Problem in Geometry

I am trying to do some simulations on these geometries, one is inside the other. There are gaps and intersections between them and I tried fixing it but with no luck. Geometry files are attached.

Any help is appreciated


  • peteroznewmanpeteroznewman Member
    edited August 15

    Your best approach is to go back the the source of the data. What software took the 3D CT Image data and converted it into segmented surfaces? Was another software used to create the STEP files?

    The software that split the surface into two or three faces has left permanent artifacts in the surface quality, which are particularly bad on the outer surface of the artery, but perhaps you can ignore that.

    The best you can do without going back the the source data is to convert the solid bodies back to facet bodies, but you have to change the default settings when you do that to 0.05 mm and 1 degree.

    The inside surfaces do not have such severe artifacts, but they are still present.

    Once you have facet bodies, there are many tools to clean up the facets while it is still in that form. Then there are many tools to fit surfaces to the facets. This is going to take many hours of work. You will need to spend many hours learning how to use the tools but in the end, you can have good quality geometry that you can make a good mesh with.

  • YawYaw Member

    Hi Newman, thank you very much for your assistance, I appreciate that greatly. I am working on the same project as Samar. We are quiet okay with the artifacts on the inner surface of the plaque. We hope to create a fluid domain out of the deformed geometries. Ideally, we would like to combine the plaque and the artery into one body so we can have one cavity defined for the fluid domain. From your analysis of the problem, we will work towards creation of the facets so we can work on the two surfaces (where the plaque meets with the artery). We will also appreciate it if you can suggest an alternative approach to offset the gaps between the two geometries so we can create the fluid domain by cap extraction. Thank you very much.

  • peteroznewmanpeteroznewman Member
    edited August 15

    When the two solids have been converted to facet body form, it is very simple to Merge the two facet bodies on the Facets tab. Then it is very simple to use the Separate All button and delete the small pieces. Finally, use the Auto Fix button to repair the facet body. Use the Check Facets button. It will come back with some defects. Just zoom in and delete an area of facets around the defect, then run Auto Fix again. After a few iterations, Check Facets will return a No geometry problems result.

    Then you can proceed to extracting a fluid volume. Add two planes, one at Z=0.1 and another at Z=49.9 mm then use the Split tool on the Facets tab to discard the two ends, then use Separate All to discard the outer wall of the artery. The image below is the inner surface of the artery and plaque. Now you can delete all the jagged facets that were created by the gaps between the artery and the plaque. With the Select tool, click them and hit the delete key.

    Move the Planes in and Split Body and Separate All then delete the overhanging bodies. Use the Merge tool to unite the pieces back into a single surface facet body.

    You might have to manually delete some more facets that represent bits that were not done above. Finally, use the Holes tool to make a repair the surface. I excluded the ends so I can smooth the surface before I use the Holes tool to make a solid facet body from the surface.

    Fill Holes left some artifacts. Use the Overhang tool to clean those up, or just delete those facets.

  • Here is a simpler and maybe better way to get the fluid volume. Use the method above but do it on the plaque and the artery separately. That way you can slice each facet body separately in order to delete the ends and the outside surface. Use the same planes used to slice the ends off the plaque to slice the artery, then delete the center portion of the artery. Now you just have three surfaces that represent the fluid volume walls. There are no nasty spikes in this version.

    There is a small radial gap between the artery wall and the plaque wall at the cutting plane.

    That gap could be closed after a surface is fitted to those facets by extending the surface up to meet the artery surface.

  • YawYaw Member

    I am so grateful Newman. Thank you very much for the assistance and all the effort. I will follow the steps you have presented and get back to you if I face any challenges.

  • At the end of the day, I was successful creating a solid body of the fluid volume that should mesh very well.

  • peteroznewmanpeteroznewman Member
    edited August 16

    The mesh above might be good to setup the problem and get a quick result.

    The mesh below includes 10 inflation layers. I couldn't get inflation working with hex elements.

  • YawYaw Member

    Thank you once again, Peter. I appreciate all the effort you put in to resolve the problem. I am yet to replicate all the steps/ outcomes you derived. I will let you know if I encounter any challenges. Best regards

  • YawYaw Member

    Dear Peter,

    I was able to create the fluid volume from the faceted bodies but I am having some difficulties translating that into a solid. I tried to merge all the elements of the facets into a whole surface…..I then tried to export the surface into a different CAD modeling tool to create a solid out of that…the file size for the exportation is very huge (~1 GB)….and this is creating some problems….the file hasn’t been opened in the other program yet. I would be very grateful if you can provide me with information on how this can be resolved. Is there a way to create the solid out of the faceted body in Spaceclaim? I have included images of the faceted body (fluid volume) and its accompanying surface in Spaceclaim. Best regards.

  • peteroznewmanpeteroznewman Member
    edited August 19

    Dear Yaw,

    You didn't answer my question, what software took the 3D CT Image data and converted it into segmented surfaces? Was another software used to create the STEP files?

    Create 4 planes, one just a little inside each end and one at the transition to the center portion.

    Use those 4 planes to slice the facets into 3 tubes, a center one and two end tubes. Delete the end cap facets.

    In SpaceClaim, on the Tools tab, use the Auto Skin tool to fit a surface to each of the three tubes.

    Now you can delete the facet bodies.

    Make copies of each of the three sufaces. Use the Split Body function to split the end tube with the center. Use the Split Body function to split the center with a copy of the end tube. Delete the trimmed pieces and the copies that were not trimmed.

    Merge the three trimmed tubes into one body.

    Use the Fill tool to put an end cap one each end. Now you have a solid you can mesh.

  • YawYaw Member

    Dear Peter,

    Thank you very much for your assistance. I am very grateful. Sorry I didn't answer your question previously. The models we are working on now were originally quite idealized (built in Solidworks) and had regular shapes but were subjected to some loading conditions in Abaqus that got them deformed. We obtained the orphan meshes of the deformed geometries from the output database in Abaqus and made surface models out of them in Hypermesh. The model we posted here is the output from Hypermesh.

    I am going to implement the steps you showed us. The part on the splitting of the tubes and their copies seem somewhat confusing. I will attend to work on that and would let you know if I encounter any challenges in that step. Once again, I am very grateful for all the time and the effort you spent to resolve the problem for us - I appreciate that greatly!

  • YawYaw Member

    Dear Peter,

    I hope you are doing great. I am having problems using the auto skin tool. The error message that appears reads "60 edges sharper than 120 degrees" .the skin thus fails to envelope the tubes I created. There are only 3 sharp edges for the other ending tube. I have tried many ways to fix this but I have been unsuccessful...(the number of sharp edges only keep changing). Should this error be expected and how can I fix that? I previously used the skin surface tool to create the surface geometry but stitching that proved problematic..(I had to define a number of control points on one of the end facets which left me with sections that need to be stitched) ...I would have to stick to the auto skin tool....can you please assist me to resolve this problem. I have attached some images to show the problems I am having.

  • YawYaw Member

    Dear Peter,

    I was finally able to resolve the problem. I used the skin surface tool with defined planes to fix that. I am very grateful for your assistance.

  • Good to hear! As I reread my directions, I noticed I left out one step. After creating three tubes with the Auto Skin tool, I forgot to say to use the Pull tool to extend the edges at the cutting plane. The end tube edge toward the center and the center tube edge toward the end. That causes the two surfaces to intersect. Then the trimming operation is done using split body as described above.

  • YawYaw Member

    Very well, Peter. Once again I am very grateful for all the assistance you provided us. The quality of time you spent on the problem, the effort and instruction are very astonishing. I am highly indebted to you!

Sign In or Register to comment.