Artifact ONLY on #3DBenchy...

Breadbin · 23 Apr 2020, 02:03

Hello all,

I built a custom cartesian printer and I love it. It prints just as well as my Prusa i3 MK3S. So far I've only had one problem with it:

It can't print 3D Benchies. Every other file prints beautifully.

At a height roughly level with where the deck (the flat part at the back? I'm not a boat expert) starts, the layers seem shrunken in the xy plane. The surface "jumps" about 1mm back out to its normal position and the jolly torture test is fine from the artifact up. I though it was a problem with Prusaslicer but I imported the gcode into an online gcode visualizer and all seems fine. It's not filament, either, because 4 different benchies (same STL)I could easily be wrong, but through deductive reasoning it seems that my problem's root lies with the board, a Duet Maestro. This issue doesn't prevent my printer from running, but it's an odd situation and I'd like to fix it. Any and all help would be appreciated.

Here are some photos of the problem, as well as a photo of the machine with an artifact-free print.
https://photos.app.goo.gl/hwchEmW9vmWFe3TS6

Thanks!

Phaedrux · 23 Apr 2020, 03:57

@Breadbin said in Artifact ONLY on #3DBenchy...:

I though it was a problem with Prusaslicer

I think it actually is an issue with prusa slicer. I think you're experiencing some form of the dreaded benchy hull line artifact.

https://help.prusa3d.com/en/article/the-benchy-hull-line_124745/

@Breadbin said in Artifact ONLY on #3DBenchy...:

It can't print 3D Benchies. Every other file prints beautifully.

Honestly that's the best kind of problem to have. Unless you need a fleet of tiny boats, I suggest you count that as a win and gleefully print on.

In all serious though, trying to understand why it happens will make you a much better printer operator.

Breadbin · 23 Apr 2020, 04:03

Interesting. Thank you! I'm curious why the artifact doesn't show in gcode previews... You're right. With a collection of ~40 benchies already, I don't really need more, so it's no big deal.

Phaedrux · 23 Apr 2020, 04:08

Because the gcode preview doesn't have to deal with the messy physics of melted plastic. The gcode is correct as far as it goes. But that's only because the slicer is also mostly unaware of the messy physics of melted plastic. If we had a perfect mental model of how the plastic would behave in physical reality, we could create a gcode command that could produce the desired end result. Unfortunately, we are mostly just guessing, and the way the plastic contracts and curls and expands ect will determine the final shape that the gcode path put in motion.

brisma · 23 Apr 2020, 06:51

Same problem to me with a custom HEVO on PrusaSlicer 2.2.0.
Now I'll retry with Cura 4.6...

dc42 · 23 Apr 2020, 08:11

Looks to me that it has skipped several layers. You could measure the height of that discontinuity from the bottom (or even count the layers), then look in the GCode file to see if the discontinuity is present in the file.

ChrisP · 23 Apr 2020, 08:36

I've never seen an error quite that pronounced in a benchy before... particularly when the rest of the model looks good. I agree with @dc42 that it looks liked skipped layers or something similar. I'd have though with such a step, if this is the cause you could simply measure the height of the whole part and check it's correct.

Out of interest, how fast are you printing - default speed & perimeters?

brisma · 23 Apr 2020, 09:31

The size is correct... very strange

I've a custom HEVO, printed with PLA at 200/60°C at maximum 60mm/s with this config:

 M350 X16 Y16 Z16 E16 I1                            ; configure microstepping with interpolation
M92 X80 Y80 Z400 E860                              ; set steps per mm
M566 X600 Y600 Z120 E900                           ; set maximum instantaneous speed changes (mm/min)
M203 X15000 Y15000 Z600 E1800                      ; set maximum speeds (mm/min)
M201 X1000 Y1000 Z240 E5000                        ; set accelerations (mm/s^2)
M906 X1300 Y1300 Z1300 E1300 I30                   ; set motor currents (mA) and motor idle factor in per cent

Edit:
DuetWifi with RRF 3.01RC9

Breadbin · 23 Apr 2020, 17:25

After reading these replies, I checked my benchies' heights using an odd method, but it seems logical enough. I don't have functional calipers, so what I did is I took 3 benchies from my MK3S (they look great, no errors there) and placed a glass sheet from a delta printer on top to define a plane at the height of a 3D Benchy:
https://photos.google.com/share/AF1QipMrmjvLZslnIE99tQ8AemXcAn1FABP5nRVXFnJq9KL96t-ZWOa-TmzHLPqFEBoXnQ/photo/AF1QipNotYOZJKBRIFohTvDiUGf1Yp45CrJaH7NV9_nf?key=YXN4bTNEVG8zVC1LaGZuTmhvMHBvNVRzd2U5anNR
Then I placed the benchies that show the artifact at the edge of the glass sheet and their heights matched the benchies from the MK3S:
https://photos.google.com/share/AF1QipMrmjvLZslnIE99tQ8AemXcAn1FABP5nRVXFnJq9KL96t-ZWOa-TmzHLPqFEBoXnQ/photo/AF1QipMXgk40WrmIWOj8URm7qpYYA09Ks_Im8JPIOFx0?key=YXN4bTNEVG8zVC1LaGZuTmhvMHBvNVRzd2U5anNR
So I think this artifact is the infamous Benchy Hull Line, and not the result of skipped layers.

Phaedrux · 23 Apr 2020, 18:42

I just checked my son's 3X size benchy and it has a bit of a visible line where the deck begins as well. Benchy Hull Line strikes again.

	M350 X16 Y16 Z16 E16 I1 ; configure microstepping with interpolation
	M92 X80 Y80 Z400 E860 ; set steps per mm
	M566 X600 Y600 Z120 E900 ; set maximum instantaneous speed changes (mm/min)
	M203 X15000 Y15000 Z600 E1800 ; set maximum speeds (mm/min)
	M201 X1000 Y1000 Z240 E5000 ; set accelerations (mm/s^2)
	M906 X1300 Y1300 Z1300 E1300 I30 ; set motor currents (mA) and motor idle factor in per cent