This paper presents an experimental investigation of Reynolds number variation on a separating, reattaching, and redeveloping turbulent boundary layer. This flow is a non-equilibrium boundary layer, meaning that it depends on one or more length scales in addition to the local viscous length scale and boundary layer thickness. These additional scaling dependencies cause the flow to be Reynolds number dependent in some regions. This implies that calculations or experiments based on low Reynolds number data can not be readily scaled up to predict high Reynolds number flows accurately. These experiments were performed using a high Reynolds number flow facility which can achieve momentum thickness Reynolds number up to 30,000 at the inlet of a test section. The mean profile recovers rapidly relative to turbulence statistics downstream of reattachment. The boundary layer separation point depends on Reynolds number, while the reattachment point remains the same at different Reynolds numbers. The flow recovery downstream of reattachment also shows a dependency on Reynolds number in both the mean velocity and turbulence statistics.