• Registered: 2012-03-08
  • Posts: 3

Topic: 'crush band' in the concrete stress-strain relation

Hello,

My question regards the 'crush band' in the concrete stress-strain relation. I would like to say that I read the topic from the user 'demexen', but it didn't answer my question.

I am using the SBeta material where you can define the 'critical compressive displacement' w_d at the tab 'Compressive' in the material Edit menu. The default value is -5E-4 m. At the ATENA program documentation part 1 theory, according to section 2.1.2.5 you are able to correspond the critical displacement to the limit compressive strain, formula 2.16: epsilon_d = epsilon_c + w_d/L_d. For the determination of the band size L_d, refer is made to the following section (2.13). The advantage of this formulation is noted as reduced dependency on finite element mesh and element orientation. When I look at the above formula, it seems to me that the epsilon_d is dependent on mesh size since the only variable is L_d and according to fig. 2-12 of the manual, L_d depends on the mesh size, isn't it?

I am using SBeta material with a compressive strain of -1.750E-03 at compressive strength of -2.330E+01. Now I want to finish the relation with a value of the critical displacement w_d corresponding to a maximal compressive strain of -3.500E-03. Can you explain me how to do this?

Best regards,

Ivar

Quote Post 1

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

Re: 'crush band' in the concrete stress-strain relation

Hello, how have you measured the maximal compressive strain, i.e., what was the length of the measurement gauge? Then, multiply your max.comp.strain by  the measurement distance to get the critical displacement.

Background: similarly to tension where crack opening is a better measure than [mesh-dependent] strain, plastic displacement is used for compression. Strains from experiments differ significantly depending on the measurement base length. In analysis, strains depend on element size (and orientation). This dependency is reduced by using absolute crack openings/compressive displacements and crack/crush band.

dpryl's Website

Quote Post 2

  • Registered: 2012-03-08
  • Posts: 3

Re: 'crush band' in the concrete stress-strain relation

Thanks for your answer. The backgrounds are clear to me now.
Though, I don't understand how the critical displacement of 0.5mm has been determined
I did not measure the compressive strain, but I want to incorporate a compressive strain of -0.35% according to EuroCode2.
My question is now; how is the critical displacement of 0.5mm determined?
In other words:
- What measurement gauge did you use to obtain the critical displacement of 0.5mm?
- And which ultimate compressive strain did you measure at the 0.5mm displacement?

And how can I incorporate a model with an ultimate compressive strain of -0.35%?

Best regards,

Ivar

Quote Post 3

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

Re: 'crush band' in the concrete stress-strain relation

The default value of 0.5mm comes from some old experimental data (one of the references in ATENA Theory - would have to ask a colleague to find the exact one). The experiments have shown the pl. displacement is more invariant than strain measures (which are sensitive to measurement base changes etc.).

If you wish to use the ideally plastic material with limit strain from EC2, you can either

A. select "Softening Modulus" (instead of "Crush Band") at the Compressive tab of the SBETA material dialog, set the softening modulus to 0, and check the 0.35% limit in postprocessing

or

B. define a NLCem2User material. If you go this way, please mind you have to define all the functions (tension, compression, shear, tension-compression interaction, ...).

dpryl's Website

Quote Post 4