index.md

Upgrading the Prusa mk3 and its influence on wall quality

⬇️ Images below! ⬇️

I upgraded my Prusa mk3 R2 build with different axis and extruder. I printed different models before and after the upgrade to see how it affects the print quality. While I did also print all-in-one test, I wanted to focus on wall quality.

Limitions

There are many things you can tweak on a 3d printer. Both hardware and software. I tried to stick with stock slicer settings as much as possible. You will also see some minor ghosting on the x axis on my stock prints, just ignore them. They could have been removed with belt tension improvements. But that is a pita on stock mk3.

Models

• 602 cube
• Single perimeter 40x40 cube
• 3d test v3
• Overhang and bridge test
• Benchy

All stl files are on this github repo.

Upgrades

• LDO blower fan (slight more powerfull)
• LDO Pancake Extruder Motor - runs cooler, even with stock current.
• Sunon hotend fan - vastly better cooling.
• Taurus Y
• Taurus X
• Taurus Z
• Skelestruder

Both stock and skele had nozzle X and a nozzle sock. Stock also had a new nozzle fan thingy from RH_Dreambox.

Slicer settings and firmware

I used slic3r PE 1.42.0-beta1 with stock 0.20mm SPEED MK3 profiles. Any per model changes are documented below. For the skele prints I edited the k value from 30 down to 20 and retraction from 0.8 to 0.6.

• 602 cube:
  • seam position: rear
• 3d test v3:
  • Detect thin walls
• 40x40 cube
  • perimeters: 1
  • Solid layers top: 0
  • Solid layers Bottom: 1
  • Infill: 0
• Ghosting / Speed test:
  • Speed: 120 mm/s (All of them)
  • Acceleration control: 1250 mm/s^2
  • Slow down if layer print time is below: 5
• fan_test_-_Part_2b
• Brim width: 5mm

Firmware was 3.4.2 before and after.

Filament

To eliminate filament as a variable, I used Prusament for some before/after prints. I could not see any difference in Prusament vs cheap local, with focus on external perimeter qualty.

• Prusament PLA Vanilla White, 210C (or 215C)
• Prusa Silver (Not Prusament, the one I got with my mk3), 200C
• Pirate Black - 3DE Premium PLA, 200C
• Some blue 3DE Premium PLA, 200C

Results

• 602 cube: Both in real life and on images I can't see any differences. Maybe some slight difference on the top? Side by side in good light
  • Blue 602 cube - (slightly lower light angle)
  • White 602 cube - (slightly lower light angle)
  • Black 602 cube - (slightly lower light angle)
  • Silver 602 cube - (slightly lower light angle)
• Single wall 40x40 cube: On skele (right) you can see some slight pattern, geoing on a 30 degree angle slighty upwards. But the stock also has some kind of pattern, harder to see on the images. Going steep down. Otherwise I can't tell a difference.
  • Blue wall
  • White wall
  • Black wall
  • Silver wall
• Overhang and bridge test: Skele with LDO blower is slighty better at overhangs, but at such a steep angle you want to avoid models like that anyway. I see no difference at bridging.
  • Comperison slider
  • Full image (!large filesize!)
• Ghosting / Speed test: I printed this, but decided it's not a good test. Had some minor ghosting on both skele and stock. But that should be fixed with messing with the belts.
• 3d test v3: I don't see a difference
  • Image (!large filesize!)
• Benchy: both look fine.

Conclusion

Shit man, IDK. Skele++ looks cool and thanks to Taurus I can change my belt tension without any tools. Many people reportet that even just sunon fan on stock makes a huge difference to them. I printed with low ambient, ~16C, maybe in hotter climate you need more fan power? I think we need more people to try and document mk3 upgrades.

Thanks

Thanks to all the peeps fromt he 602 Wasteland discord. Especially Greg, the bear, and Jltx, first to his name.