Hi,
I am trying to make a model in Abaqus of concrete column exposure to fire to all sides. I use CDP model for concrete without expansion. I import modul of elasticity in elastic material and stress-(plastic strain-stress in yield/E module) for 20,100,200,...,1100 degrees Celsius.
First step in model is vertical load and second- thermal analysis.
When I don`t import compression damage in CDP, the model convergence, but stress in core of the column is greater than the yielding stress in CDP input data. Maximum stress S33 at 20 degrees in CDP is 30.3MPa and in output plot is 38.3,MPa at 1302 sec of fire.
It is important to note that at a constant temperature of the element of 20, 100 or 500 degrees Celsius, the maximum stresses in the concrete are exactly the same as specified in the material. But once you set the fire impact, the yielding stress is larger than stress of the material.
I can`t explain this. The rough and large mesh has no significant influence on the results.
When i import values in compression damage in CDP, the model didn't convergence and write this errors:
DURING THE CONVERSION FROM CRUSHING TO PLASTIC STRAIN Abaqus FOUND NEGATIVE
AND/OR DECREASING VALUES OF PLASTIC STRAIN. VERIFY THAT THE DEGRADATION DATA
UNDER *CONCRETE COMPRESSION DAMAGE IS CORRECT
DURING THE CONVERSION FROM CRUSHING TO PLASTIC STRAIN Abaqus FOUND NEGATIVE
AND/OR DECREASING VALUES OF PLASTIC STRAIN. VERIFY THAT THE DEGRADATION DATA
UNDER *CONCRETE COMPRESSION DAMAGE IS CORRECT
The analysis ends when reaching maximum stress in the material and do not take into account the falling branch of diagram of concrete.
To accounts the softening part of the compressive stress-stain curve of concrete I made displacement-controlled simulation, but the results are the same (larger stress in the core).
If someone wants to examine the model can write on e-mail: (removed by admin)
Thank you in advance
I am willing to examine your model in order to give you ideas on how to exceed those cited problems. Indeed, for the descending branch of the behavior of concrete. you do not take automatically descendente branch of the global behavior of the structure if the RESOLUTION methode is a method step by step or a rope method. the only method allowing to have this descending branch is the secant method use the secant modulus of the concrete and steel. for more precisions. I am seeing your model and examine it
here is my email
I think that confinement of core concrete result in stress in core of the column greater than your yielding stress in CDP input data. Because CDP model with high dilation angle for concrete in Abaqus able to consider increasing of compressive strength of concrete confined with lateral reinforcements or spirals.
I'm doing fire study on concrete, i want the compression and tension behavior to be temperature dependent, there is only one temperature we can key in right now next to the stress-strain curve, i want to input multiple stress strain curves at different temperatures, any help is much appreciated.
- Badges
- Science topic