Announcing Rosalind, a parametric CoreXY 3D printer designed in OpenSCAD and licensed under the GPL

Hi all, I'm new here and I hope I'm posting in the right forum.

I just completed a first working prototype of rosalind, a CoreXY printer. The code is still a mess, but in the spirit of "release early, release often", I decided to not wait anymore.

Rosalind designed in OpenSCAD with the exception of the print fan funnel, which I had designed in blender for my previous printer and didn't re-do. My prototype is made from 40x40 aluminium extrusions and printed parts. No machining is required except from cutting the extrusions exactly perpendicular. In theory the printer should work with smaller extrusions but I didn't get to test that yet. My prototype uses a cheap-o KFB board awaiting for smoothieboard 2.0, 2 PSU's (24 and 12v) and 0.4Nm 0.9° nema 17's however nema length and electronics are your choice.

For the time being only E3D v6 and clones are supported, in single or dual bowden configuration. I used

Of course I'd like a helping hand in making this a truly configurable 3D printer. Everything is derived from the values set in `general-values.scad` but I haven't really tested for any other configuration except the default (which is the same as the prototype). I have almost enough surplus stuff to build another rosalind so after I finish calibrating and trying on various configurations, I intend to repeat the construction and make sure that if you just print the parts as configured everything will work as it should. If you build one I'd appreciate some photos.

All files except otherwise noted are GPL v2 or later. I have tried to use only free software, if you find anything proprietary or CC-NC in this repository please report.

The name of course is after Rosalind Franklin

A special shout out to schlotzz for the extruder drive and nophead for Mendel 90 from which I got various models.

Without further delay, here are some images:





Source code is on gitlab: [gitlab.com]

Edited 1 time(s). Last edit at 12/04/2017 12:50PM by qwazix.

The version I published above had issues with backlash on the x-carriage. The aluminum rods/bronze bushings combination had issues as the aluminum rods where not exactly ⌀10mm and I didn't want to do any hacks to fix it so I decided to re-design the extruder carriage.

The next version didn't have bronze bushings, it was one piece and the plastic was sliding directly on the aluminum rods, with some white lithium grease. Backlash was fixed but sliding the extruder was considerably harder than sliding the gantry so circles started getting egg-shaped. After increasing the tolerance and drilling through the hole without much improvement I decided to change direction.

I created a third version of the extruder carriage in classic linear-bearing-style so that I would first iron out all other issues and then experiment with novel motion systems. I've ordered some carbon fiber tubing to try too. (Thanks tech2c for the idea). This works nicely. I still cannot print very fast however and I suspect it's because of the extruders not being able to push enough filament. I suspect my extruders need tightening but it might just as well be that there isn't enough torque. (I'm using 0.9deg steppers)

Despite most designs now using direct drive extruders the people over at duet3d state that you need 75Ncm torque. My motors don't produce such torque so I was searching for an open-source printable geared extruder designed in OpenSCAD that fits. This turned out to be too specific but I found the Greg's B'Wadestruder which fits the bill except for not being designed in OpenSCAD. However it is designed in blender, so that one can still modify it using Free Software tools and it's mount is a separate object so by re-designing just that in OpenSCAD I can incorporate it to the parametric design without many issues. I haven't tried it yet but the option is in the repo.

I also started working on the enclosure of the LCD screen which I shamelessly stole from Josef Průša

I got a Legends Pack from E3D as the previous combination of original/chinese v6's didn't work very well as a dual extruder, not to mention the huge size of the carriage. Still haven't got around to mounting it though. The flat design looks much easier to incorporate than the v6.

I also ordered parts for another rosalind, the assembly of which I plan to record on video/photos and publish.

That's all the progress for now.

Edited 1 time(s). Last edit at 01/13/2018 11:34AM by qwazix.

Quote
qwazix
Despite most designs now using direct drive extruders the people over at duet3d state that you need 75Ncm torque.

Just to be clear, this is a recommendation to ensure that you are unlikely to be short of motor torque when extruding PLA at a reasonable printing speed, without having to increase the temperature unduly because that increases stringing. You may get away with less than this, especially if you are printing ABS which is less viscous. One of my printers is successfully printing Rigidink PLA+ using an ungeared extruder with a motor rated at about 30Ncm, although it's a SCARA printer with a low maximum print speed. OTOH one printer design that used a 26Ncm motor in a direct drive extruder had to be changed to use a 3:1 geared extruder because too many users found there was insufficient extrusion force for PLA.

Many extruder designs, including the Titan, use about 3:1 gearing. This allows a smaller motor to be used.

Edited 2 time(s). Last edit at 01/13/2018 03:17PM by dc42.

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].

Thanks,

I use the ungeared design on my old Mendel (and the new machine) with ABS and PETG successfully for a while now but I never did go over ~30mm/s printing speed because the old printer couldn't cope with higher speeds anyway (I assume it's not rigid enough) and the print quality suffers.

Rosalind seems to be able to handle the speed much better but when I do set it to ~60 there's obvious underextrusion so I thought maybe the extruder can't push the filament fast enough. I'm printing with PLA btw.

I also contemplated using a planetary gearbox but this would increase the length of the motor/gearbox assemply and would protrude out of the printer footprint so I decided against it. The plans are to enclose the printer in acrylic sheets at some point.

You better go with a geared extruder and a Nema14 motor. I use RC gears, brass plus POM, cheap and small. For the Titan they made the same choice.
Small and light so direct, no bownden and its issues.
The Wade one found on Thingiverse were fine 5 years ago but are too heavy, bulky and use printed gears.

Edited 2 time(s). Last edit at 01/15/2018 10:31AM by MKSA.

---

"A comical prototype doesn't mean a dumb idea is possible" (Thunderf00t)

The extruder carriage is modular and I will do what you suggest as an option, later on. For now, priority is to make it reliably print with two nozzles (thus bowden) because I desperately want to print buildings (which are unprintable without soluble support material, due to overhangs).

I would very much like to use something like titan but I want to build a fully open source machine and while titan assembly drawings mention GPL, the actual manufacturing drawings (or models) are nowhere to be found. I'd rather not design my own extruder right now, though it's something I might do one day.

If you have any idea for extruders give a shout.

As someone proposed it somewhere, use a Bowden for the support filament and a direct extruder fro the "build".

---

"A comical prototype doesn't mean a dumb idea is possible" (Thunderf00t)

Quote
qwazix
The extruder carriage is modular and I will do what you suggest as an option, later on. For now, priority is to make it reliably print with two nozzles (thus bowden) because I desperately want to print buildings (which are unprintable without soluble support material, due to overhangs).

I would very much like to use something like titan but I want to build a fully open source machine and while titan assembly drawings mention GPL, the actual manufacturing drawings (or models) are nowhere to be found. I'd rather not design my own extruder right now, though it's something I might do one day.

If you have any idea for extruders give a shout.

Here's one you can print yourself [github.com]

Yay!

Many thanks. This will probably get to be the default extruder for rosalind. First step is to print one though.

Quote
MKSA
As someone proposed it somewhere, use a Bowden for the support filament and a direct extruder fro the "build".

This is not a bad idea. I'll have to see if it fits with the e3d chimera.

Edited 1 time(s). Last edit at 01/17/2018 07:21AM by qwazix.

I just installed an E3d chimera on the rosalind and designed a cooling solution (repository push coming soon). Changes are live.

Here are two crude videos of the machine in action.

[www.youtube.com]
[www.youtube.com]

I've decided to stop improving this particular instance and focus on making the design truly reproducible. To that end I will start on a new one and document the process step by step.

Edited 1 time(s). Last edit at 03/15/2018 03:45PM by qwazix.

It's been a while but I finally have some updates.

I've started a wiki page which I'm updating slowly but steadily towards a nice detailed build instruction manual.

At the same time I'm recording a video log of the build process of Rosalind #2 which is on PeerTube and soon on YouTube.

Episodes #1 and #2 are now on YouTube. Stay tuned for more.

[www.youtube.com]

Hi, i am printing many parts of your nice project...

Is there an update for the Fan-Nozzle Funnel at the Extruder?
This one has too less airflow!

Any other updates of Rosalind?

Edited 1 time(s). Last edit at 01/20/2022 11:10AM by mega-hz.

The funnel for a 40mm blower fan is this one [gitlab.com]

You can assemble it like here: [gitlab.com]