Noob question: why the vertical cuts through my simply tweaked cylinder?

[Anonymous]

I'm trying to make a slight bulge to a simple cup (cylinder w/ open top, closed bottom).  It looks right while I'm sculpting, but when I end sculpting it throws vertical cuts that I'm not wanting. My 1st attempt only had 4 vertical vertices intruding into the circumferential bump, but take two got all of them involved.  Only diff was a setting when I filled the hole (made the bottom surface): 1st time=reduced star, 2nd=filled whole star (hoping I'm remembering verbage correctly - still cutting my baby teeth here).  See attached pix. 

What I'm after, oddly enough, is an STL file for my 3Dprn.  It SEEMED to be so easy to draw it, but getting that committed to an STL - another story.  Trivial little chore turned into hours of hair pulling.

Anyone see this off the top?

Thanks

pw

[jeff_strater]

It's a bit hard to tell from the pictures whether what you are seeing is just graphics effects, or real geometry. Can you post the model so we can look at it?

Or, if you can post a top-down image (from the top of the cup), that might also help

thanks,

Jeff Strater (Fusion development)

[innovatenate]

Whenever you select Finish Form, the Sculpt Body (T-splines sub-d) is converted to B-Rep Solid geometry. That's where the edges are coming from. 

Any thoughts on just using a revolve command to create the desired geometry?

See the below video, I create a cup witha spline and then export the body to STL.

http://autode.sk/16IBbzP

Hope that helps.

Thanks,

[Anonymous]

Wow - been fooling around with this for a couple HOURS now, just trying to RESPOND and provide the additional pics as Jeff req'd.  Finally resorted to trying IE browser (for some unknown raisin Chrome was not working - clicking on hot links in various places on your site accomplished NOTHING - so I couldn't get ANYWHERE!  This also meant that a popup that was asking for my feedback on the site was stuck in the center of my window and I couldn't get rid of it.  Guess what my responses would have been HAD I BEEN ABLE TO RESPOND!

Anyway, that aside, I'll try again to upload the view of the STL it produced.

...and it APPEARS to be allowing it.

So, while I will take a look at the video of cup via revolve, I remain entirely puzzled why my SIMPLE approach should be sabotaged by such voodoo-like, non-intuitive introduction of facets so clearly UNLIKE what I designed.  Honestly, would YOU have foreseen this deviant behavior, even with your experience???  To the uninitiated, even my tutorial video consumption being quite limited, I mistakenly thought I saw an application that my uber-limited grasp could actually master.  Ha!  Fooled ME!!!

Well, I need a break from this futility, but I will take a gander.

Thanks!

[jeff_strater]

Thanks, PaleWriter. Yeah, it looks like that is an actual geometry problem, not just graphics effects. I will send this to our TSpline experts. I have a theory about what is going on, but I'm not sure without asking them. Before I quit tonight, I will post a quick example of how I would go about this design, and hopefully not get those grooves.

Jeff

[jeff_strater]

OK, here is my simplistic attempt at this design.

Start with a  Form feature, then a Cylinder:

then create the bulge, using Edit Form, double-clicking on one of the circular edges near the bottom, and scaling outward:

Then, to close the bottom, use Fill Hole:

I used "Fill Star" and "Maintain Crease", to get that sharp edge.

Then, exit the form, and this is what I get:

Is that more or less the same method you used?

If you use Export, then choose "Archive file (*.f3d)" as the type, you can save your design to your local disk, and upload it here (I think you have to add it to a zip file, because this forum does not allow .f3d files...).  If you post your design, we can take a closer look

Jeff

[jeff_strater · Accepted solution]

Yep, this is a bug all right - you found a good one!

We think it is only reproducible when your cylinder has 12 or more faces to it - my example above has fewer faces.  If that works for you, it might be a good workaround.

Here is the bug re-created.  Looks OK in TSplines:

but when you exit Form, and convert to BRep:

thanks so much for posting this.  Good catch.  We will work on a fix

Jeff

[Anonymous]

Jeff - you da MAN!

Thanks for your prompt and precise handling of this!

As it happens, the steps you indicate are precisely how I went about it - the only exceptions being the parameters (height=20, lines of latitude=40, diameter=12.4, lines of longitude=16) and that I enlarged more than one latitude line (step-wise expansion, like 1.01,1.02,1.03,1.04,1.03,1.02,1.01).

I'm fairly sure that lowering the lines of longitude will still work for my porpoise - at least, the immediate one.  For the next (and much more complex) one I hoped to tackle, though, I fear this may prove to be an obstacle beyond my current abilities.  That involved trying to replicate (and subsequently modify in ways at which Fusion _SEEMS_ adept) a complicate design I've done in SketchUp.  I was looking at trying to wrap a TS cylinder around my existing STL and snapping to it.  If I knew Fusion better I'd surely start over w/ a TS form, but there are so many things to do that I don't yet see the path through the maze.

Anyway, this one should work - thank you very much.

Well did!

Back to the drawing board, as they say...

pw

(parenthetically - any idea why Chrome won't handle this site?  I don't really have IE setup appropriately on this PC, but I have to use it to check in here!  Waaaaa, waaaa, waaaaa)

[Anonymous]

Well . . . . .

Seems to be more entrenched than you're indicating: reducing longitudinal lines to 10 and even leaving the bottom unclosed I still get ONE intrusive line.  Closing w/ reduced star yielded 4, as with my original discovery.

So - we shall see what the programmers do with it.  Perhaps I can export this and fix it with tools I know.

Carry on.

pw

[jeff_strater]

oh, that's too bad. Sorry that you still get that one cut remaining... Hopefully we can fix the underlying problem soon.

It would be interesting to export this to something like IGES or STEP, and see if the geometry is bad after export.

The TSpline developers seem to think that this is a faceting issue (somewhere in between geometry and graphics), so an export will help see if that is true or if it really is the geometry.

Thanks for your help to debug this.

Jeff