We’re developing a neat little workshop based around catapults, bits of which have sneaked out into the world in test events over the last couple of months. The core of the challenge is the simplest possible catapult we could dream up, which uses a paper cup, elastic band and a plastic spoon to fire a table tennis ball at least five metres. It’s tremendously satisfying, but it’s not quite enough on its own for the workshop.

My Mark II Catapult uses the same basic ingredients, but pivots the spoon on a kebab stick bearing. It’s even more satisfying, but a little more fiddly to make and – annoyingly – slightly less effective in range.

This morning: the Mark III Catapult, which uses the elastic band as a torsion spring. It’s fiddly to make, pushes the limits of cardboard cup rigidity even more than the other designs, and…

…is almost completely ineffective.

Back to the drawing board, with one key engineering principle ringing in our ears: keep it simple.