Discussion:
[Abaqus] Properties of cohesive zone models on continuum elements
vanda.nur.ilmi@gmail.com [Abaqus]
2017-01-10 14:13:46 UTC
Permalink
Hi,

I see that interfaces are usually modelled with cohesive zone models. If I do want to try modelling it with continuum elements, for example with ductile damage, is there any specific way to "convert" the cohesive properties to the continuum ones?

Thank you in advance.

Regards
vanda.nur.ilmi@gmail.com [Abaqus]
2017-01-23 10:09:41 UTC
Permalink
Hi, let me explain more of my problem.

Let's say in cohesive surface model I have these parameters:

Cohesive behavior: Knn = Kss = Ktt = 5000
Damage initiation: max. nominal stress, normal = shear = 50
Damage evolution: type displacement linear = 0.1

After running a simple pullout simulation of two elements with these cohesive interaction properties in between, I plot the traction-separation graph. There I can recognize these parameters: K as the slope before damage, then after it reaches the max stress the damage occurs, and the graph goes linearly down. It reaches zero after the plastic displacement set for the damage evolution. Correct so far?

For later comparison I also plot the reaction force on the pullout side.

Now I want to create the same results of reaction force, but instead of cohesive interaction properties, I want to use a thin layer of element (let's say 0.002 thickness) with ductile damage material properties.

The ductile damage model requires following parameters: Young's modulus, Poisson's ratio, yield stress, plastic strain, fracture strain, stress triaxiality, strain rate, and displacement at failure for the damage evolution.

My question was how do I define these parameters to get the same results as the cohesive surface. At first I thought of using the K as the modulus and max nominal stress as yield stress (with a very low fracture strain). Yet I couldn't get the same results. I did other trials and errors but still haven't succeeded.

Thank you in advance.

Kind regards

Loading...