# Football wobble

1. May 2, 2005

### PhilG

I've noticed that a football (American) wobbles a little bit even when you throw it with a near perfect spiral. Why is this?

2. May 2, 2005

### Andrew Mason

Aerodynamic drag forces are not perfectly symetrical so they create torque on the ball. This means that dL/dt is non-zero so the angular momentum is not constant. So it has to precess. Even bullets precess slightly (bullet yaw). It can be minimized but not eliminated.

AM

3. May 2, 2005

### PhilG

Okay. I thought it might have something to do with aerodynamics, but I didn't know if the forces would be enough to produce visible motion. The football is pretty symmetrical except for the threads. What causes the imbalance in drag forces?

4. May 2, 2005

### whozum

Pockets of varied air pressure, the thread themselves too. Also, you can't throw a perfect spiral, and that has something to do with it. Also, the wind direction not being completely parallel/antiparallel will make a difference.

5. May 2, 2005

### PhilG

Right. But does the football cause those air pockets or are they already there? Also, I've seen this even when there's practically no wind.

6. May 2, 2005

### HackaB

The uneven drag happens because u throw it at a slight angle to its initial velocity. "Air pockets" don't cause a regular precession.

7. May 2, 2005

### PhilG

Okay, just to make sure I understand: if the football were thrown in vacuum, you wouldn't see the wobble, since there are no aerodynamic forces, right?

8. May 2, 2005

### DaveC426913

Well, no. The air also stabilizes the ball on its longitudital axis. In a vacuum, it would keep whatever spin you put on it, even if that were quite small.

9. May 2, 2005

### PhilG

So you don't need aerodynamic forces to make it wobble?

10. May 2, 2005

### FredGarvin

Don't also rule out imperfections in how the ball is made. The uneveness of the material and construction will move the center of mass from the longitudinal axis of the ball. There is no such thing as a perfect anything.

11. May 2, 2005

### DaveC426913

Eeeehhhhh....

Aerodynamic forces will/may impart some wobble on the ball, yes. They also will tend to keep it from going wild, yes.

If there were no aerodynamic forces, AND you could throw the ball with a spin one only the longitudinal axis, AND you had a perfect ball with no imperfections, then you would get no wobble.

A ball with imperfections would still likely wobble a bit merely because the axis of rotation about its centre of mass would not run exactly through the long axis of the ball (say, the threads add a little weight on one side, your centre of mass is off by a mm, etc.)

But if you threw the ball with any imprefection in the spin, and there were no airflow to dampen it, the ball would tumble in all three axes. It would spin quickly around the long axis, as you intended, but additionally would tumble end over end slowly - something that it would not do in the presence of air.

12. May 2, 2005

### PhilG

Thanks for clearing that up, Dave. If the ball is tumbling and spinning at the same time, isn't its angular momentum changing?

13. May 2, 2005

### HackaB

Seems like it would be.....

14. May 2, 2005

### DaveC426913

Why? It's, merely rotating on two axes.

15. May 2, 2005

### HackaB

It takes a torque to rotate something about a non-principal axis.

16. May 2, 2005

### PeteSF

Toss a pen in the air so it rotates end over end.
Now repeat, but put a spin on the pen as you toss it.

There you have it. Rotation on two axes. No torque applied after the initial impetus.

17. May 2, 2005

### Cyrus

I know bullets have spiral in them in order to keep them stable in flight. I would assume that the football needs the same thing in order to move in a predictable arc. Perhaps the amount of spin required to obtain a very stable flight is hard/usually not obtained when a person throws a football, which explains why you see (some) wobble no matter how hard you try?

18. May 2, 2005

### HackaB

It doesn't rotate end over end if you put spin on it. It does wobble though.

What do you actually mean by "rotation on two axes"? The result has to be equivalent to rotation about one axis (not necesarily a principal one). Would you say that axis of rotatoin is moving or fixed?

19. May 3, 2005

### PeteSF

It surely does. I'm doing it right now. Try it again.

I mean two components of rotation, with two different periods.
Why?
One axis (the end-over-end rotation) is moving in a parbolic arc, but always points in the same direction.