During the last years many numerical and practical investigations on sewer flushing were carried out regarding flush waves who act like dam-break waves along a sewer reach to create the needed critical bottom shear stress to lift sewer deposits and transport them to the treatment plant. An extensive summary can be found in Schaffner (2008). The usual assumption in these investigations was a dry sewer bottom downstream of the flushing gate. Unfortunately this initial condition is found very seldom in practical cases. Depending on the storage duration of the desired flushing volume the water level downstream of the flushing gate is only slowly decreasing and effects the power of the wave considerably. The intention of the present investigation was to analyze the influence of different underwater levels of the remaining dry-weather runoff on the bottom shear stresses of the flush waves and give an estimation on the potential cleaning efficiency of the wave. For practical applications like the layout of a flushing system in a new or existing sewer channel this approach gives more reliable results then the assumption of a dry sewer bottom.
Numerical model: The numerical investigations were carried out using the one-dimensional model EDWA developed by the Technical University of Darmstadt. [Steinhardt, 2008] The model is based on the Finite Volume
Method and solves the fully Saint-Venant equations by using the Godunov-Upwind scheme, the MUSCCL-Hancock method with limiter function, the HLL-Riemann solver and a time splitting procedure for the temporal discritisation based on the Runge-Kutta scheme. [León, 2007]
Investigations and Results: For the investigation a circular sewer with a diameter of 1600 mm and 1500 m flushing length was defined in the numerical model. The standard flushing volume necessary for this distance was
calculated as well as the storage time for a given dry-weather runoff. The first modeling was carried out to calculated the water levels of the dry-weather runoff after the storage time under a no inflow condition due to the closed flushing gate. These values were then used as initial conditions for the modeling of the flush wave movement along the sewer reach. This approach was then carried out for 7 different flushing volumes and underwater levels to compare the created bottom shear stresses of the different flush waves. In a further inverstigation the results of the flush waves under the influence of
downstream water levels were compared to flush waves on a dry sewer bottom using the same sewer geometry to show the loss of efficiency due to the dry-weather runoff underwater levels.