An expanding recursive mechanism, plus some variants. You can download the files to print these mechanism for yourself from www.printables.com/model/5577...
That's so funny, for me I thought it was just an engineering student/someone with a 3d printer, this seemed far too non-abstract to be a Segerman video lol. But obviously you understand something I don't😅.
These demos remind me of the Hoberman Sphere - that spiky ball toy that expands and contracts through a bunch of linked scissor mechanisms. The Hoberman Sphere is recursive too, except there the components have a network of relationships rather than just a chain of relationships.
Commonly seen on indulgence feasts in Poland - stands with cheap (often knockoff) toys, sweets and especially indulgence obwarzanki (a very light and dry kind of ring shaped bread, no idea how it's made, availble basically only on those indulgence feasts) spring up around churches on the day of that church's patron. Usually in summer. I remember having that kind of toy, but idk if it's still somewhere around in the house.
technically you could say that it's an inverted hoberman sphere. Inversion is a geometrical transformation which (in 3d) changes planes into spheres and vice versa.
I tried to put a servo on one of the joints of a plastic Hoberman Sphere so that I could remotely expand and contract it for dramatic effect. It didn't really work as my servo didn't have enough torque. I wondered if the sphere I had relied to some extent on the flexibility of the thin plastic struts or whether a solid metal construction would also work if the joints were ball / socket.
i would love to see a video about your process for creating mechanisms like this! do you use any programs or simulations when you're throwing ideas around? maybe we could see some early prototypes of some of your projects?
Did you ever even bother to check the description? It contains the weblink for the PDF download... And BTW (by the way)... THIS is basically THE VIDEO about creating mechanisms like these.
@@OmenAkumaru20223 ....why so condescending?? To answer your question, I did check the description. At the time of writing this and my original comment, it only contains a link to the files for 3d printing. If it wasn't clear somehow, what I'm curious about is the creative process behind projects like this. This video doesn't get into that at all, it just shows off the final mechanism. I want to know how he prototypes, problem solves, etc. (Henry if you are reading this strange thread, I am still very interested
@@noahgiom341 The very short answer is that these recursive racks came from thinking about Oskar's Tailspin mechanism, and wondering what would happen if we swapped out one kind of gearing (the spur gears in Tailspin) for another (sliding racks). The geared cube net is the same thing, but using bevel gears instead of spur gears. The prototyping stages are presumably not that different from how anyone prototypes - I design stuff in my CAD program (Rhino) and test it out. I implemented my own gear/rack code, then I tested out how small I could print gears before they stopped working smoothly. Having determined the sizes of my gears and racks, I then made a box with two racks and a gear, and found that I had a lot of friction. So then I modified the design to decrease the surface contact areas between parts. And there's also some stuff to do with designing for the filament based printer I have - thinking about the print orientations of the various parts and how to bolt things together.
@@OmenAkumaru20223 bruh he's asking how to *design* these things as in engineering. Instead of being a cunt you could try maybe double checking things instead of just lashing out at your own confusion.
I really want to see a machinist like This Old Tony make some of these and see if he can come up with a practical function for them, like perhaps for shelving or something. Utterly fascinating and deeply original, as always it seems. You never cease to astound me!
I swear, this will be how certain meta-materials will work some day when we can get consistent "gears" made of bio-mechanical components/layers. I imagine this could be a great mechanism for frequency combing.
@@aeremthirteen2771 Paper is called "Step-by-step rotation of a molecule-gear mounted on an atomic-scale axis". Also "A Simple Example of a Molecule-Gear Train: PF3 Molecules on a Cu(111) Surface"
Im very excited for the self-evident impending analog computer renaissance! Minimizing AI control system energy requirements is all we need now, and quantum paradigms are expanding into a new knowledge tree/growth more and more exponentially lately!
As somebody who prints and unpowders alot of your models its always great when I see them come through! Would it be possible to make a cube with recursive racks? Some thing like a Hoberman sphere or similar.
Seeing how the original worked, and then seeing the second one, I was actually pretty happy with myself for being able to work out how the second one would work relatively quickly. That being said, once you did confirm it, I immediately thought of trying to find a way to use these and other slightly modified racks to act as logic gates. Would be very interesting.
I quite like the unique way this mechanism behaves! I think the most interesting thing about these recursive mechanisms is how it made me really look at the different reference points when figuring out how it behaved :)
This reminds me of pantogarph mechanisms like used in various extension mechanism. Your last example in particular looks like a different take on a mechanism oft used in drafting machine, whereas the top bracket and the end unit must always stay parrallel relative to one another (so that lines remain parrallel). In these machines this was at first accomplished with a pantograph, with two rods comprising each segment of the arm. On later american machines, they switch to steel belts, which are put in tension so that they grip their respective wheels. The top and bottom disks are fixed, as is the elbow of the machine; so whatever movement is affected, the end still stays parallel. In very nice machines, the steel belt was welded in a manner as to never loose tension. Cheaper models used a belt with a tensioning nut on each band and their loosening could cause errors in the drawing. However the geared rack idea perhaps could be applied in keeping those two points fixed in a different manner. Either with solid racks around toothed gears, or perhaps with a toothed belt.
Shapeways is going to lose a lot of business now that Henry has developed the ability to magically bring parts into existence by simply pointing his finger. 😉
Love the video! If you take the first design, add a pin to the center box (attaching it to the surface below it), add another pin to the end of the left most gear rack, and a marker to the end of the right most gear rack. You will get a drawing copy machine! Love it!
This makes me so intrigued about this idea I had years ago to make a mechanical software interface. This would be the perfect set of mechanisms to make photoshops basic transformation tools!
I feel like the last contraption makes an involute curve It interesting to see what kind of movement you get with only racks and gear and the application they can have
I'd love to see what can be done using different gear ratios between "incoming" and "outgoing" racks of each box. Would it just change the slope of the diagonal movements of adjacent boxes or can you create more complex movements using that?
Sorry made a comment twice and deleted. Thinking about this is difficult at 4am. Especially when I misinterpret the device for 6 minutes because I was typing lol. Anyway, I think if you put gears pairs into the boxes with correct ratios it should be possible? It would basically twist the path a little if you did achieve it though, shouldn't allow any more complicated movements. Though I must confess I can't think of what other movements this system can really produce.
Played with the gear cube at opensauce and found it delightful. There’s always a small surprise when it arrives perfectly cube or perfectly flat from its intermediate folded states. The “recursiveness” of these makes me wonder about how you would formulate an energy transfer equation for n segments. That exploding outro was great 😆
The pure rack and pinion recursive mechanisms could absolutely work without the constant step up every iteration - if you take either version rotated to a vertical grid, the similarly orientated racks obviously never cross over each other, the horizontal grey racks never cross over with the vertical green racks, and the vertical grey racks never cross over with the horizontal green racks, so there are two separate sets of racks that could each occupy one vertical space without ever colliding with each other. That said, you would need to tweak the design slightly - I doubt you could just rearrange the parts you've already made to do that.
You probably dont see this a a proble to solve but, You could force the first, and I thing second, recursive rack and pinion to be flat (not stair stepped) by fixing the height of each rack to a constant value based on orientation. It would require a means of selecting the height of the rack fixed to the box to 1 of 4 values, either with spacers or custom shapes for each orientation. But completely doable
10:26 this one reminds me of watching the overcomplicated windscreen wiper mechanisms on buses in late 1980s Ireland. Mesmerising and yet almost perpetually broken...
I think it could be a very interesting exercise to see if you could use specific combinations of these to create "linkages" that trace specific paths, similar to Fourier epicycles
if sufficiently small or large it can scale pixel art within the distance that the illusion takes place. Thanks for this kind of videos Henry greetings from Venezuela
It would be fun to see what kind of fun mechanisms you could come up with by making the straight racks curved. So in addition to translation you'd get also rotation.
Ha that's hilarious. At 5:29 when you were talking about you could remove the gear as long as the box held the 2 pieces and they slid together, the first thing that popped into mind was driving them like the slotted straight driven lock/key combination shown 2 or 3 years ago in one of the lock picking videos. Then at 5:36 of course you pull out exactly that, a straight drive push gear mechanism. There's a lock based on changing the teeth of that mechanism so the matching key opens it. Fairly simple and not very secure, pretty sure they said just shoving a vegetable into it would usually open the locks. Used in Russia or Scandinavia etc for simple things like shed doors, the video showed one outside somewhere.
I think it's more accurate to call it vector racks, since recursion implies self-reference rather than reference to two parameters. Still quite awesome though.
These are wonderful! immediate thoughts are: can you make grids of any of these or do the always lock?; and that there's some connection to the wallpaper groups though that might be a miss-association. .. but that's definitely the rest of my Sunday gone :) really fun vid !
I would love to see some practical applications to these mechanisms. The parallels of the racks and of arbitrary points of the racks look like there should be ways to take advantage of it, but it is hard to envision with the video only demonstrating the simplest form of the mechanism.
