• Registered: 2010-11-28
  • Posts: 26

Topic: about interface material model

Dear Sir,

I was trying to model the cohesive zone with interface material model in ATENA 2D. My model was set up like this: the interface was defined by the common line of two macro-elements, and interface line was located at the ahead of the initial crack tip or predefined crack path. I wanted to see the cracking process along the interface under external loading. The cracking process should be reflected by the interface displacement. But the whole interface line was deformed with the same displacement at each step; this was not the result I wanted. I am seeking for your help about correctly using the interface material model.

Regards,
Teresa

Quote Post 1

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

Re: about interface material model

Dear Teresa,
please follow the recommendations from "I have problems when using the Interface (GAP) material on contacts" on our ATENA Known Problems page http://www.cervenka.cz/products/atena/known-problems

However, in the vast majority of cases, the conrete material i much better for modelling cracks compared to interface elements - may I ask what reasons do you have to use interface elements instead?

I recommend to use the NLCem2 material or SBETA. If you have to model a notch, you can do that by including the void in your geometry (effectively replacing one or several elements with empty space).

Regards,

dpryl's Website

Quote Post 2

  • Registered: 2010-11-28
  • Posts: 26

Re: about interface material model

Dear Sir,

I am doing this model to explain the fracture mechanism of concrete under tension and compression but from a micromechanics view. This model is in fact very simplified as it only considers a single initial crack originated from the aggregate-matrix bond inside the concrete specimen which is simplified to have the size of crack space. The initial crack is modeled by an opening; then I want to put an interface material which has the softening property in front of the initial crack. I am just trying to use the interface material model to explain crack growth under tension and compression from nonlinear fracture mechanics. If the smeared crack model is used, it is hard to get a right result and the result should be dependent on the input concrete material properties, but not the result caused by the growth of a single initial crack. The compression case is more complicated than the tension case. But just for uniaxial tension case, the softening process can’t be seen with the use of interface material model. It seems that the whole interface line moves together. So my question is: could the interface material model in ATENA be used for the mentioned purpose? And I have adjusted the parameters of interface material to be numerically feasible. Thank you very much!

Regards,
Teresa

Quote Post 3

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

Re: about interface material model

Dear Teresa, this sounds like either
A. the interface stiffness and strength are so high that the interface displacements are negligeably small compared to other deformations
or
B. the endpoints of the interface line are connected due to some inappropriate geometry modelling, e.g., both sides of the line belong to the same macroelement, or is an endnode shared with a third macroelement in a way that blocks the interface opening.

Please check and if you can exclude these possible reasons, send us your model.

dpryl's Website

Quote Post 4

5 Reply by Teresa (edited by Teresa 2011-04-18 15:45:35)

  • Registered: 2010-11-28
  • Posts: 26

Re: about interface material model

Dear Sir,
Thank you. I think it is the boundary problem. When I put the initial crack at the center of the specimen and applied remote tension to it, the interface line in front of the initial crack moved together. Then I tried to put the tensile stress on the initial crack face and allow rotate at the specimen’s boundary, I obtained the results I wanted: the interface line behaved like the softening fracture process zone of concrete. So the problem may not be the interface material model, but my model problem. This is a micromechanics model, so maybe I should use some tricks to get my results. Thank you again!
Regards,
Teresa

Quote Post 5

  • From: Cervenka
  • Registered: 2007-08-08
  • Posts: 35

Re: about interface material model

Dear Terese,

what you are trying to model, should work. It is difficult to judge what might be wrong in your model, if I do not see it, but one reason might be that the load, which is applied at the ends of your model is too small. You can check the stresses in the interface in the post-processor. If the stress, is lower then your tensile strength, than you should increase the load or lower the tensile strength. If by checking the interface stresses you find they are zero, then there is something wrong in your model.

With best regards

Jan

Quote Post 6