Abstract:
This research work investigates the effects of depth variation on efficiency of a selected dewatering pump for Glister Success Quarry in Alaguntan, Oyo State, Nigeria. Water samples were collected from the mine for the laboratory determination of water quality. Mine water quantity (Inflow rate) was estimated using Analytical Method, and pump head requirements were determined directly from the mine. Decision Tree Selection method was used to select a centrifugal pump. Pumping using selected centrifugal pump of capacity 360 m3/hr., input shaft power of 10.7 kW at 1600 RPM, impeller diameter of 215 mm, flanges diameter of 150 mm (suction) and 150 mm (discharge) at constant Pressure of 98 kPa was carried out at six water levels named A, B, C, D, E and F. Flow rates at the discharge of the pump were measured using flow meter. An Excel Software Calculation Tool generated by Sustainable Energy Authority of Ireland was used to determine total head, friction head and hydraulic power output. Efficiency of the pump was also determined from the values obtained from Excel Software. The result of water analysis shows that the average temperature of water is 27.130C, average pH of water is 7.66, average total dissolved solids (TDS) of water is 0.279 mg/l and average total suspended solids (TSS) of water is 16.13 mg/l and the estimated mine water quantity (Inflow rate) is 13.14 m3/hr.(219 l/min.). The determined total heads from the Excel Calculation Tool are 16.00 m, 15.00, 14.00 m, 13.00 m, 12.00 m and 11.00 m. The values of the flow rates measured from the field range between 90 m3/hr. and 259 m3/hr. Hydraulic Power Outputs determined range from 3.9 kW to 7.9 kW. Calculated efficiencies range from 36.40 % to 73.00 %. The pump is found to be efficient at water levels D, E and F; where efficiency is above 50 %, hence efficiency of the pump decreases with increase in depth (static suction). The variables measured on the field and estimated variables were also correlated using Microsoft Excel Software. The results of the correlations revealed that very strong correlations exist between the plotted variables as all there coefficient of correlations are greater than 90 %.
