Rugosidade absoluta em tubulações utilizadas para distribuição de água

Abstract

For the design of hydraulic systems used for water distribution, it is necessary to quantify the continuous loss of load along the pipes, for example, by the Darcy-Weisbach (DW) and Hazen-Williams (HW) equations. For this, one of the variables that influences this process is the roughness of the surface of these pipes, which, often, do not have updated information and exact values for a correct sizing of the systems. An alternative to this is the measurement of the roughness of the surface by means of specialized instruments that determine several parameters of amplitude of the irregularities. The aim of this study was to determine and analyze the surface roughness parameters for pipes used in water distribution. The parameters Ra, Rq, Rsk, Rku Rc, Rp, Rv and Rt, in various diameters of materials such as PVC, welded steel, galvanized steel and polytubes will be evaluated. The tests will be conducted in the laboratory using the benchtop rugosimeter and portable rugosimeter. The results obtained showed significant difference in the roughness parameters between the nominal diameters of PVC pipes, welded steel, galvanized steel and polytubes, with the exception of the parameters Rku, Rsk, Rv and Rc in galvanized steel. For PVC tubes, the parameters indicated to represent the roughness were Rp, Ra and Rp in the respective diameters of 32, 50 and 75mm. For the welded steel material, in the diameters of 70.31, 93.75, 125, 200, 250, 300 and 350mm, the suggested parameters were, respectively, Rv, Rp, Rc, Rp, Rt, Rt and Rt. Already in galvanized steel, in the same diameters as welded steel, the parameters indicated were Rq, Rq, Ra, Ra, Ra, Rq and Rq. In polytubes with diameters 100, 130, 200, 450 and 460mm, the parameters Rv, Rt, Rsk, Rt and Rt, respectively, were the most appropriate. Regarding the equipment for measuring the roughness, the portable rugosimeter presented, in PVC tubes, favorable results to its use when compared to the benchtop rugosimeter. Analyzing equations of the head loss coefficient, the formulation of Brkić and Praks (2019) presented the smallest relative errors, with values below 0.01%, and the best adjustments in the statistical indices RMSE, PMARE, NSe and LMe, in relation to the Biberg (2017) and Vatanhkak (2018) equations, in all evaluated materials. Finally, an excellent adjustment was obtained for the head loss coefficients of the Darcy-Weisbach and Hazen-Williams equations, in all materials evaluated, which can be used accurately for Reynolds number values ranging between 4000 and 500000.