Dual Direct Drive Extruders Concept

Hello everyone!

I've been thinking recently about trying to salvage the stock j-head from when I upgraded to e3d v5, and thought about dual extruders. I knew there was no way to fit 2 gear sets and 2 steppers on the carriage without losing an unacceptable print area. I found this guy on thingiverse:

http://www.thingiverse.com/thing:519320

Took 2 of these and fit them to an extended carriage. This would require the side carriages be redesigned as well to compensate, which depending on feedback here, i'll do if I move further with the concept. The following image is really just a rough idea, and I would still need to remodel the carriage and part of the extruder. But anyone think one of these 2 ideas would work?



If anyone has any ideas on how to make this work, I'm more than willing to take on the project. But if someone else gets that "spark" and wants to tackle it, by all means. We need a dual non-bowden solution.

 

I guess I don't see the purpose of two direct drive extruders. Unless you intend to print with 2 flexible materials

Definitely an interesting concept and I've considered designs where you have the motor run along the Y axis like you have on the left in the bottom right picture.

I'm thinking about making a design with a direct extruder design on one nozzle then the other is bowden. Especially since I don't see much need for a secondary direct drive. Unless you're making a combo filaflex and nylon it wouldn't really be beneficial to have 2

 

My main reason is financial. I have the stock j-head from when I upgraded to e3d, and don't have the money to buy another. My main purpose here is to make a dual direct so that the upgrade and the stock units can both be used. I'm only looking to add in the ability to use PVA on the secondary extruder, so print quality isn't that great of a concern, but in the most economical way possible.

If you ever design the direct/bowden combo, once I get the funds I am more than happy to be a tester =)

 

This is exactly what i was looking for but single.
Do you have an STL and instructions to get this single extruder installed on original base ?
Motor is too far forward and too high on ROBO, it would work much better with lower gravity center and closer toward the puleys.

I've seen this extruder allover, but not sure what its called.

 

http://www.thingiverse.com/thing:519320

That's the link to the extruder. I haven't tested the extruder yet, but hoping to soon. Not my design, just used it.

 

Yes i saw that. I think its not good enough.
Its better then nothing but its design is not printer friendly, its more of a MOLD friendly design.
Too many pieces and too many holes with sticking out parts.

I saw plenty of other simpler designes but they were for horizontal PRUSA i3 , not vertical like the ROBO is
I will still try it as soon as i finish testing my new bed mounts that i had to fix for higher precision prints that i now able to achieve after bed upgrade.

Edit:
I took that model and will modify it to make it more usable. I already found some alignment issues with it and implemented use of PTFE tubing and clean/wipe of the filament built into the lever.

Designer was not specific on what parts were used to get it working so i kinda had to figure it out inside the 3ds max measuring things.

 

Well to start with that doesn't make a whole lot of sense. If finances are the issue then PVA is NOT the way to go. PVA is super expensive, compared to HIPS, and it's rigid so you don't really need direct feed. Like I said the main benefit is flexible materials, otherwise bowden is preferred.

Secondly the E3D and the stock J-head are very different heights. It'll increase design complexity a lot to account for 2 varied nozzle heights. It's easiest to stick to one nozzle unless you have an elaborate leveling jig setup.

The nozzle leveling is probably the most complex problem to solve for dual nozzles. Some people ignore it but then you're more prone to clipping.

Fitting in a direct feed on my bowden design isn't overly complex. It's not easy because no one has done it in my particular orientation.

I'm also debating on if direct drive is going to work or if a geared drive, like greg's wade, is necessary.

 

Ah, I see. That makes a lot of sense. See, I'm still learning here. I've had my printer for roughly a year, a little over, and have finally driven myself into all of these mods and upgrades. I have to say I've learned a lot from your posts, and am in the middle of installing your auto level system. My thermistor broke last night, so printers out of service until new one comes in, otherwise it'd be done.

So, would you say, if I'm willing to take up the challenge of designing the mount and carriage, upgrade to e3d v6 direct, convert v5 to bowden, and integrate a manual leveling system in the carriage for one of the hotends?

And why, in your opinion, would geared be better over direct? Interested to learn more.

Also, question about something off-topic, but I've got some very promising ideas in mind for new filament materials. I have no way of funding a project like what I have in mind myself. I've already applied to the Autodesk Spark Fund, but being an unregistered self employed freelance artist, I doubt I'll be accepted. Any ideas of where I could look into funding a project like this? I'd ideally need a plastic grinder, filament extruder variable speed/temp, a couple different printers of different design... archetypes?, (looking to implement using fishing line to drive x and y, someone made another printer like this with unbelievable accuracy) and some ventilation for the room I'm working in, plus money for supply of raw materials and other small workshop equipment that may be needed. Just been developing ideas of ways to get these concepts in my head out there into a testable format. If any ideas, would be much appreciated.

 

Hey guys, I am "this guy on Thingiverse" who created the initial direct drive you used for your design. As another user already asked on Thingiverse for a dual version, I thought of putting a second one beside the first, both motors pointing towards the front in order to have some clearance from the smooth rods to add enough cooling, this will lead to probably 30-40mm less x travel.

Tony, for me the extruder IS good enough. The design is very well printable, simply activate the support material, mine was printed easily without a problem, sliced by curaengine. As you already saw, BOM and instructions available now at http://www.thingiverse.com/thing:519320

I like the idea of Mike combining a bowden and a direct driven solution. Very nice.

 

I already have a dual carriage design setup for the E3Dv6. It has swappable plates so you can easily change between different nozzle configurations. Though I could very quickly adapt the design to be a fixed plate if you prefer.

I like the idea of swappable plates so I can easily go between stuff like 2 bowdens, 1 bowden 1 direct, different nozzles, etc.

Well I like geared drive over direct drive because the torque gain. Theoretically direct feed would be faster at printing because there's no turndown. The counter to this as you feed filament faster you require more torque to get it fed through and a higher nozzle temperature, so it's a bit of a double edged sword.

I prefer geared because it's still plenty fast but gives you 5:1 more torque for driving filament. It also is better for slow printing as you have higher resolution per mm.

Geared is especially prefered in bowden arrangements as the bowden tube requires more torque to push filament through. Direct can skip steps or grind the filament because it didn't have enough torque to drive it.


I'm not sure on the funding thing. IndieGoGo might be your best option.

 

Mike, do you have a tip for me where to get a geared stepper? I live in Germany and it is quite hard to get parts, only few stores and the are quite expensive.

 

Mike, I just saw your dual carriage and that actually looks really nice. And it would support a v5/v6 wades and a bowden together? Cause now I'm thinking soon as I can I will upgrade to a direct v6, and get the bowden adapter for my v5.

Or, would it be better to get the bowden adapter for the v6 and keep the v5 direct? I've never worked with bowden before. Would I just print out another wades extruder and mount that outside the case with the PTFE(? or whatever it's called) tubing from end of extruder to hotend?

I also didn't realize HIPS was less expensive or preferred at all. The chemical used to dissolve it away, does it produce harmful fumes? I'm unfortunately limited to a room with no windows, though I hope to get a simple vent installed this weekend, and don't want to expose my family to anything too bad. I've pretty much been limited to PLA for a few months since I took over this space.

 

If you're looking at the one on thingiverse, that would theoretically work with the V6 but it's designed for the V5. The v5 had a bigger heatsink so I wasn't able to fit it in the preferred orientation (along y) though I did manage it with the V6. If you were looking in my printer thread than you'd see it there.

The v6 design is fantastic in that the bowden and direct share the same heatsink (aka universal). The v5 requires a totally unique heatsink with a bulky pneumatic coupling.


If you want to save some on cost here's what I would recommend if you don't want to go with 2 v6's

1) Keep your V5 heater block and nozzle, buy a v6 heatsink and heatbreak
This will cost a bit more since the heat break on the v5 has a M5 thread instead of M6 on the v6.

2) Buy a second v6

With that said the v6 nozzle is a noted improvement over the v5. Might be worth picking up another v6 nozzle to swap if you ever get a jam on the v5.

Then there's the whole ease of assembly of the v6. The heights will be the same, but the v6 will be much easier to put together.

you might find it easiest to go with 2 v6's or a v6 metal only kit.

As for HIPS, one thing to keep in mind is that this is a polystyrene, so the plastic is probably less than ideal if you're in grave concern about fumes. It shouldn't be too bad, and better than ABS, but it's still a concern.

The solvent is called D-Limonene, or Limonene for short. It's essentially an oil made from orange peals, and you'll often see it in "Orange" cleaners. You can even cook with it, at least according to the Kosher certificate i got with it. It's perfectly safe to be around, though it is pretty strong smelling. Kinda citrus-y so not unpleasant.

 

Alright, so it looks like if I just go with all the metal pieces, will cost me about $59+shipping, with just the heatbrake, heatsink and nozzle, looking at $44+shipping. Add in tubing at @ $7.50, and a coupler setup for either $3, $4, or $9 (question below), and we're looking between $54.50 to $75.50 for this upgrade, not including the second hotend.

I'm assuming an additional geared drive same as robo3d has, mounted outside is way to go then?

For the bowden adapter, which of these three am I looking at? I'm assuming these are base, stock model, mid-range easier model, and highend easier with better materials model? If that's so, is there an increase in quality by using the $9 couplers, or would the still easier to use but not as high quality $4 set work out just as well?

cheapest: http://www.filastruder.com/collecti...d-accessories/products/embeddedbowdencoupling
mid: http://www.filastruder.com/collecti...ies/products/bowden-couplings-threaded-1-75mm
high: http://www.filastruder.com/collecti...essories/products/groove-mount-bowden-adaptor

Looks like I wont be working with HIPS until I get this vent installed either haha.

Really appreciate all your help here. I'll definitely try the Indegogo thing. Really been developing a passion for this industry, and would like to make this a career, you know the whole "never work a day in your life by doing what you love" thing, but living in rural nowhere, with no 3d printing jobs within driving distance, and really no credentials apart from school and hobby experience, I figure my only option is going into business for myself and try to make something out here, but with no funding and no finances, I really have no idea where to start. There's this new, growing industry desperately seeking qualified professionals who know what they're doing, I'm out here desperately wanting in, with no real way of "touching" the industry from my location. Ugh!

One last thing:

This: http://www.thingiverse.com/thing:285851/#instructions

is different from this: http://forums.robo3dprinter.com/ind...al-bowden-e3d-on-the-robo-3d.1955/#post-18797

???

 

Didn't see this earlier. I got them from this company, seems to be the cheapest out there. http://www.omc-stepperonline.com/ge...ma-17-stepper-motor-17hs130404spg5-p-140.html

They ship from hong-kong so it probably won't be much different cost for you.

 

Mike, I'm looking at the carriage mod you have, and how would I mount the wade's extruder on there? Will I be losing x-axis build area?

 

Yeah going dual extruders is quite expensive. It's worth it in my opinion, but you just kinda need to know what you're getting into.

This is the carriage I'm currently running: http://forums.robo3dprinter.com/index.php?threads/mike-kellys-printer.1681/page-5#post-28956

It's mostly finished with prototyping and has been working great.

For the cold end I generally suggest going one of two ways.

The first is the cheaper option. Mount a secondary greg's wade extruder to compliment the one that comes with the robo 3d.

The real issue with this is that, in general, greg's wade extruders are designed to connect directly to a J-head type hot-end.

Which brings me to your bowden connectors. The cheap option are actually part of the E3Dv6 design. You should make sure that the all metal kit includes them as they're very important to the v6's design functionality.

The mid option is actually the bulky one for the v5. I forget what theads, I think they were 1/4" BSP but I don't remember exactly. In theory you could print a greg's wade extruder that connected directly to this coupling, but I'm not familiar with it.

The simplest option, is naturally the most expensive. The groove mount bowden adapters are really slick in that they'll work with any cold end that connects to a j-head. This gives you a lot more option in terms of what to print.

Once you have a design you like you just need to pick up a second nema 17 stepper motor, with close to 48oz-in torque, and a hobbed bolt. Then you'll need to determine the best way to mount to your case.


I was personally working under a height constraint, while trying to minimize the length of bowden tubing required. This led me to my current design of using two EZ Struders and geared stepper motors.

Now this design does have some issues. The first of which is that the EZStruder is designed to mount directly to a Nema17 stepper, which I didn't find very ideal. Instead I came up with a custom mounting bracket that uses flat head M3 screws to secure the motors, then the ezstruders secure to the mounting bracket as well with their normal hardware.

Combined this with the fact that a geared stepper motor has a larger shaft diameter, 6mm instead of the expected 5mm, means you'll need to bore out the hobbed pulley that comes with the EZstruder. It's a minor thing but still something to make note of.

This option is naturally more expensive in that you have to buy the 2 EZStruders at $30 each, then 2 geared steppers at $28 each. Plus I still suggest using the previous groove mount bowden adapter as I don't care for the adapter+PTC fitting that comes with the EZStruder (well i guess you buy the PTC seperate but I still don't like it very much)

I really like the groove mounts because they allow you to feed the bowden tube through so you can get it right next to the hobbed bolt and have a very seamless transition to the nozzle. They're worth it IMO.

Biggest thing I can suggest is learn the inner workings of the printer. Being able to operate it is all well and good, but if you really want experience that looks good on the resume understand the inner workings. How does a stepper motor work? What does a stepper motor drive do? How does a stepper differ from a brushless DC motor? How does it keep track of position? What does homing actually do? Etc

There's so much to learn about HOW these things work that you kinda can spend forever without really expanding on what to make with them. It's a really fascinating field.

The internet has been a great way of shortening distance. Do good work and people will take notice.

I think with a direct drive with the stepper drive shaft aligned along the X could be installed either with the motor sitting over the belts or the switch probe, similar to Stargroves design, but with a shorter "Bolt pattern"

I think if you wanted gearing the best would be a geared stepper, but naturally that costs and weighs more. The greg's wade gear takes up a lot of volume but I think it's theoretically possible to add it without losing X

 

Okay, I see. That makes sense. But a lot of this is if I were to be going with dual bowden, right? Would I need to modify the carriage to mount a wade's on top for the geared extruder for flexible filament? or should it pop right on there next to the bowden hotend?