Amin Torabi and Kortney Brown Featured in USSD Dam & Levees
Senior Staff Engineer, Amin (Mohammadamin) Torabi and Associate, Kortney Brown were recently published in United States Society on Dams (USSD) Dam and Levees Fall 2021 Issue.
The article titled, “Revisiting Historical Weir Data: Comparing Numerical Analysis and Experimental Data of Different Weirs’ Discharge Coefficient”, discusses how different variables can affect discharge rates. The authors look at historical weirs and numerically refine the discharge coefficient.
Weirs are a common hydraulic structure used to control and/or measure flow with the common types defined as the broad-crested, sharp-crested, circular-crested, and the ogee-crested weirs (Chanson and Montes, 1998). For flow measurement, it is important to understand how different variables such as weir type, shape, aerated/nonaerated, surface tension and approach flow conditions including the velocity head can affect the discharge rate. Generally, the influence of flow variables is included in a “discharge coefficient (Cd)” that is used to calculate the flow rate and is often a function of the headwater height (Ho) to weir height ratio (W). Using the correct discharge coefficient, Cd, directly affects the accuracy of the flow measurement. However, while weirs have been widely studied using physical models (Bazin, 1888), measurement error, model construction, and other errors can affect the calculation of the discharge coefficient. In a recent study, Tullis et. al (2019), compared the results of two common-geometry linear weirs (quarter- and half-round) from 20 different hydraulics laboratories from around the world, only specifying the weirs’ geometric shape (thickness, height and crest shape). The study found significant variability in the discharge and discharge coefficient as a function of dimensionless upstream head. For a 90% confidence bounds, there was an approximate 20% range across the lab data for low flows (Ho/W < 0.4) and about a 15% difference for larger flows (Ho/W > 0.4). Others have documented the potential for physical modeling errors (Ettema, 2000 and Heller, 2011).