Colin Colenso said:
Hi Jim,
I think you may have got the square inversed accidentally.
I'm not sure what you mean Colin. My formula is nearly the same as jsp's, but in a different form. We both find the cosine of the cut angle since this is the speed of the cueball along the line between the centers of the two balls at impact. Using his x (fraction of the object ball hit), 1-x is the sine of the cut angle. He then uses the inverse sine function to get the cut angle, and takes the cosine of this. My formula computes the cosine using cosine=Sqrt[1-(1-x)(1-x)]. These are equivalent. But he then squares the cosine for some reason.
But why is the cosine relevant here?
We know that at the moment just before impact the cueball's velocity can be analyzed (arbitrarily but conveniently) into two components: one in the direction of the line between their centers, which is vcos(phi), and one in the direction tangential to their surfaces at the contact point, which is vsin(phi), where phi is the cut angle. If we ignore the friction effects, we can completely ignore the tangential component vsin(phi), as far as the interaction with the object ball is concerned. The only thing the object ball "sees" is the cueball coming at it with a speed of vcos(phi) and from the direction of the line between their centers. This is true whether the cueball is colliding with a single ball or the head ball of the rack.
Colin Colenso said:
I think the property he really needs to know here is kinetic energy such that ke = m x v x v/2 . So on a half ball hit, the balls both end up with 0.7071 of th original CB velocity (assuming non rotating) and when w square 0.7071 we get 0.5. So each ball has half the original kinetic energy.
I think you mean a 45 degree hit, or slightly thicker than a quarter ball hit?
I agree that kinetic energy might be a more appropriate thing to consider but the original poster's question was about force. The impulse acting between the cueball and the head ball of the pack acts along their line of centers (ignoring friction). It's the average force multiplied by the contact time and is equal to the change in the momentum of the cueball along the line of centers. (It's also equal to the sum of the momentum picked up by the balls in the rack.) Since the contact time is nearly the same at all cut angles, the momentum of the cueball in this direction, mvcos(phi), is the appropriate indicator of the force that will develop. Just how much force is generated depends on the effective mass of the pack, which in turn depends on the approach angle, I think. If you hit the head ball at the same contact point so that the approach angle into the rack is the same, but vary the starting location of the cueball so that you vary the fullness of the hit, the mass of the pack that the cueball sees will be the same in each case. So at least here we can say, I think, that the force that develops will vary exactly, or very closely, with the cosine of the cut angle on the head ball.
Colin Colenso said:
So if, according to the cosine of 1/10th we get 0.99, then we should square this which equals less. eg. Approx 0.98 And so for a 3/4 ball, I don't have a calculator here but would be about .95 x 0.95 which is about 90% of K.E. transfered to the OB.
Or 2% loss in Kinetic energy, but just 1% loss in momentum.
I agree, but as you indicated, this would be true for a collision with a single object ball. The collision with a pack is different because the mass of the pack is not equal to the mass of the cueball.
If you still think I'm wrong, please fire away. I could be having a bad trig day.
Jim