Is anyone using a direct drive extruder on their printer?
If so, what recommendations might you have for someone considering it?
I am thinking I want to ditch the bowden and do direct. I was eyeing Walter’s flying extruder and also thinking about just putting it on the carraige directly.
Thanks in advance
Dennis
Walter said he had ringing with it directly on the carriage (though his prints were so impeccable his standards might be higher than mine). I Know some people used this design I made for direct drive:
Other than for flexibles, i have never had much issues using bowden. But I agree, direct drive extruders sure make calibration easier and more consistent across print speeds and materials.
I was looking at what Walter had published-
http://thrinter.com/cartesian-flying-extruder/
BTW: Anyone heard from Walter at all?
I am not sure why he used a jointed arm- I would think that it could add more complexity in the system. Though he does mention a bit of camber in the system to unload the carriage. I think that the height this way would remain constant with the 2 arms.
I can see the shock cord snubber that he used. It looks like he has used a piece of extrusion as a mast to counter balance the whole thing in the back.
Hey Dennis,
I loved all of Walters designs and was afraid something happened to him. After some creepy googling it looks like he found a job in a large firm, I think consulting for 3d printer manufactures. He even write a scientific paper work a few others about 3d printer. My guess is he is either really busy or is in the 3d business industry and is no longer using sites like thingiverse.
I’m using Eric’s Titan DD carriage from time to time to print PLA.
There’s certain PLA (ColorFabb black and red) that I can’t get to reliably print on the bowden setup as it keeps underextruding and then overextruding on the nozzle (not an extruder issue but filament must be getting jammed in the tube).
It works well, though I de reduce speed a little compared to the lighter bowden carriage.
I have been working on remixing Eric’s BMG carriage a bit-
I stretched in the X & Y directions to make room for LM8UU bearings to clear each other vertically. After getting some crappy bearings. I took a gamble on the EASE ones from Triangle Lab. Honestly, I am really impressed by the quality. They appear to be fully ground for squareness and length on their ends, the cylinder also look like they are This is the 4th iteration of the carriage. I am literally going to print my first prints with it this weekend.
I had some layer registration issues with the last versions that I hope will be cured with the better bearings.
I ended up opening the bearing housings because no matter what, I could NOT get them to size just right for slip fits without increasing the preloads on the bearings. I also relieved the tops of the bores so that any sag that might occur if you print without supports would not interfere with the bearings.
There is a touch of misalignement on one pair of bearings on my last print- I chased the bores with a 15mm drill bit, but did not get too agressive with it. I really want a 15mm chucking reamer for Christmas (a full set really, but I’d settle for a 15 & a 16mm). The bearings on that axis bid ever so slightly- I think that with time the PETG will relax enough that it might settle in on its own. You can feel the rod slide nicely with one or the other bearing, but with both, you can just feel a tough of binding. Half-a-blonde one as some might say.
I also modified the V6 part coller to be 8.5mm lower than the original in order to use a Volcano hot end. I also built in support blocks 0.2mm seperated from the part faces. I got some sag in the original prints.
I also modded in a 3rd screw to support the fan. The combined weight and moment arm would tweak the fan a bit on the rear screws and sag.
So far, my only dislike is that the cartridge heater wires are too stiff to route nicely.
I am hoping for the best - I will post up proof prints when I can.
Dennis
Looking great as always Dennis.
still using it? working fine?
I still haven’t tried it yet on my printer. I should really get around to setting up the carriage so it’s ready to be dropped in and swap it in just for testing so I know whether or not I should be recommending it to people.
I haven’t bothered just because I’ve got other printers with direct drive so I don’t really need to run flexible materials on my Eustathios. And when the printer’s all dialed in I don’t really run into issues even with the long Bowden in everyday printing.
@Eclsnowman lost track of your printer since G+ is gone, what are you using right now? (might be a little off topics).
I decommissioned my old Delta, one arms broke (again) and I wanted to upgrade… and also wanted a heatbed, so I decided to move to a CoreXY (and never finished) so I buyed an Ender3 for now, moved most of the Delta stuff into it (Bondtech V2, E3D, …). Just want to finish the “cube”, but not sure about the model and I wish to have it direct drive.
As far as printers I have my big HercuLien, then recent upgrades to my Eustathios, then the Talos3D Spartan, and Talos3D Tria Delta, The HobbyFab BoxR, a Printrbot Adoptabot (Highly modified Simple Pro), and a modified Ender3. I really like the Delta and the Ender 3 for quick prints with good surface finish and accuracy.
For bigger prints I normally go with the Eustathios or the BoxR. And for flexibles I almost always go with the Spartan (It probably gives me the best print finish of any of my printers). I’ll be honest I haven’t even powered on the HercuLien in a year and a half. I only bother using that when I’m going to be printing something very big. But now that I have a CNC (HobbyFab/SMW3D R7) router a lot of the big stuff I try and do and subtractive versus additive processes.
I used to do more big prints when I worked at the foundry making prototype patterns before we cut them out of tooling grade materials. For that it was great because near-net shape manufacturing like casting has a lot of variables that come into play due to chemistry and thermodynamics causing shrink and deflection. So being able to prototype the pattern and figure out what changes you want to make prior to cutting them on a CNC which might take days of process time It was really handy.
If I ever build another printer it’ll definitely be core XY, just so I’ve got the full gambit covered. My first ever printer design was a core XY back in the day. But I’ve learned a lot since then so I would definitely tackled it with different linear motion components than I did the first time around.