Thinner Filament.

Greetings to All of you!

Do any of you heard of a filament (PLA in particular) that is thinner than 1.75 mm?
Is it possible to purchase it somewhere?

I can't get away form the idea of developing very light weight direct drive extruder paired with 0.2 mm or even smaller nozzle.
It would be nice to print small parts with higher resolution.

I am aware that few tried forcing standard filaments through smaller nozzles. Most efforts were somewhat unsuccessful, due to significant strain on the extruder (and other factors).

Thank you in advance for any information.

I made my own 0.8mm filament and modified a 0.4mm nozzle with a brass insert. It works quire well with a pancake Nema 17 stepper, but the drive gear teeth bit too deep and I had to experiment with other gears. It also tends to jam when the filament guide isn't perfect.

Edited 1 time(s). Last edit at 09/20/2018 12:17AM by o_lampe.

The motor power required is mainly a function of the extrusion flow. If you reduce the filament diam., you will have to run at a higher speed to keep the same flow !
My current direct drive extruder (with a NEMA 14 l= 28mm), hot end, fan, X carriage, Z homing sensor etc... is about 200gr. To lower the mass without reducing the extrusion flow requires to go for a servo motor. The stepper motor is the heavy part and related to its power.
Now if you are OK for a lower flow, indeed you can scale down everything. 3D pen use a very small motor, yet a 1.75mm filament.

---

"A comical prototype doesn't mean a dumb idea is possible" (Thunderf00t)

... I have some ideas around "direct filament printing", where the filament is not melted in a hot chamber and extruded through a nozzle, but something like "direct" printing with thin wires (already in the "printing" diameter of e.g. 0.05mm), where the wire tip is positioned on the surface and while moving the head, melted with a laser and fed synchronously with slightly lower or higher speed to vary the diameter/volume of the melted track ...

---

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

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

Thank you MKSA.

You nailed it in your last sentence: "Now if you are OK for a lower flow, indeed you can scale down everything. 3D pen use a very small motor, yet a 1.75 mm filament."

I am OK with the fact that print speed will go down to achieve higher resolution. And I am very much embracing the point of scaling down everything.


What bothers me with 1.75 mm filament in such application is that area representing nozzle opening shrinks very rapidly:

1.75 mm diameter represents area of 2.41 mm^2
0.4 mm nozzle opening provides area of 0.126 mm^2 --------------------------------------> ~15 times smaller
and proposed 0.2 mm nozzle provides area of only 0.031 mm^2 --------------------------------------> ~78 times (!) smaller


To obtain similar extruder dynamics (as how it relates to filament melting in heated space, retraction, pressure, etc.) one would want to keep the area relationship at ~15 (as shown).
It would imply that by using 0.2 mm nozzle and ~15 times area ratio, the new filament diameter must shrink from 1.75 mm to 0.77 mm.

Finding filament between 0.5 mm to 1.0 mm would make it very much feasible to experiment with higher resolution FDM.
I hope, I was clear enough to convey my point.

Thank you o_lampe.

I just noticed the link in your post leading to your previous work - very interesting...
I will take a good look at your findings.

Excellent research o_lampe!

Regarding your conclusions on drive gears and teeth:

May be we could make assumption (based on vast empirical work accomplished by RepRap community) that popular drive gears, their geometry and tooth configurations are quite usable with 1.75 mm filament.

Could it be also true, that reducing existing drive gear by the same ratio achieved by reducing 1.75 mm filament to 0.8 mm filament would be good enough? Am I over simplifying?

Quote
MKSA
The motor power required is mainly a function of the extrusion flow. If you reduce the filament diam., you will have to run at a higher speed to keep the same flow !

I'm not good at math, but I doubt that.
I agree that melting power is proportional to extrusion flow, but from my experiments I've learned that the filament diameter has a significant impact on required motor power.
The torque of a current controlled e-motor is always max at low RPM and stays constant upto 50% of it's max. RPM.
Max RPM is proportional to V__motor. At 24V this means it can run faster than usually required for a extruder. E3D Titan is a good example.

You are right, that it has to run faster, but I say: it doesn't matter.

Edited 4 time(s). Last edit at 09/21/2018 12:18AM by o_lampe.

Quote
uticatechclub
Could it be also true, that reducing existing drive gear by the same ratio achieved by reducing 1.75 mm filament to 0.8 mm filament would be good enough? Am I over simplifying?

I don't understand what do you mean with reducing the gear?

Quote
o_lampe

Quote
uticatechclub
Could it be also true, that reducing existing drive gear by the same ratio achieved by reducing 1.75 mm filament to 0.8 mm filament would be good enough? Am I over simplifying?

I don't understand what do you mean with reducing the gear?

I mean making it smaller proportional to how much smaller in diameter the new filament became.

It would be interesting to produce thinner filament on-the-fly from 1.75 mm or 3 mm filament, i.e. remelt into thinner filament before it moves into the hotend. This remelt should be done protected from oxygen.

You could
- change the filament diameter to fit your nozzle diameter and layer height, throughput
- mix filament with other filaments to get true color or material mixes anlogue to alloys

Edited 2 time(s). Last edit at 09/21/2018 01:43AM by JoergS5.

Quote
o_lampe

Quote
MKSA
The motor power required is mainly a function of the extrusion flow. If you reduce the filament diam., you will have to run at a higher speed to keep the same flow !

I'm not good at math, but I doubt that.
I agree that melting power is proportional to extrusion flow, but from my experiments I've learned that the filament diameter has a significant impact on required motor power.
The torque of a current controlled e-motor is always max at low RPM and stays constant upto 50% of it's max. RPM.
Max RPM is proportional to V__motor. At 24V this means it can run faster than usually required for a extruder. E3D Titan is a good example.

You are right, that it has to run faster, but I say: it doesn't matter.

I didn't say it is null, of course the thicker the filament, the less flexible it is and thus more force to drive it.
Still, (unless a poor hobgear and filament path)most of the force required is to push the filament in the hothend and extrude it through the nozzle. The thicker the filament, the lower its speed to get the same extrusion rate. Power is force X speed, torque X ω (in radians/s)
That means for a 3mm fil. you may have to use reduction gears to get the higher torque required. It is seen in 3mm extruder using Nema17 while the same motor can drive directly a 1.75mm. Steppers are a bit like gas engines, you need a gearbox to get its full power

Edited 4 time(s). Last edit at 09/21/2018 04:33AM by MKSA.

---

"A comical prototype doesn't mean a dumb idea is possible" (Thunderf00t)

Quote
JoergS5
It would be interesting to produce thinner filament on-the-fly from 1.75 mm or 3 mm filament, i.e. remelt into thinner filament before it moves into the hotend. This remelt should be done protected from oxygen.

That was partly my 1st. idea, but pushing still hot filament through a Bowden tube is a no_go.

Quote
o_lampe

Quote
JoergS5
It would be interesting to produce thinner filament on-the-fly from 1.75 mm or 3 mm filament, i.e. remelt into thinner filament before it moves into the hotend. This remelt should be done protected from oxygen.

That was partly my 1st. idea, but pushing still hot filament through a Bowden tube is a no_go.

Yeah, youd have to miniaturize a a machine that is normally heavier than a 3d printer in order to do that

But i bet you could get the same effect by machining a cone shape into a volcano nozzle for example rather than the parallel sides that terminate in a your desired tiny nozzel,

'cat gut & rubber' wheels....maybe fine tooth metal,but i just like the sound of 'cat gut & rubber'
or some new cat gut variant? I know ppl have done it just not sure what the results where.

Edited 1 time(s). Last edit at 09/21/2018 07:20PM by MechaBits.

Quote
uticatechclub

Quote
o_lampe

Quote
uticatechclub
Could it be also true, that reducing existing drive gear by the same ratio achieved by reducing 1.75 mm filament to 0.8 mm filament would be good enough? Am I over simplifying?

I don't understand what do you mean with reducing the gear?

I mean making it smaller proportional to how much smaller in diameter the new filament became.

A finer pitch for the teeth would make sense, but reducing the diameter would cause more RPM for the same feed rate. Something like the Bondtech extruder would work. It would grip the filament from two sides.

Quote
o_lampe


A finer pitch for the teeth would make sense, but reducing the diameter would cause more RPM for the same feed rate. Something like the Bondtech extruder would work. It would grip the filament from two sides.

A smaller diam also means less teeth contacting the filament. As a matter of fact I made a 12mm diam hobbgear (teeth made with a M3 tap) to get more grip. A lot simpler than the Bondtech approach Stays cleaner too !

---

"A comical prototype doesn't mean a dumb idea is possible" (Thunderf00t)

if its too thin how would you stop it bunching up, would have to be a tight tolerance tube, and then as it gets near the heat would the filament have enough stiffness to create pressure?

That's why I chose 0.8mm filament. It matches 1mm ID Teflon tube. I was lucky to get a tube with 2mm OD. It fits inside the normal PTFE-heatbrakes.
It also needs proper guidance between gear and heatbrake to avoid spaghetti.

IMHO the chamber pressure can be much lower, since the cross-section of 0.8mm filament is so much smaller than 1.75 or even 3mm.
I had no problem with 1st. layer adhesion, but I have LokBuild on my printbed. PLA almost sticks too good on it.

Quote
o_lampe
That's why I chose 0.8mm filament. It matches 1mm ID Teflon tube. I was lucky to get a tube with 2mm OD. It fits inside the normal PTFE-heatbrakes.
It also needs proper guidance between gear and heatbrake to avoid spaghetti.

IMHO the chamber pressure can be much lower, since the cross-section of 0.8mm filament is so much smaller than 1.75 or even 3mm.
I had no problem with 1st. layer adhesion, but I have LokBuild on my printbed. PLA almost sticks too good on it.

Chamber pressure depends mainly on the nozzle diam and the extrusion rate (at the nozzle !). Now, the filament diam being smaller, the pressure being the same, the force is indeed reduced (prop to the diam ^2) .

An other thing, the smaller the filament diam, the more melted filament will be leaking up the heatbreak as people usually (unless you have very accurate drilling equipment) drill 2/10 mm more than the filam diam. For 1.75 I drill to 1.8mm then smooth the hole.

Anyway, buckling becomes the real killer and one has to reduce the distance from hobbgear to melt zone. This is why I don't like the E3D kind of attachment, heatbreak and heatsink that became the standard and made my own far shorter.

---

"A comical prototype doesn't mean a dumb idea is possible" (Thunderf00t)

I am thinking about whether it's possible to press the melted filament into the nozzle with other measures than the extruder.

If melted material can be pressed by e.g. Argon to avoid oxygen reactions, you could make overpressure or underpressure to control the melted filament flow.

The reason I am thinking about this is to think about printing aluminium.

Edited 1 time(s). Last edit at 09/23/2018 04:48AM by JoergS5.

Quote
JoergS5
I am thinking about whether it's possible to press the melted filament into the nozzle with other measures than the extruder.

If melted material can be pressed by e.g. Argon to avoid oxygen reactions, you could make overpressure or underpressure to control the melted filament flow.

The reason I am thinking about this is to think about printing aluminium.

Done for metal using plasma, laser, ion beam deposition.
VDX mentioned his current work, laser plus "plastic" filament.

Edited 1 time(s). Last edit at 09/23/2018 06:08AM by MKSA.

---

"A comical prototype doesn't mean a dumb idea is possible" (Thunderf00t)

Quote
MKSA
Done for metal using plasma, laser, ion beam deposition.
VDX mentioned his current work, laser plus "plastic" filament.

You mean laser plus metal powder? I mean melting metal (heat up with induction e.g.) and printing without laser.

Laser plus thermoplast is used at the FastFFF, it is a good idea.

Edited 1 time(s). Last edit at 09/23/2018 03:06PM by JoergS5.

... not powder - think about melting/fusing thin wires with the laser

---

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
VDX
... not powder - think about melting/fusing thin wires with the laser

Could you please show a picture or give my a youtube hint?

I know the HRL video, but this is powder imho.

=> I found your info on [reprap.org]
It is an interesting solution. The combination of different manufacture methods seems to open a lot of possibilities.

Besides welding I found the possibility of galvanic 3d printing, but it's very slow.

Edited 1 time(s). Last edit at 09/23/2018 03:00PM by JoergS5.

Sorry if I missed them in other threads (o_lampe), but what did prints look like, how often you use it, on what kind of parts.

Edited 1 time(s). Last edit at 09/23/2018 05:03PM by MechaBits.

... sorry, can't post more details or images as it's again an R&D project wit NDA's

---

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
VDX
... sorry, can't post more details or images as it's again an R&D project wit NDA's

I understand. Good luck with it!

Quote
JoergS5
Besides welding I found the possibility of galvanic 3d printing, but it's very slow.

... to give you a hint into this direction - think about drastically speedup (or slowdown) galvanic or etching reactions with lasers

---

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
MKSA
Anyway, buckling becomes the real killer and one has to reduce the distance from hobbgear to melt zone. This is why I don't like the E3D kind of attachment, heatbreak and heatsink that became the standard and made my own far shorter.

Here is my direct extruder. The filament path can't be much shorter. It's driven by a 1A/13Ncm NEMA17 pancake. ( forgot the exact weight )
The 28BYJ you might recon' was a fail.


Edited 1 time(s). Last edit at 09/24/2018 12:24AM by o_lampe.