Well, sorry for being absent a long time. I´ ll go with torques. Writing torques about the center of mass we get:
## \frac {L}{2\sqrt {2}} (N_y - N_x) = I_0 \alpha ##
The equation F=ma here was not useful since you added an extra unknown per equation. So I thought that I could relate the...
Let´ s call ##N_x## the magnitud of the force between the rod and the box and ## N_y## the magnitud of the force between the rod and the surface.
##N_x = ma_c##
##N_x= ma_r##
##mg-N_y=ma_y##
The following I think is to find a relation between ##a_r## and ##a_y## and that can be found by...
Though in this case it is like a combined acceleration. The acceleration of the masses depend on the acceleration of the wedge and both would accelerate immediately to the needed value.
Ahhh thank you I already found the answer with that. When thinking about this problem I thought that there had to happen some time since the ball is placed in top of the wedge until the wedge acquires the needed acceleration but would it not be strange that acquires it immediately?