"direct drive" extruder with stationary mounted motor

Hello,

Please let me know if this has already been invented. I searched around a bit, but couldn't find anything like it.

I came up with an invention to mount an extruder motor remotely and stationary (to the frame for example), with the ability to drive filament directly at the print head. This would reduce the mass of the print head, like a bowden extruder, but with the advantage of direct drive (minus minimal backlash). It will only work with cartesian style printers.

It utilizes ball spline shafts and either bevel gears or worm gears to transfer the torque from the motor to the extruder drive gear.

I attached a fusion 360 rendered drawing to illustrate the concept. It's really only meant to convey the concept, so don't mind the lack additional necessary components.

And here's an animation:
[youtu.be]

Ball splines are currently very pricey, so I don't see this being implemented on a mass scale any time soon. But perhaps comparable components could be used, or perhaps prices will eventually come down.

No idea if this is usable or even a good idea, but just wanted to share in case it is.

In response to a question on the youtube video about the mass of the drive system, I posted this:

Quote

Just looking at specifications on the THK website, 6mm shaft should be able to handle the torque. According to the specs on a 300mm 6mm linear rail (not ball spline, but it should have roughly the same weight) on amazon, one 300mm rail weighs 70 grams.

You only need to push one rail, but obviously there are the spline nuts, gears, ball bearings for the gears, and whatever frame all of these items are contained within. My ballpark guess would be 150g for a 300mm system. High torque nema 17 steppers run between 400 and 500 grams. So there should be some significant weight savings, although the cost and added complexity will still make it prohibitive for most.

Edit:

I did some math to verify the weight of the shaft.
ASTM spec 52100 bearing steel density: 7.81g/cm3

6mm shaft @ 300mm = 8.48cm3
Calculated weight = 66.25g

Just thinking out loud here, but if this design were incorporated into an ultimaker style XY, the weight of the shaft becomes irrelevant since it could be integrated as one of the (i don't know the official name) shafts that moves the print head. Come to think of it, maybe I should be posting this over there.

That's awesome...

As per @infiniteinquiry; Brings the thought of putting two acme lead screws next to each other - one to drive Y-Axis and one to drive the Extuder as the Worm gear. Think that's what he/she was after - course it would either have to slide as you've shown or compensate by firmware in order not to twist with Y-movement.

Course there's also the idea of using a speedometer/rpm cable to remotely drive the worm gear w/o having long shafts.
And also the idea of wondering how well just driving a wood screw ( as a Worm Gear tooth ) directly on filament would work as an extruder.

I've never seen any of this tried before.

This design seems to bring up all sorts of possibilities. Thanks so much for sharing

There was a kickstarter printer some time back that used something similar for the bed movement, but it didnt see the light of day.
Perhaps it could be incorporated into this [www.thingiverse.com]

There are already 2 products that use flexible cable drive to a worm gear in the extruder: Flex3drive, and Zesty Nimble.

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].

Quote
dc42
There are already 2 products that use flexible cable drive to a worm gear in the extruder: Flex3drive, and Zesty Nimble.

No flexible driveshaft here. Not really similar. The flex drives are interesting, but seem to have issues with space.

Just to add even more thought to the implementation.

Couldn't help but wonder if acme lead screws are strong enough to act as a polished rod on the X-axis of a Prusa i3 Cartesian design.

Top X-rod = A end-bearing attached; notched polished rod with a sliding Extruder gear as you've shown
Bottom X-rod = A end-bearing attached; Acme lead screw

Driving both the X-axis and Extruder through the same upper/bottom rods usually just polished rods.
Although; I'm starting to sound like I'm just restating the same thing that you've already demonstrated

Edited 1 time(s). Last edit at 03/26/2017 06:47PM by tgit23.

The idea is not new, check here and here

Chri

---

[chrisu02.wordpress.com] Quadmax Intel Delid Tools

Quote
Chri
The idea is not new, check here and here

Chri

I figured there must be something comparable out there in the wild. I searched around though, and never got any results when searching for "ball spline" and extruder. So at least that is something somewhat original.

Ball splines:
[www.nbcorporation.com]


I like the Mendel/i3 implemention with a single axis version you shared. Why isn't this more popular? I'm working on a custom build ultimaker style, so that's what I've been focused on.

Quote
infiniteinquiry

Quote
Chri
The idea is not new, check here and here

Chri

I figured there must be something comparable out there in the wild. I searched around though, and never got any results when searching for "ball spline" and extruder. So at least that is something somewhat original.

Ball splines:
[www.nbcorporation.com]


I like the Mendel/i3 implemention with a single axis version you shared. Why isn't this more popular? I'm working on a custom build ultimaker style, so that's what I've been focused on.

Because that system just has friction issues which can not be solved with in a affordable way.

Chri

---

[chrisu02.wordpress.com] Quadmax Intel Delid Tools

really love this idea! I think the guy on Thingiverse is on the right track but considering he posted that in 2013 is telling you it's a bit challenging to figure it out. One challenge will be keeping the shaft from flexing. That's a long way to carry torque on a small diameter shaft. The fluted shaft you show in your model would be muy expensivo (no, I don't speak Spanish).

But I'm with the OP, there's got to be a way to make this work. One of the biggest influences in print quality is the moving mass of the print head. Sure, bowden makes a huge difference, but there are other complications (like not being able to use flexible filaments).

Encourage you to keep thinking about this. Like Chri mentioned, focus on affordability. Consider standard, inexpensive components.

I have been thinking about something similar too... I think running shafts all the way from a stationary motor to the extruder might introduce a lot of backlash, simply because it requires multiple sliders and gears to drive one axis...

I was wondering though if similar weight-savings could not be achieved by mounting the extruder motor to one end of the X-Gantry and driving the extruder using a shaft like that, and then in turn moving the X-axis motor onto the machine frame by driving the X-axis belt using a profile shaft.
That way you don't get the extruder motor all the way to the frame, but in turn you do so with the X-motor, and you do not need the worm-gear mechanism on the end of the X-gantry...

Hi.
I have working sample . See video.

Edited 1 time(s). Last edit at 08/27/2017 03:17PM by Senimore.