Parametric Towers of Hanoi
I’ve been working on a parametric towers of hanoi set, and it’s working well. It can print the posts, or be built to use separate posts (metal, wood, plastic; cut or turned) with screws through from the underside to hold the posts in place. It can be printed with extra-high posts in order to be able to play two-color towers of hanoi.
It is, however, big enough that you might have to get creative moving things around in the slicer to fit on your printer. It just barely fits the long way on my 270mm Y bed.
I set it up for the thingiverse customizer, but that doesn’t work, so you’ll have to download it and build it in openscad if the STLs aren’t what you need. Expect it to take a few minutes to render.
https://www.thingiverse.com/thing:2994834
Parametric Towers of Hanoi by mcdanlj
Michael K Johnson July 14, 2018 19:52
for the “extended edition”
Matt Harrington July 14, 2018 21:28
It feels like you are missing an opportunity here. You have mostly flat disk. Find a way to make use of the printer’s capabilities and make something that would not be so easily made by hand.
For instance, the disks could be hollow stacking pyramids, no posts necessary. Perhaps they could be nautilus gears that mesh on the posts. Or maybe they could be lithographs. The posts could be spiral threads and the disks would spin as they are dropped down.
You have a solid foundation for customization ion here.
Michael K Johnson July 14, 2018 21:48
Search thingiverse for cups of Hanoi.
In terms of “easily made by hand” given that I have cut a bunch of the posts on a lathe and seen how many steps that actually takes, I’m happy to use the printer for the whole thing. “Easily” is easy to say. I’m not sure how I would make those rounded edges “by hand” — a custom set of router jigs and a quarter round bit? Versus waiting a few hours for the printer? I’m happy with this set.
There are plenty of Hanoi sets on thingiverse, but I didn’t find any standard sets this parameterized and done in openscad.
Ulrich Baer July 15, 2018 00:29
give each disk its own post only long as the next disk and with a thicker diameter so only smaller disks (bigger hole) will fit - also you can substract half of each following so they fit into each other .. just an idea.
Michael K Johnson July 15, 2018 07:11
+Ulrich Baer they are hard enough to pick up to move as it is; while I made sure that the model “works” (no unintended scaling) artifacts down to 3mm disk size, in practice the 5mm size is as small as I’d want to go and still be able to pick them up easily. Nesting them to half their depth would make them more than twice as hard to pick up—and little annoyances add up when you have many small moves to make! Also nesting would break the two-color variant.
For the different size posts, I’m assuming that you meant that each of the three posts has stepped sizes, so that the smallest disk sits at the top of the post, etc. That is more practical (though given a lathe, a stepped profile is a common enough task for Matt’s “made by hand” criterion), but it would break the set for the two-color variant, where there are two disks of each size.
I made this for my wife, a mathematician, to teach math with. Functionality has been my primary criterion.
Ulrich Baer July 15, 2018 07:50
+Michael K Johnson no - every disk would have a little post (cylinder) (which is used also as handle) and would fit into smaller disks but not in bigger - as those would have smaller holes. You wouldn’t need any rule for this game except that you start with the 3 plattforms and none disk is allowed outside these plattform. Your idea of stepped sized posts can’t work as a small disk would never fall down (also it wouldn’t make sense to have bigger holes in smaller disks)
..
Ulrich Baer July 15, 2018 08:08
made a model for you
Michael K Johnson July 15, 2018 08:17
Ah, picture worth 1000 words… That is indeed a nice take on the concept!
Won’t work for two color Hanoi though, and that’s one of my specific requirements. :)
Michael K Johnson July 15, 2018 08:42
+Matt Harrington have you seen the “icehouse” game? http://www.looneylabs.com/rules/icehouse
Michael K Johnson July 15, 2018 09:58
I rounded the tops of the posts and made it render a lot faster by reducing excessive segment count. Still doesn’t render in customizer yet.
Matt Harrington July 15, 2018 11:43
+Michael K Johnson , no I had not seen that game but those are the shapes I was thinking.
As for making this by hand, I like the rough handmade look so I would prolly round the disks on a bandsaw and then round the edges on a belt sander. Maybe scalloped edges. For something different, maybe I would use square posts which are easy to cut and chisel the holes. Actually, that sounds fun. Maybe I will do that :)
Michael K Johnson July 15, 2018 12:23
Hmm, fully parametric stacking pyramid set could print in two towers, printed alternately inside each other, smaller inside second larger. Not only parametric but also fully recursively implemented. I might find time to do that. It could be elegant by my idiosyncratic criteria. :)
Michael K Johnson July 15, 2018 14:11
The source is simple, and everything about it is parametric. It renders approximately instantaneously and works in the customizer. I haven’t tried printing it yet though!
Parametric Pyramids of Hanoi by mcdanlj
Michael K Johnson July 15, 2018 20:38
+Ulrich Baer unlike me, my mathematician wife instantly understood what you were proposing and why… I’m kind of out of time this weekend but I might at some point do a parametric set of the arrangement you suggest, though unlike the pyramid set it’s not instantly obvious to me how to lay out an arbitrary number of them. Maybe a spiral arrangement of some sort?
Ulrich Baer July 15, 2018 20:50
+Michael K Johnson lay out for printing? The footprint didn’t change to the normal version - so you could apply the same pattern. But a spiral would look nicer - Ü!
Michael K Johnson July 15, 2018 20:59
+Ulrich Baer well if it doesn’t need a base I don’t have that constraint to work around. I manually laid out the six in that version, but wished for an algorithm. Optimal packing is hard. :)
Ulrich Baer July 15, 2018 21:18
I would use a triangle base with 3 small pins as the main task is moving it from one plattform to another - and so nobody can outsmart you by creating 5 stacks or moving it as whole. Without the base the requirement to have only 3 stacks wouldn’t be defined in hardware.
Michael K Johnson July 18, 2018 18:47
Oh, also, the nesting stacked pyramids print great. My only initial problem was that the slicer automatic cooling algorithm thought they didn’t need cooling, and that was bad. Turning the fan on solved the problem.
Imported from Google+ — content and formatting may not be reliable