Deviation of crack 2

[albert1981]

Dear all,

I have to model a plate with a whole and a crack that start from the edge of the plate. How can I see how the crack is going to deviate?
From assembly-crack-I can modified the q vector that is the direction of the crack but how can I get the right direction?

thanks,
Albert

 

[xerf]

You have to request a CONTOUR INTEGRAL computation.

for example, keyword version:
*CONTOUR INTEGRAL, CONTOURS=5, TYPE=K FACTORS, DIRECTION=MTS

For details read:

ABAQUS Analysis User's Manual -> 11.4.2 Contour integral evaluation

In ABAQUS/CAE, you have to create a History Output Request in Step Module.

 

[albert1981]

hi xerf,

but if I moove the crack in the next step towards the new direction how can I see the parabolic path of the crack?
Since now I dealt with linear path propagation but if have to make the deviation how can I be in the parabolic trajectory of the crack?

Albert

 

[xerf]

I am not sure I understand what you're trying to do.

In ABAQUS simulation of crack propagation (if that's what you're talking about) are quite limited and occurs through debonding of previously tied surfaces. Therefore the crack "propagates" along a known path.

q vector describes the virtual crack extension direction which is used for computing stress intensity factors, not the crack propagation.

 

[albert1981]

xerf,

yes what I'm trying to do is the crack propagation.
To model this I was thinking to make different analysis with different mesh all the time but that will take me ages, even with fortran I could do an iterative procedure in order to remesh again from the input file but still is going to be a big effort.
Do you think debondig will be more fast?
Do you have any idea about it?

thanks,
Albert

 

[xerf]

If you are interested in simulating crack propagation you should look at different codes, e.g. Zencrack, FRANC2D / 3D.