This engineer claims 100 grams direct extruder. I want one

---

Viktor
--------
Aufruf zum Projekt "Müll-freie Meere" - [reprap.org] -- Deutsche Facebook-Gruppe - [www.facebook.com]

Call for the project "garbage-free seas" - [reprap.org]

As exciting as your work VDX is , it's in the Sci-Fi realm for most of us earthlings . Well, at least for me. We've deviated a bit from the sub-100g extruders we can achieve with what we can (cheaply) find now.

I am still pondering on the idea of a pre-feeder extruder and a compliant direct extruder on the printhead.
Could we explore a bit this idea if you see a value in it?

I don't know if this can be done without the bowden tube but that's how I first envisioned it. The tiny stepper would do the bowden's job of directing the filament into the hotend.
I might be totally off but I can imagine that if the tiny stepper is able to grip the filament and keep the same filament length between itself and the other stepper, the pushing force is already provided and the filament doesn't have anywhere else to go but through the hotend. The tiny stepper would only have to be in sync with the other one.

Please at least help me understand why this would or won't work so at least i'd put it at rest.

Edited 4 time(s). Last edit at 03/16/2016 08:05AM by realthor.

---

RepRap Lander concept on Concept Forge
RepRap Lander concept on RepRap Forums
My Things, mostly experimental stuff

Sadly, no, you would need something like the bowden for the remote extruder to do anything useful. The bowden is needed to transmit the force that the extruder generates, without it you just produce arbitrarily large loops of filament in the air. The exception would be if the filament between the two extruders was under tension, but that wouldn't be a useful scenario

I was afraid this could happen but couldn't clearly see it. Well...thanks

---

RepRap Lander concept on Concept Forge
RepRap Lander concept on RepRap Forums
My Things, mostly experimental stuff

... an option for a lightweight remote head could be built with some 500mW- to 1W -diodes arranged in a ring (here an example) and a thin tube feeding a dark (laser-energy absorbing) filament with e.g. 0.1mm diameter into the spot.
Then you'll only need a small motor to feed the filament and no heaters, only an aluminium plate holding the diodes to spread the heat to hold them below 40degC.

I've got some free samples of blue and black 0.3mm filament on spools - it's normally used for tooth-brushes, but there are other sources too ... for CO2-lasers even transparent nylon fishing line would be an easy to find and cheap source ...

Edited 1 time(s). Last edit at 03/16/2016 09:58AM by VDX.

---

Viktor
--------
Aufruf zum Projekt "Müll-freie Meere" - [reprap.org] -- Deutsche Facebook-Gruppe - [www.facebook.com]

Call for the project "garbage-free seas" - [reprap.org]

Wow, that's sexy! So the idea is to produce a droplet of molten plastic at a point, rather than an extruded cylinder?

hmm, very interesting. Could we use the already existing laser engraver modules more and more printers come with?

---

RepRap Lander concept on Concept Forge
RepRap Lander concept on RepRap Forums
My Things, mostly experimental stuff

... the idea is to touch the surface with the solid filament wire, and then to heat the tip of the wire and a small area around th tip of the surface, so it will melt together.

When moving the head and feeding the filament synchronous, this will lay a line of molten plastic on the surface.

If retracting the filament while heating(!), it will tear away without stringing, so you should be able to place single droplets slightly bigger than the fiaments diameter on the surface too.

It's essentially the same methode as build-up welding, but with plastic and much smaller energies, so the traces will be in the width of the wire diameter.

I've tested this some years ago with different materials and diameters - the thinnest were plastic and glass down to 0.03mm diameter and platinum with 0.01mm.

It's mostly the stiffness of the material, what's defining the thinnest possible wire diams - (hard) plastic is possible to push/feed with diameters down to 0.02mm, glass to 0.01mm, platinum or hardened steel (or tungsten) down to 0.005mm -- but here it's hard to find the sources for the wires ... and to train the ability to feed the wires into the capillary

---

Viktor
--------
Aufruf zum Projekt "Müll-freie Meere" - [reprap.org] -- Deutsche Facebook-Gruppe - [www.facebook.com]

Call for the project "garbage-free seas" - [reprap.org]

Quote
realthor
hmm, very interesting. Could we use the already existing laser engraver modules more and more printers come with?

... essentially yes, but youll best use 3 or more diodes (you can run more diodes with a single driver in series) to get unidirectional melting regardless of moving direction ...

---

Viktor
--------
Aufruf zum Projekt "Müll-freie Meere" - [reprap.org] -- Deutsche Facebook-Gruppe - [www.facebook.com]

Call for the project "garbage-free seas" - [reprap.org]

Sounds fascinating. What sort of parts cost would a suitable laser diodes + driver be?

... I'm sourcing/salvaging fibercoupled IR-diodes with 9Watts@975nm out from broken fiber-lasers (similar types can be found at ebay for 300 USD) - but they are only usable with dark/black material, which will absorb enough of the IR light.

Blue 445nm-diodes can be found for 30€ for the old 1W or 1.5W types, the newer 2W or 3.5W diodes are more like 80€ to 350€ here in Germany ... much cheaper at alibaba ...

Housings and focussing optics can be found from 15€ on.

The drivers can be easy DIY-ed with current regulator IC's - my first drivers used parallel wired LM317 or LM338 for currents of up to 9Amps ... but a single LM317 set to 1A can drive more diodes wired in serie - you only need to give enough voltage to support all the diodes +3V for the voltage drop at the chip.

Switching/pulsing is best done with a MOSFET.

So the cost for testing can be something from 50€ on with a single 445nm@1W-diode, holder, kollimator lens and simple constant current driver ...

---

Viktor
--------
Aufruf zum Projekt "Müll-freie Meere" - [reprap.org] -- Deutsche Facebook-Gruppe - [www.facebook.com]

Call for the project "garbage-free seas" - [reprap.org]

You make it sound so easy, one wonders why its not available already, was trying to think of the line feeding mechanism,
I like the Nylon fishing line idea, a couple of rubber wheels might do the job, but maybe fishing line is more expensive per kilo?
and more fiddly to feed? but I was thinking, the line could be pulled through feed tube for one arc in printed part, then fused, so the line could be one continuous line, tacked at far end then working back till it cuts the strand, but then it would have to get picked up again to repeat...if done in this way flexible parts could be possibly be made as result would be like a lattice of nylon strands....or welded all the way to make a stiff part. (just read about this picsima printer different but like something out of Viktors Lab [www.youtube.com] )

As for 2 extruders working in conjunction, I thought about that one myself a while back, but realized, I only wanted 2, because one might not work properly, but really the goal is one that does it properly and working single solutions do exist so why have 2,
what about an arrangement like the triple bearing feed screw idea, but instead of moving along a shaft, it actually pulls shaft of filament through the mechanism.

Edited 3 time(s). Last edit at 03/17/2016 01:37AM by MechaBits.

... look at the line-feeding mechanism in the 3D-printing pens (e.g doodler) - but small rubber wheels will do it too.

I've used the feeders from wire-bonders -- the feeding 'motor' is a set of piezo-clamps and actuators and capable of prezise feeding wires down to 5 microns diameter, if the material is stiff enough. Typical wires used in the wire-bonders are aluminium with around 60 microns or gold with 23 microns diameter.
I've tested them with 30 micron thick glass wires and 10 micron thick platinum for "additive welding" ... but had to develop a complete different feeding and handling tool for platinum "Wollaston"-wires with 1 micron diameter, as they were too fine for normal handling

---

Viktor
--------
Aufruf zum Projekt "Müll-freie Meere" - [reprap.org] -- Deutsche Facebook-Gruppe - [www.facebook.com]

Call for the project "garbage-free seas" - [reprap.org]

Topic Laser:
I like the laser idea, but I don't see a chance to control the filament temp in a close loop. A thermistor will be to slow to chop the laser energy. IR sensors maybe?
It gets complex pretty soon, I'd say it's difficult to DIY such a printhead.

Is it possible to feed the laser energy through fiber glass, so we could install the diodes on a fixed place away from the effector?
(4 diodes= 4 fiberwires; no knife edge array)

Topic dual stepper w/ Bowden:
I don't see a real benefit from such a setup. The "small" stepper would have to be strong enough to create chamber pressure, the "big one" would only need to overcome the friction of the Bowden tube?
All I can imagine is eliminating the backpressure in the Bowden tube and reduce retraction/oozing issues caused by the Bowden tube.

... the IR-diodes are already equipped with a glass fiber with 0.1mm core diameter ... other diodes can be fed into a fiber too - I'm doing this actually with a 6W@445nm-diode to compensate it's bad beam characteristics ...

You don't need a temp regulating - it's simply adjusting the pulse-time and energy to completely melt the wire tip, then it's a sequence of pulses correlated to the moving speed, to melt the new inserted fiber or release it from the molten blob ...

---

Viktor
--------
Aufruf zum Projekt "Müll-freie Meere" - [reprap.org] -- Deutsche Facebook-Gruppe - [www.facebook.com]

Call for the project "garbage-free seas" - [reprap.org]

How well would a laser-molten drop of filament bond to the lower layer? There would be no chamber pressure that pushes the drop onto the old layers.

Would we still need part cooling fans? If not, this would be another weight gain
What material would be suitable for the nozzle? A glass capillary tube or brass? We wouldn't need a heatsink, nor hotend fan

This new "hotend" would be soo light, how fast could we feed and melt the filament? 300-500mm/s?

With such feed rates we would need a new approach for the spool holder, since this would also be the turning speed of the spool.
(try to stop a 1kg spool for a retract... )

... look at the common build-up laser- or electrongun-welders - they melt a metal wire with the same methode, but with 'overpowering' the melting process to get a safe bond and fast build-up, so the resulting track ist mostly 3x bigger/wider than the original wire.

With finer wires and a laser beam not from top down, but more beams from the side this is much better to handle and with melting the wire tip and the underlying surface together with a single pulse this will result in a safe fusion without 'overshooting' in track width.

Cooling could be usefull when melting small/fine details, otherwise you have to wait, until the layer cools down before starting the next ... but here too a thin tube blowing air on the spot (or Argon when welding/melting metal) would be enough.

When pushing the wire into the melting spot, it will melt and fuse to the next surface in contact - so no compressing forces needed.

While FDM-printing, "compressing" is mostly used to flatten the extruded much thicker filament to e.g. 0.2mm height ... when using filament with 0.1mm diameter, the resulting molten tracks without "compression" would be not much thicker than the filament itself, so 0.1mm wide and high.

With "overextruding" while melting it, the track width/height can be made thicker than teh filament by adding more molten material on the spot ... "stretching" with higher moving speeds could be possible too, but I didn't test this when building/testing the prototypes ...

---

Viktor
--------
Aufruf zum Projekt "Müll-freie Meere" - [reprap.org] -- Deutsche Facebook-Gruppe - [www.facebook.com]

Call for the project "garbage-free seas" - [reprap.org]

As long as the talk is about lightweight hotends... Which printer design could actually benefit from it? I can only think of Delta.

Quote
Edvardas
As long as the talk is about lightweight hotends... Which printer design could actually benefit from it? I can only think of Delta.

Low mass means that you can use higher accelerations and/or that the mechanical parts can be scaled down. Delta printers gain the most, but it can help improve every kind of printer.

---

[www.bonkers.de]
[merlin-hotend.de]
[www.hackerspace-ffm.de]

Quote
Srek

Quote
Edvardas
As long as the talk is about lightweight hotends... Which printer design could actually benefit from it? I can only think of Delta.

Low mass means that you can use higher accelerations and/or that the mechanical parts can be scaled down. Delta printers gain the most, but it can help improve every kind of printer.

How would a cartesian printer benefit of it if there is still that heavy bed on Y axis rocking around?

It is not to hard to keep the printbed lightweight. If you use something like a Bulldog XL the x carriage is easily heavier than Y.

---

[www.bonkers.de]
[merlin-hotend.de]
[www.hackerspace-ffm.de]

Quote
Srek
It is not to hard to keep the printbed lightweight. If you use something like a Bulldog XL the x carriage is easily heavier than Y.

Is there any information on lightweight bed designs and materials used? I was looking into it myself and end up with an idea that dibond might work (not heated bed) but I have never had that stuff in my hands and would guess that it is not as flat as one can hope. I am currently very happy with how flat 6mm precision machined aluminium plateis but it is definitely not light.

Edited 2 time(s). Last edit at 03/18/2016 01:34PM by Edvardas.

Currently i use 3mm Dibond to support a MK2 Aluminum printbed that i print directly on. Just 4 M3 screws at the corners to mount the heatbed and 4 bronze bushings in printed holders under the dibond. Very simple, lightweight and works nicely.

---

[www.bonkers.de]
[merlin-hotend.de]
[www.hackerspace-ffm.de]

Look at the kind of bed Felix 3D printer uses...I am looking for some time to get my hands on such thing but nothing yet. All chinese cmpanies that sell corrugated aluminum panels expect orders of min 500m2 or so.

Edited 1 time(s). Last edit at 03/18/2016 04:11PM by realthor.

---

RepRap Lander concept on Concept Forge
RepRap Lander concept on RepRap Forums
My Things, mostly experimental stuff

Quote
realthor
Look at the kind of bed Felix 3D printer uses...I am looking for some time to get my hands on such thing but nothing yet. All chinese cmpanies that sell corrugated aluminum panels expect orders of min 500m2 or so.

Then stop looking and buy this: [www.felixprinters.com] I would even call it cheap.

Oh and if you do please let us know the weight.

Edited 1 time(s). Last edit at 03/18/2016 04:45PM by Edvardas.

Quote
Edvardas

Quote
realthor
Look at the kind of bed Felix 3D printer uses...I am looking for some time to get my hands on such thing but nothing yet. All chinese cmpanies that sell corrugated aluminum panels expect orders of min 500m2 or so.

Then stop looking and buy this: [www.felixprinters.com] I would even call it cheap.

Oh and if you do please let us know the weight.

) I am aware of it and the price too. I just didn't start yet sourcing parts for the bed. I'll let you know if I decide on it

---

RepRap Lander concept on Concept Forge
RepRap Lander concept on RepRap Forums
My Things, mostly experimental stuff

A lightweight head design would make it easier to use more than one or two filaments.
One could build a tiny extruder with several filament channels guided into one nozzle. No more oozing or colors mixing in the heater chamber.

... one possibility with "spot-on-melting" is to use not one, bur more filaments, pointing on/into the same spot -- with either different colours or/and different materials and diameters - this can result in a pretty small "multimaterial" printer head and some 'advanced' 3D-printing methodes

I'm tinkering since some ten years now with this sorts of manufacturing and development - sometimes in projects (then mostly for scientific research and under NDA), sometimes at home for my 'long-term' developments

---

Viktor
--------
Aufruf zum Projekt "Müll-freie Meere" - [reprap.org] -- Deutsche Facebook-Gruppe - [www.facebook.com]

Call for the project "garbage-free seas" - [reprap.org]

The lightweight extruder makes a higher acceleration easier, when the same construction for moving x/y is being used (with either corexy, cartesian or delta printers) the ringing effect is also smaller. About the Felix printers; The moving beds mass is one thing, but the problem with a horizontal moving bed arrises with a higher (less stable) or heavier print. I'm not a fan of that concept, but Felix makes a neat printer IMO, (relative slow printing though).u

Edited 1 time(s). Last edit at 03/20/2016 02:51AM by to3dornottobe.

I love the precision you've been working at, but wonder what the print times would be with filament so small?
Having a laser doing the work means speeds could be improved but all the mechanics would have to be spot on, you could use hypodermic needles for the filament(fishing line) guide throats, or would you use glass, UV Led with fibre optics? knife edge array would look like an evil genius device from a bond movie(talking of bonds [notaglue.com] though with 4sec cure...print times through the roof), surgical precision needed, different build a machine built for speed with tiny filament things would take a while, wonder about realistic speed machine would need to be designed to handle.
Fluid Pumping wonder if this would be any use [youtu.be]

Edited 1 time(s). Last edit at 03/20/2016 07:42AM by MechaBits.