Available online at www.sciencedirect.

com

ScienceDirect
Procedia Engineering 186 (2017) 69 – 75

XVIII International Conference on Water Distribution Systems Analysis, WDSA2016

Criteria of minimum shear stress vs. minimum velocity for self-

cleaning sewer pipes design

Carlos Montesa, Jessica Bohorqueza, Sergio Bordaa, Juan Saldarriagaa*

a
Centro de Investigaciones en Acueductos y Alcantarillados (CIACUA), Departamento de Ingeniería Civil y Ambiental, Universidad de los Andes
Bogotá 110111, Colombia.

Abstract

Sedimentation in Urban Drainage Systems is a huge problem that allow changes in cross-sectional area, and therefore changes in
pipe´s roughness. The increase of suspended solids deposition and consolidation in pipes affect the velocity and shear stress
distribution inside these elements [1]. Some authors have proposed methodologies to prevent sedimentation phenomena in sewer
pipes. However, such methodologies requires variables estimation that result very difficult in most cases and their estimation could
affect the resulting network pipes design. This paper aims to evaluate different methodologies and equations proposed to ensure
self-cleaning in sewer and pluvial pipes systems design. Comparison was made by graphical analysis of values established in
literature in order to assess restrictions of each design. Three methodologies were compared: Traditional criteria, Equation of
Macke (1982), and Tractive Force Methodology (ASCE). A design methodology was proposed as an optimization problem
considering a cost variability analysis using cost functions. Results shows that self-cleaning criteria during design process must be
included in cases where the relation between pipes’ flow and terrain slope exceeds the relation found by this study.

© 2016
2016TheTheAuthors. Published
Authors. by Elsevier
Published Ltd. This
by Elsevier Ltd. 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 of the XVIII International Conference on Water Distribution
Peer-review
Systems. under responsibility of the organizing committee of the XVIII International Conference on Water Distribution Systems

Keywords: self-cleaning; minimun shear stress; sewer design.

* Corresponding author .Tel.: +571-339-4949 Ext. 2805.

E-mail address: jsaldarr@uniandes.edu.co

1877-7058 © 2016 The Authors. 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 of the XVIII International Conference on Water Distribution Systems
doi:10.1016/j.proeng.2017.03.210
70 Carlos Montes et al. / Procedia Engineering 186 (2017) 69 – 75

1. Introduction

Particle sedimentation is a huge problem that affects the majority of drainage systems. Changes in cross sectional
area and in material roughness are caused by sedimentation, especially in dry periods. Roughness variations influence
the velocity distribution and the stress shear distribution. Bong [2] did a classification in three groups of criteria and
methodologies that seeks to prevent the sedimentation phenomena, as shown in figure 1.

No deposit

No deposit sediment

Limit deposit

Self-cleaning design Incipient Movement

drainage systems Movement of existing
sediment on the pipe
bed
Sediment Transport

Energy Slope

Fig. 1. Self-cleaning criteria classification.

1.1. No deposit sediment

The majority of world´s technical design norms are compound by traditional design criteria. No one of these criteria
has a formal justified background; they actually rely on empirical knowledge of designers. Table 1 shows some of
these values of minimum velocity and minimum shear stress used for wastewater, combined and stormwater drainage
systems design.

Table 1. Values of minimum velocity and sheer strees

Type of
Shear Stress
Entity Country Norm Year Drainge V min
min
System
[-] [-] [-] [-] [-] [m/s] [Pa]
Guía para el Diseño Sanitary 0.45 1.5
EPM Colombia Hidráulico de Redes de 2009 Storm &
0.75 3
Alcantarillado Combined
Sanitary 0.45 1.5
MinDesarrollo Colombia RAS - 2000 2000 Storm &
0.75 3
Combined
Sanitary - 1
IBNORCA Bolivia NB 688 2007 Storm &
- 1.5
Combined
Sanitary 0.45 -
INEN Ecuador CPE INEN 5 1992
Storm 0.9 -
Manual de Agua Potable, Sanitary 0.6 -
CNA Mexico Alcantarillado y 2007
Storm 0.3 -
Saneamietno
Normativa Alcantarillado
INAA Nicaragua - Sanitary 0.6 1
Sanitario Condominal
Recommended Standars for
Great Lakes USA 2004 Sanitary 0.6 -
Wastewater Facilities
 Carlos Montes et al. / Procedia Engineering 186 (2017) 69 – 75 71

1.1.1 No deposit

Sedimentation phenomena is not allowed in drainage systems. In order to use this model, it is necessary to identify
the particle transportation model (suspension load or bed load), sediment size, density and concentration of particles.
An equation is required to predict the sediment movement and additional data or information about minimum shear
stress that erode bed particles; also is necessary to know the equation that determines the system´s fluid resistance [2].
Studies as those done by Robinson-Graf (1972) [3], Novak and Nalluri (1975) [4], May (1982) [5], Macke (1982) [6]
and May et al. (1996) [7], tried to represent the self-cleaning inside a pipe as a function of such parameters as
concentration of sediment ௩ , pipe’s diameter d, sediment’s diameter ௦ , Darcy’s friction factor , and settling
velocity ௦ . Those authors have proposed equations like Macke’s (1982) (See Equation 1). In this equation self-
cleaning velocity ௅ must be calculated, and then compared with water velocity inside the pipe. If ௅ is higher than
the water velocity, it could be guaranteed an appropriated particle transportation along drainage system life.

ଷ ହ
௅
௩ ଵǤହ
(1)
ௌ

1.1.2 Limit deposit

This is a less conservative criterion because allows tiny accumulation of sediment at pipe’s bed. The main research
has been done by Alvarez (1990) [8], El - Zaemy (1991) [9], Ghani (1993) [10] and May et al. (1996) [7]. The
equations are obtained by multiple regression analysis, as May´s (1996) [7] shown in Equation 2.

଴Ǥଷ଺ ଶ ଴Ǥ଺ ସ ଶ ଵǤହ

௦ ௧ ௅
௩ (2)
௅

1.2. Movement of existing sediment on the pipe bed

This criterion born under the assumption of previous particles sedimentation on pipe’s bed. The equation developed
take into account sediment and pipe features. For incipient movement Novak and Nalluri (1975) [4] found
mathematical expressions at laboratory. Equation 3 is one of those.

ି଴Ǥଶ଻
௅ ଵȀଶ ௦
(3)
௦

Unfortunately, there are few studies about sediment transportation to generate a veritable knowledge in matter.
Authors as Laursen (1956) [11], and Robinson-Graf (1972) [3] developed equations as Equation 4, where a and b
are regression coefficients.
௅ ௕
௩ (4)

1.3. Energy Slope

Energy slope is a design alternative to traditional equation. Design curves have been established to determine the
minimum system slope through Tractive Force method suggested by The American Society of Civil Engineers
(ASCE). The design curves, link minimum design flow, slope and pipe diameter, in order to obtain the sand particle
diameter. Fig. 2. presents one of these design curves.
72 Carlos Montes et al. / Procedia Engineering 186 (2017) 69 – 75

Fig. 2. Self-cleaning Slope vs minimum flow. Taken from Bizier (2007) [12].

2. Graphic Analysis of Design Restrictions

Three methodologies of design were graphically analyzed: traditional criteria ( ௠À௡ and ௠À௡ ), Macke´s equation,
and Tractive Force. It was supposed a study case of a concrete pipe (n=0.013), which receives sand particles ( ௦ = 1
mm). The objective was to get curves that show the relation between the minimum pipe´s slope to guarantee self-
cleaning and the pipe´s diameter associated. Thereby, the minimum shear stress ( ௠À௡ 2 Pa) results in the more
restrictive parameter. Results are shown in Fig. 3.

Fig. 3. Restrictive Analysis.

 Carlos Montes et al. / Procedia Engineering 186 (2017) 69 – 75 73

3. Study Area and Design Methodology

The study network selected is located at north of Bogotá, Colombia. It is part of the storm drainage system of
Bogotá and includes 37 nodes, 37 pipes and 1 discharge. Fig. 4. illustrates the network.

Fig. 4. Topology of storm drainage network.

The design methodology requires a list of commercial available diameters, topography of each manhole, design
flows, topology and material features. Then, feasible pipes are generated – those that fulfill restrictions -, and Bellman
– Ford optimization algorithm is applied to obtain the lower cost design. Equation 5 and 6 shows the expressions used
to get the cost of pipes and excavation cost. In the Equations, d is the pipe diameter, V the excavation volume, and k a
conversion factor to obtain results in Colombian Pesos COP.

଴Ǥହ଻ଷ଻
௧ (5)
ଵǤଷଵ (6)
௘

4. Sensitivity Analysis of Design Costs

For rainwater drainage systems, nine scenarios were generated to evaluate the sensitivity, varying inflows and
terrain topography. Self-cleaning criteria for storm drainage is used with shear stress values of 2, 4 and 12.6 Pascal;
and minimum velocity values of 0.6, 0.75 and 0.9 meter per second. Applying the methodology described above, total
costs could be obtained for different variation of principal route of the network under each self-cleaning restriction.
As seen in Fig. 5., total cost is indifferent of self-cleaning restriction for each case evaluated.

Fig. 5. Costs for scenarios of storm drainage network.

74 Carlos Montes et al. / Procedia Engineering 186 (2017) 69 – 75

Fig. 6. shows the optimal design, as a function of diameter and minimum self-cleaning pipe slope. As can be seen
in Fig. 6., all designs, inside shadowed area, accomplish self-cleaning restrictions. Velocities and shear stresses are
higher than minimum values in technical norms.

Fig. 6.
Results Storm Draionage.

For wastewater drainage system, five scenarios were taken into account, varying topography up to 50 percent, and
fixing a critical inflow of 1 litter per second at each node of the system. In this kind of systems the self-cleaning
restrictions in minimum velocity are 0.3, 0.45 and 0.6m/s; and in shear stress are 1, 1.5 and 2 Pa. Fig. 7 shows the
results for each scenario. Designs costs for the first scenario (with terrain slope > 1.8 percent), are equal for each self-
cleaning restriction, and in second and third scenarios there is a little difference. However, fourth and fifth scenarios
present significant variation between results for different restrictions. While terrain slope decreases the total cost –
pipes and excavation- difference is higher for very restrictive criterion as v =0.6 m/s, than for less restrictive criterion,
v = 0.3 m/s. Cost differences also increases in terrain with low slopes. For high flows, the results do not show much
variation, in contrast with scenarios of small flows and low terrain slopes.

Fig. 7. Costs for scenarios of sanitary drainage network.

 Carlos Montes et al. / Procedia Engineering 186 (2017) 69 – 75 75

Different scenarios were simulated to determine the limit of a combination between flow and slope, where self-
cleaning restrictions were negligible, thus thirty different combinations of flow and slope were simulated. Potential
regression analysis let stablish a limit threshold where self-cleaning criteria ceases are a sensible variable during design
process. Fig. 8., presents the regression´s results. On the x –axes is the terrain slope and in y – axes the flow difference
between the first and last node of the main route of drainage network. If designed network is above the regression
curve, self-cleaning criteria is depreciable. Indeed, designs with a velocity of 0.6 m/s, shear stress of 1.5 Pa, or using
the methodologies of Macke or Tractive Force is indifferent.

Fig. 8. Self-cleaning threshold limit in drainage systems.

5. Conclusions

Recently, different methodologies have been developed looking for equations, obtained from multiple regression analysis, that represent the
drainage self-cleaning function of some variables that could affect the design process. Those variables, generally, are difficult to estimate
which involves making assumptions that can affect the final designs. The aforementioned makes able to think about the benefit of using
traditional methodologies, which have demonstrated well self-cleaning capacity. Simulations developed showed potential relation between
terrain slope and flow difference among first and last node. In this relation, the self-cleaning threshold limit found let establish when designs
are affected by self-cleaning restrictions or not, and when to consider them.

References

[1] Ebtehaj, I., Bonakdari, H., & Sharifi, A. (2013). Design criteria for sediment transport in sewers based on self-cleansing concept. Journal of
Zhejiang University-SCIENCE A (Applied Physics & Engineering), 914-924.
[2] Vongvisessomjai, N., Tingsanchali, T., & Babel, M. S. (2010). Non-deposition design criteria for sewers with part-full flow. Urban Water
Journal, 7, 61-77.
[3] Robinson, M., & Graf, W. (1972). Pipelines of low-concentration sand-water mixtures. Journal of Hydraulics Division, 98, 1221-1241.
[4] Novak, P., & Nalluri, C. (1975). Sediment Transport in Smooth Fixed Bed Channels. Journal of the Hydraulic Division of the American Society
of Civil Engineers, 1139-1154.
[5] May, R. (1982). Sediment transport in sewers. Hydraulic Research Station.
[6] Macke, E. (1982). About sedimentation at low concentrations in partly filled pipes. Braunschweig: Ph.D Thesis.
[7] May, R., Ackers, J., Butler, D., & Jhon, S. (1996). Development of design methodology for self-cleansing sewers. Water Science and Technology,
33(9), 195-205.
[8] Alvarez, E. (1990). The influence of cohesion on sediment movement in channels of circular cross-section.
[9] El - Zaemy, A. (1991). Sediment transport over deposited beds in sewers. Ph.D Thesis.
[10] Ghani, A. A. (1993). Sediment Transport in Sewers. PhD Thesis. University of Newcastle Upon Tyne.
[11] Laursen, E. (1956). The hydraulic of a storm-drain system for sediment-transporting flow. Iowa Highway Research Board.
[12] Bizier, P. (2007). Gravity Sanitary Sewer Design and Construction.