K8200/3DRAG CoreXY mod

This is not a mod for the faint of heart, it requires an almost complete disassembly of the printer base frame.

And probably a lot of swearing and frustration, but I am quite pleased of how it turned out.

I am marking it as a work in progress, since I plan to modify a few things to better fit readily available hardware and to come out with a bill of materials and a proper assembly sequence, eventually.

But if someone is daring enough to give it a try, here's the sources, have a go at it! :)

Here it is printing its own endstop brackets: https://youtu.be/2o1mKZOJG_4

I am intentionally only publishing the SCAD files, because you really need to know what you are doing before you tackle this. I have set up Marlin to be set in CoreYX mode (notice the exact spelling) and unless you use a recent release from the GIT repository it just won't home properly. Whenever you test movement leave the grub screws off the pinions until you are sure the motors are spinning in the right direction and the correct endstops are read. the pinions and idlers are from Anycubic, 16 teeth size which help improve both torque and resolution, and they fit over m3x30 bolts. I have some very tiny washers from my "PC bolts box" that allow them to spin freely, but probably it would be possible to make something equivalent by thinly slicing 3mm inner diameter PTFE tubing.

I am using Marlin 1.1.6 from the GIT repository, here are the main changes in configuration.h related to this mod: #define COREYX //(uncomment this, note, YX, not XY) #define DEFAULT_AXIS_STEPS_PER_UNIT { 100, 100, 2560, 640 } //(using the standard M8 allthread rod, and E steps should be tweaked) //(Note, I am using TMC2100 drivers, so the axes are ALL inverted // compared to the stock drivers) #define INVERT_X_DIR true #define INVERT_Y_DIR true #define INVERT_Z_DIR true

Updates

13 February 2022

I redesigned the Y axis smooth rod holders, to provide clearance for the bed levelling thumbwheels if you use the original small Y axis frame. One of the original replacement plates cracked after several years of use, so I have significantly beefed them up, and simplified the Y endstop mounting. The belt attachment block has been greatly simplified, and while initial belt fitting is not trivial it is a lot simpler than in the older arrangement. The trick is to loosen the tensioners completely, run the belts in the printed belt holder and fit the tie point bolts through them to close the loops, then adjust one set of belts until the distance between the bolts matches roughly the distance between the holes in the block. Cut the excess belt leaving a few mm bit for future adjustment, and screw the bolts in place. Tension can be then adjusted in the motor mounts. This is now all mounted on my machine and tested, though I am still using the larger bed frame.

29 April 2021

Further rework on the idler mounts, now they slot together and no longer need one M3x30 bolt and nut to hold the two parts together. I removed the tensioners from the belt grips under the bed, and replaced the entire assembly with a simpler structure that uses two M5x30 or M5x35 bolts as belt posts, and my GT2 cable grips from https://www.thingiverse.com/thing:3062665 .

3 April 2021

I redesigned the motor mounts yet again, to incorporate belt tensioners. I have still to redesign the belt attachment block under the bed to make it simpler and lighter now that it does not have to deal with adjusting belt tension. One of my idler mounts finally broke after several months of use, so I re-designed them a bit to make them stronger and to allow wrapping with braided fishing line and CA glue for emergency repairs

6 November 2019

On a suggestion in the comments from yuroller I switched to inverted motor mounts, to keep the footprint smaller and to reduce a bit the belt length. I only printed the parts for the moment, but I can't see why they should not work at least as well as my current ones, and there might be the potential to convert them in belt tensioners too, freeing a little more weight from the bed and making the tension adjustment a bit easier.

5 September 2018

Completely redesigned the belt attachment and tensioning system, should be easier to install and have more tension adjustment range.

Redesigned the motor mounts to be simpler and hopefully less likely to tilt, motors might still need to be braced with zip ties against the printer feet, though.

Fine tuned the belt alignment, many small changes to pretty much all parts.

Many refinements to the code, renamed a few modules, converted a few fixed values to easy to edit variables.

I printed the new parts, but have not assembled them yet. They should work fine, but can't guarantee it just yet

21 August 2018

More refinements to the idler supports, overhanging parts should print more easily as bridges. Smaller diameter spacers added, they should work but I used sections of brass tubing in their place. probably a pair of m3 nuts will work just as well to keep the idler at the right height. The belt paths are 1mm lower than with the older idlers, so I have adjusted the rest of the parts to match, and moved the Y carriage idlers 1mm toward the center. this should leave the belts a small amount of clearance, but should make the whole more robust. I am still redesigning the belt attachment and tensioning system, because the current arrangement is difficult and frustrating to assemble.

2 August 2018

It took some time, but one of my idler supports finally delaminated. I managed to print a replacement (holding the part together with JB Weld and a clamp!), but this won't stop the same thing from happening again, so I decided to redesign the idler supports to be both easier to print and more robust. The new version uses the same parts to hold the idlers on both corners (so you only need to keep one spare), allows the belts to be swapped if anyone wanted to do so, and uses less hardware and of fewer types. the downside is that two elements of the base frame have to be moved outwards by 2.5mm, but this still keeps them shy of the 5mm thick frame caps, and makes life overall much easier.

Edit 3 august 2018:

The idler spacers included in the file are too thick and cause interference problems. I am using tiny washers and two nuts to keep the idler at the correct height, but the parts can probably be printed narrower, or a piece of 3mm brass tube can be cut to size

Category: 3D Printer Parts