Impact of geometric design of hydraulic contact tanks on residence time distributions
Date
2015
Authors
Carlston, Jeremy S., author
Venayagamoorthy, S. Karan, advisor
Ramirez, Jorge A., committee member
Sakurai, Hiroshi, committee member
Journal Title
Journal ISSN
Volume Title
Abstract
The research outlined in this thesis investigates how geometric design affects the mixing efficiency of contact tanks used for drinking water disinfection. In particular the configuration of baffles and inlets are assessed in depth using both physical tracer studies and computational fluid dynamics (CFD) simulations. Two rectangular contact tanks were used, both of which are assumed to be representative of disinfection tanks for small municipalities throughout the United States. System performance is analyzed by means of residence time distributions from which the baffling factor (BF) and Morrill index (MI) performance indicators can be calculated. First, a parametric study of a 300 gallon tank used for previous research was undertaken. After model validation, 30 CFD simulations were conducted in order to determine optimal baffle configuration by altering channel width, baffle length, and flow orientation. It was found that the baffle lengths should be prescribed such that their openings are roughly equal to the width of the tank's channels. In this manner excessive constrictions or expansions, which lead to undesirable flow separation, can be prevented. It was also confirmed that high length-to-width ratios in the flow lead to a better emulation of ideal plug flow by promoting advective transport of both the water and disinfectant. Second, a 1,500 gallon, two-baffled tank housed at Colorado State University's hydraulic laboratory in the Engineering Research Center was used to investigate "sharp" inlets, which lead to poor mixing conditions. Unfortunately operating budgets prohibit many small disinfection facilities from upgrading these preexisting sharp inlets. In an attempt to provide an inexpensive solution, seven attachments to the inlet were tested for potential improvements to contact tank performance. Physical tracer studies were conducted on the prototype in order to obtain RTD curves resulting from each modification. The horizontal T-shaped attachment performed best, attaining a BF of 0.59 at a flow rate of 40 gallons per minute (gpm). This is a 74% gain over the unmodified inlet's BF of 0.34. From a hydrodynamic analysis of the flow fields obtained from CFD simulations, it was concluded that any inlet configuration leading to a quicker homogenization of longitudinal velocities (especially in the first channel of the tank) and hence better approximation of plug flow will improve system performance. Finally, seven different inlet modifications to the un-baffled counterpart of the 1,500 gallon tank were investigated. Without any alterations the un-baffled sharp inlet tank provides extremely poor disinfection as evidenced by its baffling factor of 0.05. Due to good agreement between physical and numerical tests for the baffled tank, the un-baffled system was only researched with CFD. Four equally-spaced orifices deflecting flow to the back wall of the tank resulted in the largest gain in performance with an estimated BF of 0.27. A hydrodynamic analysis again confirmed that despite lower length-to-width ratios leading to a lower ceiling of mixing performance, better imitation of plug flow results in larger baffling factors and lower Morrill indices.
Description
Rights Access
Subject
contact tanks
sharp inlets
drinking water disinfection
computational fluid dynamics