Infill Path Errors

68 views
Skip to first unread message

Daniel Wade

unread,
Jul 30, 2014, 12:45:56 PM7/30/14
to kisslicer-r...@googlegroups.com
Hi all,

I've been running into issues when slicing . My infill paths are all good, except for whatever reason at a certain layer height, the path goes bonkers and disappears or is replaced by support. I've tried repairing the STLs as well as messing with all sorts of settings, but to no avail. Anyone have any idea what is going on? 

The attached image is of 3 parts, the two smaller rectangles are identical, one was repaired, one wasn't. This happens fairly consistently and has royally screwed up a few of my prints. Any help would be greatly appreciated.




Capture.PNG

PenskeGuy

unread,
Jul 30, 2014, 10:05:55 PM7/30/14
to
On Wednesday, July 30, 2014 9:45:56 AM UTC-7, Daniel Wade wrote:
The attached image is of 3 parts, the two smaller rectangles are identical, one was repaired, one wasn't. This happens fairly consistently and has royally screwed up a few of my prints. Any help would be greatly appreciated.

OK, I've stared at those rectangles for quite a while, now, and don't see where Infill is being replaced with Support or that the two are different from each other. What am I missing?

Going out on a limb without having further information, if you're referring to the light yellow zig-zags, that would probably be Sparse Infill that is a result of Wall Thickness making up for lack of enough Loops.

Daniel Wade

unread,
Aug 1, 2014, 12:05:05 PM8/1/14
to kisslicer-r...@googlegroups.com
Ah , OK I can see your confusion. I should have included more pictures, which I will do now. Essentially those rectangles are solid. The layers above and below have standard straight infill paths. But for whatever reason, no matter what I'm printing, about 80% of the time the entire infill layer disappears and is either replaced with support (not shown) or I end up with the weird results shown above. 

Does that make a little more sense?

Thanks,


On Wednesday, July 30, 2014 7:05:55 PM UTC-7, PenskeGuy wrote:
On Wednesday, July 30, 2014 9:45:56 AM UTC-7, Daniel Wade wrote:
The attached image is of 3 parts, the two smaller rectangles are identical, one was repaired, one wasn't. This happens fairly consistently and has royally screwed up a few of my prints. Any help would be greatly appreciated.
inproper infill.PNG
Proper Infill.PNG

frozen...@yahoo.com

unread,
Aug 1, 2014, 12:22:19 PM8/1/14
to kisslicer-r...@googlegroups.com
Can you post the stl?

Daniel Wade

unread,
Aug 1, 2014, 1:34:08 PM8/1/14
to kisslicer-r...@googlegroups.com
Sure thing.
Top_fixed.STL

funBart

unread,
Aug 1, 2014, 2:26:43 PM8/1/14
to kisslicer-r...@googlegroups.com
Hi,
can you try the repaired stl, as attached?
Please report back if that solves the issue.
Bart
Top_fixed rep.stl

Daniel Wade

unread,
Aug 1, 2014, 4:08:05 PM8/1/14
to kisslicer-r...@googlegroups.com
I sliced the repaired file and it didn't solve the issue. I added in the very first version for additional comparison. The error nearly always occurs around z = 3.3 and 3.4. The lower rectangle is the repaired STL, the upper is the original. The images are labeled according to the Z height (also seen in the images)

Any other thoughts what could be causing this or any fixes?

Best,
Z 3.3.PNG
Z 3.4.PNG

funBart

unread,
Aug 1, 2014, 4:34:46 PM8/1/14
to kisslicer-r...@googlegroups.com
Unintentionally emailed directly. For those who are curious, the emails are pasted below.

Bart
I'm not sure, but isn't that the result of the 'skin thickness' on the style tab?
You define 4 loops of 0.5mm=2mm. But Kisslicer is calculating that thickness in 3 dimensions (so also 3D diagonally), so on some spots the thickness is less than expected and KS adds the zigzag solid infill. (it should give 10 layers of 0.1mm at the bottom and top, because of the 2mm thickness)

Can you try with 1mm thickness?

It puzzles me that the two stl's are resulting in different paths. Are you sure the repaired one isn't giving other results?
Bart

Daniel
hmm. I re calculated with 1mm thickness with 2 loops and the problem is gone, however this piece is being used in a liquid application, and I've had issues with the porosity of the structure with 1mm skin thickness. I''ve had successful 2mm skin prints without this issue before (even on the exact same part with only small differences in features). I'll toy around with the skin thickness settings since that seems to be partially the source of the issue. Thanks for the help!

Bart
Just put the skin thickness 0.2mm under the path width x amount of loops.
I have: 3 loops x 0.4mm = 1.2mm and thickness at 1mm, so there are not so much additional solid areas generated.
Bart
Reply all
Reply to author
Forward
0 new messages