An Idea for dual extruders

[mechneering]

Hey guys,

So I have two Max's, one at my work, and another at home. Both are running E3D hot ends and upgraded arms and both work wonderfully. But the time has come to move forward with the dual head setup.

I have spent some time designing a mounting system for the dual heads that incorporates cooling fans and all that good stuff. I know aligning the two heads has been a persistent problem, and my design doesn't really do much to improve upon that issue, I was mostly just planning on using some thin shims under the mounting screws on one of the heads if it needed adjustments.

One of the problems I can foresee with dual heads is that the one nozzle not printing will be dragging along in the exact same plane as the nozzle that is printing. Often my prints curl up a bit on edges and corners, or I get a bit of extra plastic in places. This extra head moving around in the same plane will inevitably come in contact with the printed part and cause it to break loose or disconnect one of the mag arms.

So, I was thinking how neat it would be to design a dual head mount that articulated with a small solenoid or servo motor. This would allow one of the heads to swivel up a few degrees and clear the part while the other head is doing its work. Then when its the other heads turn the solenoid switches and the other head rocks into place. This would give clearance room for each print nozzle and eliminate any problems with the other head being in the way.

I can design and build the mechanical system easily enough that allows this to happen. And would be more than willing to share my design with you guys, but my hold up is the software. How hard would it be to include a command in the G-code or update the firmware to send a signal to a servo whenever it switches from one head to another? And what pins on the Rambo could I use to do this? I hope this is a trivial question for some of you guys

Any help or direction would be appreciated.

Thanks

[Jimustanguitar]

I like the idea a lot, but I'm not the coding wizard that you're looking for.

Mechanically, I think this could easily be done with a servo or gear reduction motor of some kind.

It will be very interesting to watch this develop!

[Eaglezsoar]

I have never given it much thought but if the dual extruders are exactly the same height then the second extruder should not hit the curled part any more then the printing extruder
would. Dual extruders are becoming more and more popular and they seem to work fine without such exotic solutions. If they work without having to raise the non printing
extruder then yours should also. All printers experience the curling part problem at times including those with dual extruders and I have never read of any problems with the use
of the dual extruders such as you describe. I have never ran with dual extruders and I do not claim to be an expert but I have to question if your exotic solution is really necessary.

[dpmacri]

When using multiple extruders with Slic3r, there's a setting to have G-code that runs whenever the tool is changed. Unfortunately, it doesn't look like the change can be different based on which tool is active, but perhaps that would be an easy enough change to make. Or, depending on how you design the mechanism, maybe the same G-code would just toggle the position of the extruder (e.g. if at each end of the movement there was a switch that caused the motor to reverse or something like that).

I think this is a good idea. I setup my dual E3D configuration but haven't successfully printed anything because one hot-end was slightly higher than the other and I don't have any way to adjust that with the mount I built .

[Glacian22]

I'd love to see this on a Rostock Max...the only time I've seen something like this is here: https://www.youtube.com/watch?v=BePRUTsZh_w, where he actually has metal plates such that when a hotend is inactive, it's capped to prevent oozing.

[McSlappy]

Instead of a servo, why not mount both heads on a slight tilt like this:

[img]http://i.imgur.com/6V0CaEC.png?1[/img]

I'm showing 5 degrees but perhaps it doesn't need that much.

Then depending on the head being used, the code adjusts the arm mathemagics to tilt the effector plate (I think that's it's name) by 5 degrees to position nozzle 1 correctly or -5 degrees to position nozzle 2.

Not sure what sort of nightmares this might cause to geometry, but it's a thought.

[Eaglezsoar]

I never did get a suitable answer to my comments about there being thousands of printers out in the real world with dual extruders, how are they doing this? Are they all
experiencing the second extruder banging into the print? I do not understand. To me, it appears that you are fixing a problem which may not exist.

[enggmaug]

Instead of a servo, why not mount both heads on a slight tilt like this :
Then depending on the head being used, the code adjusts the arm mathemagics to tilt the effector plate (I think that's it's name) by 5 degrees to position nozzle 1 correctly or -5 degrees to position nozzle 2.

Not sure what sort of nightmares this might cause to geometry, but it's a thought.

The effector cannot be tilted through the 3 axis steppers. It is mechanically impossible.


I hope this problem does not exist, because I just received the Kraken today...

[neurascenic]

Instead of a servo, why not mount both heads on a slight tilt like this :
Then depending on the head being used, the code adjusts the arm mathemagics to tilt the effector plate (I think that's it's name) by 5 degrees to position nozzle 1 correctly or -5 degrees to position nozzle 2.

Not sure what sort of nightmares this might cause to geometry, but it's a thought.

The effector cannot be tilted through the 3 axis steppers. It is mechanically impossible.


I hope this problem does not exist, because I just received the Kraken today...

Tis true, the geometry of the system would prevent that... but a pivot mechanism within the effector would. An Electro magnet set up to switch N/S would do the trick.

Eagle: Astute question... might not be enough of a problem to have to solve on a well calibrated machine.

[Eaglezsoar]

I hope this problem does not exist, because I just received the Kraken today...

That is a good point. If this "problem" truly existed it would make extruders like the Kraken impossible to use.

[mechneering]

I have never given it much thought but if the dual extruders are exactly the same height then the second extruder should not hit the curled part any more then the printing extruder
would. Dual extruders are becoming more and more popular and they seem to work fine without such exotic solutions. If they work without having to raise the non printing
extruder then yours should also. All printers experience the curling part problem at times including those with dual extruders and I have never read of any problems with the use
of the dual extruders such as you describe. I have never ran with dual extruders and I do not claim to be an expert but I have to question if your exotic solution is really necessary.

You make some valid points. And there are many dual heads that are out there and are working fine with fixed heads. So I will likely just go ahead with a fixed design and see how it performs and then if any of my imagined problems surface I might continue to pursue this "exotic" idea. You mentioned that the second extruder shouldn't hit the curled parts any more than the printing extruder, this is true, but the active extruder is generally following a path that coincides with the curl, thus limiting its likelihood of catching on the part as it moves about. Plus the extruder is hot and often melts its way through the high spots. The inactive extruder would have nether of these advantages while tagging alongside the other. But perhaps I am way over thinking and making it more complicated than it is.

[Jimustanguitar]

I never did get a suitable answer to my comments about there being thousands of printers out in the real world with dual extruders, how are they doing this? Are they all
experiencing the second extruder banging into the print? I do not understand. To me, it appears that you are fixing a problem which may not exist.

I have no experience with this at all, but my suspicion is that between over-extruding, heat warping (both the printed part and the printer itself), two oozing nozzles, calibration errors, and human/slicer variables each time... that everybody doing dual extrusion is at least "scraping" the top of their print with the other head. I'm sure that Z-Hop on positioning moves and infill patterns with straight lines will help reduce the negative impact of this, but I just can't imagine that anyone is dual extruding without the other head touching their print at some point. I mean, it has to if it's set up right. If both heads are mounted in a fixed position, it's impossible not to encounter some sort of interference between the part and the inactive print head.

[Eaglezsoar]

I have never given it much thought but if the dual extruders are exactly the same height then the second extruder should not hit the curled part any more then the printing extruder
would. Dual extruders are becoming more and more popular and they seem to work fine without such exotic solutions. If they work without having to raise the non printing
extruder then yours should also. All printers experience the curling part problem at times including those with dual extruders and I have never read of any problems with the use
of the dual extruders such as you describe. I have never ran with dual extruders and I do not claim to be an expert but I have to question if your exotic solution is really necessary.

You make some valid points. And there are many dual heads that are out there and are working fine with fixed heads. So I will likely just go ahead with a fixed design and see how it performs and then if any of my imagined problems surface I might continue to pursue this "exotic" idea. You mentioned that the second extruder shouldn't hit the curled parts any more than the printing extruder, this is true, but the active extruder is generally following a path that coincides with the curl, thus limiting its likelihood of catching on the part as it moves about. Plus the extruder is hot and often melts its way through the high spots. The inactive extruder would have nether of these advantages while tagging alongside the other. But perhaps I am way over thinking and making it more complicated than it is.

Going with the fixed design and seeing how it performs sounds like a good approach. If you then discover that you do need to implement your first idea then you can have the
satisfaction of knowing you were right all along.

[Batteau62]

You guys have probably seen this: https://www.youtube.com/watch?v=CW-z6XN-1II
Wouldn't it be nice to just connect whatever hotend-filament combo you have setup on stations

[mechneering]

You guys have probably seen this: https://www.youtube.com/watch?v=CW-z6XN-1II
Wouldn't it be nice to just connect whatever hotend-filament combo you have setup on stations

I have seen that before, and yes it would be way cool to just do a tool change and swap from one extruder to the other. unfortunately it seems its the firmware/software that is holding us back from developing some of these ideas. Or maybe the software is capable, and I just dont know about it yet. same goes for things like auto bed leveling...

[artexmg]

Hey guys,

So I have two Max's, one at my work, and another at home. Both are running E3D hot ends and upgraded arms and both work wonderfully. But the time has come to move forward with the dual head setup.

I have spent some time designing a mounting system for the dual heads that incorporates cooling fans and all that good stuff. I know aligning the two heads has been a persistent problem, and my design doesn't really do much to improve upon that issue, I was mostly just planning on using some thin shims under the mounting screws on one of the heads if it needed adjustments.
[...]
Thanks

Great to read this! ... so, did you already designed a mount for the dual E3D? I already got my second one and had not find any mount yet. I started to design one, but ... if somebody had done it before, much better

For me, the major problem comes with the space of both hot ends is just enough for the diameter of the Rostock end effector, and they have to touch each other to fit in it.

Please, let me know your advance in this regard.

Thanks!

[Nylocke]

My School's MakerBot *shudder* has issues with the second hot end scraping away material. Due to the poor design of the adjustment mechanism, its extremely difficult to level the hotends, and they get out of adjustment after a couple months it seems. So it "is a problem with professional printers". Ultimaker has a much better leveling solution, shown in this great video on how to install a second hotend
http://vimeo.com/72977149 (starts around 12 minutes)

When I get my 2 Pico's this month, I'm going to try and design a system similar to how Ultimaker does it so I don't run into this issue, maybe a potential fix for the MAX's issues could arise in the mounting system? Possibly add 3 small high tension springs where the mount mounts to the effector and use thumb screws (or don't, up to you) with some extra length for a 3 point leveling system?

[Eaglezsoar]

My School's MakerBot *shudder* has issues with the second hot end scraping away material. Due to the poor design of the adjustment mechanism, its extremely difficult to level the hotends, and they get out of adjustment after a couple months it seems. So it "is a problem with professional printers". Ultimaker has a much better leveling solution, shown in this great video on how to install a second hotend
http://vimeo.com/72977149 (starts around 12 minutes)

When I get my 2 Pico's this month, I'm going to try and design a system similar to how Ultimaker does it so I don't run into this issue, maybe a potential fix for the MAX's issues could arise in the mounting system? Possibly add 3 small high tension springs where the mount mounts to the effector and use thumb screws (or don't, up to you) with some extra length for a 3 point leveling system?

Nylocke, you did it again and spelled the "printer's" name incorrectly! It is spelled Makerbutt.

[Nylocke]

I'm sorry, its just so darn hard to spell