Much of the future potential for tidal energy lies in tidal channels, where turbines can be arranged in an array in such a way as to exploit the effects of channel blockage and thus increase performance. One such strategy employed by researchers has been to have an array of turbines which only partially blocks the channel width. Previous work, employing mass and momentum conservation and considering the turbines as simple actuator disks, has shown that the local arrangement of the turbines can be altered in a way that optimises the power output of the whole array. However, real turbines also impart swirl to the flow. When this swirl imposed on the flow by the turbines is taken into account by modelling the turbines with Blade ElementMomentum Theory (BEMT) discs, further interaction effects become apparent. These effects depend on the proximity of the turbines to one another and can have a significant impact on the optimal arrangement of the devices in a given channel. This report demonstrates the key effects by comparing sets of actuator disc simulations with and without BEMT, and makes the case for taking swirl effects into account when designing tidal arrays computationally. The other benefit of using BEMT discs is that they rely only on known, physical characteristics of the turbines (i.e. lift and drag performance curves) as opposed to estimated quantities (i.e. the momentumextraction coefficient). This means that real turbines can be simulated with more accuracy. It will be argued in this paper that these twin advantages of increased realism in both the flow field and the turbine outweigh the negligible increase in computing power required to undertake BEMT simulations and so these simulations should be theminimum standard for tidal array designs.