I'm currently modelling a steel-concrete composite construction with headed studs in ABAQUS/Explicit and want to analyse the load-displacement behaviour of the headed stud.
In the first case I applied a displacement of 15mm and in the second case 10mm. How comes that the load-displacement-curve is different?
The results get more realistic when the applied displacement is close to the displacement where the headed stud fails. Is there a way to get realistic results even if the applied displacement is higher than the failing displacement?
I used the automatic, global incrementation and did not use time and mass scaling.
The material model for concrete is Concrete Damaged Plasticity and for the headed stud ductile damage.
Thank you in advance
May be an effect due to geometric nonlinearity, but impossible to say withot knowing more about the model.
I would suggest to verify the time-step curve and the displacement increment curve of the two simulations and compare them.
Since you are implementing a damage law, if you have larger displacement increment at the beginning of the plasticity pahse, Abaqus is introducing a larger amount of a damage and your material will respond as less stiff.
This is just a hints, but is too hard to evaluate the origin of the effect withouth looking at the code.
Please have a look inside my article - I worked with Ansys Mechanical.
The modelled load-displacement-curve depends on the load steps (load speed) resp. deformation steps (deformation speed). You have to find the converging curve.
Article Modelling and quasistatic simulation of the stiffness degrad...
Just a hint, you need to make sure if you don't apply the loading very fast. Based on the noisy results you get in the plastic part, something fishy is going on? You should check the change in the kinematic energy since you're using explicit. It shouldn't be considerable. Otherwise, you should reduce the rate of loading.
- Badges
- Science topic