Create a perfect-fit stand for a model

[Cadie107]

Hello!

I have recently 3D printed the 3D structure of a protein which I wanted to gift to a colleague who submitted her thesis about it. For that, I intended to print a stand, which I tried to design, that it perfectly fits the model.

My current approach for that is the following:

1. Extrude a stand (solid body) and place the protein (mesh->object) above it.

2. Lower the Protein in 1 mm steps and cut (combine tool) it out of the stand.

This method comes with 3 problems:

1. It's not perfect - some smaller artifacts which block the insertion can still remain.

2. It's perfect - there is no offset between the object and the cut. When I try to offset 5000+ faces, my Fusion 360 crashes.

3. It's tedious! 😉

So I wanted to ask, if anyone knows a way to better solve this issue of removing the overhanging regions in a mould.

I'll attach a screenshot as soon as my Fusion 360 comes back from freezing again (:

Thanks for every advice

Carroll

[jhackney1972]

Please attach both your model which contains the Protein model and your attempt so the Forum users can take a look.  If you do not know how to attach your Fusion 360 model follow these easy steps. Open the model in Fusion 360, select the File menu, then Export and save as a F3D or F3Z file to your hard drive. Then use the Attachments section, of a forum post, to attach it.

[jeff_strater]

Given that your protein is a mesh, I would tend to want to do this in the mesh world.  I believe you can solve 2 of your 3 problems in the mesh environment:

1. It's not perfect - some smaller artifacts which block the insertion can still remain.

2. It's perfect - there is no offset between the object and the cut. When I try to offset 5000+ faces, my Fusion 360 crashes.

3. It's tedious!

2 and 3 I would approach by scaling the protein mesh just a bit, and using mesh combine to subtract the scaled protein from the base.

1 is a harder problem.  There is no "push into" feature either in Solid or in Mesh that is really what you want.  You would have to do it manually - perhaps making a copy of the protein, and adding geometry to essentially create a "shadow" of the protein in the direction you want to push into the base.  That would be tedious, but could be done.

[Cadie107]

Hi! Sorry for the long delay, I had to rush home yesterday.

@jhackney1972: Thanks for the hint, the model is attached!

@jeff_strater: Okay, I was afraid that would be the case. But that's a nice advice. So far I only used the Mesh features to, well - go away from meshes 😉 The problem with scaling it larger is, that it has a cavity below it, which would then also be larger. I also thought about the 'shadow' option and tried it out by creating numerous sketches with intersections at different heights (z-stacks). But the result wasn't that great (and in the end it's the same thing which I receive with the combine/cut).

What I also thought could exist was a remesh/reduce option, which enforces, that the output includes all the volume of the input, but I couldn't find anything similar.

I will try to solve it in the mesh world (I like how that sounds - a little like RNA world) when I have time today.

Thanks a lot!

[TheCADWhisperer]

---

@Cadie107 wrote:

I will try to solve it in the mesh world

---

A couple of issues:

1. Split away everything above the stand as there is no need to calculate that.

2. Then there are some undercuts that would result from Combine-Subtract, so what is really needed is a Sweep-Cut with Solid Body (not available in Fusion 360). (Or are you going to cast the base around the print?  Do you want it to be removable from the base?)

[TrippyLighting]

I've converted the mesh model into a solid model using a workflow I've described in this tutorial. However, I don't think it is really going to solve the underlying geometry problem.  I cannot share the model here because even if zipped, at 118 MB it exceeds the max file size allowed for attachments.

[Cadie107]

@TheCADWhisperer:

That's a good idea! Don't know why it didn't come to me. Thanks! I already have the protein printed and want it to be removable. The stand should also be FDM printed. A sweep cut - thanks for that terminus. It's always helpful to be able to describe problems.

@TrippyLighting: Thanks for the tutorial, I'm at work right now but I will watch it, when home. Converting meshes to solid bodies was always something, where I couldn't fully get my head around, I hope it will help me out there (:

[jeff_strater]

The key point from @TheCADWhisperer 's description is:  "Sweep-Cut with Solid Body (not available in Fusion 360)".  Fusion does not yet have the ability to sweep a solid body along a path.  So, that is not going to help in this case.  You will have to do something like doing the cut every .5mm along the path, or construct a body which is the "shadow" of the body and use that to cut the base.

[TheCADWhisperer]

I am going to suggest to identify three protrusions that more or less form a tripod without any undercut from a Combine - Cut and then Extrude or Revolve any other interfering material to simplify the design of the base.

or

maybe something closer to >>this or similar design<<.

[evanp4509U4JZ]

An easier way to achieve the desired result might be your original process but start at the depth you want and work your way up. You will run out of interference sooner and need less computations thereby freeing up some time to take smaller steps and get closer to "perfect" fitment.

[TheCADWhisperer]

@evanp4509U4JZ 

Can you post example that demonstrates this technique?

[evanp4509U4JZ]

No, because I'm on a work computer and they frown on downloading outside files. But wasn't he lowering the protein into the base 1 mm at a time and cut/combining? If he starts at the depth he wants and works outward wouldn't there be fewer processes before he ran out of interference between the base and the protein? If he works from the top down to say 20mm deep he will need 20 processes. If he starts at 20mm deep and works up maybe 5-7 processes before there isn't any interference between the protein and the base. He could go .25-.5mm at a time and spend the same amount of time with 2-4x greater resolution.

Or did I misunderstand the end goal?

[evanp4509U4JZ]

In this simplified example with bad geometry it took 17,  .1" steps, or 1.7" to clear a 2.25" depth. This is with a much less potato looking "tool" than the protein. If the bottoms weren't flared so big it would have been less. Isn't this just basic wedge application, just with a few bumps along the way? 

[cadman777]

Hi Carroll,

I'm from the Inventor forum where inquiry about this was made a day or so ago.

Since I work in an older version of Inventor, I can't read the new ipt files.

In order top participate, I need the original mesh file, and to know the depth it will protrude into your base.

Cheers...

[Cris-Ideas]

Hi Carroll

I also saw your problem in inventor formu. 

It should be quite simple in principle as you only want to sweep it along strainght line.

Also please specify how deep do you want to move the protein body in to the base.

Can you post your file in some open format (original stl, or something else) that can be imported in to other programs, not only inventor.

Actually I believe it will be easiest to execute in autocad, so I would like to try.

Fusion file posted I am not able to open or import.

Please also have in mind that quite a few people, my self included, still run older versions of software, and Autodesk does not allow for backwards compatibility, so we can not open newer versions. Therefore open format is best.

Cris.

[Cadie107]

Hello everyone!

I tried the idea of cutting from the lowest depth first and move upwards, but my PC still didn't go through it (at some point it always got unresponsive and crashed). Also, since each process went for a few minutes and rendered my whole PC useless, I gave up at some time and just created something more or less working.

Then I got piled up in other work and this forum entry got lost in my tabs - sorry!

I added the protein (it's A. thaliana UVR8) as a mesh. It was supposed to cut about 1 cm into the stand.

Best regards and thanks for all efforts!

Carroll

[Cris-Ideas]

I am still not able to import this file.

Can you convert it to some open format and post again? 

stl, igs, step

Chris.

[cadman777]

Rhino opens 3MF no problem.

Or you can use a free online converter to get it to STL or OBJ.

[TrippyLighting]

I believe any way to make a sweep cut work in Inventor (or Fusion 360) will require a non-facetted BRep.

The approximate method by moving a facetted solid gradually out of the to-be-cut object and doing a combine-cut requires an enormous amount of single-core computations from the gemeometric modeling kernel and for this and other reasons is unlikely going to succeseed.

I had converted this molecule model into a BRep using InstantMeshes, but he resulting .f3d exceeds the file size that can be attached to a post.
  

I can share a dropbox folder to the file if there is interest.

Edit: I exported the smooth BRep to  .step format and zipped it. That gets it under the 71Mb limit. 

 

[Cris-Ideas]

Fine,

I can get it open.

Chris