Air-Water Flow on a Labyrinth Spillway

Danica Starinac1, Radomir Kapor2, Ljubodrag Savić2, Predrag Vojt1, Dragiša Žugić1, Marijana Damnjanović1, Budo Zindović2, Predrag Đajić3




1 Jaroslav Černi Institute for the Development of Water Resources, Jaroslava Černog 80, 11226, Belgrade, Serbia, E-mail: This e-mail address is being protected from spambots. You need JavaScript enabled to view it

2 Faculty of Civil Engineering, University of Belgrade, Belgrade, Bulevar Kralja Aleksandra 83, Serbia

3 Energoprojekt Hidroinženjering Plc, Bulevar Mihaila Pupina 12, 11070 Belgrade, Serbia




The design of the Beni Slimane Dam (Medea Province, Algeria) was verified by scale model analyses, conducted on two separate scale models: a partial model of labyrinth spillway and a complete model of spillway structures. Aeration of the labyrinth spillway, which was the focus of the study, could be properly assessed only by means of a large scaled model, so a partial model at a scale of 1:15 was used. The initial aeration design was improved on the basis of the model investigations. The spillway discharge coefficients, calculated from measured data, were compared to the values recommended in the literature. This paper describes the potential problems, investigation results and important conclusions, which should be kept in mind when labyrinth spillways are designed.

Keywords: scale model, labyrinth spillway, air-water flow.




A labyrinth spillway is a hydraulic structure devised as a result of the aspiration to design the largest possible spillway-crest length and ensure the highest possible spillway capacity. The labyrinth spillway is generally comprised of several segments of walls of a triangular or trapezoidal shape at the base, with a relatively thin crest (Figure 1). The upstream face is usually vertical and the crest has a Creager profile. The labyrinth spillway can be a very economical solution, given that it provides a higher discharge rate for the same total head, compared to conventional straight overflow weir and ogee crest spillways (Hepler, 1992).

The capacity of a labyrinth spillway depends on the total head, the length of the crest and the discharge coefficient. The discharge coefficient is a function of the total head, the height and thickness of the spillway wall, the spillway crest contour, and the shape and position of the side walls. The flow on a labyrinth spillway is highly complex (Paxson and Savage, 2006). Compared to a conventional frontal spillway, where the streamlines are parallel and perpendicular to the crest, and the flow two-dimensional, 2D flow on a labyrinth spillway is generally restricted only to narrow zones in the middle of side legs.

At labyrinth spillway turns, the flow is three-dimensional and there are distinct mutual influences of adjacent segments. The colliding adjacent spillway streams reduce the overflow coefficient. As the angle between the segments of the crest become smaller at the base, these influences (or the overflow coefficients) increase, while the spillway capacity decreases. The flow on labyrinth spillways is often aerated only at low discharge rates. As the total head increases, the effective length of the crest decreases, and so does spillway efficiency, which can ultimately almost be equated to a corresponding conventional straight-crest spillway.

Labyrinth spillways first came into use some 30 years ago (Hepler, 1992; Vischer and Hager, 1999). Although labyrinth spillway research (Tullis et al., 1995; Tullis et al., 2007; Khode et al., 2010) has intensified since then, the amount of empirical data is still limited and empirical formulas proposed in the literature need to be applied with due caution. Each new labyrinth spillway is a unique structure (Scarella et al., 2009; Ackers et al., 2011; Canholi et al., 2011), whose performance is determined by numerous parameters. 3D flow and the presence of air render the hydraulic conditions ever more complex, such that scale models are recommended to assess the design of such structures.

This paper presents the results of model tests of the BeniSlimane Dam, conducted at the Hydraulics Laboratory of the Jaroslav Černi Institute (Belgrade, Serbia) in 2014.