Learning Forum

[lucasppgecea]

Hello friends good morning.

I'm modeling a concrete in ANSYS using the SOLID65 element. The APDL command line is listed below:

_____

ET,1,SOLID65

R,1,0,0,0,0,0,0 !real constants

RMORE,0,0,0,0,0,0 !continuation of real constant codes

MP,EX,1,27996

MP,PRXY,1.0.2

MPTEMP,,,,,,,,

MPTEMP, 1.0

KEYOPT,1,7,1

TB,CONC,1,1,9

TBTEMP,22

TBDATA,1.0.15,0.8,2.85.50

TBDATA,9,0.5

TB,MISO,1,1,15.0 !Multilinear Isotropic Model

TBTEMP,22 !Temperature Setting

TBPT,,0,0

TBPT,.0.00025.7

TBPT,,0.0005.13.98

TBPT,.0.00075.20.88

TBPT,,0.001.27.56

TBPT,.0.00125.33.81

TBPT,,0.0015,39.37

TBPT,.0.00175.43.96

TBPT,,0.002.47.34

TBPT,.0.00225.49.35

TBPT.0.0025.50

TBPT,.0.00275.50

TBPT,.0.003.50

TBPT,.0.00325.50

TBPT,.0.0035.50

_____

However, there are no changes when changing the constant 9, which represents the relaxation factor of the concrete after reaching the tensile limit.

Thank you all for the attention.

[Ashish Khemka]

Do you want the 9 constants as indicated below?:


If yes then define TBDATA as:
TBDATA,,1,0,15,0,0.8,2
TBDATA,,0.85,50,0.5,,,
Regards Ashish Khemka

[lucasppgecea]

Hello akhemka Thank you very much for the feedback. I tried the command you oriented but was not successful.
I'll try to explain in more detail what I'm looking for:

I made a simple model of a solid structure with direct traction. I am imposing a displacement of 0.023mm which, for the mechanical properties admitted in this material, would reach the ultimate tensile strength of concrete.

Follow the image below:
With this imposed displacement, a fragile rupture occurs, proving that the Tc relaxation factor is not being considered in the tensile concrete model.

Below is a load vs. displacement graph:


According to the ANSYS manual my code should be considering the constant C9 to admit the softening of the concrete after the peak tension in tensile is reached, as you can see in the image below:
For the model being thought of, it is only necessary to input the 5 parameters C1, C2, C3, C4 and C9, as shown below

I'm having a lot of difficulties if the command I'm admitting is showing some failure, especially in keyopt(7) which is what activates C9.

Thank you once again for your attention and the contribution of anyone who can help me in this work.

[Erik Kostson]

Hi
Just a side note, that might be of help, when you study RC structure.
Having worked with some concrete, I would not trust the FE results after the failure load, and in the post cracking region.
Upto failure it (concrete model in solid65) might be OK, but needs of course to be validated/verified, but after that (failure load) it is hard to trust the results. That is fine for most times when designing, since we care often about the failure load and not the post cracking response.

Also for your reference that factor (tensile crack factor) is just there typically to aid convergence, and not really to account for post crack behaviour. This is noted in the help manual under keyopt 7 for solid65:

1--
Include tensile stress relaxation after cracking to help convergence




All the best

Erik

[lucasppgecea]

hello Erik Grateful for the contribution. I am aware of the uncertainties that exist in a numerical model, even if it is an analysis in a fractured state. However, we know very well that concrete has residual stresses, even if small, after reaching the peak tensile strength. It is in this sense that I am working, in a scientific scope, to try to predict the rupture under these conditions, establishing a certain calibration and validating it with experimental studies already carried out.

[dlooman]

Note that some of the periods in the input you included should be commas.
I confirmed your findings. There is a keyopt(7)=2 capability that was never released, but does actually do something. A solution technique I use to analyze post-cracking with solid65 is the following:
/solu
autots,off
nsub,1000 ! Specify a very large number of steps
neqit,1 ! don't try to converge
ncnv,0 ! keep going, even if not converged
outres,all,1 ! keep results history
A nice postprocessing feature of solid65 is the plcrack command. It works well with a vector display, /show,,,1

[lucasppgecea]

hello dlooman
Thanks for the answer.
I confess that I don't understand what these commands do.
Anyway I insert this command line in "solution" but nothing changed the behavior of my model.

[lucasppgecea]

Hello friends
I believe that there must be some additional command in this snippet, so that constant 9 is activated:

MP,EX,1,27996
MP,PRXY,1.0.2
MPTEMP,,,,,,,, MPTEMP, 1.0
KEYOPT,1,7,1 !<<<<<<<<<<<<<<<<<<<<<<<< this

TB,CONC,1,1,9
TBTEMP,22
TBDATA,1.0.15,0.8,2.85.50
TBDATA,9,0.5 !<<<<<<<<<<<<<<<<<<<<<<<< and this

I would greatly appreciate your help from the forum.