A number of vortex flow control (VFC) devices for urban drainage systems are investigated computationally at high flow rates, for which a confined vortex dominates the flow regime. A range of turbulence models, including both eddy viscosity and Reynolds stress closures, are compared with in-house experimental measurements of head loss and internal pressure measurements. Single-phase and multi-phase (free surface) calculations are also compared. Very good agreement with the experimental data was obtained when the swirl parameter of the device was below 3 . 14 for predictions made using the Reynolds stress closure formulations. For devices with swirl parameters above this value, the computational methodology was found to under-predict the head loss of the device. This was attributed to poor calibration of the turbulence model for swirling flow scenarios in which the pressure gradient and diffusive (turbulent) forces in the flow are comparable.