Wrecking Ball: Case Study

            This was a motion graphics project created in Cinema 4D using rigid bodies and a Voronoi fracture. It was then ported into After Effects for some post production. Everything in this project was created and modeled by me.

            The rigid bodies in this project are: an invisible wall in the building (which will work as our collision), each individual chain link, and the wrecking ball. Working with this project required physical dynamics that mimic how a wrecking ball (and falling concrete pieces of death) would work in real life.

            Here is what the menu looks like:

           

            Let’s start with the wrecking ball. I allowed it to have plenty of weight and plenty of gravity for the heaviest looking swing, as well as applied a wind effector directly behind it for extra swing:

            You may have noticed in the menu that I applied the rigid body to the Voronoi fracture instead of the ball.

            Since the ball needs to turn from a ball to a fracture at the exact moment it collides with the building, I decided to just forgo its need for a rigid body to begin. Therefore, the Voronoi fracture does not trigger until the program recognizes the ball has collided with the invisible wall.

Once it does trigger, I allowed smaller pieces towards the section of greatest impact using the Point Generator, setting the type to Exponential, and tweaking the axis affection:

 

Not perfect, but given the camera’s position, it’s doable.

            One of the issues with this design was quite a shock to me. It turns out, if dealt enough force, the chain links can slip straight through each other. This is because of 1. The chains pushing between the slots of the faces of the following chains, and 2. The force that it takes to actually push them through.

            To remedy this, I made all of the chain links have an insanely high friction level, reducing the amount of wiggle room they take:

           

If I were to design this again, I imagine that reconstructing the polygons to be interlocking triangles would do the trick, instead, leaving no slots for error.