Makergear M2: Cushwa Owl

You’ve seen the overview pictures of the half-scale cushwa Owl earlier, so here are some details…

The front view:

M2 - cushwa Owl - half scale

M2 – cushwa Owl – half scale

The left side view:

Owl - half size - left

Owl – half size – left

The conspicuous vertical lines come from the 0.10 infill honeycomb; there are no visible retraction zippers and the surface is smooth to the touch.

A closeup of the beak shows the crystal-clear drooping filament; a similar effect happened on the downward-pointing feather tips. Generally, this is a sign of too-hot extrusion, but at 165 °C I’m not convinced that’s applicable. It may simply be too much overhang at this scale:

M2 - cushwa Owl - beak detail

M2 – cushwa Owl – beak detail

Overall, it’s pretty good. The config info doesn’t include the external perimeter speed, which I’ve been dialing back from an insanely high value. I think it was 75 for this one, which might be flinging the filament off the edge of the beak below that steep overhang.

The slic3r configuration:

; generated by Slic3r 0.9.8 on 2013-03-28 at 10:28:53

; layer_height = 0.25
; perimeters = 1
; top_solid_layers = 3
; bottom_solid_layers = 3
; fill_density = 0.10
; perimeter_speed = 100
; infill_speed = 300
; travel_speed = 500
; scale = 1
; nozzle_diameter = 0.35
; filament_diameter = 1.70
; extrusion_multiplier = 0.9
; perimeters extrusion width = 0.40mm
; infill extrusion width = 0.40mm
; first layer extrusion width = 0.39mm

No source code, as it’s directly from the STL on Thingiverse.

About these ads

  1. #1 by Jason Doege on 2013-04-18 - 09:36

    Would something designed to cool the perimeter filament when an overhang was being printed be valuable? Maybe a nozzle delivering chilled air to where the filament leaves the extrusion nozzle.

    • #2 by Ed on 2013-04-18 - 12:34

      cool the perimeter filament

      I’m not convinced that’s practical, notwithstanding the profusion of extruder fans. I think it’d be better to explore what’s happening around the nozzle to understand what the variables control; perhaps investigating the effects of extrusion temperature would produce better results.

  2. #3 by Jetguy on 2013-04-18 - 10:09

    Sorry, I just don’t consider the owl to be a great test of a printer. I honestly believe I could make that print on an old MK5 DC extruder Cupcake and done slow enough, you couldn’t tell. Nothing other than overhang is impressive. The high polygon count just makes it so you never get any really fast feedrates and hides any errors anyway.
    It’s popular because I think just about any printer even in bad shape can make it look OK. Not trying to imply anything negative about your machine, just that it’s not a good test.

    This on the other hand, http://www.thingiverse.com/download:138054, is a real torture test. This will determine how well your retraction and filament control really is. Also, interesting to see if the moving bed has any effects shaking the part.

    Here’s my copy http://www.thingiverse.com/make:36313
    Printed with a MK6+ modified (well just adding a 40mm fan to cool the thermal barrier tube) and 3mm filament. That’s what makes retraction work so well. Well, that and Sailfish firmware. Also note, that is Sailfish Replicator G 40R14, with not a single change to the stock profile other than filament diameter.

    Just try printing the head, it’s a neat part anyway, and what I consider the test of a machine. It won’t take much plastic either.

    • #4 by Ed on 2013-04-18 - 12:30

      This [...] is a real torture test

      Zowie! A cylinder head is definitely on the to-do list!

      On the other paw, was that among the first half-dozen things your new shiny new printer cranked out?

      Remember where I am on this: new printer, new software, new goals. It’ll be a while before I catch up with myself…

  3. #5 by Jetguy on 2013-04-18 - 14:31

    Correct, not the first print, but it is in the first prints from the latest versions of firmware and software, both on defaults. That’s my point, you assume there’s tuning involved and I’m saying other than the basic follow the direction for hardware (belt tension, are the nuts and bolts tight), no tuning done. Just justification why I built a machine around the software/ firmware, not the other way around.
    Anyway, there’s apparently another challenge brewing in the forum around this item http://www.thingiverse.com/thing:29938

    • #6 by Ed on 2013-04-18 - 15:25

      why I built a machine around the software/ firmware, not the other way around.

      Seen that way, we’re coming from exactly opposite directions for exactly the same reason! [grin]

      I just saw a nice LED candelabra bulb at Home Depot that would look great inside that spiraling shape…