Modelling Rock using Drucker Prager Model 1

[vtmike]

Hi,

Does anyone have tips on modelling rock in ANSYS (structural analysis)? From my research so far, it seems I have to use an extended Drucker Prager model. I am having difficulty in understanding what the basic difference is between using a Drucker Prager model & the standard method of analysing metals? Can someone please explain this concept?


Thanks,
Mike

 

[TERIO]

Being another mechanical, I don't have any experience modelling rock but here is an attempt to explain the difference. Normally metal plasticity (von Mises, Tresca) is assumed to be independent of the hydrostatic stress (pressure, or average of the 3 normal stress components). For a visual, changing the hydrostatic stress (i.e., adding or subtracting a constant from the 3 normal stresses) shifts the locations of Mohr's circles along the sigma axis without changing the diameter of the circles.

Rock and similar materials respond differently in tension and compression (i.e., they can take more load in compressive stress states). Drucker-Prager is similar to von Mises but adds some dependence on the hydrostatic part of the stress.

 

[vtmike]

Thanks Terio.

So if we have two identical geometries and one of them has a Drucker Prager material model applied to it, the values of stress and deformation will be different because of a change in the yield point due to the hydrostatic part of stress?

 

[TERIO]

I wouldn't say that the stress and deformation are necessarily different. For example, a rock cylinder with 1 sq. in area and an identical steel cylinder each loaded with an axial compressive load of 1000 lb each have a stress of 1000 psi in order to satisfy equilibrium.

 

[vtmike]

So stress & deformation values are same for both metal & rock with the same geometry?
Then how does FEA take the Drucker Prager material non-linearity into account while solving the model?

 

[johnhors]

"So stress & deformation values are same for both metal & rock with the same geometry?"

vtmike, I think all that Terio was stating was that 1 square inch of ANY material subjected to 1000 lbf load will have a compressive stress of 1000 psi for equilibrium (assuming the material can withstand that stress). This is independent of material properties.

However, deformation is of course dependent on material properties.

http://www.Roshaz.com

quality, cost effective FEA solutions

 

[TERIO]

Yes. That is all I was trying to point out.

 

[mendinho]

Drucker-Prager is in some way similar to the Mohr-Coulomb criterion but with much more parameters. I will try Mohr-Coulomb (just two parameters) to have a gross number.

As the collegues pointed out, unlike Von Mises or Tresca, it is a pressure-dependent criterion. Rocks, concrete, polymers or foams are pressure-dependent materials with differente behaviour in tension and compression, metals are independent of (spherical or hydrostatic) pressure.

http://NotOnlyBridges.blogspot.com

 

[xex999]

Hi Everyone,

I desparately need to do an analysis which is basically defining a triaxial test on a rock and applying the Drucker-Prager model.

Do you have any D-P UMAT availabe to share?

Thanks a lot for your help in advance