Will Cura get *real vase mode* back?

[greens83 5]

I've started a new thread for this since I don't think its the same issue as discussed here. Thank you to all on this thread. Having the X,Y, travel/wobble solved is great. Its such a huge help having the Ultimaker community! 

[greens83 5]

 

[GregValiant 1,250]

@greens83 I downloaded that model.  Cura shows errors and the hollow shape means there is a surface on the inside that faces air, and a surface on the outside that faces air.  Those sorts of models aren't suited to Spiralize.

Your Gcode file shows the XY shuffle that is creating the seam.

 

After repairing the model in MS 3D Builder I get a true spiralize toolpath.  Here is a preview of the repaired model in Cura.  You see that it has only an outside face and is solid top-to-bottom.

 

This is in Cura X-Ray view.  Your model is on the right and the repaired model is on the left.  The red areas indicate model errors.

 

Open the attached gcode file in Cura.  When I first slice it the preview appears to show a z-seam.  Previewing the gcode in either AutoCad or in Cura shows that those shuffle moves aren't actually there.

 

 

MS3D_bezelpushfit2.gcode

Edited April 1, 2022 by GregValiant

[greens83 5]

HI @GregValiant. Once again thank you for your input! Really appreciate you taking a look at this. I'm confused though. I made the geometry a solid/water tight volume and still got the same problem:

 

 

 

Could you share the print profile/settings you used?

 

 

[GregValiant 1,250]

The 3MF is with my regular profile, No bottom layers, Spiralize turned on.

Slice it and look at the preview.  You will see what appears to be a seam.

Save the gcode file, clear the build plate, and open the gcode file in Cura.

No seam, and that is how it will print.

 

GV_bezelpushfit2.3mf

[greens83 5]

@GregValiant You are my hero! Its extremely useful to know that the preview is essentially wrong and to see what the toolpath will be you have to import the g-code after slicing.

 

I don't think it was the fact the model showed errors (as an open surface not a solid) that was the issue though. I have managed to get it to to slice perfectly as a true spiral as the original stl.

 

Copying your print settings over to my UM2 profile has made the true spiralise work even with the surface model. I'm not 100% sure which one made the difference. Its either the top/bottom thickness up from 0 which I had (even though its still set to 0 layers) or its the smooth contour. I had assumed from the preview that smooth contour wasn't fixing it though. So knowing it just doesn't preview properly is a game changer.

 

Anyway whatever it was I would not of got there without your help. Thank you. May the fishing gods smile on you! (I think I read you were going fishing once? I'm also an angler!)

[GregValiant 1,250]

I may be a fisherman but lately I haven't been a catcherman.  I do try to make up interesting lies though.

 

BTW the gcode file I produced has 6913 Z moves from the start of layer 0 to the end of printing.  That's about 200 Z moves per layer.

[greens83 5]

I always say 'that's fishing, they don't call it catching' !

That's a lot of z moves! Long live spiralize.

 

Thanks again.

[gdog 4]

I happened on this post as I was planning to undertake my first "shell" print with the vase / spiralize mode.

 

After reading up on this thread and deciding it would be a lot of trouble to figure out how to deal with the issue, I did the print anyway using an older version of CURA, specifically 3.6.0 (running on Linux) and it printed without this seam business.  So if anybody is trying to see when this corruption got into the GCODE generator, it had to be after that.

 

Now another setting in CURE strikes me as possibly implicated here, and that is the setting to hide the z-seam.  Ironically that setting is supposed to work by moving the shifting the start of each layer a bit so they don't all line up.  It's just a shot in the dark, but I could suppose a situation where that seam correction is making trouble by overseeing the the spiralize algorithm and inserting in little "hop backs" when the z-height increases by 1 layer.  In the case of a spiralized object, that would happen around the same place every time, so you would actually cause a seam to be produced.  Just an idea...

[greens83 5]

Hey Gdog, I think Ive finaly got it all sorted now thanks to gregvaliant. 

 

The thing that I think needs fixing the most is the preview that shows a seam when you have smooth outer contour selected. 

 

Here's the humdinger, that seam doesn't appear on the print, you do need smooth outer contour on to get rid of the seam.

 

I disregarded the smooth out contour on the basis of the preview. 

 

There are other settings involved but thats the problem that confused me the most.