User avatar
BaronWilliams
Posts: 159
Joined: Tue Jul 15, 2014 8:30 pm

Re: Better top surface stitching & bosses on top layers

Thu Feb 12, 2015 3:41 pm

Screen Shot 2015-02-12 at 11.55.29 AM.png
Screen Shot 2015-02-12 at 11.54.58 AM.png
These images illustrate the shell to shell bonding problem for boxes on top of a large box. They are cross section views showing the internals, with infill in the first image and no infill in the second image to help illustrate the problem. Then non-slanted box on the top right only has shell to shell bonds for 1 shell. This is the weakest piece. It will be very easy to break it off, despite having 50% infill. The boxes to the left have slanted walls that increase as we get further to the left. As the slanted box walls get closer to forming a horizontal line with the surface below it, the shells get closer to having completed bonds, and so the right walls of the boxes get harder to break off. As the slanted walls get further from forming a horizontal line, the shells begin having less bonds, and it will go all the way down to only 1 shell bonding once it's perpendicular, so in all cases above, the left walls are easy to break. A 1 shell bond is extremely easy to break.
Screen Shot 2015-02-12 at 12.29.26 PM.png
This image illustrates the problem using trapezoids on top of a large box. Again, it's a cross section view. The trapezoid on the left will break off easily because it completely runs into the Simplify3D shell bug on the right and left walls. The trapezoid on the right will be much stronger because it avoids the Simplify3D shell bug on both it's right and left walls.

Complex models can have this shell bug all over them if they have similar vertical angles present in them, producing very weak prints.

I HIGHLY ADVISE THE SIMPLIFY3D TEAM TO LOOK AT THIS SHELL STITCHING BUG! It is causing a lot of the weak print complaints people are talking about. This is definitely a weak point in the software (pun intended).

jimmybuckets
Posts: 68
Joined: Thu Sep 11, 2014 7:42 am

Re: Better top surface stitching & bosses on top layers

Fri Feb 13, 2015 8:00 am

Thanks Baron your illustrations are excellent examples. I have been in touch with Joey K from S3D. I referenced this thread even though there are others like it. There is another thread in the bugs section that I pointed to this thread with a link. Let's try to keep thus thread bumped to the top.

soa
Posts: 15
Joined: Sun Aug 03, 2014 8:09 pm
Location: Reston, VA

Re: Better top surface stitching & bosses on top layers

Fri Feb 13, 2015 4:51 pm

this is truly a huge bug in S3D. that and it's terribly handling of empty or near-empty infill. makerware has such a crap interface, but it makes much better prints.

jimmybuckets
Posts: 68
Joined: Thu Sep 11, 2014 7:42 am

Re: Better top surface stitching & bosses on top layers

Fri Feb 13, 2015 5:02 pm

In the mean time I'm going back to slicing in Kisslicer for certain prints and then just importing the gcode into S3d gcode library and then running from there. Does suck to have to do so tho

MFitz73
Posts: 16
Joined: Wed Feb 11, 2015 8:41 am

Re: Better top surface stitching & bosses on top layers

Sat Feb 14, 2015 8:58 pm

I've noticed this too... and I'm a new user!!! so, what I noticed did make some difference is if you ad a bevel to the base of the attaching smaller cube.
might help in the meantime while s3d hopefully corrects this.

User avatar
BaronWilliams
Posts: 159
Joined: Tue Jul 15, 2014 8:30 pm

Re: Better top surface stitching & bosses on top layers

Mon Feb 16, 2015 1:19 am

I am using Simplify3D for nearly all of my 3D printing. I find that Simplify3D makes the best looking prints, and I love all of the settings it has. It's very feature rich. I like the work flow. I like the way Simplify3D works in general. I seldom go back to other slicers unless I absolutely need to. To avoid this bug, I usually add a solid diaphragm where the weak point is.

I have a big handful of slicers installed that I play with occasionally. All software has bugs. There are different issues in all of the slicers out there. Each one has its strong and weak points. None of them are perfect.

Unless you slice specific shapes, it can be hard to notice that this bug is present in Simplify3D. It can be avoided by beveling, as was stated by MFitz73. A lot of professional 3D models have a lot beveled corners which avoid this problem. So if Simplify3D has been tested with shapes that have a lot of beveled corners, I could see this bug slipping through the cracks during software testing.

The very simple shapes I posted above are helpful in catching this bug.

I am sure the Simplify3D team will address this problem once they can reproduce it easily in their test environment.

soa
Posts: 15
Joined: Sun Aug 03, 2014 8:09 pm
Location: Reston, VA

Re: Better top surface stitching & bosses on top layers

Mon Feb 16, 2015 8:47 am

chamfers hardly solve the problem. if you have the freedom to make those changes to the part (aren't constrained by fit or geometry) then it sucks a tiny bit less. yes, all slicers have problems and s3d is one of the better ones. but this one is fairly egregious and it's been in there quite a while. while support has been respectful, their workarounds for this and other bugs have been, effectively, to re-engineer my parts. it may seem like small thing on their end, but every post-processing step adds a lot of time. changing the part invalidates the workaround. and for larger part counts, adds an untenable amount of time. i should be free to engineer my parts for their intended purpose and not try to game, compromise and workaround slicer deficits.

jimmybuckets
Posts: 68
Joined: Thu Sep 11, 2014 7:42 am

Re: Better top surface stitching & bosses on top layers

Tue Feb 17, 2015 9:51 am

Bump to get this bug fixed.

JD95
Posts: 34
Joined: Sun Sep 01, 2013 12:11 pm

Re: Better top surface stitching & bosses on top layers

Tue Feb 17, 2015 10:36 am

BaronWilliams wrote:
Screen Shot 2015-02-10 at 4.57.20 PM.png
Baron; thank you for illustrating the problem so clearly and thoughroughly.

S3D does a lot of nice stuff, but this issue is pretty annoying such that I've started looking at other slicers.

I don't always have the luxury of adding a chamfer, especially if it's someone else's model. I don't think I should have to. The example model I showed was from Thingiverse, but sometimes I get customer works too. I also don't think I should have to alter a model to make it work when it works much better in other slicers.

bot
Posts: 39
Joined: Mon Feb 16, 2015 6:46 pm

Re: Better top surface stitching & bosses on top layers

Tue Feb 17, 2015 9:54 pm

While I appreciate that there is a suitable workaround for some projects, and that is inserting a process to 100% infill where needed, but can this please be addressed? :D

Return to “Feature Requests”