Hey guys,
I thought I would share a new twisty puzzle idea. I thought of this idea about a year ago and I think it's time to share it! I call this puzzle the Hurricane Cube:

The puzzle looks a little strange at first, but it was inspired by the Windmill Cube. The puzzle can be explained like this:

We know that there are two versions of the Windmill Cube (a 3x3 modification). One of them looks like this from a particular angle...

...and the other one looks like this, from the same angle:

Notice that these two puzzles, mathematically, are the same; they're just mirrored versions of each other.

The thought that pondered me was: if we overlapped the sets of cuts for both of these puzzles on top of each other, what would be the result? Well, the resulting puzzle is the Hurricane Cube (which I think is a fitting name).

Designing the mechanism of this was no easy task. I decided to use a shells style mechanism, since it seemed the most obvious way to go. The result is a puzzle built up in three shells. Let's start with the first shell:

All of the pieces that you see are part of the first shell, even though some of them have been extended into neighboring shell layers.

The geometry of this puzzle is VERY ugly and has some very odd symmetry (if there is any at all!) Here's a look at one type of center piece from this shell:

This is another center piece. That's right, a center piece. (Specifically, a center piece which resides in one of the vertical layers.) And the hole there is for an M3 screw...

Let's move on to the second shell:

This shell is where we get some of the external pieces to have grooves which hold them in place. Honestly, there isn't much to say about this shell, other than the fact that, once again, the parts are shaped very asymmetrically. (This shell is actually MUCH less complex than it looks, just the cutting surfaces create extra faces on the pieces which are negligible. I haven't found a way to prevent these though).

Finally, the third shell results in the full puzzle:

Again, there isn't much to say. In this shell, however, the parts around the top center piece do get quite small and unstable. So I experimented a little and redesigned the second and third shell:

Using a slightly different curve, I was able to extend more of those tiny pieces from the second shell to the third, making them more stable. You can see this in the blue-highlighted piece.

Alright, enough with the mechanism and now for some turns! It's worth mentioning that I believe this puzzle jumbles, but in a highly unusual way. Here are some examples:




...
I've seen many jumbling puzzles, but NEVER one that jumbles quite like this. Could anyone do an analysis on the jumbling properties on this puzzle? Or prove completely that it actually jumbles? I'm very interested in knowing more about my own design!

Anyway, thanks guys! You're comments are appreciated!

Great model. Looks cool!

How did you simulate turns in solidworks?

This looks great, and actually, it has a ton of symmetry!
Excellent!

I use the "Move/Copy" feature. There's probably an easier way to do it but it allows a lot of flexibility, especially for jumbling turns.

I'm not 100% sure that this puzzle jumbles. It would be possible to unbandage just the 180 degree turns straightforwardly by adding another windmill axis to the vertical cuts, and the others might be completely unbandageable as well, it's hard to tell though.

Wow, you have a VERY good point there. But in order for it to never bandage every axis must be able to turn by 27.22 degrees (the approximate angle at which this puzzle jumbles) and that would require copying all of the cuts onto every axis of rotation, and wouldn't this exercise unarguably cut the puzzle up to dust?

My first reaction is that it is a cubic deeper-cut octagonal prism, but now I am not so sure.

-Doug

At first it looked like to me, that you basically took a windmill cube and mixed it with a square 2. But as i look at it more closely i see that the whole thing is alot more complex, and has more ways to turn then i thought it would anitially.
But anyway, very cool model.

It isn't even close to that actually. This does bring up a good question though on which regular face-turning solid it is based on. I tried to find this out myself and it turns out to be extremely close to a bandaged face-turning dodecagonal prism, but it's not quite the same:

The blue lines represent a dodecagon and as I compare them to the black lines (the cuts) they are ever so slightly off. So this puzzle is not based on any regular, even geometry. Pretty cool!

I like these kind of puzzles, the ones that make you really think about what they actually are!

Any idea of when, and if, you're going to print this? Its so awesome!!

Sorry but I doubt I'm going to print this. It will be very difficult to design it up to quality standards considering the weird geometry of the puzzle (as I've stated before) and due to the shells mechanism and high density of pieces, it probably won't be worth the cost of printing. I do understand that many more complex puzzles have been made before but I don't see this having much potential. Maybe in the far off future...

Oh okay, I've just noticed you've come up with quite a few concepts, and I'd love to see them become real (No way I could afford sponsoring )