Hello!

The Problem
A direct extruder (with the extruder motor at the nozzle) makes the end effector heavy. Due to the limited torque of the axis-motors and the limited stability of the linear bearings and the frame the printing speed is limited.

A bowden extruder (with the extruder motor at the frame, feeding the nozzle trough a tube) does not have these issues. The end effector is very light an can move with high speeds and high acceleration. However the bowden principle introduces many problems. This is mainly friction in the tube and a hysteresis due to the elasticity of the tube and play of the filament in it.
To fight stringing retraction is used. But because of this hysteresis you need more retraction. You'll also need more retraction if you are trying to print faster, because the pressure in the nozzle is higher and this pressure has to be eliminated before moving the nozzle to another position (otherwise you get strings). More retraction takes time an you get small blobs during the time the nozzle stays at one position.
Long story short: The fastest I could print with a decent quality was around 25mm/s with 5mm retraction.


The Solution
The Flying Extruder!


The extruder is suspended by rubber bands. These are running around several rolls to increase the length (each around 2m relaxed) to have more or less the same force over the whole vertical range. The extruder is connected with a very short tube (around 75mm) to the nozzle. This tube decouples the high frequencies between the nozzle and the extruder.
Long story short: I'm printing with no problems with 75m/s and 3mm retraction and still with the same acceleration of 800mm/s². But this was only a short test and I think I can bring it up with a bit optimization.
Another advantage is, that it's now easier to build an enclosure and that the printer is much quieter, because the extruder is decoupled.
(Cabling is not finished yet.)

Here is a short video: [vimeo.com]
Ask Kisslicer why it's producing such crazy moves

Kind regards,
MK

PS: This is still in a prototyping state. I haven't tested everything end some optimizations must be made.

Edited 3 time(s). Last edit at 06/16/2013 04:51PM by ähM_Key.

A great idea to find a solution to a very real problem.
A few concerns and possible solutions on your idea at [groups.google.com]
1."Does the extruder go up as the platform rises? if so, how?"
2."at the right settings/size of moves, a strong resonance will develop that will have the extrude slamming into the arms."

Possible solutions
1." to just suspend it from one point at the top middle and just use a counterbalance weight."
"possible to add a secondary positioning spring that only engages when the head falls too far out of position. Designed right, the combination should be able to be made to work such that it does not have any sensitive resonant modes, and yet, because it is a spring, nor does it violently affect movements in the end effector."

I do not understand question 1. What do you mean with "the platform"?

As I wrote, not all issues are solved yet, and I'm happy to get input

ähM_Key Wrote:
-------------------------------------------------------
> I do not understand question 1. What do you mean
> with "the platform"?
>
> As I wrote, not all issues are solved yet, and I'm
> happy to get input


What pulls the extruder up when as the build height rises?

The end effector connected via the tube connection.

So the tube bears the weight of the motor and extruder?

Perhaps the rubber bands are stretched when the effector is at its lowest point, and they retract as the effector moves upward? I don't think you want the weight to be borne by the tube because you'd still have a moment of inertia issue.

Hey ähM_Key, that is a really clever solution.

Do you adjust the tension in the elastic bands to exactly balance the weight of the cold end, so the bowden tube has close to zero weight on it?

Or do you adjust the tension in the elastic bands to hold up the weight of both the hot end and also the cold end -- so the bowden tube is in tension, supporting the weight of the hot end?

I see that currently the cold end of the "flying extruder" hangs from elastic bands that are perfectly vertical.
Would it be better or worse to re-position 3 pulleys close to the outer rim,
so the cold end hangs from elastic bands that run more diagonally from the cold end near the center to pulleys close to the outer rim?

This idea looks like it has a lot of potential. Thank you very much for sharing it with us *before* "fully optimizing" it.

-- dc

Hey David!

Thanks for your comments!

Although my post was quite old, your one is quite current, because I recently changed the system

The rubber band solution was working great at the beginning, but they wore out over time, so I upgraded to the counterweight solution.


A plastic coated metal string is running over two ball bearing down to a moving weight.

The extruder is now really "flying" - negligible force influences to the end effector no matter at what position!


(The weights are looking a bit complicated...this is not neccesar but it was - like very often - an itterative process to get it right. I would make them a bit different the next time, but it is working for now...)

I adjusted the weights, that they do not carry the hotend, because I did nat want to introduce forces, which might influence the accuracy.

A similar solution recently introduced is the Fly-n-Strude by Tricklaser (https://www.youtube.com/watch?v=bhVjtnaXD2I). But I would not recoomend it, because of the increased weight the carriages/motors have to handle.

MK

Edited 4 time(s). Last edit at 02/11/2015 02:26PM by ähM_Key.

its funny, I can print with my bowden setup at 150mm/s with no real problems. 3.5mm retraction with very very minimal blobbing, that's mainly due to variance in filament thickness and some play in the fitting I am using, I have the retraction speed set to 200mm/s.... To me the biggest source of play now comes from the fittings and not the bowden tube. The fittings need clips to ensure no play is present.....

---

My Personal Blog. Build blog.
[engineerd3d.ddns.net]

Modicum V1 sold on e-bay user jaguarking11

just imagine what it would be like with your extruder and a much shorter bowden.

The bowden tube is not making too much of a difference there for me. I run a 500-600mm bowden tube. I do not think its going to get much better unless I eliminate it all together. Then again do I really want to swing around a nema 17 at 200-300mm/s?

---

My Personal Blog. Build blog.
[engineerd3d.ddns.net]

Modicum V1 sold on e-bay user jaguarking11

Do you print in TPU filaments?

Quote
3DRapidClone
Do you print in TPU filaments?

I have not printed in anything but ABS as of now. However my extruder is designed to take just about any kind of filament. The hot end should be able to hit 280c now that I have it tweaked. I will be ordering some nylon filament as well as some ninjaflex to try out. The tolerances in the extruder ensure that I can push even flexible filament. I tried pushing some rubber sleeving with my extruder, it actually pushed it out with a higher force than I can restrain, when I tried to yank the filament out of it while pushing it tore it up and would not let it go. How will those findings work in the real world? I have yet to test them, as of now I have no use for flexible filament, I do have uses for nylon, but before trying out nylon I want to enclose my printer for fume containment as its not exactly healthy to breathe nylon fumes.

---

My Personal Blog. Build blog.
[engineerd3d.ddns.net]

Modicum V1 sold on e-bay user jaguarking11

I can't wait to see the results of your research!

Which results are you looking for?

The result for me is, that I can print without any blobbing/stringing even at long travel distances. This configuration worked for some kilos of ABS already.

Another extruder with less back pressure might have solved the issue mabe partly as well.

Edited 1 time(s). Last edit at 02/13/2015 02:41PM by ähM_Key.

It was at first directed at jagking for his extruder. Flying/Floating extruders try to mitigate bowden weaknesses while maintaining its benefits. For you, I'd want to see exactly how fast you can print. Quick moves is going to swing that extruder around like crazy, stressing the bowden tube. I want to know exactly whats the limit at where the moment of inertia starts affecting prints or the extruder bumps into the arms and what not. I am curious if a diagonal wire to pulley design with a spring attached to it would be able to resist those insane swings due to Hooke's law.

Another advantage of this flying extruder is also that (finally) Delta printer can be easy enclosed.

In any case, I'm interested to find out which are the additional tensions on the belts. This might have impact in stronger stepper motors.



Also, I just found that somebody made a "flying extruder" printer as a product, available to be bought.
More details in the below link:
http://www.personalfab.it/shop/deltawasp-20x40/

I've also built a flying extruder photos enclosed. Still testing it but the issue of throwing the motor around is not so much of a problem you throw a few motors around with a cartesian printer with often only one motor to do the lifting rather than three in a delta.

The mass of the extruder is acting fully on z moves but only slightly on x and y, in so far as the extruder twists and turns as the effector moves.

Travel moves are the fastest and longest, printing moves are often slower and shorter,so whilst the inertia of this extruder might cause issues with accurate positioning, although I haven't had any yet, there seems to be no negative effect on print quality, this is hugely offset by the shortened bowden tube. I know people like them but I have an i3 with wades extruder before I had a delta with a long bowden and the quality difference was notable, I'd use the i3 for anything that needed to be sharp, not anymore!

I have some mass resting on the bowden tube and had not considered whether bending the tube would effect filament control, it doesn't seem to be a problem but I'm at the early stages of testing it.

I intend to replace the sewing elastic with something more durable but its working and its cheap.

I'm using 3mm retraction at 100mm/s and getting very nice prints. Might put the stls for the carriages and extruder mount on thingiverse.

Edited 1 time(s). Last edit at 12/05/2015 05:29PM by DjDemonD.

Is there a direct comparison of bowden vs the flying setup?

im using the half flying setup like DjDemonD and although its working well there are some slight patterns visible in the walls that i cannot fully account for.

for the elastics i suggest model balloons. the have the right size and stretch and are cheap.

Have you guys seen S0up's sextruder? (suspended extruder): [www.thingiverse.com]
I'm using one and I've drilled out the filament path to suit 3mm filament and it works great! my one -> [www.thingiverse.com]

EDIT: I did a geared filament drive for 3mm with a nema 17 motor [www.thingiverse.com] The complete unit weighs 480grams whereas the sextruder weighs very close to 200grams ( can't remember the exact number, it was 205 grams or something like that).

Printing 3mm filament with a bowden tube doesn't work well at all with very small prints in my experience. With the sextruder, retraction behaviour is basically indistinguishable from a direct drive. I love it. It's the best of both worlds.

Edited 5 time(s). Last edit at 12/07/2015 04:26AM by TTN.

Just seen it now thanks quite cool but I might rework my extruder to have the loops I'm using attached to it as they're much easier to string the elastic/balloons with.

i have done something similar with printed plate and some rubber band from a "stretcher" :







[www.youtube.com]
It's really a great update for retract problem

Edited 3 time(s). Last edit at 12/08/2015 05:38AM by rafffel.

Probably a silly question, but what's the advantage of suspending it from the top of the carriages by elastic rather than it being more rigidly supported from the top of the carriages, much like a 'second effector'?

The idea of the suspension is to enable the extruder to follow the effector moves without adding the mass/inertia to the equation.

How would you build a fixed platform between three carriers that move up and down independently?
-Olaf

As Olaf said it's a clever twist; the flying extruder only adds it's full mass and inertia to z moves, this is easily handled by the three motors in a delta. X and y moves have only a small amount of additional mass added to them equal to the tension in the elastic and bowden tube.

The main advantage is still a low mass effector in x and y so fast printing is still possible but you have a short bowden tube and the added bonus of using the space above the effector for the extruder which is better packaging.

I don't think I can see any real drawbacks.

The other solution and its a clever one is the flex3drive which keeps the motor off the effector but allows a direct extruder using a cable drive. I've ordered one and am still keen to try it but am very pleased with the flying extruder.

Edited 1 time(s). Last edit at 12/08/2015 12:46PM by DjDemonD.

What is the flex-drive position error?
I mean how many degrees of angular position error happens in the flex-drive per 10 N.cm of torque it is transferring?
If it not specified, can you measure it for us? :-)

I haven't got it yet but the inventor mutley3d is active on this forum and I'm sure he'll fill you in. It should be very accurate as it works via a worm gear, meaning you need around 2000 steps per mm at 1/16 microstepping. Just Google flex3drive.

Edited 1 time(s). Last edit at 12/08/2015 02:40PM by DjDemonD.

Ok, I sent him the question.

XY moves have a low inertia with the nema17 flying extruder, but with a geared 27:1 nema 11 direct drive it's even better while still being plenty powerful for 3 mm diameter filament. Weighs a lot less: 480g vs 200g

Sounds good I see there is a 100:1 nema 11 being offered also why not use that one? Would need much less motor current.