sr241 said:
in the video linear momentum is conserved. and angular momentum is conserved by ball starts rotating in opposite direction of bar.
In the video, after collision, the ball and paddle on the left rotate in opposite directions; but the ball and paddle on the right rotate in the
same direction. I don't know what software you (or somebody) used to create this simulation, but it does not appear to be physically accurate. For a start, upon collision, the balls start rotating anticlockwise instantaneously of their own accord. There is no physical reason why they should do this. For the purposes of understanding this discussion, please ignore what you see in that video.
I might have confused things when I wrote, in my first post, "...the ball-paddle system possesses angular momentum." I ought to have added, "
about its (the ball-paddle system's) own centre of mass".
Please note that the above applies to your diagram in post #12, i.e. in the case where the ball-paddle (or ball-paddle-base) system possesses non-zero angular momentum about its own centre of mass.
For your final question, regarding the two balls of equal mass, we need only consider the conservation laws which
Doc Al has already stated: conservation of linear momentum, and conservation of angular momentum.
(For simplicity, let all trajectories lie in the
x-
y plane.)
Let's say ball A travels with velocity \vec u along the positive x axis. Ball A strikes ball B off-centre. Let the final velocities of A and B be \vec v and \vec w respectively. Then, due to conservation of linear momentum:
\left| {\vec u} \right| = {u_x} = {v_x} + {w_x}
{v_y} + {w_y} = {u_y} = 0
\Rightarrow {v_y} = - {w_y}
For a perfectly elastic collision, balls A and B each rebound with zero spin, and their velocities are at 90º to each other. This conserves linear momentum and linear kinetic energy.
If the balls impart some rotation/spin to one another upon collision (due to friction, deformation etc.) then a portion of the initial kinetic energy of A is transferred to rotational/spin kinetic energy of A and B. The remainder of A's initial kinetic energy is transferred to linear kinetic energy of A and B. The balls' rebound velocities will then necessarily make an angle 0º < \theta < 90º with each other. (Say if you need help proving this.)
My feeling is that if the balls do impart some spin angular momentum to one another, then the angular momentum of A will be equal and opposite to that of B. (I haven't proved this. Perhaps someone could comment...?)
The angular momentum of the system (w.r.t. some arbitrary point P) is constant. The linear momentum of the system is constant. The linear kinetic energy of the system can change. The rotational kinetic energy of the system can change.