Vertical X-axis standard !And Contest W/ Prize Updates

I think it's a cool idea. You may be right about 70 mm being too much for the horizontal distance though.

Edited 1 time(s). Last edit at 10/05/2011 09:39PM by bryanandaimee.

You might be on to something important here. Having a second X axis will be inherently more expensive than just a dual extruder head, but it would allow you to park one head while the other one extrudes, eliminating the ooze that the dualstrusion makerbots are struggling to deal with.

It also avoids all the dual extruder head patents I've found (which are good for at least another few years). Also, it would let you calibrate the extruder nozzles Z height independently, without dragging a low nozzle through the part. The Z axis could Hop to whatever height it needed while no heads were around.

Also, since the heads are separate, it wouldn't impact extruder speed. Heck, if both nozzles were perfectly leveled you could print multiple exact copies simultaneously.

NoobMan Wrote:
-------------------------------------------------------
> Right now it looks like i will go for a little
> oversized machine, and about the x axes i will
> theoretically use 4 rods in a square pattern 7cm
> each side, in order to get a second x axis with
> its own carriage. The second axis arrangement will
> get to be symmetrical to the other one, mirrored
> both ways. This way the rods can be vertical, and
> also could be used in a horizontal (classical)
> arrangement.
>
> The second axis could be used for experimental
> toolheads and such, to be toyed with, without
> having to interfere with first toolhead. It could
> let ppls have more fun with the printer.
>
> The horizontal distance between 8mm rods on my
> mendel is 50, but i increased for the square
> config at 70 horizontally aswell as vertically.
> Not sure of it yet, because at 70 the rods are
> pretty far away from Z drive and it means lots of
> extra plastic so it might not be so good, maybe i
> will revert to a smaller distance even, but dont
> know it yet.
>
> About t5 belts, probably will have 10mm. And
> certainly some mod like the "doubler accuracy
> mod", or at least have the flexibility to be
> arranged like so.
>
> This is what i came up with, so i hope its
> allright so far.
>
> Any feedback on the general picture of what i
> intend to do?

Dont know of a firmware yet to support a secondary X motor, or a gcode command to allow to switch in between them. This could be great and maybe not so hard to implement. There is a complication here, like one head gets parked, afterwards how it will remember to which position to refer to when its needed again next time, so the SF head exchange alterations might not be enough. But for the time being even if i change the wiring manually from one to another i can still try things out, which at least should let me get more fun from the printer.

About the general arrangement there are a few drawbacks or critical aspects, i think maybe good reasons for which this setup is not really preferred in general:
- the distance between toolhead "tips" depends on their arrangement, and its a distance "lost" from Y at least from the area heads could work together, e.g. for 10cm dist in between on our mendel would make the "together" distance at half; depends if the head tips are inside the bars or point towards outside, there are big differences in both versions;
- lots of extra weight, even +size -> +bed, and -speed - accel, etc; Z block needs way more plastic which isnt good (sf says the 70x70 one is like 250g of plastic haha); lots of efficiency losses in many areas;

The firmware is actually capable now, if you are willing to hack the hardware a bit.

Since marlin can control pin out values from g-code, you can trigger a 4x multiplexer which switches step/dir for the x axis and extruder. By using the start/end support gcode edits in skeinforge, you can specify the parked positions for each carriage.

You would need to skip on a heated bed since it only supports 2 heaters at the moment, but it IS possible.

I'm more than willing to add cost/complexity in exchange for being able to print practically anything in whatever orientation I want.

Edited 1 time(s). Last edit at 10/09/2011 12:40AM by Andrew Diehl.

---

www.Fablicator.com

Yup but think for it for a while.

If the 2 extruders are onto same plate, its easy. The distance between their heads is always the exact same one. So using just the start/end gcode from SF, or some small altered "extruder change" routine in hardware are both possible.

If each extruder has its own axis, then its different. One extruder gets parked for the other to work. Once it gets parked, his position is more or less "lost". So when it needs to come back and print again, there are a few lines needed to get it back to the desired position. Dont think the SF could be convinced to do that math automatically as it is now. Maybe if we use homing.gcode in a innovative way between each layer, to keep something like a "reference point". Alternatively in hardware, maybe if we calibrate head positions parked distance between each other or something like that, and use these as settings in head exchange routine, but would still need an accurate read of where the other head was, so we know what distances we have to deduct. And getting an accurate position of where the head is at any given point, in reference to a known position, that has to make things more complicated.

Adding heaters is the easiest thing to do, just a pin and a fet, so thats no issue. Older electronics had 3 fets anyways.

The position is not lost.

Here is how it works:

//Bookend
-----------------------------------------------
M142 P40 S0 //Turn the multiplexer off, select X1 axis
G162X //Home X1 axis

M142 p40 S255 //Turn multiplexer on, select X2 axis
G162X //Home X2 Axis.
----------------------------------------------
//Because we maintain power to the X1 and X2 axis, and only are switching the step/dir, we now know both are at X=0
//As long as you send the axis back to a known point (say x=0) before you switch extruder control, you always know where it is.


The only place you might get into trouble with is if you are using absolute extrusion distance instead of relative, because if you prime/wipe before each layer it will mess up future e-codes.

---

www.Fablicator.com

I expressed poorly previously. What i meant was that lets say head is at certain (x,y) position when tool change occurs. Thats the point that is a problem, e.g. where to move the new head to. Not the home. Means we need to store the previous head position before homing, so we know (-calibrations), where the other has to go exactly. Having the 2 extruders on same carriage i think already works, it only requires to change the calibration distances thats all. But this situation with different carriages where one homes first, i think needs some implementation changes, to store the current position so it can get the other at that point. Otherwise i cant see it in practice as it is now.

Or am i missing something?

Edited 1 time(s). Last edit at 10/10/2011 06:20AM by NoobMan.

I gotcha now. You are correct.

I wasn't thinking this through quite enough. Forgot splitting up the heads adds an extra variable.

So what we really need is a g-code command for store current position, and go to stored position.

I suppose I should start a new thread since we are severely off topic now, and a number of different folks might be interested in our new line of thought.

---

www.Fablicator.com

Yup well, it might be kinda soon though, i think first should get our "mechanically confident" stage. In retrospect i always had quite a number of things wrong, so gotta build this and put some stress into it and see how it behaves. I think it will work in the end though ... "smartest one always gives up" ... so this doesnt have any choice but to work.

Any news here? Just for fun, I'm planning to test out one of the vertical X axis designs; do people have a current favorite?

The contest deadline was extended to Dec 12 by vote of the contestants, and the designs are looking good. A couple more designs should be finished by the deadline. I doubt very many have been built by anyone but the designers yet, but pick one and try it out. Let us know what you think.

Bryan

[reprap.org]

As one of the contestants i have been using various versions of my design for about a month now. The idea of the vertical-x is really good, it is a much more sturdy solution than a horizontal-x. And it does reduce the amount of plastic needed by quite a bit (~30-50%)). I might switch to another design than mine, but i won't go back to a horizontal-x.
One clear disadvantage is that it is not always easy to use existing extruder and hot-end designs.
Frank

But that will become an advantage as more integrated carriage-extruder designs come out that decrease print time even more while retaining rigidity etc.

Bryan

And it's all over. Results are up on the wiki page. [reprap.org]

Congratulations to North90ty, Emmanuel and RustySpoon!

I'm happy to see this contest worked out so well. A new model for triggering collaboration in open source hardware?

---

Generation 7 Electronics	Teacup Firmware	RepRap DIY

Some very interesting designs. I particularly like RustySpoon's ability to mount other tools. I do believe I will ad(o|a)pt that design for a rebuild I'm planning.

I think it's more that while it's an old model, we now have a large enough community to make it work. And of course the brilliant design and execution of the contest itself helped.

Great job everyone! And Thanks for running everything Bryanandaimee!

Hope you guys enjoy your prizes. :-)

Thanks Bryan for managing this contest and thanks to everyone for contributing/encouraging

Thanks for running the comp Bryan. I was going to get back to my PCB Mill idea, but I was snowed under with work and then went skiing! I now have a motor, I'll get around to updating it and trying it out (once I've printed out one of the vertical X axis mounts) hopefully in the new year.

Congratulations to the victors, it's a win for all of us!

Great competition! thanks Bryan!
Perhaps we should try to integrate the good parts of each design into one great design? Having one main design might convince people easier to adapt the vertical x-axis idea.

Gratz and kudos to winners and all participants, and everybody have a good time on the holidays coming up.

Cheers

I agree that a single design will make it more likely someone will decide to incorporate it into their bot. I'm not sure how to go about that other than just suggesting the designers get together and hash out the compromise. Alternatively we could all just hack on our own designs, borrowing from each other until enough people have tried them out that a consensus emerges. I think adoption will likely be more influenced by other things though. In my mind a significant savings in print time is likely the most important factor, followed by clear documentation, followed by other things like stiffness, ease of assembly/disassembly.

bryanandaimee Wrote:
-------------------------------------------------------
> Alternatively we could all just hack on our own designs, borrowing
> from each other until enough people have tried
> them out that a consensus emerges. I think
> adoption will likely be more influenced by other
> things though.

I'm in that boat already. There are a bunch of ideas out there now that i can borrow from and evolve, which is what made the contest a win for everybody.

The one piece left out of the standard is where the belt runs, and how it connects to the x carriage. can't have everything, though. and leaving it out of the standard/contest allowed the contestants to experiment.

Today I brought the prize to our makerspace (nybi.cc)

It will spur a reprap built by/for everyone ! (Probably a Prusa, with a vertical axis ? )