Bad Slicer settings? what do I have to change?

[dorsai3d]

I have confirmed it was intentional gaps. The slicer was trying to apply top and bottom layers to the arching walls of the single perimeter walls. These partial top and bottom layers were just that. Partial leaving entire section of a layer empty when it would then rise for another layer and have no support under the unprinted sections of the partial layers. I removed the top and bottom layers to seperate processes and the errant partial supportnlayers were eliminated. Ie a nasty bug in the slicer but relatively easy to fox sinxe tou guys pointed me to what was causing it.

I think you might not be understanding what's happening here.... the "partial layers" that your video shows are only top/bottom solid layers under/over where the print is sloped in that specific area. This was being printed in addition to your one perimeter, not instead of.

Your problem is most likely that you're trying to print one perimeter only at too high of a layer height and not getting good layer adhesion because of it. This occasionally manifests as a gap where the filament curls inward because you're laying a cylinder on another cylinder and it's off center, so it's prone to not staying exactly put. Lowering the layer height will force the offset between print lines to be smaller as well as flattening each layer's perimeter more.

[nerys]

No. I watched it increase z after laying the incomplete layer where it applied a partial incomplete top layer.

If you lay a partial layer and then increase z you will have gaps 100% of the time. It wont matter what your layer height is. You will just have different sized gaps but gaps you will still have.

It lays a solid layer

Then it moves up the z axis

Then it lays a partial top layer

Then it moves up the z axis

Then it lays a solid layer

Gaps.

Watch my second video. I explicitly point out the missing layer the nozzle never goes their. Then you can watch it lay filament over the void left by the incomplete layer. If it built a top layer and then layed a solid line on that same layer i would not be having this problem.

The problem is that the slicer shows solid layers but the gcode is telling the printer to treat the partial top layer as a discrete layer.

My layer height is .25mm with a .4mm line width (confirmed by measurement the printer properly extrudes this)

[dorsai3d]

Nope. Sorry, but gcode doesn't lie. Your unmodified factory file above (that you say you're printing) gives the following with all the non-z moves stripped out:

; layer 637, Z = 159.25
; outer perimeter
G1 Z159.250 F1002
G92 E0
; solid layer
G92 E0
G92 E0
; layer 638, Z = 159.5
; outer perimeter
G1 Z159.500 F1002
G92 E0
; solid layer
G92 E0

So you see, you're printing one full perimeter, then a partial solid layer (for the exact reasons outlined in my previous post) then moving up in z to print the full perimeter.

[nerys]

It creates a partial layer then moves up. VIDEO does not lie. you can watch it do it.

maybe its not supposed to do it but it does.

It needs to create a full layer then move up.

The partial layer followed by a z movement "is" the problem.

Ie how do you account for what is obviously visible in the video.

My issue is not with the gcode but with the slicer making gcode.

It should know better than to generate a partial layer and they full layer.

Worse it does not show this in the slicer. The slicer preview does not show the gap it instructs the printer to make.

IF the gcode is correct they why it is creating an empty layer if not commanded too?

[nerys]

OK came home to the new print and WOW what an improvement. going to get rid of the bottom layer infill since its looks like crap when illuminated and instead go with 3 perimeters then 2 perimeters then 1 inside 10mm to give me the bottom interface strength I need (otherwise the single perim cone snaps off the base plate if you press to hard)

I still have a zipper up top. this one lines up with the WIPE command so I removed the wipe command and retraction during wipe all together. will reprint tomorrow and see if that fixes that (it should preview comes up clean now with no red marks)

the .2mm negative restart retraction modifier worked a treat. ZERO zits.

I still think the first issue is a flaw in the slicer that needs to be worked on but at least now I know of it thanks to you guys I can work around it.

I will post another update of the final model if it works out well.

here is todays video update.

https://www.youtube.com/watch?v=Z4eXvNVGlfE

[dorsai3d]

It creates a partial layer then moves up. VIDEO does not lie. you can watch it do it.

I did. You can pretty clearly tell that the Z leadscrew doesn't move between the perimeter and the slope's solid layer. (https://youtu.be/TteU3l2-Ibc?t=32) Like I said, gcode doesn't lie. Computers are pretty good at doing exactly what you tell them to do. You just need to know what you're telling it to do

[nerys]

Then why is there a gap right where it did not print? the gap is exactly the height of a layer??? ie what am I missing.

if it did not make a layer between the last solid and the "support" it added then there should be no gap. ???

not sure why you cued to 32 seconds that is not where the z shift occurs.?? it occurs at the END of the partial layer when it then does another solid layer.

anyway zipper is gone. took out the wipe no more zipper

one last issue. I am 99% sure it has something to do with the retraction and restart retraction setting.

I have it set for 1.5mm and -0.2mm

I was getting zits. setting restart to -0.10 reduced it -0.20 eliminated them all together. no zits.

but now I have little voids at layer shift. you can see this in the slicer too.

I am printing something else now but my next plan is to lower retraction to 1.0mm and restart to -0.15

do you think that will help? any other idea how to correct?

https://www.youtube.com/watch?v=UtPcDGOoxJw

it is coming out damned near perfect. I mean seriously gorgeous. SOOO close so close.

[dorsai3d]

Instead of using negative restart distance, I'd suggest trying coasting. I've had pretty good luck with that. I've got a direct drive printer using an E3D v6 and am using 1.5mm retract, 0 extra restart and 0.2-0.5mm coasting (depending on layer height: taller layer = shorter coast) and getting almost no perceptible seam. Similar story on an Ultimaker, but with more retract & coast (4mm/1mm) since it's a bowden style.

[nerys]

easy enough to try just print a section of the wall and try.

I was have "end of layer" gaps and coast was causing it so I turned it off.

but maybe it was coast along with something else causing it.

the -restart was to stop new layer "ZITS"

Will coast take care of new layer zits ?

[dorsai3d]

After a fashion - all these settings are related in one way or another. Increasing coasting distance allows you to decrease retraction. Many start/end of layer defects are from the combination of ooze and dwell, which is why wiping and wipe while retract can make a positive impact. Coasting is a bit like wipe while retracting, but allows you to retract less, so when you go to a new layer, you don't sit still unretracting as long (and have less pressure built up in the nozzle to ooze).

It's all balancing, but coast is nice in that it generally has less issue with small parts than negative extra restart distance, I've had some parts where the extruder basically ends up going backwards due to the retract+negative restart, then only extruding less than the negative restart distance. Gotta be wary of that.