What is your experience with dual hotends? Are they worth the trouble?

Guys, I am interested in a hot topic concerning dual hotends for 3D printers. They seem to be a tough nut in the reprap realm both because of seemingly poor firmware support and calibration.
I am interested in the working dual hotend setups you might have. What is your experience with them and if you would recommend it to others. If yes, what are the hurdles and how did you overcome them.

I want to use dual hotend to be able to have fast infills with large nozzle and detailed perimeters done with fine nozzle and perhaps exotic filaments.

Thank you for your input.

---

RepRap Lander concept on Concept Forge
RepRap Lander concept on RepRap Forums
My Things, mostly experimental stuff

I have a DaVinci Duo 2 and I just got the dual extruders working correctly with Marlin and Mega/RAMPS 1.4 recently. The setup really is not much harder than any other single extruder except the bed and nozzles must be leveled and calibrated and offsets entered in your slicer and host. I was surprised how easy it actually went considering how many hurdles I jumped to rewire the whole machine and get functional to start. It really is nice to use 2 colors, although you have to carve up your designs into separate pieces for each color. It takes some practice and patience to use correctly, but the things you want to do are certainly possible using Repetier and Slic3r, and if you have the notion to buy software get Simplify3d. With the setup I have I am limited to the same nozzle diameter, but I will design a new X car and use some E3ds' and better extruders at some point. I have used ABS and PLA at the same time but the bed temp is tricky, but it works at the average of the 2. You can tell slic3r which nozzle to use for certain parts of the print and I used it for supports/rafts, works well. The prints will take a few more minutes due to coasting the heads between colors, and you will need to clean up the whiskers on each side or use the ooze pillars in slic3r (never tried yet). Hope this helps you! Cheers.

---

Just an Artist, Musician, Designer, Programmer, Maker

Using: My H-Bot w/Mega RAMPS 1.4
M Prime 1 Custom Find me @ [www.thingiverse.com]
My SCARA Project Find me @ [www.thingiverse.com]
Zenbot 2424 CNC Router w/Mach 3

I have a CTC replicator dual (the only dual extruder I have) and like MarkBot said beside the calibration of the nozzles everything else is simple.

I found the learning curve a bit steep, and there's certainly an extra level of stuff to go wrong. I think it's a great capability to have in principle, but the extra work required to make use of it means that I haven't actually done much with it other than print test pieces to prove that I could. I guess when I've had a bit more time to work with it I'll start to feel more comfortable and won't make so many mistakes.

On machines and firmware designed for dual heads such as my flashforge creator pro and sailfish firmware it's pretty damn simple, even setting the head heights after switching to all metal hot ends was easy.
It's the slicing or design side that has stopped me using it more.
I rarely print stuff that needs support so don't use it for that either.

What I do love is the dual printing (ditto) capability for objects smaller than the nozzle spacing..... it's a feature I would love to see in the reprap firmware.

back to the OP's original questions, one big hurdle is getting the nozzles EXACTLY the same height, easy to do on the mk8,9,10 style duel extruder setups but would seem to be near impossible for slot or bolt mount setups, you really do need micro adjustable setups.

I have dual extruders on my Printrbot Metal Plus and I generally advise people to steer clear of them. I always say it squares the difficulty.

First there's a calibration. It's not so hard, but remove a hot end or extruder to service it and the chances are you'll need to set calibrate the extruder spacing.

Next there's the leveling issue. If your idle nozzle z-height is too low (and too low means low enough to catch blobs) there's a good chance you'll catch the print put down by the active nozzle and pull the print off the bed. Most printers don't have an easy way to fine tune the nozzle height.

Even when you've got the z dialed in correctly the idle nozzle is likely to drag across the print and roughen up the surface finish. I run my machine with the 2nd extruder nozzle removed to avoid these interactions.

Lastly, if you're running anything other than a Bowden set up, the extra weight on the head almost certain impacts print quality (resonances, etc.).

I love what you can do with dual extruders, and I won't give them up. But I won't buy another dual-extruder machine without independent extruders. Right now, the BCN3D Sigma looks very appealing to me.

I have no problems with the 2nd head dragging, but I did have to do a very precise leveling, I had a Printrbot Simple, the overall design is not friendly to Dual heads let alone one even. The weight of direct drives is no problem for Core XY hybrids and Box Frame machines as the Z moves and X,Y are always supported at both ends. The CAD portion isn't that bad as long as you use some type of direct modeling software ie; (SpaceClaim, SolidWorks, Rhino etc...) If you are trying with anything else you may discover why so many use these type of tools IMHO. I have used many programs and I find that direct drag-n-drop modeling beats entering formulas and manual coordinates any day if you are serious about designing and ease of use. I create separate components for all the parts then I anchor them to the grid at their current position and I save them as individual .stl and load into slic3r ignoring the warnings about too many parts etc. Center all the parts and select the proper extruder for each part and slic3r/print. I don't think the whole process is really to difficult as a whole, but if you think it will be click-n-print you will be disappointed for sure.

Edited 2 time(s). Last edit at 05/06/2016 09:51AM by MarkBot.

---

Just an Artist, Musician, Designer, Programmer, Maker

Using: My H-Bot w/Mega RAMPS 1.4
M Prime 1 Custom Find me @ [www.thingiverse.com]
My SCARA Project Find me @ [www.thingiverse.com]
Zenbot 2424 CNC Router w/Mach 3

I converted my Ormerod to dual extruder a couple of years and blogged about it here [miscsolutions.wordpress.com]. On the whole I am fairly happy with it. Some points to note:

- You really do need to design in a way of fine-adjusting the two nozzles to be at the same height.

- You need firmware that lets you set different temperatures for a nozzle depending in whether it is active or on standby, so that you can reduce the temperature of the standby nozzle to prevent oozing.

- The non-printing nozzle leaves a small mark on the print if it passes over freshly-extruded plastic. Reduce this effect by having the nozzles not too close together. A print cooling fan would probably help (my Ormerod doesn't have one). Being able to remove one of the nozzles or tilt it out of the way for single-nozzle printing would be a bonus.

My next experiment with multi-colour printing will probably be with the Diamond hot end, which avoids the problems of having two separate nozzles - but of course that doesn't allow you to use different nozzle sizes for different parts of the same print.

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].

Quote
dc42
- You really do need to design in a way of fine-adjusting the two nozzles to be at the same height.

[...] Reduce this effect by having the nozzles not too close together. A print cooling fan would probably help (my Ormerod doesn't have one).

Being able to remove one of the nozzles or tilt it out of the way for single-nozzle printing would be a bonus.

I am in the design stages at the moment, taking the parts I have designed for my build and really focusing on the details and now I realized that I have designed the dual-bowden toolhead but didn't get into too much detail with regards to micro-adjustments of the two hotends and so on. That is quit difficult to include in my current design and I would be interested how different people have managed to do that.

Besides having a tilting system, isn't it a good way to ensure both heatingblocks are at the same level by having them connected to a common plate and have no worry with microadjustments each individual hotend?

If that isn't an option I am thinking one of the hotends to be fixed and the other one to be able to move up and down a couple of mm. But I'd really like to see some actual solutions for microadjustment.

Thanks.

Edited 1 time(s). Last edit at 05/06/2016 01:50PM by realthor.

---

RepRap Lander concept on Concept Forge
RepRap Lander concept on RepRap Forums
My Things, mostly experimental stuff

Quote
realthor
Besides having a tilting system, isn't it a good way to ensure both heatingblocks are at the same level by having them connected to a common plate and have no worry with microadjustments each individual hotend?

Question: why there are long aluminum parts (@dc42: like in your blog) that are common to the two hotends? Isn't the heat from one of them bleeding into the other heater block through the common plate/part? I would think that something insulating should be used to provide perfect level between the nozzles, something like a ceramic plate or similar.

---

RepRap Lander concept on Concept Forge
RepRap Lander concept on RepRap Forums
My Things, mostly experimental stuff

The long aluminium part connects the cold ends of both nozzles/heatbreaks to the heatsink. The nozzles have independent heater blocks.

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].

Quote
dc42
The long aluminium part connects the cold ends of both nozzles/heatbreaks to the heatsink. The nozzles have independent heater blocks.

Yes I have noticed, meant to edit my post but got caught in other stuff .

---

RepRap Lander concept on Concept Forge
RepRap Lander concept on RepRap Forums
My Things, mostly experimental stuff

But even with such a block, ca you be sure that the two heater blocks and the nozzles end up at the same height?
What micro-adjustment mechanism is there to be simple enough to implement?

Edit: to me the simplest way to microadjust the level is to leave the heat-break threaded barrel sticking out of the heatsink a bit more than usual and have a nut on it to fasten it against the heatsink when level is reached.

Edit2: pic:



Edited 2 time(s). Last edit at 05/06/2016 05:12PM by realthor.

---

RepRap Lander concept on Concept Forge
RepRap Lander concept on RepRap Forums
My Things, mostly experimental stuff

My CTC printer uses a block like this one aluminum block as you can see on the second picture, the holes where the heat brake pipe goes have a locking bolt, the heat brake pipe is smooth like this: pipe so to level both nozzles you just run the leveling sequence once in position you just loosen the locking bolts and use a filler gauge to adjust the nozzles to the correct distance from the bed.

most of the Chinese dual extruders comes with threaded heat brake pipes and those are a pain in the behind to level.

Quote
ggherbaz
he holes where the heat brake pipe goes have a locking bolt, the heat brake pipe is smooth
...
so to level both nozzles you just run the leveling sequence once in position you just loosen the locking bolts and use a filler gauge to adjust the nozzles to the correct distance from the bed.

That's the approach I took and it mostly seems to work. It's not quite as easy to get them level as I'd hoped - too much flex in the X axis makes it fairly difficult to tighten the bolts without moving the hotends. The 8mm rods have to go.

With the solution i imagine (the one in the pic above), the threaded barrel would actually be an advantage.

It would go like this:
- first fasten the barrel into the heater block pretty well,
- then start screwing it into the heatsink (nut is already there on barrel, waiting to be used).
- (The first hotend is already in place, with everything fastened).
- Now all you have to do is rotate the heater block ever so slightly until a filler gauge tells you that the distance between bed and nozzle is the same with the other nozzle.
- Use nose pliers to fasten nut against heatsink.

Because of its own weight the threaded barrel will already rest on the lower side of the thread and fastening the nut against the heatsink will actually enforce that position so no "backlash" will be triggered to modify the distance already achieved.

Edited 1 time(s). Last edit at 05/06/2016 05:29PM by realthor.

---

RepRap Lander concept on Concept Forge
RepRap Lander concept on RepRap Forums
My Things, mostly experimental stuff

Look at existing work. E3d is nice enough to publish drawings of already working designs.

You could use a single heater block for both nozzles but they would have to be same size

Quote
Koko76
Look at existing work. E3d is nice enough to publish drawings of already working designs.

Which designs did you have in mind?

Quote
JamesK

Quote
Koko76
Look at existing work. E3d is nice enough to publish drawings of already working designs.

Which designs did you have in mind?

Me personally? None. I strongly considered building my new machine around a dual extruder design, mainly for the purpose of printing dissolvable support material. I ultimately decided against designing around it for a number of reasons. Mainly that to me the compromises needed to make dual extrusion happen would impact print quality with a single nozzle in ways I didn't want.
The E3D designs are the Chimera and Cyclops (the Kracken also uses a similar arrangement). They use a smooth walled heatbreak tube instead of the v6 threaded version where the heatbreak attaches to the heatsink. They use a set screw, to fix the tube in the hole and allow adjustments. They seem to hold reasonable tolerances, so I imagine the fit is fine. Were I to do it, I would wire cut a clamp block, but consumers are typically unwilling to pay the price that wire edm part would cost, so I can see why they did it that way. Something similar to a one piece shaft collar could work here. They do make aluminum ones.
Theory is one thing, but actually adjusting something like this to work in the real world is a different matter. Tightening threads against each other is generally very very fiddly.

Quote

Theory is one thing, but actually adjusting something like this to work in the real world is a different matter. Tightening threads against each other is generally very very fiddly.

Yes, I agree, I don't think I'd want to try and use a lock nut to set the final height of a threaded throat. With a smooth throat I prefer a clamp to a set screw. I cut the slot with a slitting saw which seemed to work out ok. You need a reasonable kerf to give clearance for the clamp to close around the throat, especially if your tolerances run a bit on the generous side.

I remember a thread where someone had mounted the extruders (direct drive) on dovetails which allowed fine adjustment of the Z position. That seemed like a good approach.

Quote
JamesK

Quote

Theory is one thing, but actually adjusting something like this to work in the real world is a different matter. Tightening threads against each other is generally very very fiddly.

Yes, I agree, I don't think I'd want to try and use a lock nut to set the final height of a threaded throat. With a smooth throat I prefer a clamp to a set screw. I cut the slot with a slitting saw which seemed to work out ok. You need a reasonable kerf to give clearance for the clamp to close around the throat, especially if your tolerances run a bit on the generous side.

I remember a thread where someone had mounted the extruders (direct drive) on dovetails which allowed fine adjustment of the Z position. That seemed like a good approach.

Wire cutter can do all that and more. I have also used the wire to cut a kerf concentric with the bore, entering with a radial line. The concentric cut is maybe .05" (maybe 1.5mm max) from the bore, and might cover 40 degrees of the circle. You then use a set screw to press the "flap" against the part in the bore. I've also done opposing sets of these. Nice tight fit, and non marking on the shaft. Very compact too.

Quote

I have also used the wire to cut a kerf concentric with the bore, entering with a radial line. The concentric cut is maybe .05" (maybe 1.5mm max) from the bore, and might cover 40 degrees of the circle. You then use a set screw to press the "flap" against the part in the bore. I've also done opposing sets of these. Nice tight fit, and non marking on the shaft. Very compact too.

That's a very neat idea, I'm more than a little envious!

Quote
Koko76
Wire cutter can do all that and more. I have also used the wire to cut a kerf concentric with the bore, entering with a radial line. The concentric cut is maybe .05" (maybe 1.5mm max) from the bore, and might cover 40 degrees of the circle. You then use a set screw to press the "flap" against the part in the bore. I've also done opposing sets of these. Nice tight fit, and non marking on the shaft. Very compact too.

Can you post a pic or napkin drawing for the ones that don't get it from the description?
Thanks.

Edited 1 time(s). Last edit at 05/07/2016 06:00AM by realthor.

---

RepRap Lander concept on Concept Forge
RepRap Lander concept on RepRap Forums
My Things, mostly experimental stuff

Not super useful though unless you have a wire cutter.

Quote
Koko76
Not super useful though unless you have a wire cutter.

Sorry to be thick-headed but what part is the one you are cutting the flap into? The idea is pretty neat but the flap should also not be alu i suppose otherwise the set screw might eat into it.

---

RepRap Lander concept on Concept Forge
RepRap Lander concept on RepRap Forums
My Things, mostly experimental stuff

Quote
realthor

Quote
Koko76
Not super useful though unless you have a wire cutter.

Sorry to be thick-headed but what part is the one you are cutting the flap into? The idea is pretty neat but the flap should also not be alu i suppose otherwise the set screw might eat into it.

These are things I have actually done, I don't need to suppose or speculate. What I described is a method in general to clamp objects into a bore with low runout, consistent surface contact, and non marking on the object being clamped. It would work well in the intended application, but most don't have the tools to make it work or would be willing to pay for someone to make it.

Quote
Koko76

Quote
realthor

Quote
Koko76
Not super useful though unless you have a wire cutter.

Sorry to be thick-headed but what part is the one you are cutting the flap into? The idea is pretty neat but the flap should also not be alu i suppose otherwise the set screw might eat into it.

These are things I have actually done, I don't need to suppose or speculate. What I described is a method in general to clamp objects into a bore with low runout, consistent surface contact, and non marking on the object being clamped. It would work well in the intended application, but most don't have the tools to make it work or would be willing to pay for someone to make it.

Didn't mean to offend, sorry if it sounded like that. My question was towards how can that be applied for the purpose of locking a smooth heat-break barrel against the heatsink. Basically the heatsink (alu) would have to have this flap cut into.

---

RepRap Lander concept on Concept Forge
RepRap Lander concept on RepRap Forums
My Things, mostly experimental stuff

Quote
realthor
Basically the heatsink (alu) would have to have this flap cut into.

Yes, I understand that. I understood it to start, which is why I showed a method that can do just that. Having never done this, or anything like it you suppose something will happen with the set screw. Having actually done it, it wasn't a concern of mine.

Quote
Koko76

Quote
realthor

Quote
Koko76
Not super useful though unless you have a wire cutter.

Sorry to be thick-headed but what part is the one you are cutting the flap into? The idea is pretty neat but the flap should also not be alu i suppose otherwise the set screw might eat into it.

These are things I have actually done, I don't need to suppose or speculate. What I described is a method in general to clamp objects into a bore with low runout, consistent surface contact, and non marking on the object being clamped. It would work well in the intended application, but most don't have the tools to make it work or would be willing to pay for someone to make it.

Yup, for most of us that's the impossible cut. A thing of beauty. What sort of prices are edm cutters?

RealThor, you can get set screws with brass and nylon tips which work well against aluminum, as well as steel tips with different profiles for varying amount of bite on different surfaces. One of the nice things about the design Koko shows is that even if the screw does mar the surface, the damage is on the flap and not the shaft so it won't cause problems when you come to release the part. It's a really nice approach that you won't see on anything but the highest grade equipment because of the difficulty of manufacture. Interestingly, a shape like that could be 3d printed (at an admittedly clunky large scale), I might see if I can use that somewhere just for fun