In this paper I describe the explosion of an aeroengine, the Olympus 22R, and the consequences of that explosion. Empirically, I explore both the puzzle-solving of the engineers as they tried to ascertain what had gone wrong, and the way in which this led to substantial delay in a major aircraft project, and consequent large-scale political and economic repercussions. Theoretically, I use these events to reflect on and denaturalise notions of scale and size. Instead of social and technical phenomena being seen as intrinsically different in size (a Euclidean notion), scale and size are considered to be relational effects. The aeroengine explosion is thus treated as disrupting a mathematically transitive series which was producing scale and size—and the social and technical repair work is treated as an attempt to remake scale relations so that ‘small things’, such as pieces of metal in the interior of aeroengines, were again rendered smaller than ‘large things’, such as economic and political context.