Weird Constructor / Posts / WeirdCube 3D Printer - Devlog 4 - The Core XY Kinematics

WeirdCube 3D Printer - Devlog 4 - The Core XY Kinematics

This WeirdCube 3D Printer article is about the design, assembly and test of the Core XY motion system.

Lets start off the article with a bit of electronics work. I ordered a few cable shoes to replace the ferrules for an even more secure assembly.

Crimping the cable shoes
Crimping the cable shoes
Mainboard now with cable shoes
Mainboard now with cable shoes

I also ordered a power plug socket with switch. Please don't look too closely, I got the wiring wrong in these pictures, but I found the bug once I checked it.

Power plug socket with cable shoes
Power plug socket with cable shoes

X/Y Precision Rods

I started by prototyping holders for the rods and something to grab onto the bearings.

Y rod holder prototype
Y rod holder prototype

For space reasons I decided to have only one screw for fixing the bearings inside the part. This reduces the necessary forces for precisely mounting the part to the LM8UU bearing.

Y rod LM8UU bearing prototype
Y rod LM8UU bearing prototype
Y rod parts overview
Y rod parts overview

Core XY Pulley Holders

Next up was the task of finding a design for the pulleys. I only have a limited set of screws at my disposal. So all I had that fitted the pulleys with a 5mm bore in the middle were my M5 screws, of which I had M5x10 and M5x20. So I had to design something that holds M5x20 screws. The first prototype was the following, which I later stabilized a bit due to too easy cracking.

Pulley holders first revision
Pulley holders first revision

I designed it that way, that I could use this holder design for the X gantry as well as for the pulleys in the back of the printer.

Pulley holder at the back of the printer
Pulley holder at the back of the printer

I printed it on it's side for the beginning. Later I decided to print it laying on it's back. Either way would be fine after all.

Print direction of the prototype
Print direction of the prototype

After printing I tested the pulley holders, whether it's fitting the belts properly and could be assembled.

First pulley holder prototype
First pulley holder prototype

Unfortunately it turned out, that the M5 screws can be easily overtightened and I got a crack too easy. The following picture shows the problem after over tightening the M5 screws on purpose.

Overtightened M5 screws crack the pulley holder
Overtightened M5 screws crack the pulley holder

For more stability I added pillars on the front of the pulleys too, to get rid of the U shape.

Final design of the pulley holder
Final design of the pulley holder

Designing the X gantry and Y motion system

The following parts would make up the whole motion system. You can see the Y rod mounts, the X rod mounts, the carriage for the X gantry and where the pulleys are going to be located.

Overview of the Y motion system and X gantry holder
Overview of the Y motion system and X gantry holder

I had a finished pulley holder now, and the question now was how to attach it to the X gantry properly.

Next question was how to attach the pulleys to the X gantry
Next question was how to attach the pulleys to the X gantry

I made sure to model also the belts in Blender, so I can see where they would run.

Core XY belt path on the WeirdCube
Core XY belt path on the WeirdCube

I found a primitive way to attach the pulley holder to the X gantry holder part. It was simple and could be done with a M4x10 screw.

Solution of the pulley holders on the X gantry
Solution of the pulley holders on the X gantry

Originally I wanted to load two LM8UU bearings inside one part. Unfortunately 3D printed parts are too inaccurate to hold two LM8UU bearings next to each other without skewing them. The inaccuracies were too big, so that the bearings would bind a bit when pulling them along the rod. I decided to leave out the second bearing and just have one LM8UU bearing in the middle. It turned out to be fine with one LM8UU after all.

Original plan was to use two LM8UU bearings, but that did not work
Original plan was to use two LM8UU bearings, but that did not work

Another problem became visible when assembling it though. The part I so carefully designed cracked a bit too easy under the M5 screws. Partially the 40% fan for printing the PETG was to blame for sure. But I wanted to find a better way to design the whole part.

Cracking on the X gantry holder part
Cracking on the X gantry holder part

Mounting the X gantry to the Y rods was completely overhauled. The new design would use one flat part at the top to connect the Y bearing with the X gantry directly with 3 M4 screws and two M5 screws on the X axis. This design allowed also enough play to assemble the X gantry square to the frame, while also providing enough stability.

Redesign of the X gantry holder and pulley holder mount
Redesign of the X gantry holder and pulley holder mount

The pulley holder would not be connected with the bearing part anymore, but mounted directly to the X gantry profile with three M5 screws, two on the sides and one on the bottom. Here is what the finished assembly looked like:

Final assembly of the X gantry mount
Final assembly of the X gantry mount

After printing out the parts for the two Y rods and the left and right side of the X gantry I assembled everything. I was eager to check if the Y motion would work of if I would get any binding. Problems could have come from a not fully square frame or bad part design. But it would be fine as you can see in the following video:

Test of the Y axis motion

Print Head Carriage, Motors and Frame Pulleys

The only three things that were left to do at this point would be a carriage for the print head. The pulley holders for the back of the frame and a way to attach the A/B motors for the Core XY motion to the frame.

I started out by designing a print head carriage. The belts would attach on the bottom of the carriage right in the middle. With six M3x35 screws I planned to attach everything I might need to the print head.

Print head carriage
Print head carriage

The finished X gantry would then look like this:

X gantry overview
X gantry overview

The rear pulley holders were quickly done:

Rear pulley holders
Rear pulley holders

Also the mounts for the A/B motors were rather quickly done. They would be my first layered part. That means the top would be screwed onto the bottom and the motor with four M3x35 screws. The resulting mount worked well in the end.

A/B motor mounts
A/B motor mounts

A few hours later and all parts were printed and ready for the final assembly of the Core XY motion system.

Finished parts for the Core XY motion system
Finished parts for the Core XY motion system

With all these parts finished, also my 2.3kg spool of PETG filament was finished. I managed to empty 2.3kg of filament for all the prototypes for the WeirdCube.

Finished spool, used for prototyping the WeirdCube 3D printer
Finished spool, used for prototyping the WeirdCube 3D printer

Core XY Belt Assembly And Test

Before the final assembly I made sure to grease the LM8UU bearings properly. For this I had to remix a few LM8UU greasing helpers to fit to my tube of Nigrin grease. Luckily Blender has an awesome add-on for creating threaded screws and nuts.

Blender design of the greasing helper remix
Blender design of the greasing helper remix
Finished greasing helper
Finished greasing helper

Greasing would be quickly and properly done with this helper.

Greasing the LM8UU bearings
Greasing the LM8UU bearings

Next I assembled the print head carriage, the pulley holders and the motor mounts.

Assembled parts for the last step of the Core XY system
Assembled parts for the last step of the Core XY system

I recorded myself assembling the linear rods on the X gantry here:

Also assembling and mounting the tool head I recorded:

After all the parts were mounted onto the rods I started running and tightening the belts.

First belt attached
First belt attached

After running both belts the motion system was finally done.

Finished Core XY motion system on the WeirdCube
Finished Core XY motion system on the WeirdCube

I attached the motors to the mainboard, configured a Core XY motion system in the Marlin firmware and tested it next with my dial gauge. You can see me testing the Y and X motion in the following video:

Testing the finished Core XY motion system of the WeirdCube

This concludes the Core XY motion system for the WeirdCube. Next tasks are designing a print head, assembling the heated bed and adding a few end stops for proper operation of the printer.

Weird Constructor Avatar
Weird Constructor

Nice end 30 y/o guy, geek and F(L)OSS developer that messes with: Linux, Windows, Networking, Interpreters/Compilers, Games, Audio, Music, GUIs, C/C++, Rust, Scheme/Lisp, 3D printing, electronics and more.


For updates follow me on Mastodon

All the F(L)OSS development on my projects happens in my spare time. If you find what I do useful, entertaining or just want to support me, you can do this via Liberapay:

Donate using Liberapay