With increasing of hydrological data records and the development of technologies for flood discharge estimation, as well as higher requirements on dam safety issues, a large number of existing dams require spillway rehabilitation to improve their hydraulic capacity. For such projects, the Piano Key Weir (PKW) is an efficient alternative. When compared to standard labyrinth weirs, this structure provides a longer effective crest length for a given spillway width, with the advantage that a PKW can be mounted on the top of most existing dams, due to its reduced base surface [1, 2]. Although the hydraulic design and optimization process of PKWs are today supported by physical modelling of case studies [3, 4, 5, 6], systematic basic experiments performed in laboratory channels were important to understand their hydraulic behavior [7, 8, 9, 10]. Most of the experiments consider only one sectional part of the PKW, with uniform approach flow conditions. The three-dimensional effect of the lateral weir ends, characterizing a typical reservoir inflow, is consequently not considered. As a result of these systematic tests, some procedures for designing PKWs are proposed in literature. Among these methods, a simplified formulation for calculating the hydraulic capacity of A-type PKWs [10] is based on the most important geometrical dimensionless parameters of PKWs, i.e, L/W, Wi/Wo, Pi/Po and H/Pi, (Fig. 1). The objective of this paper is to present some selected results of systematic experiments performed in a laboratory channel, focusing on the number of PKW units and the influence of the parapet walls.