vOidward
Posts: 40
Joined: Thu Feb 20, 2014 9:34 pm

Bug: Shell Convergence = ugly bulges

Mon Jun 23, 2014 8:03 pm

There's an issue with the way converging shells terminate into each other when encountering a narrowing / tapering path.
Shells will not terminate until they are completely overlap each other. When fine detailed areas have 2 layers completely overlapping, this creates a bulge instead of a smooth transition on the exterior of the print.

Here's an example of a tapering cylinder with a hollow section in the middle:
3 shells with 15% infill is used.

Shells starting to converge:
Image
Image
Image

It is only after this point, where the 3 shells are completely on top of each other when the slicer will terminate into 2 shell mode. Shells should be terminating as soon as they touch each other, or at most after a 20% overlap.

Here's a closeup of the printed result:
Image

The effect in person is almost a stair-stepping pattern. It interrupts a smooth tapering cylinder with bulging lines.
It seems like the code doesn't take into account the width of the lines themselves when determining the threshold for when to terminate into fewer shells.



-----------------------------------------------

If you're going to be looking at this area of code, I'd also recommend looking at scenarios like this:
Image
The width between the last 2 shells is less than 1 shell width so it isn't trying to put 2 of them on top of each other, but when it comes to topfill I'd rather it break the rules and try to close any gaps possible. A single line of infill in this type of scenario would be great.

Also, I recommend for any exterior surface to overwrite "minimum infill length" parameters to 0. It's a great feature, but it does occasionally leave holes on the exteriors of models where the infill length is too short.

dennisjm
Posts: 342
Joined: Wed May 07, 2014 8:51 pm

Re: Bug: Shell Convergence = ugly bulges

Mon Jul 07, 2014 8:45 am

+1 to fix this, I've had to add multiple processes with different number of shells at different heights to fix some prints because of this.

JohnB
Posts: 13
Joined: Tue Feb 04, 2014 9:33 pm

Re: Bug: Shell Convergence = ugly bulges

Mon Jul 07, 2014 6:28 pm

I see this a lot too. I think this "ugly bulge bug" is also related to the "thin wall bug" posted here: http://www.forum.simplify3d.com/viewtop ... =714#p2513

vOidward
Posts: 40
Joined: Thu Feb 20, 2014 9:34 pm

Re: Bug: Shell Convergence = ugly bulges

Tue Jul 08, 2014 2:20 am

dennisjm wrote:+1 to fix this, I've had to add multiple processes with different number of shells at different heights to fix some prints because of this.
Yeah that's what I've had to do also, manually find where they overlap and truncate the shells before then (or occasionally infill, since it tends to follow outlines and overlaps in a similar way) so it doesn't have a chance to overlap in this manner.

It's a usable solution but it prevents me from doing sequential printing of multiple parts because there's no way for it to do 1 multi-process object fully before moving on.

ashleywebster
Posts: 31
Joined: Sun Apr 20, 2014 2:12 am
Contact: Website

Re: Bug: Shell Convergence = ugly bulges

Fri Jul 11, 2014 1:42 pm

++ Agreed. I've seen this a few times now, both the overlapping paths and the gaps between paths on tapering objects.

Return to “Troubleshooting and Bug Reports”