Abstract

Increasing numbers of lines of evidence indicate that entrapment of large pockets of air during the rapid filling of stormwater and combined sewer systems may result in detrimental system behaviour. Wright et al. (2007) and Vasconcelos and Wright (2011) discuss the results of laboratory experiments in which an air pocket arriving at a vertical riser forces water standing in the riser to be propelled upwards. Lewis et al. (2011) and Wright et al. (2011) present the results of field measurements in a stormwater tunnel and conclude that the release of entrapped air must be responsible for the formation of observed geysers through a large diameter manhole. Vasconcelos and Wright (2006) discuss a number of ways by which rapidly filling pipelines can trap large air pockets. Vasconcelos and Wright (2007; 2009) demonstrate through laboratory experiments and numerical modeling that compression of the air within a trapped air pocket may alter the dynamics of the water phase flow. A related question associated with these observations of air pocket interactions is whether the presence of trapped air can lead to significant pressure transients. Martin (1976) demonstrated through numerical modeling that the compression of trapped air pockets during the filling of a pipeline closed on one end could result in large transient pressures. Although the assumed flow configuration is not particularly relevant to large sewer systems, it is possible to envision filling scenarios that result in air compression, so the possibility of significant transient pressures should be considered.

A version of the numerical model developed by Vasconcelos et al. (2006a; 2006b) to simulate the filling of nearly horizontal conduits has been applied to the design analysis of proposed combined sewer overflow storage tunnels (Lautenbach et al., 2008). In a number of simulations, the model predicts that a large volume of air will be trapped during the filling process. There are several mechanisms by which this can occur, all of which appear to be associated with the development and propagation of a hydraulic bore. Figure 1.1 indicates one of these mechanisms in which the air pocket is predicted to form following reflection of a bore off the upstream end of the filling tunnel. Since the model only simulates the water phase, the air pocket is treated as a void in the simulation and it subsequently vanishes as the filling process proceeds. The collapse of this void produces a transient like a water hammer in the simulation, which is not actually expected to occur since there would be air present to cushion the collision of the bore with the water filling from the opposite end of the tunnel. However, the mechanism described by Martin (1976) suggests that large system pressures could develop during the air compression process. Although Vasconcelos and Wright (2009) have modified the basic numerical model formulation to account for the presence of air above the water surface, this modification only applies for a continuous layer of air that extends to a ventilation location, and cannot be easily adapted to simulate trapped air pockets. This chapter presents the results of a preliminary laboratory investigation to measure pressure transients associated with a hydraulic bore impinging on a trapped air pocket. These results suggest that this situation may be a cause for concern in the design of large diameter storage tunnel systems.

This paper is only available in PDF Format:

  View full text PDF