Extruders

As you may be able to tell from my other thread, I'm designing a printer. And a big thing I'm trying to decide on is the extruder (that and build area but that's easier to customize later).

The obvious question is bowden vs direct, and my answer is I'll design it so it can use both interchangeably. What I really want opinions on are geared vs direct, and single sided (with idler) vs double sided drive (two driven hobbed gears grip filament from both sides).

I only really have experiences with a direct drive mk 8 type extruder, so other people's experiences and opinions are very much welcome. I understand that gearing the motor gives you more torque and more resolution, but what is that torque good for (able to force more filament through the extruder so faster print speeds than non geared? Or do printers never go so fast as that this becomes a problem), and is the extra resolution necessary?

Secondly, with double sided drives like the bondtech, this means you have more grip on the filament and are less likely to slip and grind the filament, but again, is this really an issue? I also imagine you'd have more precise control than with an idler and can do faster retractions and extrusions as the hobbed gears or bolts are less likely to slip when there's two of them.

Thanks in advance for any ideas, opinions or suggestions!

I can chip in a bit.
Got two makerbot clone direct drive machines and one 200x270 delta with a one sided drive. Ie: grooved wheel pushes the filament against a single hobbed bolt on a bowden system.

the bowden prints faster and with at least as much precision. It will also print ALL flexible filaments, faster than the direct drive machines.
Even ninjaflex and filaflex.

So going on recent experience - you definitely do not need a double hobbed bolt drive setup.
The really important thing is to make sure there are NO gaps anywhere in the feed system. Filament must be given no alternative route right from the extruder to the nozzle. I think that's why the bowden setup prints flexibles faster and better than the direct drive machines. I've eliminated any gap in the feed system and got the feed tube down to it's optimal (shortest length without sharp angles being able to form) length.

I fully intend to make sure there is a fully supported filament path, regardless of any other choice I make for the extruder i.e bowden/direct drive, geared or not, single or double sided drive.

Just a question, what sort of speeds do you print flexiblez at on your bowden? And what sort of speeds do you print at in general?

The double sided drive and geared extruders from what I understand only really shine with high print speeds, you dont need the extra grip and torque at 80mm/s but you might at 150/200 mm/s

with the replicator clones max speed is 65mm/s for pla and around 20-30 for polyflex - flexible pla.
Neither of them will touch ninjaflex or filaflex.

So far on the he3d k200 delta. I've gone up to 100 for pla - at 0.3mm layer height - so a fair volume of plastic going through the nozzle, 50 for polyflex and 20 for ninjaflex and filaflex.
Petg - I think - I printed at either 50 or 60 with 0.1 and 0.2mm layer heights.
Pretty sure she'll go up to 150 for the right high speed pla.

I'm hoping to make a high speed printer, definitely above 100 mm/s for pla (thinking 150+). As for flexibles, with some minor modifications to my mk8 style extruder I could get TPU printing at 60-70 mm/s at 0.25 layer height (I'm a very impatient fella haha) so I'm aiming for better than that, at least for my direct drive variant.

But yeah, I agree, at 100mm/s PLA you wouldn't need or possibly even see the difference between single and double sided drive.
This thread has a good comparison photo if you scroll down.
https://ultimaker.com/en/community/1...1#reply-120122

Flexion:
https://flexionextruder.com/

or

Titan:
http://e3d-online.com/Titan-Extruder...FSsW0wod10wE4Q

These are the top end solutions on the market right now. Flexion is a direct drive in the sense that the hobbed gear sits right on the stepper motor shaft 1:1 but it has a brushed wheel that keeps the hobbed gear clean.
The titan has 3:1 gear reduction for finer precision or more steps/mm of filament travel 3x more and also all the extra torque that comes with 3:1 gear reduction.

Have you used either/or any similar extruders? The flexion isn't too different from any other extruder, the brush is there because the deeper hobbed gear grinds harder filaments like pla and abs (non flexibles) a bit more, and it has a fixed, positive pressure idler rather than a spring loaded one. That last feature I find pretty cool and I'll make sure my hot end can do both spring pressure and positive pressure and switch between them easily.

As for the gearing, I'm not sure how useful it is, what positive effect does it have on prints and is that finer resolution noticeable in the end result? Again, I suspect it shines more at higher speeds where you need to push more plastic through the extruder.

The gear reduction will shine more at slower speeds and also in more precise retraction control. because it is 3x as many steps to travel the same distance. It also shines when you start dropping layer height and / or nozzle sizes and the filament is always moving slow. The gear reduction lets the stepper motor spin at a faster rate which is nice when it is barely crawling along. As far as experience with extruders goes ya I have a little under my belt. In fact of the 4 printers I own 3 are dedicated to never ending upgrades, mods, and conversions for the sake of collecting knowledge. Here is one printer of mine that has seen 4 different extruders: http://3dprintboard.com/showthread.p...Mixing-Printer
And it is continues here: http://3dprintboard.com/showthread.p...twork-Printers

That printer runs a titan extruder now. Here is how I have my titan configured..

http://i1050.photobucket.com/albums/...oWiz/121_1.jpg
http://i1050.photobucket.com/albums/...oWiz/124_2.jpg
http://i1050.photobucket.com/albums/...oWiz/123_2.jpg

My Tevo Black widow printer runs a proprietary extruder that has 5:1 gear reduction..

http://i1050.photobucket.com/albums/...oWiz/37_12.jpg

And My GEEETech i3 runs an MK8..

http://i1050.photobucket.com/albums/...toWiz/45_6.jpg

I've read that gear reduction actually makes retraction less precise, due to backlash in the gears. I'll look at gear reduction, and I'll probably end up making a geared and un geared version of my extruder. To be honest I quite like the idea of a double sides drive, it looks cool at the very least and the thread I linked has some evidence that it helps the extruder keep up at higher speeds, plus it would add a bit to precision since you've got a more secure grip on the filament.

My main question at this point is whether a geared or direct driven extruder can extrude at a higher rate, I understand geared has more precision for obvious reasons.

Grip does not equate to precision in 3d printing. reliability maybe. for precision we hunt more steps/mm of travel. Double drive gears do nothing for this effort so they can have no impact on precision. And I think what you read about backlash was directed more towards wade extruders with printed gears that must be set up loose so they don't bind rather than the precise mesh of the geared titan extruder that has no backlash that can be felt by a human hand. What you are obviously missing here is that stepper motors spin at a very slow speed in extruders. So low of a speed do these extruder stepper motors have to live at that companies make gear reduction extruders to the tune of 3:1 and 5:1 because even when we are printing at a fast pace these stepper motors spin at a fraction of their max potential while pushing filament.

I wont be using a metal gearbox or injection molded parts like the titan, since they are too heavy, expensive, or both for my application, so I'll have to consider the backlash of printed gears. That being said, the quality and backlash of gearboxes varies a lot based on manufacturer, and a couple of degrees of backlash, which is negligible to a human hand, is several steps of the motor.
Also, that reliability is important, if you're running at 100+ mm/s and your filament is slipping even let's say 5-10%, that, to me, is lost precision. if your motor changes e steps based on the speeds due to slippage, you no longer have a precise (or accurate, if you prefer) extruder, no matter how geared down it is.

My main question left, though, is still which extruder (geared vs not) can push plastic through an extruder faster. I suspect it is the geared extruders because as you say, an extruder doesn't push the limits on the motors speed, so the added torque is likely more beneficial than the loss of top speed which you wont end up reaching. I know the power rating of the heater cartridge in the hot end also limits your extrusion rate, but lets assume that's a non issue.

If you are worried about pushing a lot of filament fast then you will want a volcano hotend. And Just so you know the titan is one of the lightest extruder setups you can get because the gear reduction allows for a lighter stepper motor because not so much torque is needed. Here is the newest version of the titan the aero:
http://www.ebay.com/itm/3DMakerWorld...MAAOSwFJBZRCHe

But if you are looking to push a lot of filament fast then you will want this titan:
http://www.ebay.com/itm/Titan-Extrud...UAAOSw4shX5fVG
This e3d v6 with volcano hotend:
http://www.ebay.com/itm/3D-printer-A..._JbuP21GM3FbkQ
And maybe this stepper motor:
http://www.ebay.com/itm/Nema-17-42mm...8AAOSwHoFXqtAx

And you know if you are chasing speed it is a real good idea to run all this hardware at 24volts.

I'm not sure most NEMA 17 motors are rated for 24V. Also, this thread is about designing an extruder, not buying a ready made solution. I'm not really planning to design and build an affordable printer just to buy an extruder to double the price.
Also, I get that a larger nozzle lets you pump more filament through (since less force is needed for the same volumetric flow), but to be clear, I'm asking for any given nozzle, with a heating cartridge that can supply sufficient heat, where the only thing changed is if the extruder is geared or not, would the geared or non geared extruder be able to push filament faster?

Quote:

---

Originally Posted by Trakyan

I'm not sure most NEMA 17 motors are rated for 24V. Also, this thread is about designing an extruder, not buying a ready made solution. I'm not really planning to design and build an affordable printer just to buy an extruder to double the price.
Also, I get that a larger nozzle lets you pump more filament through (since less force is needed for the same volumetric flow), but to be clear, I'm asking for any given nozzle, with a heating cartridge that can supply sufficient heat, where the only thing changed is if the extruder is geared or not, would the geared or non geared extruder be able to push filament faster?

---

Ok first off the forward current I measure to the stepper motors is usually under 1v. BUT ANYWAYS, If it help you understand MOST control boards support 12 or 24v input. And most printers run nema17 stepper motors. In truth even the heater cartridges and heat beds that are optimized for 12 or 24v can be run in either scenario as they are controlled by a pwm circuit that lets you tune it in the form of ki, kp, and kd settings in marlin firmware. The reason you want the volcano hotend is because it stands the heater cartridge up on its side and gives more heated surface area to the incoming filament. See at high speeds the incoming filament wants to cool the nozzle faster than the heater cartridge can heat it and if you look at the volcano hotend it is easy to see this is what it is trying to correct for. And if you are buying many components to make up an extruder than technically you are still building it, right?

I think you're really missing the point of why I started this thread, and my previous replies.

I think the volcano also has a more powerful heater cartridge, and I understand that the heating of the hot end is a limiting factor, that's why I said before to assume it's a non issue since I'm only interested in the geared vs non geared part of the extruder. This thread was supposed to be about extruders, not hot ends.

I started this thread to get ideas and opinions on DESIGNING an extruder, and which features to include or choose (direct or geared, single or double sided drive etc. anything you can think of) for this particular design. I did not start it asking which extruder to buy and which hot end to screw into it, which to be honest is not my idea of building it. Two components hardly counts as 'many'. And again you're missing another point, the E3D Volcano and titan combination cost 70 pounds at minimum, that's the old titan and the cheapest volcano kit I could find on E3D's website. I'm not looking to design a low cost printer just to slap on a hot end and extruder that doubles the price. That last bit isn't an exaggeration, if anything it's an understatement. Even with those E-Bay suppliers the hot end and extruder combo will cost about as much as the remaining mechanical parts.

To re iterate. I am not looking for specific hot end and extruder combinations, I'm not even interested in the hot ends for this topic. I'm looking for features which people like in a hot end. Accessibility/visibility to the filament path, fully supported filament path, quick release/filament change, spring loaded or positive pressure (like the flexion), since or double sided drive (like the bondtech). I'm looking for features and ideas I should include to my own extruder design, not who else's design I should buy.

If you're building an extruder from scratch, why not build it with an auger screw, so it can handle pellets rather than just filament? Most thermoplastics come in pellet form, and are a lot cheaper that way than when extruded into filament. It would also give you more flexibility in terms of nozzle size - you can design a set of interchangeable nozzles, so you could switch between fast and thick deposition and slow and fine. If it works, let me know, because I want one...

If it was run by a stepper motor, you would be able to throw it into reverse for retraction.

Quote:

---

Originally Posted by Trakyan

I think the volcano also has a more powerful heater cartridge

---

I believe most if not all hotends use the standard 40watt heater cartridge. Someone correct me here if I am wrong. But so far as I know that is the standard. It is not a temperature issue with pushing a lot of filament. It is a surface area issue because the filament is at room temperature so as it enters the hotend it wants to cool the nozzle and maybe at higher speeds you can actually see a temperature difference between the nozzle that is being cooled by the incoming filament and the heater cartridge. Because we want a specific temperature and cant just go hotter to compensate for the temp drop from the incoming filament at the nozzle we need more surface area of the heating source to overcome the cooling capacity of the incoming filament without just raising temperatures. I hope I have explained this well and cleared up the confusion you were having about volcano hotends. And I will backout of your thread now. Sorry if I was irrelevant.

Quote:

---

Originally Posted by awerby

If you're building an extruder from scratch, why not build it with an auger screw, so it can handle pellets rather than just filament? Most thermoplastics come in pellet form, and are a lot cheaper that way than when extruded into filament. It would also give you more flexibility in terms of nozzle size - you can design a set of interchangeable nozzles, so you could switch between fast and thick deposition and slow and fine. If it works, let me know, because I want one...

---

Augers are not super precise with their deposition, filament makers use an auger screw, but they use the tension on the filament to control diameter. Augers are not positive displacement pumps so they don't really such and push like you're thinking. Augers more 'suggest' which way things should go rather than demand, like how a filament extruder pushes or pulls filament.

Anyways, onto the real issue. You'll need to use direct drive, since the plastic wont stay hot and molten through a bowden tube, and even if it did an auger would not be able to force the molten plastic through since it isn't a positive displacement pump. Next let's ramp up the heater block and cartridge, because you will need a much bigger melt zone to cover where your auger is. You'll also need a fairly big auger because of the size the pellets come in. Now add a big (and noisy) vacuum to feed this extruder with pellets from an outside source (or mount a barrel of them onto the extruder, but good luck moving it then). Basically, you end up with this.
http://imgur.com/AITgxiz
See that slightly larger than normal hotend? It's actually massive.
http://the3dprintingnerd.com/wp/wp-c...umbnail-18.jpg
Don't forget the shop vac to feed the pellets.

You wont be using that on any desktop machine, no matter how rigid you want to boast your set up is. Filament extrusion is probably one of the few ways a bowden set up would work, and it allows direct drive to work without carrying too much material on the extruder at once. I think it's as good as it's gonna get for thermoplastic FDM, a couple of decades of R&D haven't found a better method for thermoplastic FDM.

And you're right about the volcano having a standard heater cartridge, I just remember reading somewhere a while ago that they used a more powerful one. But I understand how the volcano works and why the heater cartridge is positioned the way it is and that incoming filament cools the hotend, this is why I said earlier which extruder would be able to push more filament, if we assume the hot end heating the filament is a non issue.

I'm working on the extruder right now, one of my goals is to keep it within the face profile of a standard nema 17 (42x42mm). This of course comes with some space issues as I'm trying to make it geared. I'm considering making one gear stick out slightly as a sort of thumbwheel for manual extrusion/retraction to help with loading and unloading filament, any thoughts?

Anyways, my current dilemma is how much of a gear reduction should I use? I'm wanting it to be usable with the pancake 20mm nema 17s, and I have two reduction ratios right now, a 1.8:1 and a 3.06:1. The E3D titan which people use with the stepper's I plan to use has a 3:1 ratio, is this "just right" or does it provide unnecessary torque? Would 2:1 or even direct drive work? I'm asking mostly for higher print speeds upwards of 100mm/s.