EXPERIMENTAL INVESTIGATION INTO HOT WATER SLOT JETS WITH NEGATIVELY BUOYANCY IN CROSS FLOW

Journal of Hydrodynamics,Ser. B,2005,17(4):412- 417China Ocean Press, Beijing - Printed in ChinaEXPERIMENTAL INVESTIGATION INTO HOT WATER SLOT JETS WITHNEGATIVELY BUOYANCY IN CROSS FLOW"YANG Zhong-hua, HUAI Wen-xin, DAI Hui-chaoState Key Laboratory of Water Resources and Hydropower Engineering Science， W uhan University,W uhan 430072, China， E mail:y_工bua@ sina. com(Received Aug. 12, 2004)ABSTRACT: An expstudied the behavior of the dense jets with the slotnear-field behavior of negatively buoyant planar jets in flow-diffusers. Slot jets are created by the discharge ofing environment. Hot water jet was projected downwards atan effluent into a receiving- water body from a sin-different angles from a slot into a uniform cross flow. Microgle orifice of slot geometry. The characteristics ofAcoustic Doppler Velocimeter ( Micro ADV) system is usedto measure the velocity and turbulent fluxes of Reynoldsnegatively buoyant jets of particular interest arestresses. The whole field temperatures were measured with the geometric coordinates and dilutions at the mxi-fast response thermocouples. Pure jets experiments were mum penetration of the jets and at the point of bot-made also to study the effect of buoyancy in negatively buoy-tom impingement-6.7l. Many studies have beenant jets. It is found that the influenced area of hot jets is lardone by numerical simulation in recent years' 8 10.ger than which of pure jets when the jet angle is 90° and theWhen a jet of heated water is dischargedinfluenced area of hot jets is smaller than which of pure jetswhen the jet angle is 45°. The difference is not obviousdownward into an ambient region of cold water60° angle jets. This means that the rising of tempeature has negatively buoyancy decelerates the jet flow andeffect not only on negatively buoyancy, but also on the in- this ultimately reverses to produce an upward flowtensity of turbulence. The contrast of these two influences surrounding the central downward flow. Exampledominates the trend of jet flow.is hot water from cooling tower of power plant is .KEY WORDS: negatively buoyant planar jets， Micro A-discharged downward into the surface of river.coustic Doppler Velocimeter ( Micro ADV), Reynolds stres-The specification of the temperature and velocityesdistributions and the depth of penetration of such a .flow is basically an elliptical problem which hasnot yet been solved. It is also important to deter-1. INTRODUCTIONmine the dependence of these quantities on the in-Study on negatively buoyant jets in cross flowflow velocity and temperature， characterized iris mainly focused on dense jets in which the nega-terms of the inlet Reynolds number ( Re ) andtive buoyancy is caused by the difference of con-Grashof number ( Gr ). It is found that the flow .centration. Example is brine discharged upwardsinto environment. The effluent is denser than themainly depends on the mixed-convection parameterreceiving water and the jets are pointed upwards atGr/Re2，which is the inverse of a F roude numbersome angle to achieve a trajectory resulting in suf-( Fr )，over the parametric ranges considered.ficient dilution. The negative buoyancy causes theThe velocity ratio of discharge and ambient is alsojets to reach a terminal rise height and then fallan important parameter in study the behavior ofback to the lower boundary where it spreads as a jets中国煤化工regative buoyancy, thedensity current. Diffusers have been designed with jetMYHCN M H Gm because of the flownozzles with different angles-4]. Shahrabnif'] has ambient. So the interaction of jets flow and cross* Project supported by the National Natural Science Foundation of China (Grant No: 50279037) and Open ResearchFund Program of State Key Laboratory of Water Resources and Hydropower Engineering Science (Grant No: 2005C011).Biography :尔民落塘nong -hua (1977-), Male, Ph. D.，Lecturer413flow determines the form of pure jets. For nega-tively buoyant jets in cross flow，the Froude num-ber and velocity ration are two main parameters.This paper reported a detailed experimental studyof hot water jets in cross flow, and includes the meanvelocity and temperature as well as the intensity of tur-bulence and Reynolds stresses. The effect of buoyancyto the flow is analyzed expressly.Uo.T。H三2. EXPERIMENTAL ARRANGEMENTSThe problem under consideration is illustratedschematically in Fig. 1. The two -dimensional neg-Fig. 1 Sketch of hot water discharged downward in crossatively buoyant jet was realized in the experimentsflowby introducing a constant downwards discharge ofhot water from a specially designed injection boxA series of parametric experiments has beeninto a deep rectangular flume ( W =8.0m，H =0.carried out for different values of cross flow mass4m，B =0. 3m) with cold water flow through it.fluxQ。and the jet angleθ (θ =90°，60° and 45*).The flume itself was fabricated with glass plat Qau =7. 0X 10* m*/s andQ.: = 7. 0X 10* m2/s withsides built on a steel frame to allow a better flow the velocityoa =6. 06X10*m/s andva2 =8. 67Xvisualization when view from the side. The flowin 10*m/s. The jet mass fluxQ。=1. 0m'/s with thethe flume is pumped from the reservoir to be recy-jet velocity V。=0. 37m/s. So there are two veloci-cled. The reservoir is large enough to ensure thty ratios selected in the experimentstability of ambient temperature. The electricflowmeter is used to measure the mass flux. TheVo__ 0. 37= 6.10hot water was from two tanks with electric heatingual 0. 0606pipe inside and the tank is set highly to give the jetpressure and the injector box was supplied with a .R,=0_0.37= 4.271 0. 0867suitable constant-head arrangement and pressure .valve to exclude the presence of air bubbles insidethe starting negatively buoyant jet. A recycle de-The hot water temperature T。=55C and the am-vice was set to ensure the steady water level of the .bient temperature T。= 15C. Pure jets experi-tank. The two tanks were always overflowed. Thements were made also to study the effect of buoy-water overflowed from the two heating tanks wasancy in negatively buoyant jets( T。= T。=15C).Measurement of mean and fluctuating veloci-collected to the recycle tank and was pumped intothe heating tanks subsequently. The overflow ofties was accomplished by means of Micro ADVthe heating tanks should be controlled to minimumsystem. The ADV measures the velocity of waterif possible to minish the loss of quantity of heat.using a physical principle called the DopplerThere was an automatic control instrument to con-Effect. The transducer sends a single pulse ofsound and measuresthe frequency change of thetrol the temperature of the water inside the tank.The injection box was fixed on the flume and thereturn signal. The ADV uses a technique calledjet angle can be changed easily. There are grids in-pulse- coherent processing. T he instrument sendsside the injection box to ensure the uniformity oftwo pulses of sound separated by a time lag; itjet velocity. The discharged hot water was intro-then measures the phase of the return signal fromduced through a rectangular slot of width B = 2.each pulse. The change in phase divided by the5mm and length W = 300mm positioned at the bot-time中国煤化工ly proportional to thetom of the injection box, with the geometrical di-YHC N M H Ge water. Pulse coher-mensions of the slot being chosen in order to main-ent processing is used because it provides the besttain the two dimensionality of the buoyant jet. Thpossible spatial and temporal resolution. The Mi-slot was positioned below the receiving water sur-cro ADV probe has a 0. 05m nominal distance tothe sampling volume. The maximum sampling rateface so that tr境t was submerged.414is 50Hz and the sampling volume is a cylinder 4mm decrease faster in the vertical jets than in 60° anglein diameter and 6 mm tall, for a total measurement jets. In the downstream of the hot water jets， thevolume of 0. 08X 10*"m’. The field processor uses temperature of surface water is high than the watera default baud rate of 19200.downward which causes the stratified flow. It isThere are two factors for the reliability of ve-also found that the hot water can spread for a longlocity measurement. One is that if the movement distance.of the particles in water can represent the velocity 3.2 The reattachment eddy center and the sepa-of water correctly. Sampling analysis confirmedrate pointthat the particles in experimental water are mainlyOf particular interest is the position of the re-clay granule and the particle diameters are from attachment eddy center and the separate point5μm to 50μm. The particles have good concomi- where the velocity is very small. Relations be-tance with the water. Detailed descriptions of the tween the position and the velocity ratios or the jetparticle analysis are given in Ref.[11]. The other angle are analyzed. It was found that the size offactor is the Doppler noise. The Doppler system the reattachment eddy of the hot water jets is lar-has an inherent measurement noise that is a result ger than which of the pure jets in vertical jets (seeof the physical process by which the sound waves Fig.2 and Fig.3)，in the 45° angle jets the size ofare scattered from particles in the water. In the reattachment eddy of hot water jets is smaller thanexperiment，the Signal to Noise Ratio (SNR) that of the pure jets (see Fig. 6 and Fig. 7), and inranged in 14dB-25dB and the correlation coefficient the 60° angle jets there are no obvious distinguishranged in 85%-98%，which means that the experi- between the two reattachment eddies (see Fig. 4ment was under good operation conditions (SNR>and Fig. 5).15dB and correlation coefficient greater than70%). The correlation coefficient is a data qualityparameter that is a direct out put of the Dopplervelocity calculations.T emperatures were measured with fast re-ξ_01sponse thermocouples whose precision was con-0.2firmed to be 0. 1C. The measuring system in-020.6cludes signal collction portion and data processingportion. The temperature probe is made up of aFig.2 Streamline of pure jets ( θ=90°)metal tubule and can be made to different shape forthe experimental requirement. The head of theprobe is the temperature inductor，which is about3mm in length and is connected with the probe bythe adiabatic part. Average temperatures wereE0.1measured in the experiment.Detailed descriptions of the experimental e-4x/nquipment and techniques are given in Ref. [11].Fig.3 Streamline of hot jets ( θ= 90°)3. EXPERIMENTAL RESULTS3.1 Flow pattern and mean tem peralureFigures 2-10 are streamline and temperature3. 3 The effect of the negative buoyancy andturbulencecontour of hot jets withR =6. 1 and streamline ofIt is obvious that the direction of buoyancy isthe pure jet with R =6. 1. Similar flow patternsupw中国煤化工cause the jet tempera-are shown in pure jets and hot water jets. The jettureMHC N M H Gnt temperature which .turns to the downstream due to the cross effect andmeans that the density of jet is lower than the den-there is a reattachment eddy under the surface ofsity of ambient. The reattachment eddy of the hotwater. The size of the reattachment eddy is differ-water jets should be smaller than the reattachmentent in different jet angle and temperature. Theeddy of pure jets. In this experiment, different re-temperatatettrases fast near the spout and it415perature out of the reattachment eddy in these figures. So the turbulence caused by the rise of tem-perature may enlarge the reattachment eddy of thehot water jets that is contrary to the effect of nega-E0.03tive buoyancy. The contrast of these two influ-ences dominates the trend of jet flow. In vertical-0.00.jets，the influence of turbulence is stronger thanx/mthe influence of negative buoyancy that the size ofthe reattachment eddy of the hot water jets is lar-ger than which of the pure jets. In inclined jets,Fig. 4 Streamline of pure jets ( θ=60°)the influence of turbulence is weaker than the in-fluence of negative buoyancy that the size of the re-attachment eddy of the hot water jets is smallere-0.04than which of the pure jets. When the intensity of^0.08the two influence factor is comparative ( 0 =60°)，0.2there are no obvious distinguish between the tworeattachment eddies (see Fig. 4 and Fig. 5).Fig.5 Streamline of hot water jets ( 0 = 60*); 0.1-0.01CC四E002020.4-0.030.040.12xhFig.8 Temperature contour of hot water jets (θ =90°)Fig.6 Streamline of pure jets (θ=45°)~24三0.04E0.020.080120.1Fig.9 Temperature contour of hot water jets ( θ=60°)Fig. 7 Streamline of hot water jets ( θ=45°)lationship of the hot water jets and pure jets isfound in different jet angles. For hot water jets(different from density jets upward)， the factors230that influence are not only buoyancy but also the一25turbulence caused by the rise of temperature. The-2temperature is higher, the flow is more turbulent中国煤化工.and the influenced area of the jet is larger. This .TYHCNMH G .can be seen from the temperature contour (Figs. 8-10) where the influenced area of temperature is Fig. 10 Temperature contour of hot water jets (0=45°)much larger than the reattachment eddy. The tem-perature行包熬隔lly reduced to the ambient tem-416be seen in these figures.3.4 Turbulent fluxes of normal stresses(2) The normal stresses of x direction are lar-The u;? represent fluxes of the normal stressger than the normal stresses of y direction in mostand the standard fluxes of the normal stress can bearea expect the position near the spout. So the tur-gained if divided by the mean velocity of jets Vo .bulence mainly occurs in the x direction for jets inFig.11 is standard fluxes of the normal stress incoming flow.(3) In the section of reattachment eddy centerthe section of coming flow， the spout, the centerand the separate point， the fluctuating velocitiesof reattachment eddy and the separate point ( R =are 0. 05-0.1 times of the jet velocity Vo. The tur-6.1.0=90). (i2)12/0 and (02)12/0 represent bulence is very strong in this experiment.the standard fluxes of the normal stress in x and y(4 ) It is the turbulence isotropy if thedirection respectively. It was found that:(u2)1/2/o。and (v2)1/2/00 are equal. In this experi-ment Anisotropic turbulence can be found because(u2)/2/0o and (v)2 )1/2 /vo are different in these sec-tions.(5) The fluctuating is similar as the normalexpresses are invariable along the section. In the后section of coming flow and outlet， the normal ex-presses are almost changeless which means that thefluctuating is similar in the two sections. In thesection of spout the fluctuating is similar at y/BOE-8≤- 30.3.5 Turbulent fluxes of the Reynolds stresses(a) The section of the(b) The section of theincoming flowspoutThe correlations u' U' represent fluxes of theReynolds stresses. In this paper, u∪' (divided by↓0,1%。% in the section of coming flow，the spout, thecenter of reattachment eddy and the separate point( R = 6.1，θ =90°) are analyzed. Relation betweenuv' / and the y/B are given in Fig. 12. Itwas found that Reynolds stress is small in the .40f罗coming flow section. In the section of the center ofreattachment eddy and the separate point， the3of80fReynolds stress is larger which means that theflow is much turbulent. In the spout section， th(c)The section of the(d)The section of separateReynolds stresses are large near the spout and de-reattachment eddy centerpointcrease fast until y/B≤- 30. The maximum Reyn-olds stresses are at the interface of main flow ofjets and the reattachment eddy.Fig. 11 The normalized fluxes of the normal stress indifferent section ( R= 6.1, θ=90°)4. CONDLUSIONSExperiments were performed to measure the(1) The normal stress is small in the cross velocity and temperature of the hot water dissection which means the turbulence is not strong.charged downward at different angles to the hori-The maximum of the normal stresses are in the zont中国煤化工:h jets are frequentlyspout section where the velocity gradients are very used|YHCNMHGotwaterintothenat-large. At y/B≤- 30，the normal stresses de- ural environment. The Micro ADV system is usedcrease obviously. There is also a maximum of nor- to measure the velocity and turbulent fluxes ofmal stress at the reattachment eddy center and the Reynolds stresses. The whole field temperaturesseparate有数拥he distribution of turbulence can were measured with fast response thermocouples.417The turbulence is weak at the coming flow andthe outlet flow and is strong at the reattachmenteddy center and the separate point. The maximumReynolds stresses are at the interface of main flowof jets and the reattachment eddy.0.010.02REFERENCES[1] CHU V. H. Turbulent dense plumes in laminar cross30 fo80flow[J] . J. Hydraulice Res.，1975， 13:263-279.[2] ANDREOPOLOUS J. Heat transfer measurements ina heated, jet pipe flow issuing into a cold cross -stream(a) The section of cross flow(b) The section of spout[J]. Phys. Fluids, 1983, 26 :3201-3210.u1唱[3] ROBERTSP. J. w.. Toms G. Inclined dense jets in001flowing current[J] . Journal of Hydraulice Engineering,1987，113(3) :323- 340.[4] LINDBERG W. R. Experiments on negatively buoy-。ant jets, with and without cross flow[A]. Recent Re-search Advances in the Fluid Mechanics of TurbulentJets and Plumes [ C ]. Dordrecht， The Netherlands,1994，131-145.[5] SHARABANID. M. and DITMARS J. D. Negatively(C) The section ofthe(d) The section ofbuoyant slot jets [A].13rd Conference on Coastal En-reattachment eddy centerseparate pointgineering[C]. Honolulu， Hawaii, USA，1976， 2976-2993.[6] KAMORANI Y. and GREBER I. Experiments on aFig. 12 The normalized fluxes of the Reynolds stress inturbulent jet in a cross wind[J]. J. Fluid Mech. ,different section ( R= 6.1, θ=90°)1972，15:481-496.The mean velocity and the temperature， normal- [7] SUBRAMANYA K. and POREY P. D. Trajectory ofized fluxes of normal stresses and Reynolds stres-a turbulent cross jet[J] . Journal of Hydraulic Re-ses，and other properties were analyzed.search, 1984， 22:343- 354.The negatively buoyant jets in coming flow[8] HUAI Wen-xin and LI Wei. Prediction of characteris-tics for vertical round negative buoyant jets in homoge-are much turbulent and the streamline is flexural.neous ambient[ J] . Journal of Hydrodynamics, Ser. B,There are reattachment eddy and separate point in2001，13(1): 103-110.this flow. The temperature decreases fast near the[9] HAN Huiling and ZHANG Hong-min et al. Meanspout and hot water on the surface can spread for abehavior of three dimensional line buoyant jets in crosslong distance. Because of the negative buoyancyflows[J] . Journal of Hydrodynamics, Ser. B, 2003 ，and the turbulence caused by the temperature ris-15(2): 32-36.ing，the characteristic of hot water jets in coming [10] HAN Hui- ling and LI Wei. Numerical prediction offlow are different from the density jets upward.line buoyant jets in cross flows[J]. Journal of Hydro-The influence of the negative buoyancy and thedynamics,Ser. B, 2001, 13(1): 12-16.turbulence are contrary that the contrast of these [11] YANG Zhong- hua. Study on the behavior of negativelybuoyant jets[D]. Ph. D. Thesis, Wuhan: W uhan Uni-two influences in different jet angles dominates theversity，2004. (in Chinese)trend of jet flow.中国煤化工MHCNMH G .