stress (of the interface volume) at the various timesteps. The best way would be a maxmin monitor for the interface volume but there is not the data attribute interface stress. Is there another possibility to create such a diagram?
I am using GiD postprocessor with which I can create a diagram at any node but the problems are that 1. GiD can not create a value-table which I could edit with excel and 2. it is possible that the maximum interface stress does not occur at the same node in all time steps.

Quote Post 5

  • dpryl
  • CC software developer
  • Offline
  • Registered: 2007-07-13
  • Posts: 1,107

Re: Interface Failure Test

Probably the easiest way is to define a MAXmonitor for "STRESS" on the interface volume in GiD, and before running the analysis, edit the .inp file and change its definition to "INTERFACE_STRESS".

dpryl's Website

Quote Post 6