Altair RISE

A program to recognize and reward our most engaged community members

Nominate Yourself Now!

Boundary Condition Selection of Interface Surface

Hello community members,

I have a question. When we've seperated a solid partition from the main volume. Which boundary condition should be applied to the interface surface of two solids in Altair CFD 2022.3.

To summarize; I'm first running my model with MRF in Steady simulation. Then projecting my result to initiate Transient simulation. In both simulations, I'm using the same mesh.

For both MRF (Steady) and Rotational mesh motion (Transient) I'm doing the same definition by selecting surrounding volume and the rotating surfaces.

My curiosity is if I'm doing the right thing by leaving the interface surface as default wall.

Thanks in advance!

https://help.altair.com/hwcfdsolvers/acusolve/topics/tutorials/acu/acu_5000_intro_cfd_c.htm

https://help.altair.com/hwcfdsolvers/acusolve/topics/tutorials/acu/acu_5001_intro_cfd_r.htm

Find more posts tagged with

AcuSolve

HyperMesh CFD

Accepted answers

All comments

acupro

Correct - those surfaces between the fluid volumes can be left in the default wall set as in the tutorials. They are written with type = auto_wall, and the auto_wall processing determines whether they should be a wall, or internal (with no BC applied).

FKALAY

Dear Acupro, thanks a lot for your fast response.

Unfortunately, I'm facing with a problem.

The steady simulation finalized with no error and a logical residual/solution ratio. The flow doesn't get interrupted at the interface surface.

However, transient flow finalized in 4th step with an assertion error (Shared error code below - Inflow detected at outflow boundary)

When I check the result (Write initial condition activated). I see that the flow is seperated near solid partition. It doesn't flow through interface surface.

I wonder if I'm doing something wrong while projecting results of steady simulation to transient.

I'm exporting velocity / pressure / kinetic_energy / eddy_viscosity by using below command in AcuSolve CMD prompt. Then I'm inserting .nic files as Nodal Initial Condition to .inp file of Transient Simulation.

acuProj -pb Transient_Name -crd .\MESH.DIR\Transient_Name.crd.B -fpb Steady_Name -fpdir ..\Steady_Name

Error code: *** ASSERTION in Function <lesGmres> File <lesGmres.c>

Thanks in advance,

Regards.

acupro

Can you attach the transient run input file (.inp) and Log file (.Log)?

FKALAY

Unfortunately not, I need to open a ticket for that. What I can do now is that I can share the rows I've inserted for nodal initial conditions.

Is it correct to enter eddy_frequency for tomega.nic?

NIC.txt

acupro

Those NIC commands look correct. Yes - best to open a support ticket (via your professor if using Academic license) if you cannot post those attachments (at least the complete Log file) here.

FKALAY

Nice, so I'm doing the right thing with NIC. What I understand from this is, that I'm exporting results of steady simulation according to node coordinates and projecting those results to initialize my transient simulation. I'm doing this, because restart option doesn't work properly when mesh motion activated.

As next step. I've modified the transient, by defining the interface as "Interface surface" from Motion>Constraints. Good news is, that I've got rid of the inflow at outflow boundary. However, I still have the same ASSERTION error.

It's like a rolling cylinder on the ground. Therefore, volumes are cut by the ground (Defined as moving ground).

Rotating surfaces are connected to ground by very thin surfaces (Defined as default wall).

Now, I've started from scratch. Applied the interface surface to steady simulation too. The background is to have exact same conditions in both steady and transient simulations. Only difference will be the activation of mesh motion in transient.

In the transient simulation with Mesh Motion, should I keep MRF activated? (Rotating surfaces are defined as No Slip Wall).

When it finalizes, I'll project it to transient again and check. Accordingly, I'll open a ticket.

Best Regards

acupro

It's difficult to comment without seeing at least a simplified version of what you're trying - image, etc. It sounds like the motion you want to try is not possible, but without an image cannot say for sure.