Seams in spiral mode

[bondus 2]

I have noted that even if using spiral mode there are seams in the model. They can be seen in the preview and the printed model. 

 

Moving around the model on the build plate will move the seam to different places.

 

It makes no sense that spiral mode have seams.

[burtoogle 514]

Could you please attach the model to this thread so I can take a look? Thanks.

[bondus 2]

The model is https://www.myminifactory.com/object/3d-print-woman-body-optimised-for-vase-mode-84682

[bondus 2]

I sliced a simple cone that I exported from Fusion 360. In that case the place of the seam is always close to where the first travel move enters the cone.

 

 

[burtoogle 514]

Thanks for the model.

 

Firstly, ignore the seams shown in the Cura layer view as it doesn't really do spiral mode correctly. It often shows the seam even when the model shows nothing.

 

I sliced that model using my Cura and I can't see any seam looking at the gcode.

 

This is the Cura layer view, seam is visible...

 

 

And this is the gcode visualised in S3D, seam not visible...

 

 

However, it may be that UM Cura doesn't do such a good job as my Cura so you may wish to try one of my releases.

 

You can find the releases at https://www.dropbox.com/sh/s43vqzmi4d2bqe2/AAADdYdSu9iwcKa0Knqgurm4a?dl=0

 

Please read the README.md file there before using.

 

[bondus 2]

Got a github source link for that? I do not install random dropbox things on my computer.

[burtoogle 514]

The github source links are in the README.md file. You will also need my versions of the cura-build and cura-build-environment repos.

[bondus 2]

Alternative CURA forks, interesting. I had no idea.

 

Please push the fix upstream. I am a professional software developer, but in this case I prefer to stay a user. 

 

But it is very nice to see that open source works 🙂 

[burtoogle 514]

18 minutes ago, bondus said:

But it is very nice to see that open source works 🙂 

 

But only after a fashion because a lot of the stuff that I offer to UM gets rejected which is why my fork is now so different from UM Cura.

[bondus 2]

Poking around in the code I see there is python in there. No wonder that UI is so sluggish and takes random long pauses, just like java but 10 years later.

[burtoogle 514]

Yeah, IMHO, using Python for the Cura front end was a big mistake along with Qt/QML.

[bondus 2]

Printing the same model, different slice, on another printer gives different results. 

First printer, Reprap Firmware (Duet3d)  shows the seams.

Second printer, klipper, do not show the seams.

It's unclear if it's the GCODE is bad or the interpretation in wrong.

Just like burtoogle showed that the CURA and S3D rendering was different.