This is simulation of one pendulum attached to other pendulum. Observe strange behaviour. Real pendulums behave (probably) the same. Have fun! Green flag to start. Space - random parameters 0-9 - presets, a - circle preset (@Crystall), r - reset t - toggle trace, l/h - change gravity arrows - slightly change rods' speed - experiment
I added the formulas for kinetic and potential energy. I guess if the pendulum rods are supposed to be massless there is no rotational energy, so it doesn't seem like energy is actually conserved...?