Concrete strain dependent - tension damage isn't correct

[Batzo]

Hi all
I'm trying to model reinforced concrete subjected to blast,
but when i try to run the model, i get an error:
DURING THE CONVERSION FROM CRACKING TO PLASTIC STRAIN Abaqus FOUND NEGATIVE AND/OR DECREASING VALUES OF PLASTIC STRAIN FOR MATERIAL CONCR. VERIFY THAT THE DEGRADATION DATA UNDER *CONCRETE TENSION DAMAGE IS CORRECT

i know what it means, but i can't find my mistake...

https://files.engineering.com/getfi...f7de25ba490b&file=CONCRETE_DAMAGE_TENSION.png

https://files.engineering.com/getfi...44ee-98b8-cb96d377bbac&file=Tension_Graph.png

I used for:
Concrete Tension - exponential function in ACI 318-02
Concrete Compression - Thornfeldt's model (1987)
DIF - CEB-fib Model Code

 

[FEA way]

Can you attach a picture showing the table in Material Editor with this data specified or data lines under *Concrete Tension Damage keyword in the input file ?

 

[Batzo]

Sure. Here:

https://files.engineering.com/getfi...73391e7f8a84&file=30.06.2021_13.40.50_REC.png

https://files.engineering.com/getfi...4709a53f0b9e&file=30.06.2021_13.41.09_REC.png

 

[FEA way]

Actually, if it’s not a problem, it would be best if you attached part of the input file, just with material definition. This way I could check the correctness of the whole material data specification and run some tests using basic single element model to see when the error occurs.

 

[Batzo]

Hi =, only the material prop.:

 

[FEA way]

The material data generally looks fine but the specified tensile damage curves must be incorrect and require thorough check. Here's the equation used by Abaqus to convert the cracking strain to plastic strain:

If the calculated plastic strain is negative / decreasing with growing cracking strain then Abaqus displays the error message mentioned in your first post.

You can find more details in the documentation chapter "Concrete damaged plasticity".

 

[Batzo]

But when I use one of the curves (single graph) it is working

 

[FEA way]

Did you check all of them ? Maybe only selected ones cause the error here. Strain rate dependence used in your case shouldn't be a problem - when tension damage is deleted (only compression one is left) the model passes data check. And both types of damage are defined in the same manner.

 

[Batzo]

When only compression is left, it passes with all strain rates?
I will check each tension with each strain rate. But if it wont work?

 

[FEA way]

Yes, when I deleted the tension damage (while leaving the compression damage without changes) the data check run completed successfully.

For now, make sure that each curve works correctly in the analysis. There’s a good chance that this check will indicate the source of the error.