Effective testing of Wave Energy Converter models in multi-directional wave tanks requires accurate control of the generated incident wave fields. Since the commissioning of the Edinburgh Curved Wave Tank in 2003, studies have been conducted with the aim of assessing and improving its performance [Cruz et al., 2006; Lucas et al., 2008]. This paper presents the latest stage in these studies, focusing on the tank's beaches reflection coefficients and the tank ability to accurately generate wave fronts at different angles. Using the Mansard & Funke  method, the beaches reflection coefficients were measured for a range of incident wave amplitudes. Monochromatic and polychromatic waves were recorded. Guidance over minimum wave amplitude (7 mm) for a monochromatic 1 Hz wave is given in order to keep the reflection under 8%. On the polychromatic case, statistical analysis of the results suggests that the amplitude of each wave component do not have an influence over the reflection coefficient. This observation is significantly different than results published in earlier studies. An assessment of the tank's capacity to generate wave fronts with a precisely defined direction of propagation is underway. Based on linear wave theory, a method to measure the direction of wave fronts in a multidirectional wave tank was developed. It utilises the phase differences between three wave probes and the wave velocity as inputs to estimate the direction of propagation of an incident wave front (PTPD method, Fernandes et al. ). The probe layout has been optimised to get the best possible accuracy within the tank's geometrical constraints. The directions of wave propagation have been systematically measured over a 60° range of incident wave fronts. Initial results show that the tank reaches a 1° precision for waves in the [-20°;+20°] range. Following the results of this study, corrections will be applied to increase the tank's precision.