Snail Ball
November 16, 2010 6:00 AM Subscribe
The Geometry of the Snail Ball [pdf] - an interesting article (with some DIY advice at the end) about a toy shop curiosity you may have encountered.
Couldn't you make a snail ball a lot more simply by just putting internal vanes in that have to go through the glycerin? The off-center weight makes it go back and forth, which isn't very snail-like. More illusory, though, and maybe that's the point.
posted by DU at 6:45 AM on November 16, 2010 [1 favorite]
posted by DU at 6:45 AM on November 16, 2010 [1 favorite]
What do you mean? The off center weight is what makes the ball work.
posted by CaseyB at 6:51 AM on November 16, 2010
posted by CaseyB at 6:51 AM on November 16, 2010
I still have a strong desire to see the snail ball cut in half.
you WILL tell me your secrets, snail ball...
posted by orme at 6:57 AM on November 16, 2010 [3 favorites]
you WILL tell me your secrets, snail ball...
posted by orme at 6:57 AM on November 16, 2010 [3 favorites]
Another good trick: When I was a kid we'd string an elastic band between the lid and base of a coffee can and suspend a weight, like a bolt, in the middle. When you roll the can along the floor the bolt will not turn, so the elastic winds up and the can eventually stops and then returns to you. Your kids will love it.
posted by weapons-grade pandemonium at 6:58 AM on November 16, 2010 [5 favorites]
posted by weapons-grade pandemonium at 6:58 AM on November 16, 2010 [5 favorites]
Oh, I get it. I didn't even realize that there was a second ball, I figured the interior was half-filled with glycerin, as you describe.
My guess is that there is a greater difference in mass between the interior ball and the glycerin filling than there would be between glycerin and an air bubble, hence moving the center of gravity farther away from the axis and accentuating the overall effect.
posted by CaseyB at 7:03 AM on November 16, 2010
My guess is that there is a greater difference in mass between the interior ball and the glycerin filling than there would be between glycerin and an air bubble, hence moving the center of gravity farther away from the axis and accentuating the overall effect.
posted by CaseyB at 7:03 AM on November 16, 2010
If you want to really jam people's radar, let them try it but give them the Non-Snail Snail Ball.
posted by weapons-grade pandemonium at 7:06 AM on November 16, 2010
posted by weapons-grade pandemonium at 7:06 AM on November 16, 2010
I guess it depends on what "the effect" is. Is it the slowness (as indicated by the name) or is it the jerkiness? If it's the slowness, then you don't need a "difference in mass" you just need the vanes and viscosity. If it's the jerkiness, then yes, you need the mass.
posted by DU at 7:14 AM on November 16, 2010
posted by DU at 7:14 AM on November 16, 2010
Also, I guess the interior weight version is a lot easier to make, especially for a 2D roller.
posted by DU at 7:15 AM on November 16, 2010
posted by DU at 7:15 AM on November 16, 2010
2D rollers don't want to stand up on edge, DU.
posted by weapons-grade pandemonium at 7:18 AM on November 16, 2010
posted by weapons-grade pandemonium at 7:18 AM on November 16, 2010
The off center weight is what makes the ball work.
That's half of it. The other half is the viscous fluid which puts a delay in how fast the inner sphere can roll around. If the fluid is too viscous and the inner and the outer balls will lock up and move a single mass. If the fluid is not viscous enough, the delays will not be very long. The break on the inner ball, what's slowing everything down is the dissipation of heat in the viscous fluid.
This is exactly the same effect as dropping a ball bearing into a pot of honey. Just imagine rolling the pot of honey as the ball falls through it. The bearing will trace a circular path through the honey. That's what's happening inside the outer ball.
The paper details motion of the centre of mass of the system, but doesn't describe the dissipative fluid dynamics involved (and I'm too lazy to work them out). But, if glycerol or corn syrup is what you need, that's a viscosity of 10 to 15 Pa-s, thicker than honey.
It's a neat effect when done on a cycloid.
posted by bonehead at 7:35 AM on November 16, 2010 [3 favorites]
That's half of it. The other half is the viscous fluid which puts a delay in how fast the inner sphere can roll around. If the fluid is too viscous and the inner and the outer balls will lock up and move a single mass. If the fluid is not viscous enough, the delays will not be very long. The break on the inner ball, what's slowing everything down is the dissipation of heat in the viscous fluid.
This is exactly the same effect as dropping a ball bearing into a pot of honey. Just imagine rolling the pot of honey as the ball falls through it. The bearing will trace a circular path through the honey. That's what's happening inside the outer ball.
The paper details motion of the centre of mass of the system, but doesn't describe the dissipative fluid dynamics involved (and I'm too lazy to work them out). But, if glycerol or corn syrup is what you need, that's a viscosity of 10 to 15 Pa-s, thicker than honey.
It's a neat effect when done on a cycloid.
posted by bonehead at 7:35 AM on November 16, 2010 [3 favorites]
Couldn't you make a snail ball a lot more simply by just putting internal vanes in that have to go through the glycerin?
That will slow the system down and make the ball roll more slowly, but it won't have the start-stop motion that a cycloid system does.
posted by bonehead at 7:39 AM on November 16, 2010
That will slow the system down and make the ball roll more slowly, but it won't have the start-stop motion that a cycloid system does.
posted by bonehead at 7:39 AM on November 16, 2010
Imagine a fresh egg rolling on it's side. That's essentially what you're describing, DU.
posted by bonehead at 7:40 AM on November 16, 2010
posted by bonehead at 7:40 AM on November 16, 2010
That will slow the system down and make the ball roll more slowly, but it won't have the start-stop motion that a cycloid system does.
Exactly, as noted here and here.
posted by DU at 7:55 AM on November 16, 2010
Exactly, as noted here and here.
posted by DU at 7:55 AM on November 16, 2010
Actually, you could get a non-regular motion with just a fluid by making the interior of the sphere non-uniform. An elipsoid interior or non-uniform obstructions inside a spherical ball would have a similar effect, though the period of the effect would not change with rotation. The neat thing about the ball-within-a-ball is that the periodicity of the start-stop effect appears to be random, unrelated to the rotation of the exterior sphere.
posted by bonehead at 8:18 AM on November 16, 2010
posted by bonehead at 8:18 AM on November 16, 2010
Neat. I didn't even know they had balls.
posted by Ratio at 8:57 AM on November 16, 2010 [2 favorites]
posted by Ratio at 8:57 AM on November 16, 2010 [2 favorites]
Viscous fluid inside. Similar to the effect of pushing a half-filled coke can. Ah, with an inner ball... nifty!
posted by mrgrimm at 9:03 AM on November 16, 2010
posted by mrgrimm at 9:03 AM on November 16, 2010
Snails have more balls than we do.
(At least on a per capita basis they do, seeing as they're hermaphrodites.)
posted by nebulawindphone at 9:04 AM on November 16, 2010 [1 favorite]
(At least on a per capita basis they do, seeing as they're hermaphrodites.)
posted by nebulawindphone at 9:04 AM on November 16, 2010 [1 favorite]
I can think of a few variations on this idea.
What will a nonuniform roller do on the 2D version. Would a triangular bar cause it to beahve differently? Or a square bar?
The other thing I'd like to try is a non-newtonian fluid, which would change viscosity with shear. What would shampoo do, it's gets less viscous as the shear increases so it would possibly start out jerky and slow but would then speed up and get smoother. Or, the opposite, say...ketchup, would start out smooth but get jerkier as it sped up, which would slow it down.
Would be fun to try all thses.
posted by Confess, Fletch at 9:13 AM on November 16, 2010
What will a nonuniform roller do on the 2D version. Would a triangular bar cause it to beahve differently? Or a square bar?
The other thing I'd like to try is a non-newtonian fluid, which would change viscosity with shear. What would shampoo do, it's gets less viscous as the shear increases so it would possibly start out jerky and slow but would then speed up and get smoother. Or, the opposite, say...ketchup, would start out smooth but get jerkier as it sped up, which would slow it down.
Would be fun to try all thses.
posted by Confess, Fletch at 9:13 AM on November 16, 2010
If you want a picture of the future, imagine a ball suspended in a semi-viscous fluid inside of another ball, rolling down a gradual incline. Forever.
posted by kaibutsu at 9:41 AM on November 16, 2010 [4 favorites]
posted by kaibutsu at 9:41 AM on November 16, 2010 [4 favorites]
If it's the slowness, then you don't need a "difference in mass" you just need the vanes and viscosity.
The maximum angle down which you can do a slow roll (or a "stop") is dependent on how far the center of gravity is displaced from the axis.
Are you suggesting that merely filling a sphere with fluid and vanes will do a slow roll? You'll get a barely noticeable amount of eccentric behaviour from the difference in inertia between the shell and the fluid (like the old egg trick), but it'll roll right down a ramp.
posted by CaseyB at 10:11 AM on November 16, 2010
The maximum angle down which you can do a slow roll (or a "stop") is dependent on how far the center of gravity is displaced from the axis.
Are you suggesting that merely filling a sphere with fluid and vanes will do a slow roll? You'll get a barely noticeable amount of eccentric behaviour from the difference in inertia between the shell and the fluid (like the old egg trick), but it'll roll right down a ramp.
posted by CaseyB at 10:11 AM on November 16, 2010
This gave me a neat idea for a toy. That someone has already gone ahead and built, of course. An opaque sphere would be more fun for confusing random passersby though.
posted by CaseyB at 10:20 AM on November 16, 2010
posted by CaseyB at 10:20 AM on November 16, 2010
Not only have they built it, they've given it a brain and it's brought friends.
posted by bonehead at 10:34 AM on November 16, 2010
posted by bonehead at 10:34 AM on November 16, 2010
Are you suggesting that merely filling a sphere with fluid and vanes will do a slow roll? You'll get a barely noticeable amount of eccentric behaviour from the difference in inertia between the shell and the fluid (like the old egg trick yt ), but it'll roll right down a ramp.
I think they're suggesting half-filling a sphere with fluid and vanes. In this version, the mass of the fluid in the bottom half of the ball is what retards the ball's motion; the ball can only roll as quickly as the fluid can flow down the sphere's inner wall. Adding some vanes would impede this flow rate, therefore slowing the sphere down even more. This was my initial guess for how these balls worked.
(Of course, you'd eventually get most of the viscous goo in a fairly even coating of the inside of the sphere, reducing the mass available to retard motion and required to move, so the longer you used it in a given session, the faster it would go. It'd be a different, but still cool, puzzle/toy).
posted by metaBugs at 10:36 AM on November 16, 2010
I think they're suggesting half-filling a sphere with fluid and vanes. In this version, the mass of the fluid in the bottom half of the ball is what retards the ball's motion; the ball can only roll as quickly as the fluid can flow down the sphere's inner wall. Adding some vanes would impede this flow rate, therefore slowing the sphere down even more. This was my initial guess for how these balls worked.
(Of course, you'd eventually get most of the viscous goo in a fairly even coating of the inside of the sphere, reducing the mass available to retard motion and required to move, so the longer you used it in a given session, the faster it would go. It'd be a different, but still cool, puzzle/toy).
posted by metaBugs at 10:36 AM on November 16, 2010
With respect to robotically controlled hamster balls:
The swarm orbs are a more fully realized example of the same idea.
I do love these old dime-store curiosities, though. Snail balls, gyroscopes and the like just make me happy.
posted by poe at 11:57 AM on November 16, 2010
The swarm orbs are a more fully realized example of the same idea.
I do love these old dime-store curiosities, though. Snail balls, gyroscopes and the like just make me happy.
posted by poe at 11:57 AM on November 16, 2010
« Older Going Dutch | I'm not familiar with that address. Would you... Newer »
This thread has been archived and is closed to new comments
posted by 256 at 6:41 AM on November 16, 2010