Discussion Forum

Hi

if you can use the Parasolid CAD format, that at least is a true 3D code, as IGES is not really fully 3D compatible and might leave "holes" in the 3D representation of complex solids, particularly swept ones.

Normally it works quite ok in COMSOL to draw a sweep line and sweep an Area along, even better if you have the advanced CAD modules of COMSOL, these complex tools comes in "slices" ...

--
Good luck
Ivar

Thanks Ivar, I will try this.
Sorry for late reply, I was not notified in my email that you had answered.

Perhaps you can assist me on another issue.
I have made a poroelastic 3d Square plate with a centered hole (axis of hole is in z-dir., and the extend of the plate is in the xy plane). I use this to model borehole instability of an oil well at approx 2km depth, i.e. analyze the stresses around the borehole and the pore pressure development inside the plate.

In order to simulate the inward stresses acting on the borehole, and the outward pressure from the hole's drilling fluid I have Applied to the plate:

50MPa (compression) in the +y-direction on one edge that is parallel to the x-axis
-50MPa (compression) in the -y-direction on one edge that is parallel to the x-axis
45MPa (compression) in the +x-direction on one edge that is parallel to the y-axis
-45MPa (compression) in the -x-direction on one edge that is parallel to the y-axis

-60MPa (compression) in the z-direction to simulate the wei´gth of the overburden (however I know this will not act when plate is constrained in z-dir)

10MPa as a pressure in the borehole (only pressure -and not boundary load works on This poroelastic model)

In initial values in the poroelasticity (poro) I have inserted a value of 1MPa to the pf -as I assume this is where I apply a pore pressure to the plate?

Boundary conditions:
Spring Foundation on all egdes with a spring constant of x=50MPa to the edges parallel to the y-axis, and y=50MPa to the edges parallel to the x-axis. Leaving all spring constants z=0.

To obtain plane strain condition I have constrained the two large surfaces (parallel to the xy-plane) against displacement in the z-direction.

I have Applied Roller to the borehole as a BC or else the model acts crazy.

No my problem:
I create several diff 3D Plot Groups with surfaces or volumes and plot different stresses.
I then wish to make a 1D graph of e.g. the stress distribution poro.sx along a defined 3D Cut Line which I made inside the plate along +y direction. This Works fine. But what does not Work is that comsol cannot detect the 10MPa which I Applied to the borehole. And at the very wall of the borehole every graph that starts at the borehole wall should detect the 10MPa.
Apparently there is a problem with the BC on the borehole so that the 3D Cut Line can't detect the 10MPa.
Any suggestions please?
Thanks

Hei

Well all Norwegians are not in the Oil industry, as I'm based in Switzerland, my domains are mostly micromechanics, ACDC, RF, optics, mxieofluidics, and small scale MEMS, nano-science and multiple physics therearound ;)

But from what you describe, it looks indeed as if the fluid pressure is not correctly applied to the "structural" porous media, could it be the gauge pressure settings ? (pressure head), I assume COMSOL is using an offset pressure value in PF too as for the other fluid physics, if not it's really that your physics equations are not coupled correctly either a BC is missing or a reference in one of he sub nodes of the physics (I do not have those physics so I do not know if they use the separate "multi-physics" nodes yet, neither the details of their settings). Check carefully the equations tab normally you will see therein which variables apply to what.

I would also suggest, make a simpler model (we call them "toy-models", where we check ONE phenomena at the time, just to be sure we master all the settings, as there are so many ticks to check, I often make a dozen per full model ;) and check how to get the fluid pressure onto the structure.

And do not forget that "support" are the only that really knows all the details under the hood, that might explain misunderstood issues, and they are rather good and really responsive ;)

--
Good luck
Ivar

Hi Ivar.
Thanks for your reply.
I solved it by applying loads as pressures, but it was not straightforward because I did try to alter the Boundary load from a load in N/m2 to a pressure in N/m2, and this did not work. I deleted the Boundary load and instead applied a pressure from darcy's law in the model tree. Now it registers the pressure on the edges of my plate.

Another little Q:
What is the difference between pf and poro.pm?
I have read from the www.comsol.co.in/multiphysics/poroelasticity that pf might be the fluid pressure in the fluid filled pores of a porous rock, and that poro.pm might be the pressure of the solid part (grains) in a rock. Is this true?

There is a difference in magnitude between poro.pm and pf, and I must use the effective stresses for my thesis which I find as Sigma'=Sigma-Biot*pore fluid pressure.
So I need to know if pf is the pore fluid pressure or it is the poro.pm?

Thanks again.

/L

Hei Lasse

Sorry, I do not know exactly the correct answer there, as I do not have (or have not used ;) that module. The best is really to ask directly Support

--
Good luck
Ivar