Jump to content
danimur

Error fluid boundary condition

Recommended Posts

Hi guys,

I am doing a CFD analysis with AcuSolve on a centrifugal pump. I realized a 2D-mesh with trias of 2mm and the "check elements" doesn't show any kind of problem. For the 3D-mesh I used CFD tetramesh where I selected the carter on the comps of "With BL (fixed)", the section of Inflow and Outflow in the comps  "W/o BL (float)" and the solid impeller in the comps "With BL (float)" selecting "swap only" and "All are fluid volumes". In the BL parameters I have chosen 3 layers with first layer thickness of 0.010 and BL growth rate of 1.200. When I realised this 3D-mesh I can see in the check elements that some tetra elements collapsed and some elements give me problem in the Jacobian; is this important for the convergence of the analysis? How can I solve this problem?

However, if I continue the process, I set in the solver a "steady state" analysis, the model "Spalart allmaras" and in the mesh type I chose "fully specified" so I can insert the rotation of the impeller. For the fluid I added the default "Water", for the Inflow I chose a pressure of 1e5 N/m2 and the viscosity of 1e-6 m2/sec. In the outflow I selected a "Wall" type and for the Impeller I chose "Wall" type putting in a Mesh Motion where I set the rotation around 1 nodes of 314 rad/sec. When I launch it, AcuTail shows: warning: Potential errors in boundary conditions; if I open the warning file appears a list of nodes and it says: WARNING: Missing nodal/element BC for fluid boundary node. How can I solve this problem? 

 

Thank you very much!

Share this post


Link to post
Share on other sites

What is the problem with Jacobians ? what kind of values are there? If there are no mesh intersections generally AcuSolve will be able to solve. Accuracy might suffer due to some bad elements. 

 

It seems there are multiple issues with your setup. Do you have two volumes, one for the moving reference frame and one for the rest of the fluid? 

Also do you have an interface surface between these two volume? 

 

As a start I would recommend working through tutorials, that are available in your installation. Check \HW2018\acusolve\win64\doc folder

 ACU-5000: Blower - Steady (Rotating Frame) 
ACU-5001: Blower - Transient (Sliding Mesh)

 

Missing nodal BCs for fluid node is not good. Fluid nodes at any boundary should have BCs. 

Also "wall" at outlet does not make sense. The outlet surface should have outlet BC! 

Share this post


Link to post
Share on other sites

thanks ydigit

I was wrong to write: the outflow has an outlet BC, the carter has a Wall type. I've only one volumes of fluid so now I will introduce the fluid  for the moving reference frame in the impeller, but which kind of setup I will introduce in this volume?

thank you so much 

Share this post


Link to post
Share on other sites

The details of assigning Moving Reference Frame to the inner volume are described in  ACU-5000: Blower - Steady (Rotating Frame) 

You will assign rotational velocity to the inner volume (Reference Frame) and the surfaces (of blades) in the inner volume

Share this post


Link to post
Share on other sites

I've realized the volume for the rotational fluid and I've assigned it the reference frame, the same that I've assigned to the impeller. When I try to launch solver I obtain an error: No valid 3d element found in the neighborhood of tria element. Does it depends on the quality of the mesh? Would I introduce a BC on the interface between the 2 volumes of fluid? I left the impeller as a solid and not as a surface, is it a problem? If I delete the solid, CFD tetramesh fill the volume of the impeller with fluid.

thanks a lot for the help!

Share this post


Link to post
Share on other sites

×
×
  • Create New...