I want to model creep rupture of D9 tubes using Abaqus. From experiment I have the minimum creep rate and rupture life. If I am modelling with the time hardening law then I am getting the creep strain values. But how do I get the rupture life from model.
Hello Romali Biswal ! To analyze the creep-rupture you may apply the following approaches. The simplified approach is: you compute the stress distribution in a steady creep state of the pipe with your time-hardening model. Then find the maximum von Mises equivalent stress and estimate the rupture life from the stress vs. time to rupture data.
A more accurate approach is as follows. Calibrate a creep-damage model from your experimental data. To this end you should apply a damage evolution equation. Then write a usercreep subroutine inside ABAQUS. With this approach you can visualize, how creep damage develops in the tube.
Hello Romali Biswal ! To analyze the creep-rupture you may apply the following approaches. The simplified approach is: you compute the stress distribution in a steady creep state of the pipe with your time-hardening model. Then find the maximum von Mises equivalent stress and estimate the rupture life from the stress vs. time to rupture data.
A more accurate approach is as follows. Calibrate a creep-damage model from your experimental data. To this end you should apply a damage evolution equation. Then write a usercreep subroutine inside ABAQUS. With this approach you can visualize, how creep damage develops in the tube.
Thank you Professor Konstantin..I have tried the first approach and obtained the stress and strain distribution in the pipe. The deformation pattern obtained in the model (a uniform expansion was seen throughout the length except at the ends, as I had given fixed boundary condition) is not matching with the experimental value ( where a highly localized zone had bulged before the rupture).
I calculated the von mises and hydrostatic stress and obtained the equivalent stress and the time to rupture using LMP. But can I get the same data from the model?
If a uniform tube is subjected to the uniform stress state, for example due to internal pressure, then any known creep-damage model would predict a uniform damage growth, since the damage rate is related to the stress and/or creep strain rate. However, one may expect imperfections, for example, as a result of processing of the tube. Heterogeneous distribution of carbides, residual stresses or something like that may be the reason of strain localization you observe in experiment.
Dear Romali,
In my opinion also the first approach explained by Prof. Konstantin should give a reasonable approximation of time to rupture. But as you are telling that the results are not matching with experiment please reconfirm that you have simulated the correct boundary conditions and you are using the correct stress field for you calculations.
Also from your creep analysis as creep strain reaches to 1% global (deformation) and / or 3-5% local may be taken as a crude approximation for the end of life of component.
Regards
hello Romali,
I am having the problem with the Creep ruputure test in Abaqus. I am using Chaboche model to identified the parameter. i have UMAT for it. but i am seeking help to analyse in Abaqus environment.
Can you share some of your inputs regarding this?
thank you and regards,
Anil Patel.
- Badges
- Science topic