Compression test using CDP- Stress-strain curve obtained does not follow the defined exper. curve

[Matte StrEng]

Dear all,

I modelled a concrete cylinder (diameter:150mm,height:375mm, fixed at the base) for a compression test and applied 2 mm displacement under 2 sec and 4 sec separately, using both dynamic implicit and explicit methods. The obtained results were the same. However, the problem is that the stress-strain curve of a concrete element does not follow the experimental stress-strain curve defined. I tried different analysis options that included decreasing tolerance by a scale factor of 0.1 in one case, using the CDP model without the compression and tension damage features in another case. Unfortunately, none of them fixed the problem of following the defined stress/strain curve.

Could you please explain what can be the reason for this situation?

The CDP parameters used are as follows:
Dilation Angle=40
Eccentricity= 0.1
fb0/fc0=1.16
K=0.6667
Viscosity Parameter=0
E=20111 MPa
fyield=24.5 (Inelastic Strain=0)
fcu=29.1 (Inelastic Strain=0.000638)
fy=3.6

All the other yield stresses and inelastic strains were defined as explained in the manual.

I really appreciate any help you can provide.

 

[FEA way]

Try with a single element model first. In the Verification Guide, you can find exemplary input files with such tests for various material models. Download one of them and replace the material properties with your ones. Then run the simulation and check outputs.

How and at which location do you measure the stress and strain in postprocessing ?

 

[bkal]

I fully agree with FEA way. I recently did a similar exercise and the best way to test the constitutive model is to use single element tests. Make sure that you provide minimal boundary conditions to avoid multiaxial stress state which would affect your force-displacement corve.