Recently, I have been researching several CoreXY DIY designs for an upcoming 3D printer build. The primary advantages of CoreXY printers over Prusa-style Cartesian configurations are:
- The cubic enclosure provides additional rigidity
- The bed is stationary along both the X and Y axes (and sometimes Z). This allows the bed to be made thicker, heavier and more stable, and the print itself does not get flung at high speeds along the Y axis.
Keep in mind that there is nothing inherently wrong with Prusa kinematics, and they have been fine-tuned over the years to produce excellent results. But aside from the advantages listed above, a CoreXY might appeal to you just to build something fun and different, and learn new things along the way.
My goals are as follows:
- Reliably and repeatably print PLA and PETG at 60-100mm/s
- Use a direct-drive extruder to print flexible filaments as well
- Roughly 300mm3 build volume
- 24v wiring
- 32 bit control processing
- Stay within a ~$600 budget
The ability to print ABS in the future would be a bonus, but not a requirement.
Could be your end-game 3D printer
All belts and linear rails
Slick build manual
Designed from the ground up to be enclosed and filtered (ABS)
Most complicated to build
Good illustrations, but not much text documentation
Be prepared to print LOTS of ABS parts during build
“Standard” BOM is 12V by default
|A nice iteration on previous designs (HEVO, D-Bot)
Avoids potential problems with bent shafts
|Mains voltage attached to a moving bed makes me nervous
Smaller/newer community, not many example builds
Not easily enclosed
|Large community, lots of build examples
Many examples of high-quality prints
Can be made very rigid
Lower than average number of printed parts required
|Bed may have wobble issues in single Z-motor configurations (double Z has been added as a default option)
8mm X-axis shafts may have too much flex in larger builds
Possibility of sourcing bent/poorly machined shafts
|Large community, long track record, lots of build examples
Good iteration on the tried and tested C-Bot printer
Can scale very large
|Cantilevered bed design may cause problems with heavier prints
Not easily enclosed
Relies on printed parts for frame rigidity
The Voron 2.1 is the successor to the highly acclaimed first-generation Voron, designed by Maks Zolin. It is a high-end DIY 3D printer with a fairly high price tag to match. As an upgrade to the Voron 1.6, it adds a fixed bed as well as belts and linear rails on all axes.
I originally read that the Voron 2.1 could be built for roughly $600-800. I generated a BOM spreadsheet and set to work sourcing and estimating parts with the help of a Google Sheets sourcing guide created by community members. Well, I tried pretty hard, but nothing I could spec out–even with primarily AliExpress parts–would bring the theoretical build cost below $1200. Perhaps most Voron builders have already built several 3D printers and happen to have a lot of spare parts on hand.
Do I like the Voron? Yes, I’ve studied the plans, and it’s an awesome piece of engineering. But if a $600 CoreXY build could lay down plastic just as well, at $1200+ you’re really hitting the law of diminishing returns. If printing ABS is really important, you’re space-constrained, and you like that the enclosure is built into the printer frame, then the Voron may have more value for you.
Another issue I have with the Voron is what seems to be a gap in the documentation. There is a slick PDF build guide that reminds me of Lego instructions, but it is light on text. There is also a firmware repository on Github that might provide some clues, but for the beginner to intermediate-level builder, the only real option is to go on Discord and wade through hundreds of search results or start asking your own questions. The Discord community is helpful, but I can’t help but feel there needs to be a better way to bridge the gap between the physical build and the final product.
The Voron 2.1 uses mains current to heat the stationary bed, and 12v everywhere else. 12v may seem like an odd design decision as most newer printers are moving to a 24v configuration, but many of those are also counting on those same 24v to heat the bed. Using 12v reduces the need for some adapters, for example, if you choose to use stock Noctua cooling fans.
The Railcore II is also in this price class, but build costs average around $1500, so it was already out of my initial scope of consideration.
A relatively recent design, the V-King, developed by Roy Berntsen, aims to overcome the percieved limitations of round shafts and linear bearings by using V-slot extrusions and wheels. While it’s true that many things can go wrong with the shaft and bearing design, the same argument could be made for cheap wheels.
“Leave the v-wheels on the $150 Chinese printers,” someone posted in the Voron Discord. While I don’t necessarily agree with this sentiment, it is important to source quality motion components no matter which method you choose. Bearings, V-wheels, and linear rails could all potentially have issues, and by the same token, all have the potential to work extremely well.
The V-King community, primarily active on Facebook, is very friendly and helpful, but relatively new. There aren’t a lot of builds in the wild as of January 2019. Roy’s sample prints look very good, but it seems like he’s still fine-tuning a few things.
My primary concern with this design is the use of mains current attached to a movable bed. Granted, the Z axis moves much more slowly than X and Y, but since you will be wiring this yourself, you will want to be very certain your connections are solid before running 120 or 240 V through it.
The HyperCube Evolution (also known as the HEVO), developed by SCOTT_3D, is an iteration on the excellent HyperCube 3D printer designed by Tech2C. It adds 3030 extrusions around the frame for added thickness, as well as a few more upgrades. It has a decent Wiki as well as a couple of different BOM generators, online and in Excel format.
Really, there’s not a lot that I don’t like about the HEVO. There are many successful builds in the wild, and a lot of high-quality print examples as well.
Having a stationary bed like the Voron would be nice, but the dual-Z motor configuration seems to have solved most of the issues with bed wobble from the intial design.
I have heard reports that the 8mm X-axis shafts are not stiff enough for larger builds, but they seem to work fine spanning a 300x300mm build area. Some people have gotten around this issue in larger HyperCube Evolution builds by using 10mm shafts on the X-axis, sometimes using hollow aluminum tubes or carbon fiber to reduce the X gantry weight that the Y-axis needs to sling around.
HyperCube Evo Links:
The D-Bot, by spauda01, is the oldest of the 4 designs compared here. It is a remix of Carl Feniak’s Core-XY C-Bot, and can definitely be considered a tried and true design. There are hundreds of makes and remixes on Thingiverse, not only of the D-Bot but its predecessor, so there should be no shortage of documentation and examples to help with the build. Like the HyperCube, there are many good print examples as well.
In my mind, the major disadvantage of this design is the cantilevered bed. Heavier and taller objects could potentially place uneven stresses on the unsupported side of the bed, causing wobble and stuttering along the Z axis. Although a lot of workarounds have been applied to fix these issues, I would prefer a design that avoids them in the first place.
Another common criticism is that printed parts hold a lot of the frame extrusions together. This may or may not cause squareness and rigidity issues, depending on the tolerances of your printed parts, the materials used, and how carefully you assemble everything.
The D-Bot, as well as the HyperCube and HyperCube Evolution, have some ready-made kits available on Aliexpress and elsewhere to help you get started.
After reviewing the various options, I have narrowed down the options to the HyperCore Evolution and V-King. This is not to say these are the best CoreXY printers, or even the best for the money, just the ones that fit my current build goals the best. Your mileage may vary.
I will make my final decision within the next few weeks and begin sourcing parts.
Have you built any of the designs above or have any other CoreXY designs you would like to share? Let me know in the comments.