Local surface failure in safety calculation

Dear Vinko, 

Please refine mesh in the «decent looking fail surface», uncheck «Apply strength reduction» option  for all clusters and try to use Max steps more than 100 (eg. 500 or even 1000). Also it is interesting to see the whole model to inspect model size (boundary conditions influence).

Might be interesting:

— Article Safety analysis and displacements
— Webinar Using PLAXIS to Determine Factors of Safety

Hope it helps.

Dear Vinko, 

Please refine mesh in the «decent looking fail surface», uncheck «Apply strength reduction» option  for all clusters and try to use Max steps more than 100 (eg. 500 or even 1000). Also it is interesting to see the whole model to inspect model size (boundary conditions influence).

Might be interesting:

— Article Safety analysis and displacements
— Webinar Using PLAXIS to Determine Factors of Safety

Hope it helps.

Thank you for your suggestions Rushan Gizzatullin.

I have just finished refining mesh, unchecking clusters and performing safety analysis. I'm getting Msf = 8.6 for yet another local fail surface.

Seeing these results made me think i have overestimated strength and stiffness parameters of my soil models. After checking those, I ensured  that they are within reasonable values. 

Slope height is 30 meters and slope angle is 30 degrees. I would appreciate it if you could take some time and look at the model to share your observations. 

https://www.dropbox.com/s/g591p601b9yddme/p11.p2dx.7z?dl=0

Dear Vinko,

It is very difficult to estimate the problem and find the only one reason. The best variant is to submit a service request, and the PLAXIS Support team will help you in detail: https://apps.bentley.com/srmanager/ProductSupport. In my experience, the PLAXIS team has high quality specialists that definitely help you.

What concerns your model, I will suggest you check the geometry firstly.

As I understand, you have imported geometry from dxf. In most cases it is not needed to import points, because your polygons already have correct coordinates.

Boundary interfaces are not needed as well. Probably it is connected to the previous experience from classic PLAXIS 2D versions, where we should create fixities via a special button. In new (from 2D AE) versions boundary fixities are created automatically and interfaces on boundaries are needed only for dynamic calculations.

Mesh quality is very low (10^−9). The reason is in the imported geometry. After deleting unnecessary points and interfaces I suggest you check polygons coordinates. Some polygons do not have common points, so PLAXIS creates very-very small finite elements between them. The best way is to check origin dxf geometry for such collisions or fix it manually in PLAXIS.

There are some other questions:
— model size (model boundaries influence on results or not?)
— water conditions (is it a real condition, that water level is on 26.8 m, so the slope is under the water?)
— soil parameters (why Undrained A; why these soil models; how soil parameters have been obtained, from what lab tests and so on);
— phase sequence (why GL, not K0; why do you deactivate big clusters and unload the soil on Phase 1 and not on Initial phase or do it gradually etc.)

Each model and project in PLAXIS is a complex problem with a lot of variables. Each of these variables should be taken into account by an user. That's why I recommend submitting a service request via https://apps.bentley.com/srmanager/ProductSupport.

Wish you to find a good and right solution =)

Heya, I just wanted to write you a follow-up. It was geometry problem. As you said I had tiny tiny tiny polygons and after removing those and maximizing mesh quality I got much better results. Once again thank you!

Dear Vinko,

Thank you very much for the feedback =)