converting to .STL

[Anonymous]

Hi,  I can create threads in Fusion and 3d print the parts good using Cura with our Ulitmaker 5s.  When I use Inventor to create the threads and convert it to a .STL file then use the Cura with our Ultimaker 5s,  the threads come out wrong.  I have included a pic.  You can see the shadows of the 2 orange pieces show the threads leaning to the left and not like the top sample straight up and down.

Is this a problem with .STL converter in Inventor?

[mcgyvr]

Please post your ipt file so we can ensure the geometry is correct there and see how you created the threads. 

[johnsonshiue]

Hi! I don't believe it is STL's issue. It is the geometry was not created correctly in Inventor. Please share the Inventor part here.

Many thanks!

[Anonymous]

Here are the sample pieces.

[swalton]

Have you imported the STL back into Inventor or other viewer?  That might be a way to see if the STL looks correct, or if there is an issue with the export.

[mcgyvr]

@Anonymous  What output resolution did you use when creating the STL file (see options button before you hit save.. assuming you just exported to stl)?

Can you also post your STL file that printed incorrectly?

Do the previews of your STL look ok in Cura (I use Cura with my Ultimaker)? 

Have you looked at the layer views in the slicer to see if they look correct?

I saved out a high resolution and brep in Inventor from your files and they look just fine in Cura for me.. 

[JDMather]

Attach the resulting *.stl files here.

Not sure if I missed where you included this information, but what Options did you use when saving?

[cbenner]

@johnsonshiue 

Careful, Johnson.... you never know who might have been doing the modeling.  😉 Or whose tutorial ( @JDMather ), they might have used.

j/k, but we did model the threads just like JD showed in one of his videos, and we have also tried taking a part out of Fusion into Inventor, and printing it.  Whatever combination of modeling and importing we have tried.... we get good threads from Fusion and bad threads from Inventor.  Even on the same part model.  So, maybe we just don't know the correct way of setting up the .stl files.  From Fusion you just hit print, and it goes directly to the slicer... badda boom we get good threads.

[JDMather]

---

@cbenner wrote:

j/k, but we did model the threads just like JD showed in one of his videos

---

Uhmm, not exactly.
I fully define my sketches.

[johnsonshiue]

Hi Chris,

When I said the geometry was not created in Inventor, I did not mean it was users' fault. I have seen Coil command creates poor geometry. It depends on the chosen settings. I will take a closer look at the parts.

Many thanks!

[johnsonshiue]

Hi! I took a look at the two parts. I don't see any modeling issue. The parts look correct and the geometry checks clean. One thing about Coil command is the cross section may not be normal to path. To have consistent cross section, you will need to use Sweep and create the profile on a workplane at the start of the path and also oriented normal to the path. But, it should not affect your case.

Attached are the STL files exported from 2021 using Brep Option. They both look clean in Inventor as mesh. Please share your STL files so we can take a look.

Many thanks!

[cbenner · Accepted solution]

@JDMather  we were aware of the model's shortcomings, it was a very quick test...., I was trying to pay you a compliment.  

@johnsonshiue  I'm sorry, my comments were meant as humor.  I figured you didn't know I was anywhere near this.  That's one more I owe you at AU.  😉

Everyone....

We solved this issue, so far, by using an app in the CURA marketplace called Inventor Integration.  This app basically grabs the data straight from the .ipt file without going through .stl conversion.

Using this app we have (again, so far) been able to produce very good threads using JD's modeling technique (and YES... we fixed the sketches).  We have also worked up a workflow that uses coolOrange thread modeler and addresses the known issues surrounding end chamfers.  I am going to create a Screencast video of that as soon as we have test printed the first piece.  I'll post the video here.  Thanks everyone for the input.

Not bad for two people working 30 miles apart due to social distancing!  Thanks @Anonymous for all of the test prints.

[cbenner]

As promised, here is a quick video of the coolOrange workaround for chamfers.  Initial printing test was a great success.