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


by Johann, published

Vert-X-Belt-Truder by Johann Dec 29, 2011



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

11294Views 554Downloads


Early concept for a fully symmetric belt-driven extruder carriage for upgrading Prusa Mendel. Belt-driven means there's no stepper motor on the X-carriage, so it can be much lighter and move much faster.

It has two X-belts that are controlled by one stepper on each side.

If both belts move in the same direction, the X-carriage moves in that direction too. If both belts move in opposite directions, the extruder moves but the X-carriage stands still.

x_motion = (stepper1 + stepper2) / 2
extruder = (stepper1 - stepper2) / 2

We'll need new firmware to control x_motion and extruder independently by setting the correct speed for each stepper motor for the two belts.

stepper1 = x_motion + extruder
stepper2 = x_motion - extruder

This design attempts to reduce the effects of the torque that's required to move the extruder:

  • It is based on the excellent VertX (vertical x-axis) by watsdesign with LM8UU linear bearings: http://www.thingiverse.com/thing:10711
  • It uses a pretty large timing belt pulley on the back of the carriage.
  • The two belts are entering the carriage very close together and perfectly parallel.


This is an early concept. Feedback is welcome!

All Apps

Upgrade this Thing with Thingiverse Apps

No results.

May 5, 2017 - Modified May 5, 2017

In case someone is still interested this is my experimental off-carriage extruder, similar to garyhodgson's proof of concept: https://www.youtube.com/watch?v=6C-Jhj3amho

Comments deleted.

Great idea to remove the motor off the axis and a novel solution, but there may be a simpler way. keep the x axis as current prusa, but turn the extruder 90 degree around the z axis. Drive the extruder using a motor mounted at the end of the axis with a square or hexagon bar mounted horizontally. The extruder drive gear moves along the shaft when the x axis moves, but can still be driven by the motor at any point.

Another alternative would be a bowden motor drive.

Keep up the work on this one though as I think its a neat solution

That's a great idea too! I made a similar proof of concept to take the X motor off the carriage (which isn't actually that useful) but it uses a similar idea by using a square rod to drive and the carriage slides through it: http://garyhodgson.com/reprap/2011/09/experimental-off-carriage-x-axis/http://garyhodgson.com/reprap/...

I'm certainly no EE, but how hard would it be to make an interface board that would "add
amp; subtract" the X motor steps from the extruder steps? If that could be plugged in between the command signals to the stepper drives and the stepper drives themselves, so no software changes would be required, and this could potentially work with any machine.

The faster these machines get, the more important it will be to remove the moving mass from the end .

An alternate thought may be to make a shaft parallel to the X axis rotate, perhaps with a key-way in it, or a spline, or other method to transfer rotation to the extruder drive gear. This would n
ot affect the X axis belt tension, except for the additional drag of the spline, which should be a constant.

The actual electronics would be dead simple. The actual promblem is getting the step commands to send at a regular rate (hint: you need the firmware to do that). Let's look at X moving slowly and E moving somewhat faster, at a constant velocity. These are their "step" and "direction signals: (^ is supposed top be a high signal, or step command, _ a low signal)

Xs: |^||^||^||^|____


So if we take these two and say "stepper1 = x_motion + extruder", this is what stepperdriver 1 would see:


Looks wild? It is, yor stepper will most likely not like it.

Same thing for stepper2, stepper2 = x_motion - extruder:

2s:|^|^||^|^||^||^|^||^|^| (same as 1s)

And that's just downright impossible for a stepper. At higher speeds, this would reverse th
e motor from full speed to full speed backwards many, many times per second.

My concern with this would be how much will the extruder drive affect the accuracy of the X positioning. Having the extruder motor on the carriage like the traditional design means that extruding doesnt cause any (significant) forces on the carriage. In your design it will apply torque to the carriage so the rigidity of the carriage mounts and rails will be much more significant. The existing 8mm rails dont require much force to deflect by a mm or so, and that would cause signigficant quality issues with a print.

You're making a good point about torque. I have improved the design, and added torque considerations in the description.

I'd be happy to help with the coding if needed. At least for getting this to a working proof of concept. I'd prefer to play with Teacup, hopefully that doesn't pose an issue ;)


Concept is solid. The hard part will not be in hardware though. You will probably have to branch your own version of the Prusa firmware, or get the developers to buy into the concept.

My Repstrap has a similar problem... Cross that bridge when I get to it.