fttcnatUcRs esearctt \Ahlingrford PERFORMANCOEF AEMTORS FOR DAU SPILLWAYS ByRWPMayandAPDeamer Report SR 198 March 1989 Registered Office: Hydraulics Research Limited, I{allingford, Oxfordshire OXl0 8BA. Telephone: O49l 35381. Telex: 848552 This report describes work funded by the Department of the Environment under Research Contract PECD7 /6/46, It is published on behalf of the Department of the Environment, but any opinions $q)ressed j-n this report are not necessarily those of the funding Department. The work was carried out in the River Engineering Department of Hydraulics Research, Wallingford headed by Dr W R l{hite. The nominated project officers were Dr R P Thorogood for DOE and Dr W R iihite for HR. @ Cro*n copyright 1989 Published by permission of the Controller of Her Majesty's Stationery Of,fice ABSTRACT This report describes laboratory tests on the performance of spillway aerators carried out as stage 2 of a research contract funded by the Department of the Environrnent. In stage 1, a comprehensive review was made of the available literature on cavitation and aerition in hydraullc structures; results of the review were presented in an earlier Hydraulics Research Report No SR 79. on the basis of this review it was decided to carry out a systematic erqperimental study of ramp aerators which are used to prevent eavitation dgmageo n dam spillways by entraining air into the high velocity flows. A 4tt: was specially built for the study; the test seciion is 4m-l ong and the width can be varied up to a maximumo f 0.3m. The fh:roe can be set it angles between horizontal and 45o, and the pumph as a flow capaci.ty of O.2msls ind can produce velocities of up to l5m/s. Initial tests vere carried out rsith a nitrogen gas injection system in order to study the convection and diffusion of gas in turbuLent flows. Measurementso f velocity profiles and gas eoncentration profiles were made downstreamo f the injection point for a range of flow depths and velocities. An air supply system was installed in the flume for the tests on the aeration ramps. This enabled the flow over a ramp to create a low-pressure air cavity which drew in air naturally from the almosphere. A largl number of tests was carried out in order to determine the effect on the air demand of the following factors 3 water velocity; water depth; height of aeration rarp; slope of, aeration rampi slope of channel; and head Loss characteristics of air supply system. Results were analysed both in di-rnensional and non-dimensional form and compared with ftrmulae from other studies The e:q>erirnents were designed to provide data for I proposed numerical model of cavitation and aeretl.on in dam spillways. The rnodel wouLd be developed !1om an existing computer program (SWAI[)f or spillway flows, and would identify the risks of cavitation damagea nd facilitate the design of suitable aeration systems. Descriptions are included of the existing SWAN progran (produced by Binnie &,P artners) and of a proposed convection-diffusion modeLo f . two-phase f1ows. SYI'IBOLS Aa Cross-sectional area of air ducts at outlet a Constant in Equation (34) B Width of channel C Local air goneentration e Depth-averaged air concentration c Speed of sound in air c1, c2 €tc Constants d Depth of flow measured normal to channel invert dc Depth of flow at vena contracta q, Mean size of voids E Euler number (Equation (37)) e Constant in Equation (38) F l{odified Froude m:mber (Equation (18)) Fk Critical Froude number at start of air entrainrnent Fr Froude number (Equation (20)) g Acceleration due to gravity h Height of ramp measured vertically h1 Height of rarnp measured normal to channel invert i Energy gradient of flow J Head loss parameter for air-supply system (Equation (9)) k Air entrainment coefficient (Equation (15)) ks' Nikuradse equivalent sand roughness tc Length of air cavity Lcm Value of L" measured from tip of ramp to centre of mass of reattaching jet Lr Length of ramp measured parallel to channel invert m Coefficient in Equation (34) Nv Number of voids in sampling period n Manning roughness coefficient p Pressure Ap Pressure difference below atmospheric Qa Volumetric flow rate of air q Volumetric flow rate of water per unit width of channel qa Volumetric flow rate of air per unit width of channel R Hydraulic radius (= area./wetted perimeter) R" Relmolds number r Vertical rise velocity of bubble s Height of offset in channel floor, measured vertically s1 Height of offset in channel f1oor, measured normal to invert T Temperature T. Time probe in conducting liquid T., Time probe in vqids t Time u Mean velocity paral1e1 to x-axis V Mean velocity of water V" Mean velocity of air Vd Velocity of water at downstream end of air cavity Vk Effective minimum velocity of water for air entrainment (Equation (38) ) vr Root-mean-square velocity fluctuation v Mean velocity parallel to y-axis }Ie Weber number w Overall step height \ Parameter defined by Equation (26) x Distance measured along channel v Distance measured normal to channel c Energy coefficient p Aj.r demand ratio (= qulq) €x Coefficient of turbulent diffusion for air in water in x*direction ty Coefficient of turbulent diffusion for air in water in y-direction o Angle of channel to horizontal I Darey-Weisbach friction factor U Kinenatic vj-scosity of water ua Kinematic viscosity of air p Density of water pa Density of air o Surface tension for air-water interface 0 Ang1e of ramp relative to channel CONTENTS Page INTRODUCTION 1 DIMENSIONALA NALYSIS 4 PREVIOUSS TUDIES 10 EXPERIMENTALA RMNGEMENT t7 4.L Flume L7 4.2 Air supply system 20 4.3 Aeration ramps 2T 4.4 Measuring equipment 23 TESTS WITH GAS INJECTION SYSTEM 27 TESTS WITH AERATIONR AUPS 31 ANALYSISA ND DISCUSSION 35 CONCLUSIONS 38 ACK}IOWLEDGEMENTS 40 10 REFERENCES 41 TABLES 1 Test data for aerator no. I 2 Test data for aerator no. 3 3 Test data for aerator no. 7 4 Test data for aerator no. 9 5 Data analysis for aerator no. T 6 Data analysis for aerator no. J 7 Data analysis for aerator no. 7 8 Data analysis for aerator no. 9 FIGURES 1 Geometry of aeration system 2 Layout of high-velocity flurne 3 Air supply system 4 Measured and predicted velocity profiLes at x = 0.195m 5 Measured and predicted velocity profiles at x = 0.5m A Measured and predicted velocity profiles at x = 1.0m -7 Measured and predieted veloci_ty profi_}es at x = 1.5m X Measured and predicted velocity profiles at x = 2.0m 9 Gas concentration profiles at x = 0.08m, 0.195m, 0.5m and 0.75rn 10 Transverse variation in gas concentration profiles at x = 0.08m CONTENTS( CONT'D) FIGURES (CONT'D) . 11*36 Air demand versus velocity Figure No Fh::ne slope Air valve Aerator No setting 1379 15.50 0 11 18 24 31 15.50 2 12 19 25 32 15.5" 5 13 20 26 33 15.5. 4 L4 2t 27 34 45.30 0 15 22 28 35 45.30 16-29- J 45.30 4 17 23 30 36 37 llead loss characteristics for air supply : valve setting 0 38 Head loss characteristics for air supply : valve setting 2 39 Head loss characteristics for air supply : valve setting 3 40 Head loss characteristics for air supply : valve setting 4 41-66 Air demandr atio versus Froude number Figure No Flume slope Air valve Aerator No setting 1379 15.5. 0 41 48 54 61 15.50 z 42 49 55 62 15.50 3 43 50 56 63 15.50 4 44 51 57 64 45.30 0 45 52 58 65 45.30 3 46 59 45.30 4 47 53 60 66 APPENDICES A Numerical convection - diffusion model B Numerical model of cavitation and aeration INTRODUCTION Cavitation in hydraulic structures is usually associated with high-velocity f1ows. If the pressure within flowing water drops close to the vapour pressure of the water, cavities will form in the flow and be carried along by it. Regions of low pressure can typically be caused by a general increase in flow velocity (e.g at contractions in closed conduits), by flow separation at sharp edges (e.g joints, s1ots, surface irregularities), or by velocity fluctuations due to high turbulence. The cavities are normally occupi.ed by a mixture of water vapour and air, and will ocpand in size if they remain in an area of lov enough pressure. When the cavities move into an area of increasing pressure, they can collapse violently and generate very high velocities and pressure impulses in the fluid; pressures as high as 15,000 atmospheres have been recorded by Lesleighter (1983). If cavities collapse up against a solid boundary, they are capable of damaging materials as hard as stainless steel. In the case of concrete, cavitation attacks the sand-cement component and loosens the aggregate, which is then plucked out by the flow. A surface danaged by cavitation usually presents a pitted appearance, and concrete exposed to severe cavitat'ion can somelimes be eroded to depths of several metres. Cavitation is therefore capable of causing serious damage or even failure in tunnels and spillways carrying high-velocity flows. As a result, the worldwide trend towards the construction of larger dams with higher heads has led to an increased awareness of the dangers posed by cavitation. In 1985 the Construction Industry Directorate of the Department of the Environment (DOE) cornmissi-oned Hydraulics Research (HR) to carry out a research project on caviLation damage in major civil engineering works. It was forseen that the study would be rnainly experimental, and that it would be necessary to build a new test facility for the r+ork. However, before embarking on major capital or.penditure, it was decided that the available l-iterature on cavitation should be revi.ewed in order to identify in which area new research would be most beneficial. The literature on cavitation is large so it was decided to extend the scope of the review and produce a document which would be of use not only to researchers but also to engineers designing hydraulic structures. The review t/as published in 1987 as Hydraulics Research Report SR 79, aad includes sections on the mechanism of cavitation, the factors governing its oceurrence in various types of hydraulic structure, the resistance of different materials, the use of air entrainment to prevent damage, and, the problems of modelling and instrumentation. The review identified two nain options for the experimental part of the research study; these were to study at reduced scale either: 1. the occurrence of cavitation at features of hydraulic structures such as transitions, slots and surfaee irregulari-ties t 2. the performance of systems for entraining air into high-velocity flovs in order to prevent cavitation damage. The first option required the use of a vacuum test rig because, in order to reproduee cavi_tation correctly at reduced scale, it i-s necessary to lower the ambient pressure below atmospheric. Such a test rig would have been very expensive to build and maintain. New