Low Cycle Fatigue Analysis in ABAQUS using direct cyclic approach

[waqashf]

Hi friends,

Hi Friends,

Considering there are alot of people who are working on low cycle fatigue, i may need your help for one of my numerical model. I constructed a model in ABAQUS 6.10-2 and simulated a quarter model of a pipe that was subjected to a rigid indentor under zero pressure conditions. The indentor was then removed and elastic rebound was calculated. The pipe was then subjected to some pressure (65% SMSY) and then the pressure was ramped to zero and inelastic rebound was calculated. The model worked fine upto this point. After that, i applied 25,000 cycles to the pipe ranging from 36 -72% SMYS followed by 32,000 cycles ranging from 6.5 - 78% SMYS. Accoridng to literature, the pipe should have generated a crack by now, but when i checked my SDEG (stifness degradation) values in output, it reports 0 at all time increments. Furthermore,i checked element status and it was 1(which means element was fully active). I am using an elastic-plastic material model (X-52 pipe)I am unsure what might be wrong with the model. If not failed, the model should have atleast reported some stiffness degradation values but it said 0 at all cases. I am unsure what to do now as i m stucked with this problem for quite a while. I would really appreciate if someone who has worked on LCF in ABAQUS can help me in this regard. It will be highly appreciated. Thanks alot

P.S Does this might have to do with the C1,C2,C3 and C4 material constants used in damage criteria of LCF in abaqus?



Waqas

 

[bxtsafe]

Hi wagashf,

Unfortunately, I'm in a stressful hurry and won't be able to write a big answer... but let's go...

Any fatigue analyses using the S-N or the epsilon-N approach uses a key parameter to account the structure fatigue life analyzed. This is called DAMAGE and it's symbol is the letter D. The structure fracture occurs when D=1. The period when 0<D<1 means that there is a crack growing (but it can also stop if D is stop growing also) in the structure. However, both approaches can not account for stiffness degradation, as it happens in crash analyses. In fatigue analyses, all finite elements remain in the structures contributing for the structure stiffness. The unique fatigue analyses that uses the modification of the mesh is the fatigue analyses by fracture mechanics.

Well... I think your problem is not related to abaqus, but with fatigue theory. And analyses of idented pipes are very complex, hard and risky to do... I know only one guy that is studying a MSc in fatigue and has been studing this problem for 2 years.

Regards,

bxtguard.