"The easy answer is just to quote a term at you: "pitch-roll coupling".

The more technical answer (in layman's terms) goes like this:

Have you ever done the gyroscopic motion experiment where you hold a spinning bicycle wheel by handles on the axle, and then tilt it to one side? It reacts by tilting 90 degrees to both the direction you tilted it and to the axle.

Think of your long, over stable rocket as a gyroscope initially rotating around two axes: around its "roll axis" (the vertical centerline of the Rocket) due to fin misalignments and asymmetrical airfoils, etc., and around its "pitch axis" (horizontally into the wind around the CG due to weather cocking).

Due to the way gyroscopic forces work, the pitching moment combines with the roll to result in a torque at 90 degrees to either (around the "yaw" axis, at right angles to the other two axes). Since the rocket is still rolling, the yawing moment causes a similar torque around the pitch axis again, this time at right angles to the pitch and roll axes, in the opposite direction from the initial weather cocking tilt. Of course, now the yawing moment combines with the roll to cause a torque around the pitch axis....

The end result is that the rocket starts swinging its tail and nose around in circles, describing a moving cone, with its apex at the rocket's CG. This coning movement naturally reinforces the roll, leading to the problem continuing as the rocket ascends.

What can be done about it?

Well, first off, make sure your fins are nice and vertical. If you sand airfoils, make sure they are symmetrical. You don't want the rocket to roll much, if at all. The less it rolls, the lower the magnitude of any gyroscopic torques.

Secondly, try to reduce the "moment of inertia" of your rocket. Your fins are trying to keep the rocket straight. They apply a limited torque due to their force being applied out at the end of a lever arm with the fulcrum at the CG. Now, if your rocket has most of its mass near the CG, it will have a lower moment of inertia, and snap back into line quicker than a rocket with most of its mass out near the nose and tail. The CG might be in the same place, but imagine spinning a three foot dowel with a 1 lb lump of clay at the center, or a 3 ft dowel with 8 ounces of clay at each end to see the difference.

Third, keep the rocket stable, but not overly so. Remember: the rocket rotates (including coning) around its CG. The farther forward of the rocket's center that the CG is, the bigger the circle the tail end will describe in the sky when (not if) it cones.

Making over stable rockets by adding lots of nose weight exacerbates the coning problem two ways: it increases the moment of inertia, which means you get bigger pitching moments before the fins can counteract them, and also means that the gyroscopic effects are larger. You also move the CG forward, so the magnitude of any coning is much worse.

Submitted by Rick Dickinson