infill overlap not working

[oppass 0]

Hello,

Any Cura ideas on how to tackle my strange gap problem outside this rectangle hole shell would be greatly appreciated. I really need this part to be 100% solid. (I know, the rest of the infill isn't there either but I can fix that)

Infill overlap #1 is 10%

 

Infill overlap #2 is 50%

 

Result: no difference

 

Using rev.14.07. I tried changing the nozzle from 0.4 to 0.3 but the gap around the hole persisted. (Did give me a much better infill elsewhere tho )

 

Cheers.

 

 

 

[Dim3nsioneer 557]

Do you see the gap in the layer preview of Cura?

If not, then it's rather a backlash problem on the printer itself and you should check belts and pulleys.

[oppass 0]

You're right 3, this has nothing to do with Cura (moderator: can we move this topic to the correct forum? sorry). But I think it's due to material shrinkage not belt tension.

Here is the latest output (in Cura 12.08, not that it matters) with nozzle 0.3 to give more flow and solidity. Also now have 10mm wall vs. 0.8mm yesterday. Here is the key in my opinion: the material is nylon:

 

 

Nylon = shrinkage. Solid nylon = extra shrinkage. The interior rectangle "part" shrinks away from the round outside part on cooling causing these (now more dramatic) gaps.

 

OK, if that is the problem, what might be the solution? I'm using a non-heated garolite platform which I realize isn't helping. Printing at 250C and 50mm/sec speed..but maybe a higher temp or lower speed would help?

 

[gr5 2,167]

I don't think it's shrinkage. I think it's backlash. But possibly not what you think "backlash" means. Shrinkage causes different problems than you are describing.

There's 2 things that can cause backlash on UMO:

1) Loose belts - especially the ones to the motors. Tighten them by sliding the motors down while the 4 screws are loose. And make sure the long belts are tight enough to produce a tone/note of low pitch. But audible (>50Hz).

2) Friction. If the head has too much movement friction then when the motors stop the head hasn't moved quite all the way to the desired position.

In both #1 and #2 above the head isn't moving all the way to the desired position.

#2 is most often fixed by loosening the end caps. The head should move quite easily by pushing the side blocks only with the smallest finger on each hand (push both blocks together).

#2 can also be caused by long belts too tight (crazy huh? backlash can be long belts too tight *or* too loose!).

#2 can also be helped a bit with some light oil (preferably with no additives).

 

[gr5 2,167]

Why did you choose nozzle .3 by the way? And what is your shell thickness? 1.0? .8? I don't understand what the 10mm is about - did you mean 1.0mm?

If you set, say, nozzle to .3mm and shell to .8mm it will print assuming .4mm nozzle and do exact same thing as .4mm nozzle and .8mm shell. But if you do .3mm nozzle and 1.0mm shell it will assume .33mm nozzle and make the passes of ALL INFILL and shell traces .33mm apart (instead of .4mm) and it will purposely undrextrude by 33/40 or equivalent of 82.5% flow - but this is not underextrusion because the traces are closer together so that these 2 changes work together and give you properly solid traces and shell.

 

[oppass 0]

Thanks UU. I will go through your tips on healthy belt tension and lubrication. (would be great if someone could upload an mp3 to youtube maybe so we can hear happy belt plucking music!).

I switched from .4 to .3 nozzle just to follow a 9/16/12 Joergen NYC tip. Willing to try anything! He said then:

"It is a known issue (if you have rules out belt slack and good enough infill/overlap) with cura and slic3r. less so with cura, more so with previous releases of slic3r.

you can try a couple of things: reduce the number of outlines or the wall thickness, to trigger cura to use infill at that point. or increase the wall thickness and reduce the nozzle size until you see an extra inner loop filling the gap.

kisslicer doesn't have this problem, and I can't remember seeing this issue from netfabb, so you could give those 2 a try. "

Will go back to .4mm now since .3mm didn't seem to solve my issue anyway.

Yes, wall is infact 10mm (you can see in photo too) not 1mm. Having super high wall thickness is my way of having concentric circles in some parts and avoid any transition to infill at all. Doesn't work with this geometry tho.

Following up on my own slow printing idea for reduced delamination of the nylon. I started printing at 10mm. So slow... Happened to walk by after only a couple of layers and saw this:

 

Cura seems to build in these gaps which allow/promote the shrinkage problem that I see as the key.

 

UU, I respect that you don't think shrinkage is really the core problem. But how to explain then that the gap was worse when the part was printed more solidly (via 0.3mm nozzle setting)? Thanks.

 

[oppass 0]

"And make sure the long belts are tight enough to produce a tone/note of low pitch. But audible (>50Hz)."

Do your belts sound like 50Hz here? Seems really low to me.

http://onlinetonegenerator.com/

 

[oppass 0]

It seems that there's a Goldilocks number for shell/wall thickness. In the case of this part 16mm was "just right" to merge the two shapes with no gaps.

Here's the results (Cura 14.07, Layer 0.1, Shell 16, Fill 0%, Speed 90):

 

 

 

Thanks 3 and UU for your input. I did spruce up my belts which certainly didn't hurt things, thanks. Hope my log here might help others too.