Available online at www.sciencedirect.com ScienceDirect Energy Procedia 85 (2016) 266 272 Sustainable Solutions for Energy and Environment, EENVIRO - YRC 2015, 18-20 November 2015, Bucharest, Romania An Experimental Approach Regarding the Sewage Self-Cleansing Conditions Marius Iliescu a, Mihnea Sandu a*, Ilinca Nastase a, Elena Iatan a, Florin Bode a a Technical University of Civil Engineering in Bucharest, Building Services Department, 66 Avenue Pache Protopopescu, 020396, Bucharest, Romania Abstract The paper focus on the self-cleansing conditions which have to be satisfied in order to ensure the good exploitation of a sewage system. The data used in the literature refers mainly to open channel with coehsionless (sand) deposits on the bottom. The study is very useful because of the lack of data regarding the domestic sewage. An experimental setup has been conceived in order to measure velocity profiles and shear stress velocities which occur in free surface flow in a circular sewage pipe, with deposit material at the bottom. The experiments have been conducted using a PIV measurement system and the results have been compared with others works. An experimental procedure on how to measure the parameters needed to characterize the selfcleansing conditions for a sewage pipe has been defined. The paper defines also the premises for the further studies to be conducted in order to achieve final results regarding new self-cleansing conditions in a sewage pipe. 2015 2016 Published The Authors. by Elsevier Published Ltd. This by Elsevier is an open Ltd. access article under the CC BY-NC-ND license Peer-review (http://creativecommons.org/licenses/by-nc-nd/4.0/). under responsibility of the organizing committee EENVIRO 2015. Peer-review under responsibility of the organizing committee EENVIRO 2015 Keywords: Sewage self-cleansing; PIV measure ; surface free flow 1. Introduction One of the most important parameters involved in wastewater design is the minimum discharge that will maintain the pipe self-cleansing. There are several recommendations found in the literature depending of the author/place/country. According to Vicari [1] there are two conditions that must be satisfied for the minimum discharge in sewers: minimum flow depth hm = 3 cm and minimum tractive force Sm=2.5 N/m2. Ackers et al. [2] conducted an extensive study about the aging of the sewers and concluded that the sewer coating development varies strongly with the material content of the sewage. In their study they have recommended equivalent roughness to ks = 1.5 mm for sewer coatings smaller than 5 mm thick in normal condition, for good condition ks = 0,5 mm, and for bad 1876-6102 2016 Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). Peer-review under responsibility of the organizing committee EENVIRO 2015 doi:10.1016/j.egypro.2015.12.244
Marius Iliescu et al. / Energy Procedia 85 (2016) 266 272 267 condition ks = 3 mm. These values become as high as about 25 mm for crusted conditions and even about ten times higher for sewers with gravel deposits. Smith [3] concluded that a sewer should maintain self-cleansing for the minimum discharge and he has provided a design procedure based on a minimum velocity value. Yao [4] has shown that the minimum sewer velocity that maintains the self-cleansing depends on the characteristics of the flow boundary surface, the deposited material and the depth of flow. He has replaced the minimum velocity concept by the minimum wall shear stress concept. The minimum shear stress relates to the initiation of motion and it is estimated from the Shields diagram (Graf [5], Raudkivi [6]. Yao have proposed for separate sewer systems for particles of diameters from 0,2 mm to 1 mm a minimum value for bed shear stress τom= 1 to 2 N/m2 and for combined system sewers minimum values between 3 and 4 N/m2. The mean shear stress τ0 for uniform flow in open channels is τ0 =ρgrhs0 [N/m²] (1) Using the formula of Manning and Strickler equations that expressed a relationship between the bottom shear stress τ0, the sewer diameter D and the velocity v corresponding to different part full conditions of the sewer can be established (Hager [7]). Correlations between the pipe diameter, the slope and the minimum velocity for different part full ratios of the channel have been established. They concluded that velocities values from 0.5 ms-1 for the smallest possible diameters to 0.80 ms-1 for the largest size diameters may be adopted. Researches on the minimum slope required to avoid deposition of sand and gravel in sewers have been conducted by Novak and Nalluri [8], Mayerle et al. [9], Butler et al. [10], Nalluri and Ab Ghani[11]. The German standard ATV [12] is based mainly on the works of Macke [13][14] and Sander [15] and provides a table for various sewer diameters at 50% part full flow containing the minimum velocities vm and the corresponding minimum bed slopes S0m to avoid deposition if material. According to this standard the relation between the diameter D and the minimum velocity vm can be expressed by: vm=0.5+0.55d [m/s] (2) For 10% part-full flow, the minimum velocities obtained from eq. 2 should be increased by an additional amount of about 10%. Another approach was given by Sander [15] who observed an average particle size of d= 0.35 mm and concluded that the minimum filling ratio should be 10% and the minimum bottom shear stress should be 0.9 Nm-2. According to Romanian rules a minimum velocity of 0.7 ms-1 have to be ensured in order to maintain sewage self-cleansing, no matter the pipe diameter, the pipe material or the waste water provenience. In order to better understand the mechanism of the sediment motion Thus, we can conclude the following: The main parameters used for defining the self-cleansing conditions are the flow velocity and the shear stress; Most of the conclusions regarding the self-cleansing conditions are based on experimental research; Most of the studies are focused on open flow channels with sand or other cohesionless particles and not on domestic sewage systems.
268 Marius Iliescu et al. / Energy Procedia 85 (2016) 266 272 Nomenclature h m S m k s τ om τ o ρ ρ s g R h S 0 S 0m D v v m d minimum flow depth corresponding to minimum discharge minimum tractive force absolute equivalent roughness minimum value for bed shear stress mean shear stress for uniform flow in open channels water density particle (deposit) density gravity acceleration hydraulic radius bed slope of the pipe minimum bed slope of the pipe to avoid deposition material pipe diameter flow velocity minimum flow velocity to avoid deposition material particle size 2. Some theoretical aspects In order to better understand the mechanism of the sediment motion one may consider a plane stationary bed consisting of loose cohesionless sand particles of uniform size and a layer of water over it. As the liquid start flowing, hydrodynamic forces are exerted upon the solid particles of the bed. For a particular stationary bed, when the particles in the movable bed are unable to resist to those forces, they are dislocated and afterwards they eventually start to move. The condition of the initial movement of the bed is determined by observation and it is called critical stage or critical condition or initial scour. This critical stage can be explained in several ways: With critical velocity equations, considering the impact of the liquid on the solid particles; in this case force equilibrium on the moving particle is considered and several formulas for the bottom critical flow velocity or for the average critical flow velocity can be deducted Luca & Tatu [16]. With critical shear stress equations, considering the frictional drag of the flow on the particles; in this case there is the Shields diagram and the parameter c s gd is often referred to as Shields parameter. The shear stress is strongly related to the shear stress velocity which may be classically estimated for example by the logarithmic method (Katul et al. [17]) or by Reynolds stress method and not only. With the lift force criteria, considering the pressure differences due to the gradient of the velocity. Given these entire one may conclude that by measuring the velocity distributions, the shear stress on the bottom of the channel, or the friction velocity and by correlating these measured values with the beginning of the deposit or sediment motion new critical conditions regarding the sewage self-cleansing might be provided. 3. Experimental Setup In this study it has been considered a system which generates a free surface flow in a circular sewage pipe, with deposit material at the bottom. The interior pipe diameter is Di=144 mm, pipe length is 3,9 m made of two Plexiglas pieces. The coupling between the two pieces is very smooth in order not to introduce additional local losses. The experimental setup is presented in figure 1. The water is pumped from the tank (1) towards the tank (5). This one is a free surface tank and from here we obtain a free surface flow through the main Plexiglas pipe from the tank (5) towards the tank (1). The flow rate is determined with an ultrasonic flow meter (6) and it may be varied using a regulation valve.
Marius Iliescu et al. / Energy Procedia 85 (2016) 266 272 269 Fig. 1. Experimental Setup An artificial roughness has been created in order to simulate the deposits on the bottom of the pipe Calibrated spheres (4,5 mm diameter) have been introduced in perforated Plexiglas plates. The result was a plate with hemispheres created to simulate the consolidated deposits on the bottom of the pipe. This method chosen for creating the artificial roughness has been considered important because of the comparison with other experimental data [ ]. The plate with hemispheres is presented in figure 2 (a) and its characteristics are shown in figure 2 (b). a) b) Fig 2. (a) Artificial Roughness, (b) Hemispheres The main goal was to measure velocity distributions field above the area with artificial roughness. The system used was PIV system having the following characteristics: one CCD camera type Nanosense MKII, provided with a Nikon 50mm focal lenght and 1.2 objective aperture. The frequency acquisition for this camera was maximum 5000 Hz but the flow images have been taken at 300, 500 and 1000Hz. The number of captures varies between 1000 and 3000 frames/measure. The laser used for particle lightning was an infrared Nanopower one, with795 nm wavelength and 4Wpower. Fig. 3. PIV System
270 Marius Iliescu et al. / Energy Procedia 85 (2016) 266 272 The tracer particles introduced in the fluid flow were glass spheres covered with silver with diameters equal to 10 μm. These particles have been injected at the bottom of the pipe as shown in figure 4. Fig. 4. (a), (b) particle injection principle The measurements were performed for three flow rate values at 0.7 l/s, 0.9 l/s and 1.1 l/s. On the fig 4 (a) it can be also observed the two configurations studied, the first one considering the laser plane in the middle of the pipe between two rows of hemispheres and the second one considering the laser plane on the first row of hemispheres adjacent to the middle plan. 4. Results and discussions Given the complexity of the fluid flow pattern over the hemispheres, the observed domain had to be split in two subdomains in order to complete the particles images treatment. These two subdomains are characterized by different values for the image correlation parameters. This is the reason for the existence of two areas as shown below in figure 5. Fig. 5. Subdomains of the studied region Velocity distributions for different case studies have been obtained, for the two case studies configurations, the first one with the laser plane in the middle of the pipe between two rows of hemispheres and the second one with the laser plane on the first row of hemispheres adjacent to the middle plan. The data have been compared with those measured by Agelinchaab and Tachie [18] for a fluid flow in an open channel over an artificial roughness very closed to the hemispheres shape used here. In this paper we are presenting the results only for the second configuration with the laser plan over the first row of hemispheres. The velocity vectors which describe the fluid flow are shown in the figure 6 for the area over the hemispheres and also for the area between the hemispheres. One can observe the vortex created in the small area between the hemispheres.
Marius Iliescu et al. / Energy Procedia 85 (2016) 266 272 271 Figure 6. (a) area over the hemispheres, (b) area between the hemispheres Velocity profiles are provided for the three case studies flow rates, respectively 0,7 l/s, 0,9 l/s and 1,1 l/s on over the top of two consecutive hemispheres. Regarding the figure 7 below where the velocity profiles for 0,7 l/s and 0,9 l/s flowrates values are shown, one can observe that there is not a big difference between the velocity profile corresponding to the first hemisphere and the velocity profile corresponding to the second hemisphere. They match pretty well and the conclusion is that it is not very important the choice of the hemisphere on which the velocity profile is traced. Fig. 7. Velocity profile over one hemisphere (a) 0,7 l/s flow rate, (b) 0.9 l/s flow rate The shear stress velocity presented in figure 8 is calculated for the velocity profiles measured on a hemisphere using the Reynolds stress method for a 0.7 l/s flow rate. Fig. 8. Shear stress velocity
272 Marius Iliescu et al. / Energy Procedia 85 (2016) 266 272 5. Conclusions and further studies An experimental setup has been conceived in order to measure velocity profiles and shear stress velocities which occur in free surface flow in a circular sewage pipe, with deposit material at the bottom. The experiments have been conducted using a PIV measurement system and the results have been compared with others works. The main goal of this study was to define an experimental procedure on how to measure the parameters needed to characterize the self-cleansing conditions for a sewage pipe. This might be useful to know the flow patterns which allow the appropriate design for the sewage pipes. Furthers studies have to be conducted from now one in order to identify the self-cleansing conditions for particular sewage pipes or for particular points of interest in a sewage system as the pipe joints, the elbows or the intersections. Acknowledgements This work was supported by project of the Romanian National Authority for Scientific Research, Sectoral Operational Programme "Increase of Economic Competitiveness" - financed by the European Regional Development Fund, Priority 2 Competitiveness Research -Development and Innovation, D2.2: Investments in RDI infrastructure and related administrative capacity O2.2.1: Development of existing CD and creation of new infrastructure (laboratories, research centers) ID 1883, SMIS-NSRF code 49 161". References [1] Vicari M, Minimum bottom slopes for sewers. Gesundheits-Ingenieur (1916);39(51):537-540 [in German]. [2] Acckers P,Crickmore MJ, Holmes DW. Effects of use on the hydraulic resistance of drainage conduits Proc. Institution Civil Engineers(1964) 28: 339-360; 34:219-230. [3] Smith A A, Optimum design of sewers, Civil Engineering and Public Works Review, (1965) 60(2): 206 208: 60 (3): 350-353; 60 (9): 1279-1283. [4] Yao K M, Sewer line design based on critical shear stress, Journal Environmental Engineering Division ASCE (1974) 100(EE2): 507-520; 101 (EE1): 179 181; 101(EE4): 668 669. [5] Graf W H, Hydraulics of sediment transport (1971), McGraw-Hill: New York. [6] Raudkivi A J, Sedimentation Exclusion and removal of sediment from diverted water, IAHR Hydraulic Structures design Manual 6 (1993), Balkema, Rotterdam. [7] Hager W H, Wastewater Hydraulics Theory and Practice, Second edition (2010) Springer Verlag, Berlin Heidelberg. [8] Novak P, Nalluri C, Sewer design for no sediment deposition, Proc. Institution Civil Engineers (1978) 65 (2): 669-674; 67(2): 251-252. [9] Mayerle R, Nalluri C, Novak P, Sediment transport in rigid bed conveyances, Journal Hydraulic Research (1991) 29(4): 475-495. [10] Butler D, May R W P, Ackers J C, Sediment transport in sewers, Proc Institution Civil Engineers Water, Maritime & Emergy (1996) 118 (6):103 120. [11] Nalluri C, Ab Ghani A, Design options for self-cleansing storm sewers, Water Science and Technology (1996) 33 (9): 215 220. [12] ATV, Guidelines for the hydraulic design of sewers, Regelwerk Abwasser Abfall, Arbeitsblatt A110, (1998) Abwassertechnishe Vereinigung: St Augustin [in German]. [13] Macke E, On sediment transport for low concentrations in partially filled pipes (1980), Mitteilung 69, Leichtweiss Institut fur Wasserbau, TU Braunschweig [in German]. [14] Macke E, Design of depositionless flows in sewers, (1983), Korrespondenz Abwasser 30 (7): 462 469 [in German]. [15] Sander T, The design of depositionless sewers under particular attention of new results relative to sedimentation, (1994), Korrespondenz Abwasser 32 (5): 415 419 [in German]. [16] Luca O Tatu G, Environmental Impact of free surface flows evaluation and protection, (2002), Orizonturi Universitare, Timisoara. [17] Katul G,Wiberg P, Albertson J, Hornberger G., A mixed layer theory for flow resistance in shallow streams - Water Resources Research 38 (2002) [18] M. Agelinchaab, M.F. Tachie, Open channel turbulent flow over hemispherical ribs - International Journal of Heat and Fluid Flow 27 (2006) 1010 1027