STL Files and Working with Surface Models

[RunningMan0505]

Hello,

I have a CT Scan of a fracture inside a rock (void space). This file has been formatted to an STL File.  I am trying to figure out a way to take this file and subtract it from a generic solid block, then split this block in half to make it printable, then print this manifold and assemble for flow testing.

I am having issues finding a solution to working with this Surface Model.  I initially tried to bring it into 3ds Max and create a block and do a boolean subtraction, but the surface model has too many errors and needs repairing.

Anyone have any suggestions??

Thank you.

[TheCADWhisperer]

---

@RunningMan0505 wrote:

I have a CT Scan of a fracture inside a rock (void space). Anyone have any suggestions??

---

Do you observe that stl is rubbish.

No curves, only triangular planar surfaces.

Can you Attach your *.stl file here?

What version of Inventor are you using?

[RunningMan0505]

It doesn't necessarily have to be STL right now, is there a better format you'd suggest? It will eventually need exported as an stl though for printing compatibility.

I apologize, but I'm actually not allowed to attach the file and I am using Inventor 2021.

Thank you.

[TheCADWhisperer]

---

@RunningMan0505 wrote:

I apologize, but I'm actually not allowed to attach the file…

---

In that case you will have to start your learning process by finding other stl files that are not proprietary.

Autodesk MeshMixer and Fusion 360 have stl editing tools.

[RunningMan0505]

Thank you for the input!

[cadman777]

If you have a really bad surface model, you can much more quickly edit it in Rhino3D.

Using Rhino3D is similar to using AutoCAD, except it's a 3D surface modeler.

I've fixed many 'bad' surface models over the years using Rhino3D, that would be unbearably painful and time consuming to do in Inventor's CopyToConstruction environment. I recently did one for a guy in here that was impossible to do in Inventor, using Rhino.

But if you have a mesh file, then Rhino can fix it, but it's a bit more work. I'm currently using Rhino6, but maybe Rhino7 has more tools for fixing meshes. They have QuadMeshnig, which is what most animation modelers use.

I've used MeshMixer before, and it's pretty good, but not good enough for production level work. It's more of a hobby software. There are some other 'free' mesh editors that are on the about the same level that you can use to edit meshes. Some of them are easy to use, and others have a very steep learning curve.

Mesh editing and surfacing are different kinds of modeling than solids modeling and have their own learning curves which are pretty steep compared to solids modeling.

[Gabriel_Watson]

After repairing your surface anyway you can, you could extract the cavity using the solution here (or Derive component):
https://forums.autodesk.com/t5/inventor-forum/how-to-fill-internal-volumes/m-p/5620082#M545472

[RunningMan0505]

what about bringing this mesh into Fusion 360 and converting it to a solid Brep and then working from there.

The end goal is to be able to subtract this rock fracture void from a solid block and be able to 3d print it. So I can then set up a test to run fluid through it to see how different rock fractures behave.

[TheCADWhisperer]

---

@RunningMan0505 wrote:

what about bringing this mesh into Fusion 360 and converting it to a solid Brep and then working from there.

---

How will you learn the software?

[RunningMan0505]

I've used Fusion 360 in the past, but not a lot of exposure to it, not really sure how to answer that question.  Learn as I go I suppose.

[SharkDesign]

Maybe I'm wrong, but if you are talking about errors and repairing surfaces, potentially you are going to end up with a model that is not accurate to the original and therefore not worth testing?

I don't know how precise you need to be?

[TheCADWhisperer]

There was a good Autodesk University class posted a couple of weeks ago (search Google) on using the newest mesh editing tools in Fusion 360.

My guess (not having seen even a picture of your data) that you will have difficulty scaling the learning curve.

[RunningMan0505]

Very good point, currently the goal is to find a path forward, then improve on that. Smoothing of some of the surfaces and losing some of the minor splinters of the fracture aren't too high of a concern at the moment.

and obviously the counter-argument to that is if the path forward doesn't produce somewhat accurate results why bother going down that path. Right now it's more of a proof of concept.

[RunningMan0505]

thank you for the suggestion!

[cadman777]

Can you post the pointcloud?