Solid65 Properties after Cracking

[Researcher123]

Once the element reaches the limit in tension and "cracks" does it have a strain softening portion of the curve or does the element immediately stop providing force in the direction normal to the crack?

In other words, after cracking, does the Coefficient of Elasticity immediately go to zero, or does it slowly decrease?

Thanks!

Will
Rose-Hulman Institute of Tech.

 

[RahulLeslieAlexander]

As far as I know, it goes immediatly to zero. I have learned from papers that in ABAQUS, in goes to zero in a (straight) ramp. Of course, the actual property curves to zero.

Regards,
Leslie

 

[RahulLeslieAlexander]

Well... I'm sorry that I'm backtracking from my view on ABAQUS. I'm not sure about the feature in ABAQUS. I've used only ANSYS.

Regards,
Leslie

 

[ansysfreak]

Hi

Yes, after cracking the stress strain relation in a certain dir goes to 0. This is why this element has so many convergence problems. Take a look at figure 14.39 in the ANSYS help.
There is a keyopt(7)=1 that should make to analyse more stable. When you want to use stress relaxation there is a undoc. keyopt(7)=2 (it works)

Regards
Garry

 

[RahulLeslieAlexander]

Dear Garry,

Can you please suggest what to do with:
[1] What to set for Keyopt(3), and what does it mean?
[2] What vakue to give for "Stiffness multiplier for cracked tensile condition" for Keyopt(7)=1

Regards,
Leslie

 

[ansysfreak]

Dear Leslie

[1] ANSYS provides "incompatible" modes" formulation (also referred to as "extra shapes") for modeling bending applications. If your problem is predominantly bulk deformation, then you may choose to turn extra shapes off to reduce CPU/storage requirements and enhance convergence. However, doing so precludes the ability to model any bending
[2] That depends how much relaxation you want:
Figure 14.39: Strength of Cracked Condition
where:
ft = uniaxial tensile cracking stress (input as C3 with TB,CONCR)
Tc = multiplier for amount of tensile stress relaxation (input as C9 with TB,CONCR, defaults to 0.6)

Regards
Garry

 

[omar111]

Dear All,
i simulated a simple concrete beam in ansys as a volume and the element defined as a solid 65 and i simulated the reinforcement as lines and the element defined as link 8. i used a material multilinear (5 points for the stress-strain curve), concrete and elastic linear materials for the concrete and bilinear material for the reinforcement. i applied the loads at the mid of the span but the beam failed very early by cracking as if the reinforcement does not exist however i merged the nodes between the reinf. and the concrete . please help me, i could not find any solution for this problem.

omar

 

[ansysfreak]

Dear Omar

Try using the initial prestess in the link8 element; this is a real constant.

regards