Inverted Delta

[Trakyan]

Hi All,

Some of you may have heard of or seen the GUS Simpson printer from a couple of years ago. Personally I think it's awesome. However it suffered from some issues. I'm currently working on a revision of the GUS, trying to fix or improve anything I can think of.
Just wanted to start a thread here for anyone who wanted to collaborate, suggest ideas or improvements, or point out where I've done something stupid.

I have a few major goals.
-User friendliness
-Usability
-As many standard or readily available parts as possible, and as few different parts as possible.
-Improve mechanics and print ability
-Make it cheap, make it scale-able. Using only rotating joints is a big plus to the GUS style printer.

I have a bunch of improvements I've implemented or plan on implementing, but more is always better. Right now I'm trying to decide on the build area for the printer, since the GUS style printers have an a typical build volume (approximately a trigonal base pyramid with bulged sides) it's not as simple as saying "200 mm cubed". Besides, who uses all of that build volume (especially the Z?) to begin with?

The design will be fully parametric and open source. I'm hoping I'll be able to slim things down to ~5 parameters that the end user can tweak to customize the machine to their needs (desired build volume, how large of a printer they have available to print the parts), input the details of the hardware they want to use (if it's different from what I'm using), and everything else will be calculated to fit automatically.

Feel free to ask questions/make suggestions. I'll document as I go if people are interested.

[curious aardvark]

http://reprap.org/wiki/GUS_Simpson

Interesting. Looks like you'd struggle to get any kind of decent height build out of one - which is kind of what delta's are known for.

But a very cool looking printer :-)


Actually looks like three robotic arms fighting over the extruder :-)

Massively complicated compared to a standard delta, can't see any advantages.
So you building it just for fun, or did I miss something ?

[Trakyan]

It's not massively complicated actually, no more than a standard delta, the kinematics are almost identical to a linear delta, you just change the radius rather than center point in your kinematics, and the drive system is a block and tackle, not complicated at all, you could make it even simpler and just have the fishing line just pull either side of the gear joint shut, the block and tackle is just for mechanical advantage (you could use this with belts, too, but not really any point). If anything it has fewer parts than a standard delta and the construction is less fussy and more rigid than getting three really tall towers evenly spaced and perfectly vertical, orthogonal and rigid.

Anyways, linear deltas are known for height, that's only because people build them ridiculously tall, it's not a required or inherent feature of deltas. You could just as easily make a Cartesian printer tall. Personally I don't print tall objects so this doesn't bother me, most practical prints are not that tall, just random rockets and vases that are tall. While this printer is much more compact, you still get a decent height. With a 250 mm bed diameter you get just over 210 mm as a max height, which is plenty for me and many others. That being said, that's only if you make it so the maximum extension of the arms is equal to the bed diameter, just print longer arms and your print volume gets taller (and wider). You could keep all your hardware the same and just print longer arms to get your desired height.

The main advantage is it's a heck of a lot cheaper than a linear delta. Compare the cost of long linear rails, linear bearings, long belts and long aluminium extrusion to a couple of 608 bearings, fishing line and some bolts which just about the only non printed part of this printer, bar electronics and motors etc which all printers need. If you want a different sized printer you can reuse all of your hardware to do it, on a delta to change the print volume you need completely new linear rods, extrusions and belts. The kicker? The bigger arms for a bigger printer can fit inside the original printers build volume.

Also, it looks awesome.

Don't jump to conclusions against this design, if you have any more questions/concerns feel free to ask.

[curious aardvark]

Didn't jump to conclusions - that's why i asked :-)

So what electronics do you need ?

And yes I am now seriously thinking about making one :-)

It does look seriously cool.

[Trakyan]

It needs four steppers (three arms+extruder), three end stops and your choice of electronics (ramps and a mega should work fine). It's about as bare bones as you get for a 3 axis CNC machine.

If you do want to make one, I wouldn't suggest going straight with the STLs Nicholas Seward published, he noted a few issues with his model and couldn't get the accuracy any better than +-0.4 mm according to him. So I'd recommend doing a re design or waiting for me to finish mine (I'll be making sure accuracy is at least comparable to a standard i3). I suspect part of the issue he had with the fact the gears were printed vertically so teeth, especially those in the middle, wouldn't have had smooth curves, but there may be more to it.

EDIT: I should have my first revision out in a couple of weeks. I've done the hardest part (the gear joint, this one is designed so the gears are printed flat), I just need to get around to designing the motor mounts, and the extruder hub. Shouldn't take too long but I've got a busy week next week and won't have time to work on it.

[Viscephala]

I've previously seen a few other iterations of this type and it certainly looks like a worthwhile build. Great price to WOW! ratio. Looking forward to your designs. Couple of questions.
Would this in practicality require ABS parts, or would PET-G or even PLA suffice?
What firmware(s) would/could drive this with a RAMPS board? The ones I saw were using custom compiled Marlin on a Smoothie board.

[curious aardvark]

no rush :-)

[Trakyan]

It does have a great wow factor, it looks cool in general and I find the gear hinge type joints awesome. It's also cheap in general as it only uses rotational joints, no linear rails or rods required. so good price to printer ratio.

Materials wise, PLA and PET would actually be better than ABS. PLA is noticeably more rigid and stiffer than ABS, and the temperature resistance isn't too much of a priority. Maybe for the hot end mount you'd want something more heat resistant and rigidity wont be as important as in the arms. For the arms particle filled PLA might even be a good idea, from what I've read even though they break more easily than regular PLA, they are also more rigid. That being said I'll have to test all this and see.

The kinematics to run this are very similar to a standard linear detla, so I'll be modifying marlin or repetier so it can run on RAMPS/any other printer controller. The smoothieboard variant you're talking about I believe I've seen, and uses smoothieware rather than marlin so it only runs on smoothie based boards. There was someone planning to modify the standard marlin but I don't think they ever finished/published. Since the RAMPS/Mega combination can run linear deltas it should be able to run this, the only difference is you change the radius rather than center point of your 'spheres', but the calculations are otherwise identical, same number of addition/subtraction, same number of squares and square roots.

There have been some other iterations but there are always things I would have done differently, I figured instead of complaining I'd do it myself now that I've learnt a bit of CAD. The main issues I've seen are print ability, the arms are really hard to print them both flat and without supports, You can get around both of these by running the string along the outside of the arms, but it introduces more parts and complexity, that and I don't like the look of routing the string on the outside. My main goal for my design is to make it print flat, without supports and route the strings internally.

I also want to work on some other aspects of the printer, I'm wanting to design my own extruder that has all of the features I'd like in an extruder, and work on some cheap bed leveling solutions. If people have any features they'd like to see in the printer, please let me know and I'll do my best to include it. Any common issues you guys have with printers that you'd like to see fixed or avoided in this printer?

[Trakyan]

Ive been thinking of parallelizing the gus simpson design so that it can have an effector platform like a typical delta, rather than an effector which rests in the center of a set of bearings. This will make multiple nozzles/different toolheads much easier to implement, change and adjust. Right now you're pretty much limited to a bowden design with the bowden tube going down through the bearings. It's possible but I imagine would be wonky to mount a direct drive with a stepper on top of the end effector.

Any thoughts on this? It would add more places where two parts need to mate and join together, which in my mind introduces more room for error. On the plus side, you have a much more versatile end effector. You can mount different or multiple toolheads, more space to mount cooling fans, sensors, direct drive extruders. I'll also be working on making the arm motors stationary to reduce the moving mass. In the current design the motors are mounted low on the arms, so while they don't move drastically they are still moving. I don't know if this is really a problem to begin with but it may appeal to some people and the elegance of it appeals to me.

I've been toying with the idea for a while, but it only recently clicked how to implement it (both the parallelization and making the motors stationary). I'll probably make the motors stationary regardless of the decision on parallelization or not, but would like input from others as it does add cost/parts and some complexity.

[curious aardvark]

use a much smaller stepper motor for the extruder.
Just got some nema 14s, which might do the job. Easy to mount on an effector.
nema 14 & 17 side by side.

nema14and17_266x480.jpg