Physics of a spinning M&M

[Skott]

Take an M&M (plain) or similarly shaped piece of candy and give it a good spin. If you balance it right, it will spin on its bottom center for a second or two and then it will start spinning on its edge! Why is that?

[Squink]

Well, an M&M’s only got one axis of symmetry passing through its rounded faces, and it’s got zillions passing through its edges, and the sharper arc of the edge makes spinning on that axis less susceptible to perturbation by irregularities in the table surface. I think nutation (wobbling) gets handled less catastropically in the upright position too.

For more spin fun, get yourself a rebellious celt:

a 10-centimetre-long plastic toy with a base shaped like the hull of a boat. When you spin it one way, it turns a few times before the ends start to rattle up and down. The more it pitches, the slower it rotates until it stops spinning altogether. Then the most intriguing thing happens: the celt starts to spin in the opposite direction. It’s as if some unseen hand is playing a joke on the laws of physics.

[Mr.Blue_Sky]

I used to take my class ring and spin stone side down. Eventually it ended up spinning on the shank side before rattling to a stop.

[Skott]

@Squink

Well, an M&M’s only got one axis of symmetry passing through its rounded faces, and it’s got zillions passing through its edges, and the sharper arc of the edge makes spinning on that axis less susceptible to perturbation by irregularities in the table surface. I think nutation (wobbling) gets handled less catastropically in the upright position too.

So, basically the reason that it makes the transition from one state to the other is that when spinning on its rounded face, it encounters enough minute bumps in the table surface to tilt it on to its edge?

My hypothesis had been that as the M&M sped up, the force being directed to the edges begins to outweigh the gravitational force keeping the M&M on its face. Or something like that

It just seems odd, like your other example, since it takes a second or two for the transition to occur. I would think that the fastest spinning would take place at the initial twist one gives it, and thus that would be the most likely time for it to pop up on its edge.