Performance assessment of vortex settling chambers

• Published in: ISH journal of hydraulic engineering Vol. 20; no. 3; pp. 324 - 338
• Main Authors: Ansari, Mujib Ahmad, Khan, Mohd Amir
• Format: Journal Article
• Language: English
• Published: Taylor & Francis 02-09-2014
• Subjects: geometrical configuration; sediment removal efficiency; vortex chamber; water abstraction ratio
• ISSN: 0971-5010; 2164-3040
• DOI: 10.1080/09715010.2014.925330

In present investigation an attempt has been made to study the variation in sediment removal efficiency of vortex settling chamber with the locations of the inlet and outlet channels and to develop a new model for predicting the removal efficiency of the vortex settling chamber. Existing equations for removal efficiency are checked for their accuracy using experimental data collected in present study along with data available in literature. The computed efficiency by existing equations was found to be inconsistent with the corresponding observed ones. Therefore, a new equation is proposed in this study. Experiments were conducted on two geometrical models of the vortex chamber. In the first type of the extractor model, both the inlet and outlet overflow channels were kept in an alignment following a straight line tangential to the vortex chamber. In the second type of extractor model, the straight inlet channel joined the vortex chamber tangentially at its one side. The straight outlet channel was taken off tangentially from the upstream end of the chamber at 90° from the inlet channel. It was observed that the efficiency of the geometrical Model-II has been obtained higher because in this model the sediment particle may travel long helicoidal path and thus, resulting in higher settling length, smaller turbulence, and large residence time in comparison to geometrical Model-I. The proposed equation is found to produce results with a maximum error of ±35% for about 100% of the total data. The qualitative performance of the present predictor indicated that it has lowest MAPE (25.03), RMSE (0.150), and highest R
2
(0.753) as compared to other existing predictors.