2.4 Beta - what happens to the spiralize feature?

[berndjm 16]

Just noticed, that I get a good visible Z-Seam on a spiralized object (a simple soild cylinder).

Inspecting the gcode explains where the Z-Seams comes from, but why is the gcode generated this way?!?

Here an excerpt from the code around one "layer-change"

 

G1 X84.653 Y88.444 Z4.785 E89.28759G1 X85.095 Y87.794 Z4.787 E89.29991G1 X85.563 Y87.191 Z4.789 E89.31188G1 X86.065 Y86.621 Z4.791 E89.32378G1 X86.621 Y86.065 Z4.793 E89.33611G1 X87.191 Y85.563 Z4.795 E89.34802G1 X87.794 Y85.095 Z4.797 E89.35998G1 X88.444 Y84.653 Z4.8 E89.3723G0 F5400 X88.659 Y84.526G0 X88.659 Y84.526 Z4.6;TIME_ELAPSED:248.010741;LAYER:23G0 X88.444 Y84.653 Z4.8;TYPE:WALL-OUTERG1 F924.4 X89.093 Y84.27 Z4.801 E89.38412G1 X89.792 Y83.913 Z4.803 E89.39642G1 X90.484 Y83.613 Z4.805 E89.40824G1 X91.214 Y83.35 Z4.807 E89.42041G1 X91.956 Y83.135 Z4.809 E89.43252G1 X92.702 Y82.968 Z4.811 E89.4445G1 X93.464 Y82.847 Z4.813 E89.45659G1 X94.232 Y82.775 Z4.815 E89.46868G1 X95 Y82.75 Z4.817 E89.48073G1 X95.768 Y82.775 Z4.819 E89.49278

 

As you can see Z increases smoothly to 4.8 than makes a move back to 4.6 and to 4.8 again than continue with the smooth incresement ... the backmove is one layer-height (0.2)

This behavior repeats on every "layer-change" ....

[ultiarjan 1,223]

And on top of that I think also the ability to get a solid bottom is broken ....

[bagel-orb 61]

Cura 2.5 will handle spiralize a lot better!

Thanks to Mark Burton.

[burtoogle 514]

Hi,

Yes, you can expect the spiralize to work quite a lot better in 2.5. As a test, I have just printed out a randomvase_2 in PETG and it looks stunning. Furthermore, it's completely watertight (you can actually use it as a vase!)

I would be grateful if when the 2.5 betas appear (or if you can build from the git repo) that you try out whatever models you wish to spiralize and if there any problems post here or better still open an issue at https://github.com/Ultimaker/CuraEngine

Cheers,

Mark

[ultiarjan 1,223]

Yes, you can expect the spiralize to work quite a lot better in 2.5.

 

You made me curious, what is it you changed to improve it ?

[burtoogle 514]

You made me curious, what is it you changed to improve it ?

The two main differences are:

1 - it now ensures (or at least tries to!) that the start point of each layer follows on from the end of the last layer in the direction of the spiral. Should be no more gaps between where one layer ends and the next starts.

2 - to avoid a kink where one layer ends and the other starts, it now smoothly blends the x/y coordinates of each layer with the coordinates of the previous layer.

As I said before, I am hoping this will get some good testing and I will receive feedback (both +ve and -ve).

Cheers,

Mark

PS - I don't actually have a big interest in the spiralization function but it's always nice to fix problems and make improvements. I have some other contributions that have been accepted into 2.5 that I am far keener to use. That's another story...