Hey! This thing is still a Work in Progress. Files, instructions, and other stuff might change!

Drew's Rostock variant

by Ellindsey, published

Drew's Rostock variant by Ellindsey Oct 5, 2012

Thing Info

55626Views 6895Downloads Found in 3D Printers
Report Thing


This is a printer I am currently building based on Johann's amazing Rostock printer.

This design is modified to use Makerslide rail instead of smooth rod and linear bearings. I chose to use Makerslide partly because I found a pile of V-roller bearings at a junkyard for real cheap, but also because I thought the Makerslide would be more rigid and would allow me to hide wiring inside the channel.

This printer is also being designed to be a single self-contanied piece of furniture, with all wiring and electronic components hidden inside it as much as is possible. I intend to eventually equip it with Bluetooth, a SDcard reader, and a full LCD interface so it can print completely stand-alone.

The original version of thie printer didn't work bery well due to structural issues. I don't have access to a CNC mill or laser cutter or any other proper machine tools. I'm basically building this thing by hand with a table saw and power drill. The wooden structural parts for the original version weren't accurate enough, so I've redesigned it to be possible to build without haveing proper machine tools.

The revised version uses 3/8 inch threaded rod to make triangular frames at the top and bottom ends of the printer. The position of the rails can be finely adjusted by turning the nuts on the screws, so you can tweak it until everything is properly lined up after building it. The wooden frame is now reduncant, although I'm still using it to mount secondary parts.


Each drive axis is made from a lenght of Makerslide rail. I'm using three 666mm pieces (cut from a single 2000mm length) but the length could be changed for a larger or smaller print area.

The top ends of the rails each have a 5mm hole drilled in them, and one "railtopblock" part secured to the rail by a 5mm bolt and nut. This bolt also goes through an idler pully built around a 525 bearing. The rail top blocks also have pockets for the upper end travel limit microswitches, and channels for the wiring to those switches and other stuff on the top end.

The three top end blocks are joined to each other with three 3/8" threaded rods and nuts. The "topfoot" part allows you to mount an upper plate to the rods, which is convienent for somewhere to put the extruder drive, filament spool, lights, etc.

The bottom end of each rail has one "railbottomblock" part secured to it with a 5mm bolt and nut. Like the top, the bottom blocks are joined by 3/8" threaded rods, and the "bottomfoot" part lets you secure the rods to the bottom plate.

The rail bottom blocks have four thru holes for #10 bolts, with hexagonal pockets for #10 nuts. The two closer to the rail are to mount the motor blocks, with bolts that can be turned from above to adjust the belt tension. The two further away are for longer #10 bolts to support the plywood plate that supports the print bed. I use springs to support the bed support plate so that the print bed has some give to help prevent motor skips if the print head bumps into the printer part.

The "motormountblock" part has a mounting pattern for the motor, and a pocket for a 525 bearing to support the end of the motor shaft. Two #10 botls attach this block to the rail bottom block and can be turned from above to adjust the belt tension. I also put in slots that can be used to attach grounding wires to the motor face screws if desired.

The rail carriages are each made from one "carriagemid" and two "carriageside" parts. Each carriage side holds two V-groove bearings. Two #6 threaded rods go through the back side of the carriages to join the sides together. The drive belt is pressed through a groove in the carriage mid section, and then one #6 threaded rod and one M3 threaded go through the two sides and the center to join them together. Adjust the nuts until the piece slides smoothly but without any ability to shift sideways or rotate, and then lock-tite all the nuts.

Yes, I'm using a mixture of english and metric hardware. It's hard to find metric hardware around here, so I use english unless I specifically need to use a metric part.

The M3 threaded rod also goes through the two connecting rods on each carriage. The connecting rods are made from Traxxis rod ends and carbon fiber tubes.

The "headmainbody" part holds the three M3 threaded rods that secure the other ends of the connecting rods. It also has channels for wires and pockets for LEDs and three 20mm cooling fans. The mini J-head is secured in place with a little triangular wooden piece secured up inside the head main body with three screws. On top of this one "invertedextrudeclamp" and one "tighteningcone" secure the Bowden tube.

The extruder drive is a heavily modified Airtripper, which I'm not really happy with and intend to redesign.

I have stuck the original part from my first attempt in a zip file. I don't recommend using them, the new design works much better.

More from 3D Printers

view more

Thing Info

55626Views 6895Downloads Found in 3D Printers
Report Thing

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

Hi Drew,

I really like you Rostock build and I’m currently building one up. I was initially going for my own design however after seeing yours I’ve decided to build a clone as I’m new to 3d Printing.

I shall start the mechanical build in the next couple of days once the final bits arrive (waiting for the makerslide) however there are a couple of things I would like clarifying. What length parallel bars are you using; mine are 250mm between the centres of the bolts. How long are the 3/8” (8mm) threaded bars between the rail tops or bottoms? I am aware that there will need to be a little tweaking to ensure that the 3 rails are perfectly collinear but a close starting point would be good.

I’m, however, less confident with the firmware setup. I assume using Marlin would be the best option however what are the Delta settings that you have ended up using? Specifically I’m looking at the Delta settings section of the configuration file (copied below) Is there anywhere else that I need to alter?

Any other hints and suggestion would also be gratefully received.

Many thanks in anticipation

Rob Fallows

//============================== Delta Settings =============================
// Enable DELTA kinematics
//#define DELTA

// Make delta curves from many straight lines (linear interpolation).
// This is a trade-off between visible corners (not enough segments)
// and processor overload (too many expensive sqrt calls).


// Center-to-center distance of the holes in the diagonal push rods.

define DELTA_DIAGONAL_ROD 250.0 // mm

// Horizontal offset from middle of printer to smooth rod center.

define DELTA_SMOOTH_ROD_OFFSET 175.0 // mm

// Horizontal offset of the universal joints on the end effector.

define DELTA_EFFECTOR_OFFSET 33.0 // mm

// Horizontal offset of the universal joints on the carriages.

define DELTA_CARRIAGE_OFFSET 18.0 // mm

// Effective horizontal distance bridged by diagonal push rods.


Man , you're killing me. I already have all the parts printed out for a rostock, but this thing of yours is too elegant not to build.. Oh well back to printing.

I'm working on a variant using makerslide to connect base and top instead of 5/8 threaded rod. I know you said you designed in Autocad feel like posting some design files? ;)

My build got close enough this week for me to start firing up bits and start working with the tuning. It is a 1m high MakerSlide based machine with a hex shaped Misumi frame, 12" heated bed with a 3-point leveling mount. My guess is that with the current rods and carriages it is probably an 11"x11"x18" envelope. With some small changes, I think it could be 12x12x24.  While everyone else seems to aim at making everything, I just wanted to see what could be made with a minimum of fabrication with available parts. As soon as I have it all running, I'll post a BOM with detailed instructions and pictures.

I used a modified version of your motormountblock (had to cut out the bearing floor to fit my pulley closer to the slide to prevent clearance issues with my carriages), a modified version of your headmainbody (took off the top 3/8" so I could mount a pneumatic fitting - hope that doesn't come back to bite me) and a newheadclip.

The filament spool rollers and extruder are the only other printed parts I used.

I am finally:     > moving stuff around (not yet sure I understand why homing only does short upwards jumps of all stages in sequence or why the motors shut down after 30 seconds of inactivity)
     > endstops are showing leds but I'm not sure the software is reading them right yet
     > heating bed (my 12" bed is going to require a dedicated fan for the RAMPS 1.4 fuses as they aren't really happy with 12V and 15A)
     > cooking cartridge heaters (they like 12V and not 13.8V)
     > and ringing out my cabling (if you follow all the documentation for the motor's colored lines and what the RAMPS board wants you get motors that just lock in place. If you just do a color-blind 1 for 1 connect across the motor and cable connector pins you get a cable that works. Now I can't even say which document was wrong.)

Hi!  Glad you're having good luck so far with your assembly.  The homing only doing short jumps is probably an issue with with your endstops, either not working or not being configured properly so that your code is thinking the endstops are already pressed when they aren't.

The motors shutting down after a period of inactivity is a feature of the Marlin firmware.  There's a setting you can change in the code to change the timeout period.

The color coding on motor wires varies by manufacturer, so you really do need to meter them out before hooking them up.

And yeah, I'm not surprised the RAMPS board has trouble with a 15A heated bed.

This is probably obvious to everyone, but I have printed headmainbody.stl and with the part in one hand and my hot-end in the other, it looks like the top of the hot-end goes into the round cutout area inside the part. What holds it there?!?

 Initially, there was hand-carved wooden block which clamped around the top of the J-head and was mounted to the inside of the headmainbody part.  Since then, I have determined that a printed plastic part will work for that piece.  I have now uploaded that part as newheadclip.stl.  This part presses around the groove on the J-head and is secured into the head main body with three 3mm screws.

What partnr for the traxxis rod ends and what diameter carbon fibre rods?

 I used Traxxis 5347 for the rod ends, you can buy them in packs of 12 which is just enough for one printer.  You should use 4mm ID carbon fiber tubes, so you can easily use a M4 threaded rod to join the tubes to the rod ends.  I used 3.5MM ID tubes and had to drill them out slightly to get the threaded rods to fit, don't do that.

 I'm building mine with Misumi rail for the superstructure and MakerSlide for the stages (note that they make 60 degree corner extrusions). A small metric die and tap set should be on your tool wish list as it would be much easier if you tap the M5 holes on the end of the MakerSlide to attach the top and bottom blocks. Your parts are some of the best my poor old MakerBot has made. Until I can get your designs to modify, I'm going to drill two 5mm holes on the back of the blocks to attach them to the extrusion slots on the back of the MakerSlide.

Very impressive work!

Great job Drew.  If you run short on MakerSlide parts I can give you some through my MakerShip program.


Nice use of MakerSlide. I have a similar design in progress with MakerSlide that has bearings at every pivot point and full housing design.

Looks like you added a lot of weight to the y-axis by using makerslide rails and 4 bearings plus all the nuts & bolts, I think that is gonna come back and bite you in the ass when you start printing

stick with the LM8UU/smooth rod, it's a proven concept, less is more :)

 The weight of the Makerslide itself isn't relevant because it doesn't move.  IMHO parts of the printer that don't move should be heavy, that helps with the overall stability.  The only  real difference in weight is the V-roller bearings versus the linear bearings, and I'm not convinced that's enough to be an issue.  I have started printing, and it looks like the only performance problem I'm facing at the moment is due to the flex in the crappy two-part plastic connecting rods, which I'll be replacing with carbon fiber tubes soon.

If we all stuck with proven designs, the Rostock would never exist at all.

Thanks loads for posting - I don't suppose you did all this hard work in OpenScad, did ya?

I've gone with a simpler carriage design, but I'm running into problems and it looks like I'll have to re-think it - I really like yours!


 Sorry, I don't know how to use OpenScad yet.  I really should learn now that I have a working printer.  All my design work was done in AutoCad 2006.

I'm actually redesigning the carriages to have adjustable bearing spacing, which makes them a lot more complicated.

 I'll see if I can dig up the original link I found regarding the 3 wheel, offset design using the eccentric nut - that part seems to be working fine for me - I think I overcomplicated my life trying to use Spectra rather than belts! :P