When a flexible object is bent, then it will oscillate about its equilibrium position for a while, but eventually the motion will stop, being converted into heat, that is motion of its internal molecules. The applet below was written to look at this process. It models a collection of masses in the form of a triangle, connected together with springs. The equilibrium position (without gravity) would be an equilateral triangle. At the start it is distorted to form a right angled triangle and released. The applet shows its subsequent motion.

There are a number of choices you can make. The strength of gravity can be varied (gravity helps to stop the springs bending back to form a confused mess). The spring constant can be varied and a damping factor can be introduced. These choices take effect as soon as you change them, but the following require the restart button be pressed before they take effect. There is the option of having the masses and springs in the form of a Sierpinski triangle, and of having the masses take random values between 0.8 and 1.2 times the normal value. The number of small triangles along each side of the main triangle can be varied in powers of 2 from 1 to 128.

The applet tends to use a lot of processing power, so there is the option to pause processing.

The graph in the lower right corner shows the average kinetic energy of the masses. When the system is undergoing a large scale oscillation, this energy will tend to oscillate with it. When the energy has been dispersed amoung the individual masses, the average will tend to be constant over time. The Sierpinski option does seem to speed up the dispersal of energy, but surprisingly choosing random masses doesn't seem to make any difference - I'd expected it to disrupt the regularity of oscillations.