Air traffic controllers use a radar separation minimum when keeping aircraft safely apart. The choice of minimum needs to ensure, to a very high degree of confidence, that radar inaccuracies are not putting the aircraft into a risk of collision – controllers can of course be required to use larger minima for operational reasons. A particular separation minimum necessarily feeds back into requirements on radar design criteria, data processing and performance. The estimation of the radar separation minimum is significantly improved by the use of a new methodology. Previous derivations were a major step forward, but were ‘static’, in that they did not attempt to take properly into account how controllers use separation minima dynamically. An analysis on the use of separation minima in en route airspace by controllers concludes that they are generally likely to use the radar minimum to pass aircraft, rather than to use the minimum in a ‘strategic’ sense (e.g. ‘in trail’ separation). The new ‘Event Model’ methodology recognises that collisions cannot happen instantly: the aircraft pair concerned must move from a safe configuration to a hazardous one. Collisions are therefore events caused by flawed flight paths, dependent on both the initial positions of aircraft and their subsequent movements and velocities.