Strange patterns when printing with spiralize

First of all: must admit that I am extremely satisfied with my LulzBot Mini. Rock solid and nearly all prints perfect from the first one and so on.

But now I have a couple of questions regarding some prints I have done with the spiralize setting in Cura.

I’ve tried to print a vase that I downloaded from Thingiverse and a lamp shade I designed in Fusion. Printing in HIPS. I base my prints on the LulzBot Mini setting for HIPS High Quality in Cura. But add the “Spiralize the outer contour” and uncheck “solid infill top”.

I have two questions. Both the vase and the lamp shade have the same problems:

  1. The outer stripes that you can see in the image. What are causing them? They are not in my model.

  2. The very uneven top layer inside the shade. I have seen that in other prints as well. Both with HIPS and ABS. A very uneven top layer. Do you have any ideas why this is happening?
    2016-03-11 17.23.54.jpg
    2016-03-11 17.29.56.jpg
    2016-03-11 17.23.00.jpg

In a way it is rather pretty, but I understand your frustration since you didn’t expect it.

Have you checked the layer preview in Cura? Does it show the printhead making any moves that would account for those lines? The shape isn’t that complicated but it isn’t simple either. Have you tested with a simple cone or cylinder to see if there is the same effect?

Can you link to the thingiverse file? I’m curious and interested to see if I have the same result. Running it on two printers should be informative at least.

I tried ‘spiralize’ once for a six sided vase, but after I saw how it was printing seemingly random short little strips back and forth on the layer to spread the heat around, I never tried it again. :cry:

What was your print speed, wall thickness and layer height? Looks like there was serious vibration, at least for the top layer!

I’ve now published the model here: http://www.thingiverse.com/thing:1411441

I printed with:
Layer height: 0.18 mm
Shell thickness: 1.0 mm
Bottom thickness: 2.0 mm
Print speed: 50
Printing temp: 240
Bed temp: 110
Spiralize the outer contour: ON
Solid infill top: OFF

I tried to print this one with the same settings: http://www.thingiverse.com/thing:687931

And got similar stripes there as well.

Is it really printing one continues spiral? On the last picture it looks like there are shifted start-end points at the scars.
1mm line with is quite high,it may work but for a test i would recommend to set it to 0.7mm.

I had a short look at the gcode created by the spiralize feature of Cura, I think it’s broken. Here is a sniplet of what it is doing:

G1 X207.024 Y172.236 Z17.665 E5514.06029
G1 X205.781 Y174.851 Z17.666 E5514.13974
G1 X204.468 Y177.383 Z17.667 E5514.21801
G1 X203.064 Y179.875 Z17.668 E5514.29650
G1 X201.569 Y182.323 Z17.670 E5514.37521
;LAYER:97
G0 F10500 X201.606 Y182.315
;TYPE:WALL-OUTER
G1 F1500 X200.044 Y184.688 Z17.671 E5514.45316
G1 X198.398 Y187.005 Z17.672 E5514.53115
G1 X197.424 Y188.336 E5514.57641
G1 X197.021 Y188.802 Z17.673 E5514.59331
G1 X197.004 Y188.832 E5514.59426
G1 X196.706 Y189.167 E5514.60656
G1 X196.485 Y189.375 E5514.61489
G1 X195.923 Y189.851 E5514.63510
G1 X194.948 Y190.570 Z17.674 E5514.66834
G1 X193.036 Y191.957 Z17.675 E5514.73316
G1 X191.118 Y193.265 Z17.676 E5514.79686
G1 X189.145 Y194.528 Z17.677 E5514.86114

As you can see, there are G0 (travel) moves at layer shifts that shouldn’t happen in spiralize mode as there are no layers in the original definition. There are also G1 print moves without Z, which means it is printing in one plane at this steps. This may be due to the Z increment being so small that Z is constant for some mm, but it also looks strange…

Here is a sniplet of code generated by spliralize function from Slic3r for comparison:

G1 Z77.409 X93.748 Y199.428 E4996.44220
G1 Z77.411 X91.166 Y198.714 E4996.53880
G1 Z77.411 X90.640 Y198.545 E4996.55874
G1 Z77.411 X90.204 Y198.355 E4996.57588
G1 Z77.411 X89.875 Y198.181 E4996.58931
G1 Z77.412 X89.564 Y197.980 E4996.60268
G1 Z77.412 X89.273 Y197.755 E4996.61594
G1 Z77.412 X88.955 Y197.456 E4996.63167
G1 Z77.412 X88.694 Y197.166 E4996.64575
G1 Z77.412 X88.402 Y196.761 E4996.66375
G1 Z77.413 X88.185 Y196.384 E4996.67942
G1 Z77.413 X87.979 Y195.922 E4996.69770
G1 Z77.414 X87.340 Y193.717 E4996.78050
G1 Z77.415 X87.098 Y192.973 E4996.80870
G1 Z77.416 X86.430 Y190.529 E4996.90007
G1 Z77.417 X85.786 Y188.056 E4996.99225
G1 Z77.418 X85.171 Y185.561 E4997.08493
G1 Z77.420 X84.584 Y183.046 E4997.17807
G1 Z77.421 X84.026 Y180.512 E4997.27164
G1 Z77.422 X83.480 Y177.876 E4997.36876
G1 Z77.424 X82.881 Y174.754 E4997.48340

No travel moves here, just continous printing moves until the top of the model.

Thank you. I tried again with the “Shell thickness” set to 0.5, since that matches the nozzle size on my Mini. And that turned out much better.

So I guess that keeping the thickness close to the nozzle size is a good idea for prints like these. The bottom inside of the bowl also turned out nicer with this setting.

And also: based on the code you show, it seems like Slic3er might be a better idea in general for prints like this then…
2016-03-13 14.22.11.jpg
2016-03-13 14.21.55.jpg

When examining the last print (and the other ones as well) it seems like I have one last problem to solve: The first couple of layers. They are quite uneven on the outside. Any ideas why this is happening?

Given your (still) uneven inner top layer and your squish on the first layers, I guess your are over extruding. You don’t have the errors when the shell-only part begins because then the excess material can simply result in a thicker than expected line width.
You should do an e-step calibration.

OK. Do you have a link to a good description of how I do an e-step calibration on a Lulzbot Mini? (Or are there any quick settings in Cura where I simply can do a test with a slightly more moderate extrusion?

There is a official guide how to do it:
https://ohai.lulzbot.com/project/extruder_calibration/calibration/

But don’t use the 100mm/s as they tell you, a feed speed as used in your prints later will give you a better result. 50-60mm/s should work well.