Welcome! Log In Create A New Profile


Independent X-carriages for improved dual extrusion

Posted by buildrob 
Independent X-carriages for improved dual extrusion
July 19, 2013 09:05AM
This is my effort for improving dual extruder performance - independent x-carriages - I think it works pretty well.

If you're interested I wrote it up here: [www.thingiverse.com]

My next change will be to also extend the firmware to allow it to print two copies of an object at once. I'll post some more video when I get that working.

Edited 1 time(s). Last edit at 07/19/2013 09:56AM by buildrob.
Re: Independent X-carriages for improved dual extrusion
July 20, 2013 09:57PM
i saw the videos of your printer the other day, i like your ideas, it has me thinking a lot about trying to build one my self.

Re: Independent X-carriages for improved dual extrusion
July 21, 2013 10:44PM

nice idea,

this gave me an idea to this, when using a closed belt, it could be posible to add an belt gripper to the x-carriages, which is controled by a servo
so it would be posible to run N x-carriages with 1 Steper and N Servos (belt gripper)

with x-carriages-servo-gripper you could avoid big changes in firmware,

also open belt would be posible with a firmware change (the firmware have to avoid to drive the belt opening over the idler or pully)
Re: Independent X-carriages for improved dual extrusion
July 22, 2013 03:58AM
Yeah, that was my initial plan for implementing the independent x-carriages, i.e., to either use a gripper or clutch mechanism. I even bought some relay and solenoid parts for doing this but decided to change my plan to adding an additional stepper instead.

I still think that approach is possible but I decided that that the design was a little too finicky/difficult to get right, i.e., absolutely no sideways movement when clamped (even when being accelerated quickly backwards and forwards), no interference when not clamped, robust and reliable, and in a small form factor that mounts neatly on the x-carriage.

You would also likely need to rehome the x-carriage each time you change to it as there is nothing which holds it perfectly in position when inactive (e.g., vibration could move it slightly). [Not that this is a major problem.]


Re: Independent X-carriages for improved dual extrusion
July 30, 2013 10:30PM
Given this is the mechanics forum I thought that the following subsequent change to my dual x-carriage design might be of interest to some people.

This dual-x carriage design requires that the distance between the two endstops is properly calibrated to ensure that the printing from the two extruders is aligned. I found that that the variability in triggering of the mechanical microswitches led to inconsistent alignment of the two materials (the X and Y endstops in a normal printer don't need to be very precise). I solved this by replacing the microswitch endstops with Hall-effect non-contact magnetic endstops ( [reprap.org] ). Search ebay for 'Hall Endstops' - they are around $10 each.

I also replaced the Z-axis endstop with a Hall-effect endstop which is completely unrelated to dual x-carriage support but gives a nice improvement in first layer height consistency. [ I'm sure I could design a better Z-axis endstop holder but some Polymorph/Plastimake/InstaMorpg/Polycaprolactone (great stuff - get some if you don't have any, the uses are endless) and a cable tie is working fine for the moment. ]

The endstop already includes a trimpot for fine adjustment of the trigger point.

Pictures here: [bukobot.com] [bukobot.com] [ I used bigger trigger magnets than the ones supplied but that was simply for ease of positioning. ]

Edited 2 time(s). Last edit at 07/30/2013 10:34PM by buildrob.
Re: Independent X-carriages for improved dual extrusion
August 02, 2013 09:53PM
Another update on my dual x-carriage printer project: I've just finished implementing a software enable-able "duplication mode" for the Marlin firmware.

This allows you change into a mode where you can print two arbitrary sized objects using the the independent x-carriages (as long as they fit on the print bed of course). I added a video of it in operation here: [youtu.be].

Thingiverse link is here: [www.thingiverse.com].
Re: Independent X-carriages for improved dual extrusion
September 08, 2013 02:23PM
Very cool. I've been doing a fair bit of dual-extrusion printing with my MendelMax lately, and having independent carriages would be a great help. Maybe I'll try to set up something similar.

Help improve the RepRap wiki!
Just click "Edit" in the top-right corner of the page and start typing.
Anyone can edit the wiki!
Re: Independent X-carriages for improved dual extrusion
September 08, 2013 02:58PM
Check out my x-carriage for PrusaMendel [www.thingiverse.com] - maybe can fit into MendelMax.
I plan to use buildbob's marlin rom update to run it.
I can't get a time to build my second prusa for tests.

Re: Independent X-carriages for improved dual extrusion
October 27, 2013 07:11AM
HI BuildRob,

I love your independant dual X axis,

is it very difficult to modify the marlin code to extend this to independant Y and Z axis?

I am asking this because i am working on a project for 6 axis movements on a machine.. being.. X1,Y1,Z1,X2,Y2,Z2 or X,Y,Z,A,B,C

how do you GCODE second x axis to move?

thks for your great help.
Re: Independent X-carriages for improved dual extrusion
October 27, 2013 07:33AM
Hi Conrad,

The Marlin code has two primary modes of operation in dual X-carriage mode (ignoring the duplication mode):
Mode 0 (M605 S0) requires slicer support. Each time you do a tool change it reassigns the current position of the "active" toolhead. As the toolheads can move independently the slicer must also have the same understanding of the position of each of the two tool heads. This mode provides the optimal level of control and routing of the toolhead path after a tool change.
Mode 1 (M605 S0) does not require slicer support. It works by automatically (in the client) restoring the previous position each time a tool-change occurs. It works quite well because it also does a lift in the Z position when moving in and out of the new position. [The actual logic is more complicated than this but that's the general idea.]

Having independent tool movement in Y & Z would not be too difficult in concept to support but I did not make these changes because:
a) it would have more changes to the Marlin code than what I was wanting to make
b) I had no way to test it (and testing of this stuff is crucial)
c) it would have required more configuration as you have to configure which axes are linked.

Also remember that this is still not actually a "4-axis robot" in the normal sense. The toolhead is only ever moving in 3 axes (and so requires no G-code extensions). Its just that each time you do a tool change you are driving a different set of steppers for XYZ movement - but again still only moving in 3 axes.

So it may not be what you are wanting anyway.

Hope that helps,

Sorry, only registered users may post in this forum.

Click here to login