I have never crashed while doing a wheelie also..... but I did read this before I started...maybe thats why???????
For a wheelie to occur, torque must be applied to the bike such that it rotates around the rear axle, the speed at which the bike accelerates around the axle being determined by the amount of torque applied.
Figure 1
Torque is defined as mass x distance. Figure 1 shows the torque derived from gravity and acceleration acting on the mass of the bike. The diagram shows the center of gravity (CoG) of the bike which we'll use to pinpoint the bike's mass. Note that this looks higher than it should - what's missing from the diagram is the mass of the rider. Regardless, the principles remain the same wherever the actual CoG should be.
Fg is the force asserted by the bike's mass as a result of gravity. The torque - Tg - around the rear axle that this produces is Fg x Dg where Dg is the distance from the axle at which the force acts.
Fa is the force asserted by the bike's mass as a result of acceleration. The torque - Ta - that this produces is Fa x Da.
Fe is the effective force that results when Fg and Fa are combined. De is the distance from the rear axle at which this force is applied, so the resultant effective torque - Te - is Fe x De.
Put another way:
(Fe x De) = (Fa x Da) - (Fg x Dg)
or:
Te = Ta - Tg
Animated figure
In Figure 1 the effective torque around the rear axle is torwards the front, so the net effect is to keep the front wheel on the ground. Figure 2 shows the situation if the bike accelerates harder which, in turn, increases Fa and, hence, Ta.
In Figure 2 the effective torque around the rear axle is towards the rear so the front wheel will lift. But this isn't the end of it:
Figure 3 shows the situation a short while later with the bike doing a small wheelie. The bike's acceleration is kept the same as in Figure 2, but now the rotation of the bike has reduced the torque due to gravity and, at the same time, the torque due to acceleration has increased. The net effect is that the effective torque around the rear axle has increased by a large amount, thus increasing the rate at which the bike wheelies. This positive feedback means that, unless the bike's acceleration is quickly reduced, the wheelie will rapidly get to the point where recovery is not possible.
Engine power, per se, is not a significant factor in wheelies. Let's be pessimistic and suggest that the torque due to gravity acting on the CoG is a mere 150kg/m. A reasonably powerful 1200cc bike might produce 70lb/ft or torque, which sounds a lot but translates to 9.7kg/m. Enough to get the suspension working is all it amounts to.
Where engine power is important is in its ability to accelerate the bike. On smaller bikes the sustainable acceleration isn't sufficient to apply enough torque to lift the front wheel. However, using tricks such as 'dipping' the clutch or tugging on the handlebars, a brief burst of extra torque can be applied for a short time. This extra torque is not sustainable but, as we can see above, once the CoG has moved up and back a much lower torque (that is, acceleration) is required to sustain a wheelie.
That should clear up all the confusion :thumbup