Just when you think you’ve seen it all in the 3D printer world, something new pops up! [Nicholas Seward] posted a video of RepRap Simpson, his latest project. Simpson is a delta robot – but unlike any delta we’ve seen before. Previous offerings vertical rails on which the arms travel. As you can see, this design mounts three articulated arms directly to the base of the printer, using steel cables as part of the joint mechanism.
Judging by [Nicholas’] posts on the RepRap forums, Simpson’s grounded delta design has already gone through a few revisions. The basic geometry though, has remained the same. [Nicholas] calls this edition a “Proportional Gear Drive Joint Simpson”. The name may not roll off the tongue, but the movements are incredibly smooth, organic, and fast.
As with any delta design inverse kinematics play a huge role in the software. [Nicholas] is trying to simplify this with an optical calibration system. For the adventurous, the equations are posted on the forums, and a python Gcode preprocessor is posted on Thingiverse.
Even Simpson’s base received special attention. It’s built from a water jet cut piece of basalt. We like the use of opposed helical gears on the large joints, as well as the guitar machine heads used to tension the cable drive. One thing we are not sure of is the longevity of system – will cable stretch play an issue? Will the printed parts suffer wear from the cables? Only time will tell.
[Thanks Theo!]
OK, that is really neat. I would love to see someone take it the next step and stiffen it up, right now it is about as rigid as a 5-1/4″ floppy.
Looks like it’s printing pretty well, don’t see why it needs stiffened up.
With a 360p video I cant tell what it is doing, it is just an overexposed orange blob.
It has comparable stiffness to other traditional printers that I have such as a Printrbot Plus. (Comparable as in about 1/2 as stiff but still stiff enough.)
But the Simpson refuses to print anything but donuts. D’oh!
Seriously though, at least from the sample video, it appears to have pretty good accuracy despite so many potential sources for error. Will be following comments to see what people with actual mechanical knowledge say about it.
Water jet cut piece of basalt? What did that cost and can it be heated?
No idea what they cost – but it does look like the beds are heated from below. I’m curious how long the stone takes to get to temperature.
Who knows, but once it gets there I bet it’s solid as a rock on keeping a constant temperature. Think of stone bake ware and how long it remains hot after coming out of the oven.
(This was a serious comment, but I couldn’t resist chucking a pun in there in honor of [Josh Marsh] and his way with puns)
The beds cost me $40 for a short run without shopping around. It takes about 5 minutes to get to temp. I need to put an insulator and heat director under the heating pad.
Nick – 5 minutes! What kind of heaters are you using (I’m guessing 120v?)? Sounds really good – but I’m not sure my Prusa could muscle one of those on the Y axis!
CIRCULAR 24V SILICONE RUBBER HEATER BED 200MM DIAMETER (8″)
Basalt is an igneous rock with a melting point around a thousand degrees Celsius. There’s pretty much no chance he’s going to get it hot enough to cause a problem as long as it’s not, like, heating it to hundreds of degrees and then spraying cold water.
It looks awesome! As far as accuracy, a lot depends on how accurately those inter-meshing chevron gears are made (and how they withstand use – plastic rubbing against plastic? Hmm…)
I would love to see how it prints something square and regular shape – the only way to assess print quality.
FWIW, they are called herringbone gears.
Even those herringbone gears could be replaced by a couple of steel cables each, running on two drums and crossed so that the drums won’t slip. It would also get nicely rid of any backlash using just one spring and adding little to no friction.
I’ve been following this for about 3 weeks (just lurking on the reprap forum). Actually, that idea you propose was seriously discussed at the beginning of the new arm design here. http://forums.reprap.org/read.php?178,233674,page=1 I think it was passed over mostly due to the herringbone gear design being printable and therefore achievable quickly. I hope they go back to the steel band idea eventually.
Looks like the actuator of the extruder lays on the table!
I think Nick hadn’t firmly attached his extruder stepper yet – it is Simpson’s first print after all.
I support creativity and applaud those who create and spend a lot of work on their projects…
But regarding 3D printers, am I the only one here who:
– wants to see LESS new 3D printers
– BETTER accuracy on the exisiting 3D printers
– CHEAPER 3D printers
???
I definitely understand what you are saying. The 3D printing world is in its awkward teenage stage. The countless deficiencies in current 3D printers is driving the flood of attempts to solve them. The dream of buying a cheap machine that you plug in and hit print is far from reality.
If want better accuracy printers that are cheaper then I think more new 3D printers will be needed. On the accuracy and the cheaper front, Simpson can provide both. All other printers have to use semi-expensive linear rails and bearings.
” All other printers have to use semi-expensive linear rails and bearings.”
They are not that expensive. Furthermore, those “joints” on the Simpson printer will also wear out eventually, just like everything else.
BTW, check out the Zim 3D printer on KS. The makers claim it can be run by children.
That looks an awful lot like a Canfield joint: http://www.youtube.com/watch?v=dvFukC3kLRA
I have talked with Dr. Canfield because I was also struck with the similarity. He told me that the joint isn’t protected by patent law anymore so I am trying to find a way to use it.