Random cavity in 5.7.0

[PBG 0]

Ok, so Cura is adding this strange cavity to one of the four models after slicing.

 

The cavity has walls but no skin, and also adds infill but only on the layers where infill exists elsewhere.

Sometimes it exists from layer 2 and up, sometimes in just the upper few layers, but only on the front right part. (EDIT: Any part oriented in that direction)

It always appears when the part is oriented in the 1st direction and moving it slightly in the XY directions does not fix it.

It does not appear if the part is rotated.

 

Any ideas what could be causing this?

CE5_Power rack Bracket 1 ERROR.3mf CE5_Power rack Bracket 1 ERROR.gcode CE5_Power rack Bracket 1-ERROR Reoriented.gcode

Edited May 23 by PBG

[Carbon 28]

This will be a "Floor" in the model itself....Have you checked the part?

Edited May 23 by Carbon

[PBG 0]

I'm not sure what you mean. I made the part in Fusion360 and they are all the same, but this only occurs in that one orientation.

Also, just as an update, I got it to slice properly by moving that part beyond a certain threshold on the XY axis (which did not work earlier). Moving it less than that causes the cavity to change shape and sometimes appears beneath the topmost layer.

[Slashee_the_Cow 382]

@PBG if you could post a Cura project file .3mf, not just the model (set it up in Cura then go to File > Save Project) that would really help. The model slices fine for me so it could be your settings.

[Carbon 28]

This is what I can see.....

 

[Slashee_the_Cow 382]

7 minutes ago, Carbon said:

This is what I can see.....

 

Is that from the model file or the gcode? Because the gcode will show the problem but it's almost impossible to work backwards from it to figure out why it happened.

[PBG 0]

I attached the .3mf in the op. I can provide the .stl as well if you want, but there's no anomaly in the model itself.

[Slashee_the_Cow 382]

18 hours ago, PBG said:

I attached the .3mf in the op. I can provide the .stl as well if you want, but there's no anomaly in the model itself.

The .3mf you posted is just the model (though it has been in Cura) - did you go to File > Export with the model selected? We need the file you get from File > Save Project.

[PBG 0]

Sorry, you're right, that was from File > Export.

I've recreated the setup here and the error still occurs, but in doing so I think I've found the culprit. I'm printing with a 1mm nozzle, and having the line width set to 1mm is what appears to cause this issue.

CE5_Power rack Bracket 1.3mf

[Slashee_the_Cow 382]

4 hours ago, PBG said:

I've recreated the setup here and the error still occurs, but in doing so I think I've found the culprit. I'm printing with a 1mm nozzle, and having the line width set to 1mm is what appears to cause this issue.

I can't figure out why it only happens in one section, because looking at the mesh it seems pretty damn symmetrical, but your line width isn't the root of the problem.

Edit: There's a better fix down below, with the pictures.

Set Top/Bottom > Skin Removal Width to 3.8mm.

 

For some reason it thinks there's a bit there thin enough that it's worth taking out and replacing it with infill, and since your infill density is so low it wouldn't generate any in that area.

 

Edit: Still no idea why and this just makes me even more 🤔 but we start with this:

Use the mirror tool and flip it on the X axis:

Then use the rotate tool and rotate it 90° around the Z axis:

Bam! Slices fine.

Now if only I could figure out why... 😕

Edited May 25 by Slashee_the_Cow
added bit about transforms

[PBG 0]

I'll mark that as the solution in case anyone else has the same issue.

It's a strange one for sure, seems to follow that hole around. Occasionally, one random skin line would also be super thin for no reason, though that could just be a graphical glitch rather than the actual gcode.