1

Topic: New extruder concept

Next year, I have it in my goals to build my own printer from scratch. As part of that, I had been looking at bowden Vs. Direct drive and the advantages and disadvantages of each. Then a thought occurred to me. Why not combine them?

So, what I am envisioning is a flex shaft system where the filament is fed down the center. The twisting of this tube feeds the filament by "screwing" just like the Z axis on our printers. The filament is free to move in one direction and the flex shaft rotates.

The result is that the filament is driven at the point of extrusion like a direct drive, but the stepper can be located elsewhere to reduce weight. Also, the space an extruder like this could save would allow for several extruders in a small space and low weight.

And additional thought to go along with this is a driven hub on the spool. The flex shaft could be a tap off this to keep the whole feed system synchronized and smooth.

Right now, it is just a concept and seems simple enough. I wanted to put the ideas out here for discussion before I start experimenting with it.

2

Re: New extruder concept

Already exists:

http://mutley3d.com/Flex3Drive/

Cons:

Increased backlash relative to direct drive, only ~150g weight savings.

3

Re: New extruder concept

elmoret wrote:

Already exists:

http://mutley3d.com/Flex3Drive/

Cons:

Increased backlash relative to direct drive, only ~150g weight savings.


That's not even close. That's just a flex drive shaft to typical extruder.

Maybe I'll have to draw something up, but what I am talking about is the filament being driven by the twisting of the flex shaft. Imagine an internally threaded screw biting into the filament. By turning this screw, the filament is driven forward or backward. This happens along the entire length of the shaft which results in the filament getting full engagement all the way through the drive system.

So, flex shaft is basically like a nut, filament is the bolt. Turn the nut, and the bolt moves forward/backward.

The only moving parts are the stepper drive on the hub of the filament reel with a tap-off to the flex cable. At the hot-end end of things, nothing is moving and no added weight or gearing. I imagine backlash would be minimal since the filament is fully engaged the entire length of the flex shaft. It would also facilitate autofeed systems for quick filament changes.

lol. Can you picture what I am talking about now?

4

Re: New extruder concept

Good idea in concept...but I think very difficult to execute.  Two of the major issues are the variability of filament diameter affecting engagement, and the inevitable twisting of the filament in the flex drive which would constantly alter the feed rate.

Not trying to be a negative Nellie, but the bowden setup already accomplishes the end effect of what this would give you if it works...

SD2 - Stock - Enclosure - Heated Bed - Glass Plate - Auto Fire Extinguisher
Ord Bot Hadron - RAMPS 1.4 - Bulldog XL - E3D v6 - 10" x 10" PCB Heated Build w/SSR - Glass Plate
Thanks for All of Your Help!

5

Re: New extruder concept

Not feasible. It would chew up the filament, and flex shafts are lubricated. Running filament down the middle would both contaminate the filament and remove the lubricate from where it belongs, in the flex shaft.

Not to mention the tremendous backlash and friction you'd be introducing.

EDIT: and what IronMan said.

6

Re: New extruder concept

IronMan wrote:

Not trying to be a negative Nellie, but the bowden setup already accomplishes the end effect of what this would give you if it works...

I appreciate the responses. That's why I thought I would bring it here.

I don't have experience with bowden systems. What are the drawbacks in that system?

7

Re: New extruder concept

Drawback to bowden is nonlinearity/backlash, and inability to print flexibles.

8 (edited by wardjr 2014-12-02 17:42:46)

Re: New extruder concept

Mostly retraction settings/issues not really drawbacks.
Flexible filaments don't work well if at all.

What Tim Said wink

Printit Industries Model 8.10 fully enclosed CoreXY, Chamber heat
3-SD3's & a Workbench all fully enclosed, RH-Slic3r Win7pro, E3D V6, Volcano & Cyclops Hot End
SSR/500W AC Heated Glass Bed, Linear bearings on SS rods. Direct Drive Y-axis, BulldogXL
Thanks to all for your contributions

9 (edited by Retroplayer 2014-12-02 17:56:08)

Re: New extruder concept

Well, I don't think I should abandon the idea outright, honestly. But good to identify what/why it wouldn't work and determine ways around that.

We know from direct-drive that the main advantage is more pressure as close to the hotend as possible which gives us the most control over feedrate. The drawback is the extra weight from the steppers/gearing and pulling on the filament can sometimes affect accuracy. The extra weight also limits how fast we can print and change direction.

So, this is where I am coming from. Remoting the extra weight, but allow the filament to be driven right at the hotend.

Instead of the flexshaft engaging the filament the entire length, what about only at the very end of it?

Instead of hard threaded drive, we might be able to do a spring like coil sharp enough to bite into the filament. This would maybe help with varying filament diameter. Having the driving area smaller like the last 10mm might help reduce twisting.

10

Re: New extruder concept

I know! A hydraulic powered extruder - you'd get water cooling and power from the same hose :-).

11

Re: New extruder concept

The main issue I see is that you would just twist the filament until it breaks. Here's what I mean: If you just turn a nut on a screw without holding the screw in place (stationary), the screw will just rotate together with the nut. In order for the screw to move, you need to counter act the torque you apply on the nut. So there are two issues: 1st, how do you intend to keep the filament from just rotating with the threaded shaft. 2nd, even if you can somehow prevent it from rotating, it will most likely just twist off.

12

Re: New extruder concept

The main issue I see is that you would just twist the filament until it breaks. Here's what I mean: If you just turn a nut on a screw without holding the screw in place (stationary), the screw will just rotate together with the nut. In order for the screw to move, you need to counter act the torque you apply on the nut. So there are two issues: 1st, how do you intend to keep the filament from just rotating with the threaded shaft. 2nd, even if you can somehow prevent it from rotating, it will most likely just twist off.

13

Re: New extruder concept

Most extruder are quite good already, we shold put more focus on the slicing program and extrusion mechanic, these are the factor that can cause problem quite many times.

www.3dfilamenta.com - Affordable filament