Remember Me?

 Straight Dope Message Board Remember Me?

#1
01-14-2019, 12:10 PM
 filmore Guest Join Date: Aug 2002 Posts: 4,197
What would happen if you released marbles into a *spinning* coin funnel?

At places like kid museums, you may have seen large coin funnels where you drop a coin into it and spirals around the funnel until it finally drops through a hole on the bottom. They look like this. The funnel is stationary and the coin spins round and round.

But what would happen if the funnel itself was spinning rather than stationary and you dropped marbles into it? Would the marbles be able to find stable orbits around the hole in the center? Since the funnel itself is spinning, it would provide some outward force on the marbles to counteract gravity. If the funnel was spinning really fast, the marbles would be thrown off. If it was spinning very slowly, the marbles would spiral down the hole. But it seems like if the funnel was spinning just right, a marble couuld find a sweet spot where the outward centrifugal force matches gravity and would end up in a stable orbit around the hole.

I tried looking for such an experiment, but I couldn't find anything. There's plenty of videos of marbles spiraling around funnels, but the funnels are always stationary.
#2
01-14-2019, 12:21 PM
 CalMeacham Member Join Date: May 2000 Location: Massachusetts Posts: 43,120
That would probably work. The coins fall into the center -- as do marbles used in some museum versions of this -- because friction quickly eats up the energy the coin or marble starts out with. But if the funnel itself is rotating, it's essentially providing the energy needed to keep the coin or marble "up", so that even in the presence of friction you'll get a stable situation.

I've often thought of something more ambitious -- construct a "double funnel" model that duplicates the gravitational potential of a two-body system, with one body being significantly larger than the other. Now rotate THAT model, and see if a marble placed at the L4 or L5 Lagrange points maintain equilibrium. Even cooler, disturb them slightly from that equilibrium and watch them execute their kidney-shaped orbits around the Lagrange points. (You could also demonstrate how the other Lagrange points are un stable.)
__________________
"Blue, Navy Blue, he's as blue as he can be;
'Cause my steady boy said "Avatar!" and joined the Na'avi."
#3
01-14-2019, 12:22 PM
 md2000 Guest Join Date: Feb 2009 Posts: 14,318
As an alternative, consider those circus acts where one or more motorcyclists are going in circles on the interior of a big mesh ball (cage). What would happen if the motorcyclist sat still and the cage spun under him? He would not upside down hang in mid-air.

The motion of marbles, pennies, or motorbikes depend on their movement. The coin spirals (and would work the same with marbles or anything that moved smoothly) rely on the motion of the object - it gains momentum then wants to go in a straight line. the surface of the spiral chamber forces it to curve, and that is the apparent centrifugal force that keeps the coin spiralling.

Of course, the small amount of friction could begin to drag the marble along until it is spinning with the funnel container. The closest real-world analogy would be a roulette ball; by hitting the spinning wheel surface and being impacted by the irregular surface, it gains momentum that causes it to spin with the wheel, until it has more momentum that the wheel (less friction between rolling metal ball and surface than between wheel and table.) At that point it continues rolling while the table under it slows down.
#4
01-14-2019, 01:35 PM
 rat avatar Member Join Date: Dec 2009 Location: Seattle, Wa Posts: 5,304
You may be able to find a stable point with marbles or coins if you could impart just the right forces for a little while. It would be non-trivial though and when the marble's orbit would actually decay faster than in the typical example.

Centrifugal force is a "fictitious force" which always proportional to the mass of the object on which it acts. In other words a feather and an elephant will see the same "acceleration". The issue is that under General Relativity the force we typically call "Gravity" is also a "fictitious force". Orbiting bodies under GR are traveling in a straight path with zero acceleration, but spacetime is curved in just the right way that results in an orbit.

So the important thing to remember is that if you built this contraption it wouldn't be an actual model of gravity but just another visualization tool that approximates the behavior.

The math to figure it out involves inertia, rotational energy, angular momentum, gyroscopic motion, and rotational stability so it is a bit complicated to figure out exactly. But getting it to slip in just the right way would be the trick and may guess is that precession and nutation would make it unstable.

The constant torque and angular momentum vectors in the marbles frame precess about the angular velocity vector. As a rolling marble will act like a rolling gyro and if the speed was just right with some slip it would orbit but in the reference frame of the spinning funnel the marble would still slow down leading to nutations to cause a stall or the rotation of the marble slowing down and dropping in he hole.

Remember that the coins stay up because they are gaining forward speed as they drop due to the conservation of momentum (remember to include angular). An orbiting body need to have a faster sidereel speed to have a closer orbit. So when that marble slows down the orbit is going to degrade.

Consider that Wall of Death motorcycles have a throttle lock or you have to manually close the throttle. You don't really gain anything from spinning the funnel as the relative speed will still drop due to air resistance and rolling friction so a motorized unicycle would probably be easier to accomplish your goal. While the rotating funnel would have a centrifugal/centripetal component that force is going to work to maintain the distance from the origin and not forward momentum.

Last edited by rat avatar; 01-14-2019 at 01:36 PM.
#5
01-14-2019, 02:50 PM
 OldGuy Charter Member Join Date: Dec 2002 Location: Very east of Foggybog, WI Posts: 5,102
Quote:
 Originally Posted by md2000 The closest real-world analogy would be a roulette ball; by hitting the spinning wheel surface and being impacted by the irregular surface, it gains momentum that causes it to spin with the wheel, until it has more momentum that the wheel (less friction between rolling metal ball and surface than between wheel and table.) At that point it continues rolling while the table under it slows down.
IANAGambler, but it's my recollection that the ball in a roulette wheel is directed opposite to the direction that the wheel itself spins. Also most of the time the ball is revolving it is not actually touching a spinning portion of the wheel but is above that. Friction slows the ball then it falls into the spinning portion, but at that time it's motion is mostly chaotic.
#6
01-14-2019, 03:23 PM
 Marvin the Martian Member Join Date: Jun 2015 Location: Phoenix, AZ, USA Posts: 1,082
When you roll a marble (or a coin) into a stationary funnel you will impart an initial velocity which can be broken down into a vector towards the center of the funnel and one tangential to the circumference of the funnel at the point where the marble first hits the funnel. The path of the marble through the funnel will be determined by this initial velocity.

If the funnel is spinning, the tangential part of the initial velocity is increased or decreased by the rotation of the funnel, but once the marble is rolling the path should be the same as if it were launched into as stationary funnel with an increased (or decreased) initial tangential velocity. If you launch the marble opposite the direction of spin with a tangential velocity exactly the same as the rotational velocity, the marble should fall straight down the hole.

Some variance will occur from different wind resistance. Also, friction between the marble and the funnel will be different initially, at least until the marble transitions from skidding to rolling.
#7
01-14-2019, 03:59 PM
 Melbourne Guest Join Date: Nov 2009 Posts: 4,442
If you mounted the funnel slightly eccentric, you could use the motion to pump the marble. You wouldn't have to depend on rolling friction to drive the marble.

The problem, as I see it, with using rolling friction (imagine a very slightly rough surface), is that the marble would spin/roll backwards in it's stable (forwards) orbit. If it dropped down slightly, it is spinning too fast for it's forward motion, but instead of the spin kicking it faster (and up), it kicks it slower (and down).

(Intuitive science, not engineering: I haven't done an analysis).
#8
01-14-2019, 04:01 PM
 Chronos Charter Member Moderator Join Date: Jan 2000 Location: The Land of Cleves Posts: 80,734
Cal, I strongly suspect that your Lagrange funnel wouldn't work. Stability of the Trojan points is actually a fairly subtle effect, and the funnel is only an imperfect analogue for gravity. But I'd nonetheless love to see someone try.
#9
01-14-2019, 04:01 PM
 DSYoungEsq Member Join Date: Jul 1999 Location: Indian Land, S Carolina Posts: 13,997
My question is this: is the funnel spinning as fast as the marble is rolling?
#10
01-14-2019, 04:46 PM
 rat avatar Member Join Date: Dec 2009 Location: Seattle, Wa Posts: 5,304
I have posted this video before, but seem to always credit MIT and not University of Toronto. Despite being from 1960 it is the best example of frames of reference impact this.

https://youtu.be/bJMYoj4hHqU?t=1024

I tried to link to the relevant time period of 17:04 to 18:10, but it will clarify a very common misunderstanding on the nature of forces in a spinning frame.
#11
01-14-2019, 06:02 PM
 Sam Stone Member Join Date: Jun 1999 Posts: 27,376
Quote:
 Originally Posted by OldGuy IANAGambler, but it's my recollection that the ball in a roulette wheel is directed opposite to the direction that the wheel itself spins. Also most of the time the ball is revolving it is not actually touching a spinning portion of the wheel but is above that. Friction slows the ball then it falls into the spinning portion, but at that time it's motion is mostly chaotic.
Correct. The ball runs around the track because the dealer started it spinning, not because it's spun up by the wheel. The track is slightly angled such that when the bacll drops below its 'capture velocity' it will fall into the rotating part of the wheel, which as you say rotates in the opposite direction. There it bangs into 'randomizers', or little metal plates that stick up in various places on the spinning surface to make sure the ball doesn't move predictably.

That said, they aren't perfect randomizers, and if you know the exact position and velocity of the ball in the track and the velocity and position of the spinning wheel, you can beat roulette by a pretty good margin. This is why computers are illegal at the roulette table.
#12
01-14-2019, 06:06 PM
 Sam Stone Member Join Date: Jun 1999 Posts: 27,376
As for the marble dropped into a spinning funnel, the most important factor is friction. The ball is going to want to go straight down the hole, but will be slowly accelerated by the spinning wheel under it. If the spin is too low or there isn't enough friction, the ball will just follow an arc down into the hole. If the table is spinning fast enough, the ball would actually trace a hyperbolic trajectory and be thrown completely clear of the rotating cylinder. If you get the speed just right you could get the ball to stay, but it would be tricky.

Best bet - set up a stepper motor and a camera, so that if the ball moves towards the hole the funnel speeds up, and if it moves away the funnel slows down. You should be able to set up a nice feedback circuit that keeps the ball just about where you want it in the funnel. That would be a fun project to try.

Last edited by Sam Stone; 01-14-2019 at 06:06 PM.
#13
01-14-2019, 07:04 PM
 rat avatar Member Join Date: Dec 2009 Location: Seattle, Wa Posts: 5,304
OK I tried to make the math work with a simple y = -1/|x| hyperbolic funnel and couldn't get the math to work with even added momentum.

I then found this paper:

https://scholar.dickinson.edu/faculty_publications/771/
Quote:
 For the first surface—the hyperbolic funnel—we found that the marble does not mimic planetary motion, except in the special case of circular orbits for which Kepler’s third law holds. In general, however, the bound orbits traced out by the marble are not closed. For an inverse-square surface, an analytical solution for r(t) can be obtained and we can show that all trajectories asymptotically approach either zero or infinity; here the circular orbits form a line of equilibria, but they are unstable against perturbations
Basically these funnels kind-of work for circular orbits but don't obey Kepler’s third law, which should be the inverse law or:
T2∝R3
On the funnel is:
T2∝R
Basically no cylindrical symmetric surface at this scale exists that would be capable of reproducing the orbits of an object subjected to uniform gravity.

While circular orbits can get close they are unstable if you have a "closed" orbit and are only stable for these devices because none of the "orbits" are closed. Looking at other papers spandex or other elastic surfaces and mimic an orbit more accurately but a ridged surface would have to be shaped in a way where the situation really resembles spinning a bucket of water over your head than an orbit.
#14
01-14-2019, 09:38 PM
 Chronos Charter Member Moderator Join Date: Jan 2000 Location: The Land of Cleves Posts: 80,734
That's not what the paper you quoted says.
#15
01-15-2019, 11:03 AM
 rat avatar Member Join Date: Dec 2009 Location: Seattle, Wa Posts: 5,304
Quote:
 Originally Posted by Chronos That's not what the paper you quoted says.
Which part? About the paper, the link or my math conclusions? I do agree it didn't say "Ring-ding-ding-ding-dingeringeding! Gering-ding-ding-ding-dingeringeding!Gering-ding-ding-ding-dingeringeding!"

I provided the cite that will let you find other papers, and the math is there but that paper I provided did error on the side of finding pedagogical value which is appropriate for their main use case.

https://pennstate.pure.elsevier.com/...arious-funnels
Quote:
 We investigate the trajectories of a small marble constrained to roll on frictionless surfaces, called funnels, of varying shapes. Actual coin funnels have a hyperbolic surface, and here we disprove the common claim that the orbits of the rolling marble or coin are the same as the Kepler orbits for planets revolving around the sun. In fact, it is straightforward to show that for no funnel surface can the Kepler orbits be recovered. Furthermore, we find that the types of trajectories that can arise depend heavily on the funnel shape.
And here is a random other paper who agrees with my math, which was based on the above cite, on the Third law, look under this text to find that.

https://arxiv.org/pdf/1506.03342.pdf
Quote:
 We further note that although Eq. (3.14) yields the desired Nε orbits, these orbits do not, in general, obey Kepler’s third law
If you look on page 9 at Figure 1, note how the "κ = −1 surface" at 1 is approaching a cylinder at r =1, exactly as I found based on the original paper.

"The square of the period of a planet in a circular orbit is proportional to the cube of the orbital radius"

But I think you will find that this part of the original paper:

Quote:
 we found that the marble does not mimic planetary motion, except in the special case of circular orbits for which Kepler’s third law holds
Remember those papers are considering if it is close enough to use as a teaching tool, and it is, I was trying to use those same assumptions to see if one could figure out a way to maintain stable orbits per the OP. Any torque which would cause the "marble" to climb is problematic in the case of this thread but not in the case of a physics demo.

As my math related to Kepler's 3rd law, related to T and the previous paper didn't solve for that solution obviously the paper and my math wasn't written in the text but if you look at the math in section "III. A SURFACE PROPORTIONAL TO r-1" it is trivial to come to the same conclusions using high school level mathematics.

As the period is just T = 2πR/v it is all there. As I stated above the effects of climb and added torque are the easiest place to see the issue.

Last edited by rat avatar; 01-15-2019 at 11:05 AM.

 Bookmarks

 Thread Tools Display Modes Linear Mode

 Posting Rules You may not post new threads You may not post replies You may not post attachments You may not edit your posts BB code is On Smilies are On [IMG] code is Off HTML code is Off Forum Rules
 Forum Jump User Control Panel Private Messages Subscriptions Who's Online Search Forums Forums Home Main     About This Message Board     Comments on Cecil's Columns/Staff Reports     General Questions     Great Debates     Elections     Cafe Society     The Game Room     Thread Games     In My Humble Opinion (IMHO)     Mundane Pointless Stuff I Must Share (MPSIMS)     Marketplace     The BBQ Pit

All times are GMT -5. The time now is 07:21 PM.

 -- Straight Dope v3.7.3 -- Sultantheme's Responsive vB3-blue Contact Us - Straight Dope Homepage - Archive - Top

Send questions for Cecil Adams to: cecil@straightdope.com