My first printer–a Pruse i3 Hephestos–arrives tomorrow.
I’ve just learned why bowden extruders produce higher quality prints and I’m wondering if anyone has tried combining the bowden with the direct drive extruder to gain the advantages of both (direct drive being better for soft filaments).
ermm not to sure where you are heading with this! a bowdon has a direct drive gear but it is fixed away from the head, this reduces the weight of the moving head and allows for faster printing, a direct drive should be able to achieve just as good quality, just slower! one advantage of a direct drive is that it is better for flex able material like ninja flex, this is due to there not being a massive long tube where compressive force swell the diameter of the material and cause it to jam! Bowdens can print this but very slowly.
Maybe give a more detailed explanation of your idea, as if you could come up with a better way of driving the material and maintain and good quality print you would have a very unique product!
Please explain, how you would like to make a bowden-direct-extruder, I don’t get it. Bowden means, that the extruding motor is in a position where it doesn’t move. Because the printhead is moving, the motor has to be connected to the hotend with a PTFE tube, also called bowden tube, hence the name. A direct drive extruder sits on the moving printhead, as close to the hotend as possible. This reduces the pressure in the tube between the hotend and the motor and allows printing of flexible filaments, like ninjaflex or Brassfill. Direct drive extruders also can work with smaller retractions, but can’t print as fast as bowden systems. Those systems are different in their concept and can’t really be combined. Fast speeds require a motor, thats not on the printhead, a motor that isn’t on the printhead has to be connected to it with a tube and that makes flexible filament printing very difficult. Cheers, Marius Breuer
One comment, you can print Brassfill/Bronzefill/Copperfill very nicely on a Bowden, I have had a job printing for the last 24 hrs in copper fill on my UM2 with. I have also got a total of nearly 120-150 hrs of printing bronzefill.
I do agree with the Ninja flex though, or any flexy filament these are very tricky to get going on a Bowden.
I assume, you’ve also printed in brassfill before, yes it’s relatively easy to print with bowden systems, but it depends on the printer. Especially the 1,75mm filament, that I have to use, is flexible and the leapfrog creatr HS doesn’t have a very great extruder… There’s quite a large gap of circa 1cm between the “contact zone” of the drive gear and filament to the typical 2mm hole that goes to the bowden tube. Generally speaking: 2,85mm or 3mm filament is a little bit less precise (0,05mm diameter tolerance effects larger diameters more, then smaller ones), BUT thicker filament isn’t that flexible and elastic so it prints easier on a bowden printer. Also brittle materials like PLA won’t break that easy. The only downside is, that the reduced flexibility can be a problem, when your bowden tube makes crazy turns. In that case the filament won’t follow the path of the tube that easily and can cause extra pressure in the bowden system. Cheers, Marius Breuer
ahhh. I see where you are coming from with that, my um2 uses 3mm… ;0). lol, also my drive gear has a clip over guide that eliminates the gap between the gear and the start of the tube.
I have been thinking about something like this for a while. It would Work by placing the motor away from the hotend. Connecting a flexible axle to the motor, and to a “standalone” extruder gear placed just above the Hotend, functioning as a direct-drive.
Main drawback, and the only one I have thought about so far, is the inherent flexibility of the axle, as it would mean accurate retraction would be hard to achieve - I havn’t tested it in practice, so retraction might Work with a bit of firmware tweaking.
Interestingly I’ve made up similar thoughts for a pellet extruder. Due to a driving screw (instead of a drive gear, because pellets are used) the printhead is already heavy. I was planning, to get an lulzbot printer (similar to the i3). I could mount the motor on the X Axis and with a “D” shaped axis, it would be possible, to transfer rotational movement on a gear, that also van slide on the axis. The “D” shaped rod has to be parallel to the X axis and the gear on this rod, can bemused for pellet or filament extrusion. However this is fairly expensive. You don’t only need a matching printer, but also at least two customer milled gears and the rod. The gears and the rid have to be very precise. 3d printed gears would… a) have too much friction to slide on the rod b) have too wide tolerances, they would either be too tight or have a lot of backlash c) plastic is weaker then metal, you can’t transfer a lot of torque via those gears If you have a single extruder printer, you should go with a direct drive system. For dual extrusion a bowden system is recommended as two NEMA 17 are quite a mass to move. To ensure proper quality, you don’t want to print faster then 150-200mm/s and NEMA17 can handle such speeds with a single extruder direct drive system, for dual extrusion you would have to swap to NEMA 23 or bigger, but these require more amps. Which normal drivers can’t supply. You’d have to get stronger stepper drivers and most likely a stronger PSU, so all in all the investment isn’t worth it. Cheers, Marius Breuer
You might have an idea using the flexible drive shaft idea, you can find them pretty cheap like this http://www.amazon.com/Stalwart-75-TJ9975-Flexible-Drive-Shaft/dp/B003JBWUJ6 117. However there is inherent slop in the cable which will make retraction similar to a bowden system. Also not sure how much torque you are going to be able to achieve with that. Finding a way to secure each end of the flexible shaft housing will be important. Definitely worth a try though and would make an interesting prototype. You could just adapt a standard wade extruder to accept the flexible shaft output and mount the motor to the top of the printer. That would keep it out of the way of the print area and give it enough room to move.
I allready own a flexible drive shaft from my Dremel. Compairing this to Bowden is not fair for bowden designs.
I imagine these flexible shafts to be much worse as it takes some spins before it starts moving in the other end - might me compensated in firmware though.
I just imagined using some standard 3x5 “clamps” to connect the motor to the shaft to the gears. Clamps as in the ones used for many z-Axes to connect motor to vertical threaded rod (english word for it eludes me atm).
Regarding torque. The shafts has ok torque but a small geared extruder setup might be needed.
Directly combining the two would be tough as the pull from the two systems would fight each other if not perfectly synced and probably cause stripping of the filament at one end or another. Some of the indirect drive systems below are probably a good solution (no additional weight at the extruder head, but still provides extrusion and retraction right above the hotend), or you could potentially have a system where you disengaged one or the other depending on the filament you were running, but then it would be an either/or benefit, not necessarily the combined benefits of both.
Found that it has allready been “invented” http://www.mutley3d.com/Flex3Drive/ 226
edit: also this one: https://ultimaker.com/en/community/view/7987-flexible-shaft-the-best-of-both-worlds 157
Thanks for this. I just ordered one, and if it works, I may order a second so I can run dual extruders on my Prusa i3 Hephestos.
