Available online at www.sciencedirect.

com

ScienceDirect
Energy Procedia 00 (2017) 000–000
www.elsevier.com/locate/procedia

Sustainable Solutions for Energy and Environment, EENVIRO 2016, 26-28 October 2016,
Bucharest, Romania

Study for selecting the solution of a dam’s discharge chute canal

Felipe Cisnerosa, Erwin Barahonaa, Cornel Ilincab, Cristian Anghelb

a
Universidad de Cuenca, Ecuador
b
Technical University of Civil Engineering, Bucharest, Romania

Abstract

Some hydraulic studies to find an adequate solution for a discharge chute canal are presented in this paper. Tha chute canal is
appropriate work of the Lechugal 2 earth dam (H=20 m), which is part of a large irrigation system from Ecuador, presently under
construction.
Studies have consisted in experiments on hydraulic model of the chute canal built at geometrical scale 1:35 and by numerical
simulations in one-dimensional steady flow running HecRas software. In first stage numerical model was calibrated based on
experiment results and in second stage other three constructive variants of the chute canal were analysed numerically in order to
improve the solution settled by hydraulic model.
The technical performances of each variant settled by these experimental and numerical studies follow to be completed with
economical analyses which will allow designer to choose the best solution for discharge chute canal.

© 2017 The Authors. Published by Elsevier Ltd.

Peer-review under responsibility of the organizing committee of the international conference on Sustainable Solutions for Energy
and Environment 2016.

Keywords: dam, discharge canal, hydraulic model, numerical simulation

1876-6102 © 2017 The Authors. Published by Elsevier Ltd.

Peer-review under responsibility of the organizing committee of the international conference on Sustainable Solutions for Energy and
Environment 2016.
 2 Felipe Cisnerosa, Erwin Barahonaa, Cornel Ilincab, Cristian Anghelb/ Energy Procedia 00 (2017) 000–000
1. Introduction
Surface spillways and bottom outlets are appurtenant works having essential function in dam’s safety.
When these works are not functional it is possible to provoke uncontrolled water discharge from reservoir with risk
for dam’s breaking generating catastrophic effects for environment.
Presently in Ecuador, there is under Definitly Studies a large irrigation system (Proyecto Pacalori) which
will ensure irrigation of a potential surface of 112900 ha in De Los Rios province. Lechugal 2 hydraulic
development consisting of an earth dam with 20 m maximum height, chute canal, bottom outlet and a reservoir
storing 70 million m3 at normal retention level (NRL) is part of this irrigation system [1].
This paper presents studies which preceded the selection of the design solution for dam’s surface spillway
of chute canal type. Studies were particularly extensive consisting in experiments on hydraulic model of the chute
canal at scale 1:35 and some numerical simulations of the flow through canal designed in different constructive
variants [2], [3], [4], [5].
Numerical simulations are performed accepting the hypothesis of the one-dimensional steady flow running
HecRas software.
The objectives of these studies were the followings:
- to ensure the transit through canal of the design flow rate (326 m3/s) taking into account the safety of the
dam and chute canal safety;
- to limit the velocity of the flow through canal and to reduce cavitation phenomena in order to protect
concrete faces of walls and apron chute canal against erosion;
- to have a good dissipation of the flow energy in different transit conditions through type II USBR
dissipation basin;
- to protect against local erosions the banks and the bed of the connection canal to the natural river
(restitution canal).
Experimental and numerical studies have evaluated the technical performances of each variant. They
completed with economical analyses will allow designer to choose the best solution for discharge chute canal.

2. Experiments on hydraulic model

Fig 1 shows a general view of the hydraulic model of the Lechugal 2 discharge chute canal. The canal is
made up of the following areas from upstream to downstream: access platform, curved spillway with free flow,
transition zone, rapid canal no.1, transition canal, rapid canal no.2 with baffle blocks in the upstream half of the
canal, stilling basin type II USBR, connection canal to natural river. Figure 2 presents geometrical elements of the
hydraulic model.
Hydraulic modelling is based on the similitude of the general laws of mechanical systems. Two fluid
currents are similar if they are geometric similar and if for all homologous points are fulfilled the similitude
conditions of their cinematic and dynamic characteristics.
 Author name / Energy Procedia 00 (2017) 000–000 3

Fig 1. General view of the hydraulic model.

Fig. 2. Hydraulic model of the discharge chute canal: layout, representative cross sections and longitudinal profile through canal axis
showing measuring points of the flow parameters.
 4 Felipe Cisnerosa, Erwin Barahonaa, Cornel Ilincab, Cristian Anghelb/ Energy Procedia 00 (2017) 000–000
In practice, in order to have a satisfactory similitude between prototype current and model current are
necessary to carry out the following conditions:
- geometric similitude;
- similitude of the initial conditions and boundary conditions;
- equality of the dynamic similitude criteria, which for experiments carrying out must be settled
taking into account the main forces which determine the prototype current.
The similitude criterion used to build the hydraulic model of the Lechugal 2 chute canal was the equality of
Froude number (Fr) from model (Frm) and prototype (Frpr) as follows:
Fr = v2 / (gl) αFr = Frm / Frpr = 1 (1)

where v is current velocity, g – gravity, l – length and αFr – scale coefficient of Froude number.

Usually, in hydraulic construction modelling, the liquid for model is water with same density (d) and
viscosity (vs) as for prototype. Also, the experiments are performed in gravity field (g). Consequently, density scale
coefficient (αd), viscosity scale coefficient (αvs) and gravity scale coefficient (αg) are all equal to unity:
αd = dm / dpr = 1 αvs = vsm / vspr = 1 αg = gm / gpr = 1 (2)
The scale coefficients of main parameters are expressed function of the length scale coefficient (αl):
αl = lm / lpr (3)
lm being length in model and lpr homologous prototype length.
In the case of Froude number criterion used in hydraulic model of the Lechugal 2 chute canal the main
scale coefficients are the followings:
area scale coefficient αl2 acceleration scale coefficient 1
volume scale coefficient αl3 flow rate scale coefficient αl2.5
time scale coefficient √αl force scale coefficient αl3
velocity scale coefficient √αl
The length scale coefficient for Lechugal 2 chute canal hydraulic model was 1:35. The model has not been
distorted by reproducing the prototype geometry. Area of the model reached 45 m2 versus 55125 m2 homologous
area of the prototype.
The flow rate scale is a derived scale from αl. It resulted 1:7246. The maximum flow rate conveyed through
chute canal model was 45 l/s which corresponds to 326 m3/s on prototype, this being the design flow rate for
prototype.
Experiments on hydraulic model were performed in two variants, respectively with and without artificial
roughness on the upstream half of the rapid canal no.2. Introducing the artificial roughness (baffle blocks) had the
function to reduce the velocity at the entrance to stilling basin.
Table 1 presents the velocities measured in representative points on model with and without artificial
roughness. It may remark that in variant with baffle blocks the velocity in R4 was reduced to 16 m/s which is
acceptable.
 Author name / Energy Procedia 00 (2017) 000–000 5

Table 1. Velocities measured in representative points on model

Measurement Velocity (m/s)
points
Without baffle blocks With baffle blocks
R1 11 11
R2 15 15
R3 16 16
R4 21 16

A special attention was paid to restitution canal which is an unlined earth canal. From Fig 3 it may remark
that at short distance from entrance in restitution canal there is a curve on the route where the risk of erosion of the
right bank is very high.

Fig 3. View of the hydraulic model to the end of stilling basin and entrance in restitution canal

According to velocity values recorded in restitution canal on hydraulic model and presented in Table 2, it is
clear that velocities are too high and during transit of the design flow rate are expected erosion in restitution canal.
Local, right bank of the canal curve can be protected against erosion for instance through an Incomat mattress, but
for the rest canal solutions must be found to reduce the velocity up to 2 m/s. This issue follows to be studied by
numerical simulations for other constructive solutions of the chute canal.

Table 2. Velocity values recorded in restitution canal on hydraulic model

Measurement Velocity (m/s)
points
Without baffle blocks With baffle blocks
Vel. at stilling basin entrance 21 16
Vel. at stilling basin exit 6 3
Average vel. in restitution canal 5.5 2.8
Vel. at restitution canal exit 4 2.4
 6 Felipe Cisnerosa, Erwin Barahonaa, Cornel Ilincab, Cristian Anghelb/ Energy Procedia 00 (2017) 000–000
3. Numerical simulations with HecRas software
HecRas (River Analysis System) is a public software elaborated by US Army Corps of Engineers –
Hydrologic Engineering Center.
HecRas is a software allowing to perform one-dimensional steady and unsteady flow river hydraulics
calculations, sediment transport – mobile bed modelling and water temperature analysis.
Steady flow water surface profiles, used in present chapter, is intended for calculating water surfaces
profiles for steady gradually varied flow. The system can handle a full network of canals (channels), a hydrologic
basin (river + tributaries) or a single river reach. The steady flow component is capable of modelling subcritical,
supercritical and mixed flow regime water surface profiles.
The computation procedure is based on the solution in finite differences of the one-dimensional energy
equation. Energy loses are evaluated by friction (Manning’s equation) and contraction/expansion (coefficient
multiplied by the change in velocity head). The momentum equation is used in situations were the water surface
profile is rapidly varied. These situations include mixed flow regime calculations (i.e. hydraulic jumps), hydraulic of
bridges, and evaluating profiles at river confluences (stream junctions).
In figure 4 are presented the constructive variants analysed by numerical simulations.

Variant 1 corresponds to the hydraulic model as constructive composition, but at the prototype scale. The
baffle blocks cover upstream half of the rapid canal no.2. It can be remarked from Table 1 the efficiency of baffle
blocks reducing the velocity of the current flow to the entrance in the stilling basin (point R4) at 16 m/s. The
objective of the numerical simulation in the variant 1 was to check if numerical model provided output results
having satisfactory accuracy versus their homolog results from hydraulic model. The answer to this question is
found in the table in figure 5. There is a good correlation between the experimental values and those provided by the
numerical model. Accordingly numerical model is performing and can be applied to analyse other variants.
In variant 2, baffle blocks are extended to all length of the rapid canal no.2 (fig.5).The aspect of the flow in
variant 2 is illustrated in figure 4 and the values of the velocity in representative section are presented in Table 3. It
may remark that in variant 2 the main hydraulic jump starts on the slope of the rapid canal no.2, about 3 m upstream
from its downstream toe. Also, the jump is not drowned in stilling basin. Second jump born after the current crosses
over threshold is drowned by downstream water level. Accepting the lengths of hydraulic jumps in variants 1 and 2
are equals, then the length of the stilling basin in variant 2 can be reduced with 3 m versus variant 1. This means a
reduction of the concrete volume poured in stilling basin body, saving which needs be compared with the cost of
additional baffle blocks for checking if variant 2 was recommended.
In variant 3a, the chute canal is gradually widened from 17.50 m to 24.25 m starting since the middle of the
rapid canal no.2. The purpose of this composition is to reduce the current velocity at stilling basin entrance and
along of restitution canal. Figure 5 shows the aspect of the flow in this variant and Table 3 the velocity in
representative points. The results are not those that were expected. The reduction of the velocity in R4 section is
 Author name / Energy Procedia 00 (2017) 000–000 7

quite small one (17.34 m versus 17.52 m in variant 1) and the reduction of the velocity in restitution canal (2.64 m/s
versus 2.79 m/s) is insufficient to avoid the erosion of the canal bed.

Variant 1 for calibration with hydraulic model

Variant 2

Variants 3a and 3b

Variants 1, 3a – longitudinal profile

Variant 2 – longitudinal profile

Variant 3b – longitudinal profile

Fig. 4. Constructive variants of the chute canal analysed by numerical simulations running Hec-Ras software: layouts,
longitudinal profiles through the axis and input/output data sections.
8 Felipe Cisnerosa, Erwin Barahonaa, Cornel Ilincab, Cristian Anghelb/ Energy Procedia 00 (2017) 000–000

Fig. 5. Measurement cross sections in numerical simulations and comparison of output data on hydraulic model
and numerical simulation variant 1.

Fig. 4. HecRas copies of screens showing water surface in numerical simulations of the design flow transit
through chute canal, versions 1, 2, 3a and 3b
 Author name / Energy Procedia 00 (2017) 000–000 9

A new composition of the chute canal in order to improve the performance of the variant 3a is tested in
variant 3b where the part widened is gradually deepened down to 0.50 m. As may be seen from Table 3, the results
are not satisfactory.
Table 3. Velocity results

Measurement points Velocity (m/s)

Hydraulic model HecRas software HecRas software HecRas software HecRas software
Var.1 Var.2 Var.3a Var.3b
R1(42) 11 10.52 10.52 10.52 10.52
R2(40) 15 15.53 15.53 15.53 15.53
R3(39) 16 14.58 14.58 14.58 14.58
R4(4) 16 17.52 15.46 17.34 17.60
(1.25) - 5.74 5.74 4.53 4.53
** 2.4 2.79 2.79 2.64 2.64

** – average velocity to natural river connection channel

Table 4. Water surface results

Measurement points Water Surface (maSL)

Hydraulic model HecRas HecRas HecRas HecRas software
software Var.1 software Var.2 software Var.3a Var.3b
R1(42) 35.27 35.07 35.07 35.07 35.07
R2(40) 27.1 26.98 26.98 26.98 26.98
R3(39) 27.02 27.05 27.05 27.05 27.05
R4(4) 16.68 16.73 17.73 16.75 16.31
(1.25) - 21.59 21.59 21.04 21.04
* 22.23 22.23 22.23 21.47 21.47

*–connection channel to natural river

Concluding, the variant 2 seems to bring some advantages versus the solution found by hydraulic model
and it deserves to be analysed from economical point of view.
The numerical simulations have not identified a satisfactory solution for restitution canal with current
velocity up to 2 m/s. The solution seems to be by segmenting the restitution canal in steps with reduced longitudinal
slope. This means to build thresholds associated with dissipation systems at intervals of about 100 - 200 m along of
restitution canal completed with restitution canal banks protection..
HecRas software supplied also the water surface of the current along the chute canal during steady transit
of the design flow rate (326 m3/s) which is presented in Table 4. These information can be used to optimise the
elevation of the chute canal lateral walls crest versus free guard.
 10 Felipe Cisnerosa, Erwin Barahonaa, Cornel Ilincab, Cristian Anghelb/ Energy Procedia 00 (2017) 000–000
4. Concluding remarks
Extensive studies by hydraulic and numerical models were carried out in order to find an optimal hydraulic
composition of a discharge chute canal, appurtenant work of Lechugal 2 earth dam (H=20 m). This dam is part of a
big irrigation system from Ecuador, presently under construction.
Experiments on the hydraulic model at 1:35 geometrical scale of the chute canal and numerical simulation
running HecRas software for four variants concerning composition of the canal provided the technical performance
of each variant. These information completed with economical analyses will allow designer to choose the best
solution for chute canal. However, the restitution canal which is an unlined earth canal has not yet a satisfactory
solution found by above mentioned studies. In this stage seems that a satisfactory solution, according to restrictions
of velocity along the restitution canal to maximum 2 m/s is by segmenting the restitution canal in steps with reduced
longitudinal slope completed with restitution canal banks protection.
The study pointed out the capability of the numerical method based on HecRas software to analyse a lot of
variants of the discharge chute canal with minimum effort in preparing input data and with results obtained very
fast.

Acknowledgements
The authors express their gratitude to the authorities of the Secretary of Water (SENAGUA) and the
Empresa Publica del Agua (EPA EP) Ecuadorian entities of water issues in charge, for permission through
PROMAS University of Cuenca to use in this paper some data from Pacalori project whose ownership belongs to
these institutions.

References
*** Proyecto Pacalori – Plan de aprovechamiento y control del agua en la provincia de Los Rios. Estudio de
prefactibilidad (in Spanish). PROMAS – Universidad de Cuenca, Ecuador, 2013.
*** Vertedero en Abanico Lechugal 2. Informe de Modelo Físico (in Spanish). PROMAS – Universidad de Cuenca,
Ecuador, 2015.
*** Design of Gravity Dams. A Water Resources Technical Publication. Denver, Colorado 1976 (page 156-201).
*** Hec-Ras software, USACE, Hydrologic Engineering Center DAVIS, CA 9561-4687, USA.
*** ICOLD Ad-Hoc Committee on Computational Aspects of Analysis and Design of Dams
Computational procedures for dam engineering . Reliability and applicability. Bulletin ICOLD No.
122, Paris, 2001
Popovici, A. Baraje pentru acumulări de apă (in Romanian) Editura Tehnică, Bucuresti, 2001 (page 331-346).
Arthur H. G., Jabara M. A. (USA) Problems involved in operation and maintenance of spillways and outlrets at
Bureau of Reclamation dams. Proceedings IX-th International Congress of Large Dams Q33, R5, Istambul, 1967.
Chanson H. Hydraulics of Open Channel Flow. The University of Queensland, Australia. Elsevier Butterworth
Heineman, Linacre House, Oxford, 2004.