Dual CoreXY setup that uses LME12UU and 12mm shafts

by Thump2010, published

Dual CoreXY setup that uses LME12UU and 12mm shafts by Thump2010 Aug 29, 2015


Liked By

View All

Give a Shout Out

If you print this Thing and display it in public proudly give attribution by printing and displaying this tag.

Print Thing Tag

Thing Statistics

9190Views 1596Downloads Found in 3D Printer Parts


This is a modified CoreXY setup I have been working on, I have to do the corner bearing holder and the motor mount yet, this is for a printer I am creating, it uses LME12UU bearings and 12mm shafts, I went bigger as this is also used in a much larger printer also, that and I like tanks, I don't like things that are flimsy and flex.
It does not cross over it runs at two different levels so everything stays straight.
I used 686Z bearings on the ends and the flanges I made are designed to be paused then the bearing added then printing to continue so they are encased in plastic (a blast of a heat gun before hitting resume makes a strong bond for the continuing layer.)
I made it around this hotend that I got from amazon:
Signswise Assembled J-head Hotend 0.4mm Nozzle Printer Head for 1.75mm Filaments

It shouldn't take much to change the clamp to fit your hotend.

Things to do:
Motor mounts.
Corner bearing mounts.
Spot for a small magnet on the Y carriage for the X carriage home and another for Y homing.
Spot for a hall effect sensor on the X carriage to line up with above
Spot for a Z probe.

My thanks to Jayftee and Make-a-FaceI used Jay's bearing holder and part of Makes belt older.

Update 9/12/2015:
I had to redesign the X carriage, when I went to put it together it wasted almost 2 inches of space so I flipped the Y axises upside down to lift it up higher and to put the belts above the printer on top of the frame which solved the problem of no where to mount anything.
it is functional now, I was going to put new picks but I am almost done mounting everything and running the belts I just need to print out some more bearing flanges, on those I stop at height with cura, and pop the bearing in and resume printing.
I have made a motor mount that mounts to the outside on the sides with the belts passing through, I will have picks up hopefully by tomorrow as soon as I get the rest of the bearings done everything else is mounted, lines up and is ready to go.

Update 9/15/2015:
I finished assembling the mechanical parts, all seems to look good and moves as it should, I could have brought the corner bearings in more and use a t-nut in the slot and had them in line but I wanted more support than that offered and tapped the corner holes of the extruded aluminum and used nice long 1/4" bolts to hold it together, so it causes the belts to angle out to the corners but it shouldn't matter as it gets pulled away on one corner it gets pulled tighter on the other so there is never any slack and the other thing look at the serpentine belt on your car, it does not make a perfect square. ;-)
Pictures added, I just need to wire the motors to the smoothieboard now.

Update 9/25/2015:
I worked out where to put the endstops, I made them separate to glue on (acetone for the win) now that I have how I want them designed and best placement for them I will put them in the design of the parts so that they are built in.
I have everything hooked up but the hotends and the extruders and told it to print, told it to do a print speed of 150mm/s and a move speed of 300mm/s and it didn't blink an eye, I cranked it to 999% and it flew, I doubt it could push the plastic out that fast especially as I normally use a 0.3mm nozzle but the mechanism handled it just fine and flew, it was a glorious sight.

Update 10/9/2015:
This is fully operational now.
Endstops were added, print and enjoy.

Update 10/23/2015:
Raised the bottom of the X carriage (hotend holder) by 1mm to give more space at the bottom of the hotend heatsink.
Uploaded the fan shroud, it is designed to use blower fans for the plastic active cooling and a 40mm x 40mm x 20mm normal 12v pc fan to cool the hotends, a standard 10mm fan does not move enough air to cool both heads at a time but the 20mm fan works great.
Uploaded a in progress fan duct, I have printed it a couple times for fit and am printing the final version of this style now for testing tomorrow.
I thought about using two fans on the hotend one to push one to pull, the ones I found though blow a lot of air and with both nozzles at 220c you can barely feel a warm breeze at the bottom and quite cool at the top, but it is getting cooler now, the test will be in the middle of a hot summer.

Here are the fans that I used.
2 - http://www.ebay.com/itm/321760495643
1 - http://www.amazon.com/gp/product/B006ODM76C

Update 11/2/2015
I have decided to do a dual Z lift system with a hardened linear shaft in each corner, that should eliminate all Z wobble and issues, and I am switching from a 1/4" thick aluminum build plate to a 1/8", the 1/4" would probably be fine with the dual Z but might as well lighten the weight and help the build platform to heat up faster.
I expected to have it all done last weekend but I forgot to get more M5 nuts and bolts as I ran out and in the US it costs me $1.50 for 2 bolts/screws from the hardware store since they are metric. :-(

I was looking at everything and have decided to change the XY parts, right now they clamp the LME12UU bearing in place, sometimes that can throw the bearing out of alignment, I found that to be the case on the Z axis, it was bad. I decided to make a circular mount like you would get if you bought the brackets in metal, the metal ones you slide the bearing in and put a locking clip in, my new ones is completely encircled in plastic and you put two stop screws in the corners to keep the bearing from coming back out.
I left the v5 clamp ones posted and included the new v6 as a separate item for those that may not like that style, I printed it out and it aligned perfectly and worked really well, I am working on the X axis part now.

Update 11/16/2015
Uploaded all the new parts, it is working so much better, I will get pics up soon, just really busy.
I gave up on the bearing flanges, they were more trouble than they are worth, unless your printer prints perfectly which my original one didn't then you have imperfections in the inside of the flange where the belt rides and I had them pop apart and then the belt fall off several times, maybe PLA wouldn't pop apart but I was concerned about the temperature issues if I enclosed it.
What I did was to just do a bushing that the bearing sits between so the belt sits right on the perfectly round, and smooth bearing itself.
changing that has changed my prints to perfectly aligned perimeter layers, I am extremely happy with the prints.

I am going to tweak the XY motor mounts so that the bearings are a bit further in, finish the active cooling ducts, and I may add a z probe but lack of space it will probably be a switch driven probe and then I am going to consider this finished. (until I remember something else)

Update 11/23/2015:
All of the new parts are installed and everything is working beautifully.
All I have left on this project is the improve the active cooling ducts for the fans to cool PLA.
Uploaded new pictures.
see some print videos here

Update 11/29/2015:
Finished the airduct.
I tried many different styles but due to the tight spaces most were too close to the heaterblock and eventually melted and sagged.
I finally hit on a design that did not blow too much towards the bed or too much on the nozzle cooling it down.

Update 12/2/2015
I forgot to post the end stop magnet and hall effect holder for the Y axis.
The fan ducts still blew on the nozzle too much I modified it a little, and realized one fan didn't align due to the fact that the fan outlet is offset so I adjusted the duct for that side so print both a and b and then you will know which goes on which fan when the center lines up with the nozzles.

As far as I am concerned this is finalized, it works beautifully.

Update 1/10/2016:
I just had a scathingly brilliant idea.
Aligning the heads to the same height is an absolute pain in the A** you have to loosen the clamp and twist the heatsink with some needle nosed pliers until the height is obtained, and trust me it is not fun.
I just had the idea of the clamp itself, on one side I will leave the clamp as is, on the other hotend I will shrink the inner part of the clamp by 1mm top and bottom so that you will have 2mm wiggle room, set the bed until nozzle one is touching, let nozzle 2 drop and touch the bed and then tighten the clamp, done.
Simple and easy, I guess there is a reason for the term KISS and trust me, this is so simple I really do feel stupid..... ;-)
With this change it is important to put the clamp in properly now, match thin side to thin side and thick to thick.

I do have some add on features for later but the core is done.
those features being a cable/boden holder and a z-probe.

I am working on a single head version of this setup, the X carriage and Y carriage will be different, I thought about making them universal but with a single nozzle I can squeeze extra build space out of it if I try, like 2" in one direction and 1" in another.
But I will do a Drop in single nozzle X carriage replacement for this setup as well.

Also I will be working on an 8mm version of both the dual and single head version

Update 3/19/2017:
Version 8 of the Dual Hotend:
I have switched to physical switches instead of hall effect as I was having issues with them and went back to old school.
One head is held snug, the other head can adjust +/- 1mm up and down to help to level the heads, loosen the clamp, bring the bed to just touch the set hotend and let the other hotend drop to the bed and then tighten the clamp, you are now aligned.
Added a BLTouch version.
Added a Inductive probe version with a lock screw to secure it in place. I used an LJ18A3-8-Z/BX

NEW option for the upper motor mounts.
either old mount or new will work but the new motor mount has been extended to hold a bearing to hold the top of the 12mm ball screw stable, I used 12mm x 28mm x 8mm 6001ZZ bearings, the bearing holder is adjustable.

Update 3/26/2017:
There was a problem with the belt mounts being too loose in one corner, fixed all of them.
I finished the fan shroud, it holds 2 30mm fans to cool the hotends, and 2 40mm fans for part cooling, I used 40x40x20mm fans for part cooling, the more air the better.
I moved the bearing screw holes as they were too close to the bearing throwing off alignment a little.
I modified the inductive probe version to us a clamp, a setscrew was just not cutting it, it would throw the probe into an angle when tightened.
Special thanks to Saldot and his Fan holder/mount for E3D v6 http://www.thingiverse.com/thing:780379 which I butchered into my fan shroud as I have been using his fan holder for quite a while and really like it.

Coming up next will be a tweak to the bearing holder motor mounts to make them a bit smaller and eventually be a single head version.


I used ABS but I think that really is only important for the parts around the nozzle, everything else PLA should work fine.
0.3mm nozzle.
0.2mm layer height

Set your slicer to pause right before the top flange of the bearing, when the print pauses then insert the bearing and continure printing.

Parts List:

- The number on the left is the total number of that item needed, not the number
of times to order the included link, so if it says 4 - and the part is a 4 pack then
just get one 4 pack -

4 - 688Z Metal Sealed Deep Groove Radial Ball Bearings 8 x 16 x 5mm
10 - 698Z 8mm x 19mm x 6mm Sealed Deep Groove Radial Ball Bearings
8 - LME12UU Linear Motion 12 mm Ball Bushings
2 - Nema 17 Stepper Motors
4 - A3144 A3144E OH3144E 3144 Hall Effect Sensor
1 - (depending on the size of your printer, you do the math)
Thomson QS 12 MM 2000, Quick Shaft
4 - Neodymium Magnets 1/4 x 1/16 inch Disc N48
(That is the size I used but they are not available on amazon, check your local
hobby store or ebay)
2 - GT2 20 Tooth 5mm Bore Pulleys (bought them with belts, it was cheaper than without
and just didn't use the belt as it was too short)
1 - 5 Meters GT2 2mm pitch 6mm wide Timing Belt
1 - Sunon 40x40x20mm 3 pin fan #KDE1204PKVX
2 - 12V 0.13A 40mm Turbo Blower Fan 4020 40x40x20mm
2 - 3D Printer J-head/E3D 0.3mm Extruder Printer Head for 1.75MM Filaments

8 - small m3 screws to hold in the bearings, I used some small computer screws
for holding in cd/dvd drives that I had lying around, they just need a flat
bottom and a big enough head to cover a little of the bearing edge.
22 - 4-40 bolts and nuts, I just got long 1.5 inch ones and cut them down to various sizes with a dremel as I hate hunting for screws
10-14 feet of 3 wire cable for the sensors, fans, thermistors. More if you have a
bigger printer.

A spool of ABS plastic in your color of choice to print all the parts.

More from 3D Printer Parts

view more

File Name



All Apps

Auto-magically prepare your 3D models for 3D printing. A cloud based 3D models Preparing and Healing solution for 3D Printing, MakePrintable provides features for model repairing, wall thickness...

App Info Launch App

Kiri:Moto is an integrated cloud-based slicer and tool-path generator for 3D Printing, CAM / CNC and Laser cutting. *** 3D printing mode provides model slicing and GCode output using built-in...

App Info Launch App
KiriMoto Thing App

With 3D Slash, you can edit 3d models like a stonecutter. A unique interface: as fun as a building game! The perfect tool for non-designers and children to create in 3D.

App Info Launch App

Quickly Scale, Mirror or Cut your 3D Models

App Info Launch App

I know that it is a little bit too late for asking, but I am building a printer like yours, with dual extrusion, corexy, and using an smothieboard. I have been like two months with the same problem, and it is that I cannot print with two colors because my offset doesn´t work properly, and it prints where it wants to, so I would be very thankfull if you could tell me which gcode generator you use, or uploading yor setting, or sending your config file of smothieware, in fact whatever would be great.
Thank you beforehad

The belts do not look like they are positioned correctly. The variable length segments need to be parallel to the X and Y axis guide rails. As shown in the photos, the belt tension will vary depending on the extruder carriage position in XY space. That may cause the mechanism to bind, or belts to slip, and will also result in distorted prints.

It has worked flawlessly for 2 years with ultimaker quality I am happy with the results.

Hi I planing do this printer. I have open beam 4040 and 2020 and a heat bed of 300x300 mm. Can use this open beam? And for the bed ?

the corexy parts should work.
the corners I used to build the actual printer wouldn't but that is why I separated them. ;-)

I worked out a simple way for head alignment, I feel so stupid for not thinking about it sooner.
See the instructions/notes for the details.

Work area dimensions?
What was the total cost sans 3d printed parts?

Nevermind, Just noticed this can be adjusted for any size.

Correct, I am just showing what I did with the other parts I created and tweaking them and working them out.
these same parts will be going into a 15"x15"x20 build platform printer.
I am going to make another corexy setup same parts but designed down to fit 8mm shafts, in building my current printer I realized the 12mm is mega overkill in the XY axis's and is just added weight in smaller printers.

It'll work great for me though, planning to start building a 24" cubed build area printer.

I havent posted the fan shroud yet as I was still working on it but I will get it posted so you have at least the fan for cooling the heads.
If you wouldn't mind some advice from what I have encountered so far with a printer that size use 4 shafts and 2 threaded rods for your z axis set opposite each other for three main issues that I have encountered/noticed with my bigger printer.

  1. the bearings blow out with just one sided support as it pulls them at an angle which eventually throws the bearing balls out of the bearing.
  2. it would be too much for one stepper to lift.
  3. it will give you a rock solid build platform that will not bounce/move at the unsupported end.

Good luck, let me know how it goes, message me if you hit any issues or need some help.

Personally I am probably going to use bronze bushings rather than bearings. Less to worry about.

I had thought about that but was not sure about the tolerances, so I cant say how well the would or not work as I have not tried them.
I have had 0 issues with the bearings though, if you use a plated hardened steel shaft there is no wear and tear on the shafts.
Also it is important to pack the bearings well, how I do it is I fill the bearing with grease (super lube) but my finger or thumb on one end of the bearing tightly and then slowly (don't rush it) push the shaft into the bearing, it will push the grease throughout the bearing as it displaces the air, when you cant push anymore or have grease coming past your finger it is nice and packed, it will accept about 85% of the grease you put into it if you fill it.

I'm curious, how large is your build area and what size stepper motors are you using? I'm planning on 600x600x600 and I'm thinking nema 23 for mine but I'm debating between that or a geared dual nema 17 setup.

my current build area is 9" x 9" x 12"
nema 17s
even for your size you are thinking of you really don't need to get 23s as you are not moving a whole lot more, normally your Y would be moving more as it moves 2 shafts and the head BUT you have 2 motors moving it.
I would go 23s if you are only going to run 12v but if you fed your steppers 24v then it would have plenty of torque to move everything around quickly.
with that big of a build plat form I would do a dual or tripple Z axis, it would flop around WAY to much otherwise.
If doing dual Z I would use a marling board with dual Z and dual Z endstops so it levels it the same every time.
If you did triple Z axis I would position them in a triangle formation and the 2 on the same side I would link together so their enable, step, and dir pins shared the same input, and the motor on the other side set it up as dual Z with dual Z endstops with marlin, again that way when marlin homes it will home both sides of the large build area making sure you are good to go every time.
I am waiting for my new marlin board as my smoothie board does not allow for dual Z endstops so I always have to adjust the other motor by hand if I shut off the motors or power off the printer as it will either be too close or too far away as smoothie has no stops for the other motor so doesn't care

This is fully operational now.
Endstops were added, print and enjoy.