Convergence of SIF results for a Composite

[Vishakh_Rajendran]

Hello,

I have a query with the Contour integrals capability in ABAQUS.

1. When I request Stress Intensity Factor (SIF) results for an Isotropic material, there is very good convergence in the SIF results between contours

2. I try to request SIF results for a Composite (material defined using 'Engineering constants' in ABAQUS). The SIF results do not converge even after 15 contours.

As can be seen, the SIF results keeps oscillating.



Is there something I need to do or any solution to rectify this please?



Thanks in advance.

Best Regards,

Vishakh Rajendran.

 

[SWComposites]

Images don’t show up.
Post a picture of your model showing the crack and composite layers.
Is the crack between layers?
Or is the frack thru the thickness?
Problem is likely due to the orthotropic layer properties or difference in properties between layers leading to numerical issues.

 

[Vishakh_Rajendran]

Thanks very much. It just seems to be a very bad choice of the material properties in the transverse directions which made the stiffness matrix non-positive definite (maybe).

With acceptable material properties in the transvserse directions, the resutls converge fine.

Best Regards,

Vishakh.

 

[SWComposites]

good to hear you solved it.

 

[FEA way]

made the stiffness matrix non-positive definite (maybe)

Did you get any "negative eigenvalue" warnings in the message file ? They would indicate it.
Abaqus checks and prevents some relationships between the specified constants. Stability conditions for anisotropic elasticity are given in the documentation chapter "Linear Elastic Behavior".

 

[Vishakh_Rajendran]

Hello,

Thank you for the reply. There is no negative eigen values in my case.

Thank you for sharing the documentation chapter. I was working with a demo model and ofcourse, I seem to have made a very bad choice of the material properties. Thank you.