Stats

  • State Not Answered
  • Replies 5
  • Subscribers 6
  • Views 2194
  • Members are here 0
More Content

Create a Block between inflation layers

MS20240828216

Hello everyone,
My problem is creating a block between the inflation layers. The reason for the failing of the block creation are the domains on the outer surface of the vane-block, which were created during the extrusion of the vane in rectangular z-direction. The problem here is that these domains of the vane are too long and go directly into the wall layers (problematic domains marked in white). That's why it's not possible to create the block in the empty space between the inflation layers of the walls and the vane.

My question: is it possible to shorten the domains that they fit in the space between the two wall layers and they keep the same height so that there is no gap between the block I want to create and the vane block.
I’ve already tried to create the domains manually by creating connectors with the length of the distance between the two wall layers. This allowed me to create a block, but there was a gap between the created block and the block of the vane. The reason for the gap is that the manually created domain was not exactly on the outer surface of the vane-block.

I hope you could follow my explanations
Thanks for your help

  • 0

    Hi,

    Thank you very much for sending your question. 

    I can understand that you have tried creating domains via extrusion and that you have also created them manually. From those domains, you were able to assemble the desired block(s) but there were gaps with the existent block. Unfortunately, it is a bit hard to understand the entire mesh from a few images with different orientations. 

    In order to offer the best possible recommendation, I would need to see the actual Fidelity Pointwise file with the grid to better understand what is going on. Would this be possible? Keep in mind that this is a public forum. 

    Please note that you could use, for example, a Path extrusion in order to have more control over where the extrusion starts, stops, and what kind of grid point distribution the final domain will have. You should also be able to manually create domains that fit exactly the space you need them to fit (via the precise definition of the bounding connectors). That being said, which approach is best would depend on the particulars of your grid.

    Path Extrusion: 

    Looking forward to your additional comments. 

    Best regards,

  • 0 in reply to Claudio M Pita

    Hi,
    Thanks for your reply.
    The model is divided into 3 sections (inlet, outlet and vane). The aim is to have inflation layers around the vane and inflation layers from the shroud and the hub (wall layers) that extrude into the body.  The empty spaces should then be filled with blocks and initialized. Furthermore, the case should be solved with mainly quad cells.

    I can only upload a file containing the database and the connectors. The original file is too large to be uploaded here.

    Linear_Periodic_Case_Con.zip

  • 0 in reply to MS20240828216

    Hi, 

    Thank you very much for your reply with the information. 

    If your entire Fidelity Pointwise project can be shared in a public forum like this one, please use the FTP account below to upload the entire Fidelity Pointwise file. In addition, please indicate the name of the problematic domain and/or use a distinctive color on it. 

    Could you please describe the process you are using the create this mesh?

    • Based on your comments and what I see in the provided file, it seems as if the connectors and domains are created manually. Is this the case? Although this process is valid, I am wondering, is there any particular reason why you have decided not to use the Automatic Surface Mesh functionality to create the surface grids automatically?
    • Along the same lines, it seems as if the boundary layers are being created via regular Normal extrusions. Is this the case? Even though this is a valid approach, is there any reason why you decided not to use the T-Rex algorithm? 
    • Using the connectors you have created, I was able to assemble a single block (i.e., volume grid) and finish the grid using T-Rex in just a few minutes. If you are OK with it, I can upload the grid in the FTP account shown below. Please remember that this is a public forum.

     

    FTP Server : ftp.cadence.com
    FTP Account : xcftp1180
    FTP Password : X56cmRqW
    Customer : MS20240828216
    FTP Account Expiry Date : 09/03/2024

    Looking forward to your additional comments. 

    Best regards,

  • 0 in reply to Claudio M Pita

    Hello Claudio,
    Sorry for the late reply.

    I uplouded the original file.
    My mesh creation process:
    -     Importing the CAD model and assemble the two vane quilts into a single quilt
    -    Create connectors on database entities
    -    Add a 2-point connector at the top and bottom of the vane
    -    Manually create the vane domains
    -    Extrude normal to the vane domains
    -    Assemble special domains for the for the walls (shroud and hub)
    -    Extrude the wall domains normal
    -    Manually create the domains for the rest of the surfaces
    -    Create a block for the inlet- and the outlet-section

    I tried your advice with the automatic surface mesh but the leading edge has big holes

    I also tried to create the mesh with T-Rex but I cannot find a way to create the mesh with mainly hexahedra cells. Is there a way to create the mesh with T-Rex and with hexahedra inside?

    Thank you very much for your help

  • 0 in reply to MS20240828216

    Hi, 

    Thank you very much for the update on this case. 

    Here my comments:

    • The important thing is to ensure that the entire geometry is assembled into a single watertight database model (i.e., no gaps). Once this is done, you will need to assemble the quilts as needed to define the needed meshing regions. Note that the vein needs to be assembled into two separate quilts to nicely define the leading edge where a connector will be created. In order to better understand the Fidelity Pointwise meshing process, I would recommend reviewing he following videos:
    • If the leading edge is not meshed correctly, is likely due to the fact that the entire vein is assembled into a single quilt; as indicated, you will need two quilts: one on the pressure side and one on the suction side. Please take a look at the videos above for a nice example. 
    • If the surface grid is quad dominant (i.e., triangles and quads), T-Rex will create as many hexahedral cells as possible. In addition, you can use the Voxel algorithm to place cartesian cells in the isotropic section of the volume grid (seen in the third video above). 

    I have uploaded a new file called Linear_Periodic_Case_Con_PWI.pw with the final mesh I obtained using the connectors that you have manually created (note that no quilting was done!). You can see the hexahedral cells in the boundary layer. In this case, I used the Delaunay algorithm to populate the isotropic regions but, as indicated, you can change this to the Voxel algorithm if needed. 

    As indicated, the manual procedure you are following is not invalid; you will likely get to a valid good mesh. However, it is a more time consuming way of getting to the final result. What I am trying to show you is a faster way to obtain the mesh you are looking for. 

    Note: I will be on work travel this week; my responses may be delayed. 

    I hope you find this information useful. 

    Best regards,

Cadence Guidelines

Community Guidelines

The Cadence Design Communities support Cadence users and technologists interacting to exchange ideas, news, technical information, and best practices to solve problems and get the most from Cadence technology. The community is open to everyone, and to provide the most value, we require participants to follow our Community Guidelines that facilitate a quality exchange of ideas and information. By accessing, contributing, using or downloading any materials from the site, you agree to be bound by the full Community Guidelines.