Ow Regimes Research Articles

Investigation of residence time distribution and local mean age of fluid to determine dead-zones in flow field

Residence time and local mean age of flare important parameters of the fl uid fl ow, with which we are able to detect the functioning and non-functioning zones. It is a novel approach to estimate the effect of dead-zones on chemical and biological processes. In this paper, we introduce a numerical tracer study, which is calibrated by fi measurements and which is applied in different cases such as settling tank and aeration tank of wastewater treatment. Using an advective-diffusive passive scalar transport equation, the age of the fl uid can be calculated. It also shows the short-circuits and dead-zones of the fl uid and this corresponds with our observations. This method helps eliminate potential hydraulic problems at the early stage of design. Settling tanks in the wastewater treatment have a key role in the separation of suspended solids and sludge fl ocks. The design of these tanks is based on the ATV-A131 standard (1), which takes into account only the global parameters of the fl ow, e.g. average residence time, but in many cases the inequalities inside the tank may have serious effect on effl uent quality. In this paper, we are fi rst and foremost interested in the detection of dead-zones, where the mixing is adverse due to low velocity, high residence time and fl uid age. In aeration tanks, for example, anaerobic zones may develop because microorganisms consume the oxygen. There is mass transfer between the fl ow and dead-zone because of the presence of turbulence; how- ever, this is much less than the effect of convective transport. On the other hand, hydraulically inactive zones are redundant and these increase the cost of operation. Computational fl uid dynamics is widely applied in wastewater treatment tank design (2-4)) because with the knowledge of hydraulic phenomena, the understanding of biochemical processes could be improved. One signifi cant linking parameter between the hydraulics and biology is residence time; therefore, it is essential to know the precise value for every point. Residence time dis- tribution (RTD) can be measured experimentally (5). In pulse experiment, the tracer is mixed in a small volume of fl which is introduced to the reactor in a very short time interval (Dirac-delta function) and the response function is measured at the outlet. The tracer should be a conservative material without any background concentration of the wastewater. The response function shows the changes in tracer concentration in the course of time, from which we can estimate the hydraulic behavior of the tank. RTD calculations can be performed numerically, and they can be compared to the results of the fi eld measurement. This paper deals with circular fl ow reactors, one aeration tank and one radial fl ow settling reactor. The fi rst is designed as a fully mixed reactor, but the RTD investigation demonstrated that there is a short-circuit, where the nutrients could fl ow through the reactor rapidly and did not have enough time for biodegradation. Circular radial fl ow settling tanks generally have problems with short-circuiting and the wind also could have an infl uence on the fl ow regimes.

Structure of Flow over a Hypersonic Inlet with Side Compression Wedges

The e ow over a hypersonic three-dimensional inlet with ramp and side-swept compression wedges was experimentally studied. Two identical inlet models were tested in the blowdown wind tunnel T-313 at Mach numbers M1 =4 and 6 and in the hotshot wind tunnel IT-302M at M1 =6 and 8. Optical visualization of the e ow, near the inlet entrance and in the inlet duct, as well as oil-e lm visualization of the streamlines on compression surfaces were performed. Distributions of static pressure and heat e uxes on these surfaces in typical cross and longitudinal sections, and pitot-pressure distributions at the inlet duct exit, were obtained. Patterns of the supersonic e ow over the compression wedges and in the inlet duct were identie ed. A characteristic feature of these e ows, in particular, is what the compression wedges form shock waves glancing relative to the wedge surfaces. These shocks induce oblique separations of the boundary layer with intense helical vortices propagating far downstream. As a whole, a complex system of multiple shock and expansion waves and vortex structures occurs over the inlet section of external compression, so that the e owe eld at the inlet-duct entrance, in the duct, and at the exit is signie cantly nonuniform. Tests with a natural development of the boundary layer on compression surfaces and with tripping were carried out. “ Grater-type” trips were used, which demonstrated the high efe ciency at high Mach numbers M1 =6‐8. The unit Reynolds numbers in wind tunnels at the same test Mach number M1 o6 were different, Re11 o20£10 6 1/m in T-313 and Re11 =(2‐2.5)£10 6 1/m in IT-302M, respectively, but the boundary-layer tripping ensured identical e ow regimes determined by thestateof the boundary layer. Experimental data obtained in wind tunnels of different operation principles are in good qualitative agreement even with taking into account some quantitative difference mainly attributed to the unit-Reynolds-number difference.

Dynamic Modeling of Ring Nutation Dampers

Introduction R ECENTLY published analyses of nutation dampers are not widely available, but modeling damper performance is an interesting problem combiningmany disciplines. This Note presents a new perspective on nutation dampers such as the one examined in Ref. 1. There, the damper consisted of a hoop of metal partially Ž lled with a viscous  uid and mounted in the spacecraft spin plane, with both horizontal and vertical offsets from the spacecraft center of gravity. For small nutation angles the  uid slug is held intact and toward the outeredgeof thehoopby thecentrifugalforce.Small oscillations are induced at the nutation frequency, and viscous forces dissipate energy. Large nutation angles produce rectiŽ ed torques that propel the slug around the hoop. Although there is a component of steady  ow in this case, the dissipative torques at harmonics of the spin and nutation rates can play a signiŽ cant role in the overall damping of the nutation. The most important parameter for modeling the action of a  uid-based nutation damper is the kinetic Reynolds number.2 For unsteady  ow this quantity plays a similar role to the standard Reynolds number of steady  ow theory in deŽ ning regions of laminar and turbulent  ow. Unsteady  ow regimes typically dominate for low viscosity  uids. However, one low viscosity  uid that has been used extensively in nutation dampers, mercury, has the disadvantage of exhibiting highsurfacetension.Surface tensionhasbeendemonstratedto cause small angle “lock-up,” in which the damper ceases to dissipate energy below a particular nutation angle. This Note does not model effects of surface tension, but evaluation of the potential for small angle lock-up is recommended in general applications. The shear stress in an unsteady  ow regime is generally much larger than that in a steady  ow situation,and this fact can be used to advantage in designing faster nutation dampers.One salient feature of the unsteady  ow regime is that the damping turns out to be proportional to the square root of the  uid viscosity, rather than directly proportional as might be expected. A simple experimental apparatus is sufŽ cient to demonstrate the validity of the analysis.

Automated CFD Analysis of Two-Dimensional High-Lift Flows

Validation results are described for an automated process for Navier ‐Stokes analysis of multielement high-lift cone gurations in two dimensions. A systematic study of grid requirements for resolving high-lift e owe elds is presented, and guidelines are provided for generating grids suitable for resolving e ow regimes containing phenomena characterized by widely varying scales, for example, free and cone uent wakes and boundary layers, separation and reversed e ows. Sensitivities of computed solutions to variousgrid parameters aredocumented. Theautomated softwarepackageis validated for slat and e ap effectiveness studies, Reynolds-numbereffects, and gap sensitivities. Results obtained so far show promise of a capability for routine and automated analysis of complex high-lift e ows.

Subcritical Flow Around Bluff Bodies

Results of experimental investigations of vortex-shedding frequencies and surface pressures of circular and square cylinders with the same hydraulic diameter are described. Pressure measurements on circular and square cylindersandhot-e lm measurementsinthewakeswereconducted. TherangeofReynoldsnumberswas1 :3 £ 10 4‐ 2 £ 10 5 , which is characterized as subcritical e ow regimes. Also, to reveal the combined ine uence of Reynolds number and angle of incidence on the pressure and spectral density distributions of velocities, experiments were carried out on square and rectangular cylinders for the angle of incidence in the range 0 < ®< 45 deg. At zero incidence,pressurecoefe cientsonthefrontandsidefacesofthesquarecylinderchangeconsiderablywithReynolds number, whereas as the angle of incidenceincreasesto 30 deg, theeffectoftheReynolds numberseemsto diminish. Spectral density distributions in the wake and surface pressures depend on Reynolds number and the angle of incidence.Thevariationsofthestrengthofthespectraldensitieswith Reynoldsnumberinthewakesofcircularand square cylinders show different characteristics. Variations of spectral density distributions and vortex shedding with Reynolds number, geometric dimensions of models, and angle of incidence are presented.

CFD-Based Aeroelastic Eigensolver for the Subsonic, Transonic, and Supersonic Regimes

We describe a linearized computational e uid dynamics (CFD) method for computing an arbitrary number of eigensolutions of a given aeroelastic problem. The proposed method is based on the reengineering of a three-way coupled formulation previously developed for the solution in the time domain of nonlinear transient aeroelastic problems. It is applicable in the subsonic, transonic, and supersonic e ow regimes, and independently from the frequency or damping level of the target aeroelastic modes. It is based on the computation of the complex eigensolution of a carefully linearized e uid ‐structure interaction problem, relies on the inverse orthogonal iteration algorithm, and reutilizes existing unsteady e ow solvers. This is a validated method with the e utter analysis of the AGARD Wing 445.6 for which experimental data are available.

Three‐dimensional Spherical Simulations of Solar Convection. I. Differential Rotation and Pattern Evolution Achieved with Laminar and Turbulent States

Rotationally constrained convection possesses velocity correlations that transport momentum and drive mean —ows such as diUerential rotation. The nature of this transport can be very complex in turbu- lent —ow regimes, where large-scale, coherent vorticity structures and mean —ows can be established by smaller scale turbulence through inverse cascades. The dynamics of the highly turbulent solar convection zone therefore may be quite diUerent than in early global-scale numerical models, which were limited by computational resources to nearly laminar —ows. Recent progress in high-performance computing tech- nology and ongoing helioseismic investigations of the dynamics of the solar interior have motivated us to develop more sophisticated numerical models of global-scale solar convection. Here we report three- dimensional simulations of compressible, penetrative convection in rotating spherical shells in both laminar and turbulent parameter regimes. The convective structure in the laminar case is dominated by ii banana cells,ˇˇ but the turbulent case is much more complex, with an intricate, rapidly evolving down- —ow network in the upper convection zone and an intermittent, plume-dominated structure in the lower convection zone and overshoot region. Convective patterns generally propagate prograde at low lati- tudes and retrograde at high latitudes relative to the local rotation. The diUerential rotation pro—les show some similarity with helioseismic determinations of the solar rotation but still exhibit signi—cantly more cylindrical alignment. Strong, intermittent, vortical down—ow lanes and plumes play an important dynamical role in turbulent —ow regimes and are responsible for signi—cant diUerences relative to laminar —ows with regard to momentum and energy transport and to the structure of the overshoot region at the base of the convection zone. Subject headings: convectionhydrodynamicsstars: interiorsSun: interiorSun: rotation ¨ turbulence

Confined Swirling Flows with Heat Release and Mixing

A simplie ed model is used to illustrate global features and e ow regimes encountered in swirling e ows with combustion, as an approach to providing an overall interpretation of the behavior of primary zone e ows in leanpremixed-prevaporizedgasturbinecombustors.Asoneexampleofthis, thephysicalmechanismsthatcharacterize theinteractions between swirl and heataddition aredescribed, including the differing effects on recirculation zone formation for low and high values of swirl. The model, which is based on a quasi-one-dimensional control volume formulation, contains many approximations for the relevant e ow processes. Nevertheless, comparisons of this simplie ed model with results from an axisymmetric Navier ‐Stokes code support the utility of the approach for gaining physical insight into the overall behavior of these parametrically complex e ows.

Differential rotation of lithosphere and mantle andthe driving forces of plate tectonics

Uncertainties regarding the relative importance of basal drag and boundary in plate tectonic models are a consequence of ~awed assumptions imposed by the use of the hotspot reference frame[ Velocities of lithospheric are in~uenced not only by lateral boundary forces but also by basal drag resulting from Earth rotation[ Drag is exerted on the base of the and motion is transmitted upwards to the lithosphere[ However as the transmission of stress in the mantle is viscosity! dependent the reduction in viscosity through the results in su}ering a net westward lag[ This {di}erential rotation| causes continental to be more strongly coupled to the deep mantle as they are separated from the mesosphere by only relatively thin regions of asthenosphere[ For such plates the calculated drag are of the same order of magnitude as the boundary forces[ Intraplate volcanism during continental rifting and in opening is related to transverse convection cells set up where topographic structures in the lithospheric root oppose mantle ~ow[ The motion of oceanic is dominated by conventional plate boundary which may either reinforce or oppose drag from an eastward mantle ~ow[ Reinforcement e[g[ Nazca plate# gives rise to Couette ~ow in the asthenosphere[ Opposition e[g[ Paci_c plate# results in counter~ow[ Shear stresses in both regimes are concentrated in the upper and lead to melting of concentrations of hydrous minerals {wetspots|# derived from eroded continental mantle introduced by lateral asthenospheric ~ow[ Under a counter~ow regime melt collects in a stationary layer at shallow depth in the asthenosphere at the crossover point between plate! and mesosphere!induced ~ow regimes[ Release of melt to the surface is governed by lithospheric stress trajectories set up by convergence along plate boundaries[ Intraplate volcanism thus has a common source although due to di}erent interactions between the boundary and drag forces ~ow pro_les will di}er between basins giving relative motions between melting of a few centimetres per year[ The overall e}ect however is to give an illusion of a series of quasi!_xed melting anomalies though in reality all these and the lithospheric are moving relative to the deep mantle[ Þ 0888 Elsevier Science Ltd[ All rights reserved[ Corresponding author[ Tel[] 99775 5164 6464^ fax] 775 5163 9174^ e!mail] mochinmÝmail[ncku[edu[tw A[D[ Smith C[ Lewis : Geodynamics 17 0888# 86Ð005 87

Mixing and Spray Formation in Coaxial Jets

The e ow regimes and e ne-scale structure of the mixture resulting from the destabilization of a dense, slow jet by a fast light annular jet are discussed. From the primary shear instability between the streams, it is shown that the momentum ratio M = of the annular to the inner stream is the key parameter 2 2 r u /r u 2 2 1 1 on which the inner potential core length and the condition of a recirculation transition depend. The instability analysis shows that the initial wavelength of the disturbances l is proportional to the vorticity thickness d of the fast stream at the nozzle exit l ; d(r1/r 2) 1/2 ; this further provides, by an original scenario, an estimate of the droplets’ size formed by the capillary instability of the sheet’ s rim formed from the initial interface disturbances. Liquid viscosity is not expected to play any signie cant role in practical conditions in liquid rocket propellant engines. These e ndings are put in relation with a selected review of known or yet unexplained results on the subject. HERE is a frequent need to realize a uniform mixture from two initially segregated streams in many practical instances. A signie cant example is the case of liquid rocket propellant engines because the length of the combustion chamber is limited by the ability of the injecting devices to fragment and mix the reactants down to a sufe cient level of homogeneity for evaporation and combustion to completion at the desired distance from the injector outlet. For technological reasons one of the reactants is usually available in the liquid state and the other in the gas phase; for historical reasons the coaxial geometry is commonly used to merge the two streams. 1 The relative e ow rates of the reactants must be adjusted so that the global stoichiometry is at least respected, or such that the gas phase is in excess for all of the liquid to vaporize and burn, this implies that the gas stream is usually much more rapid than the liquid stream at the injector outlet. This is the situation encountered in H 2/O2 engines, where a slow, dense liquid oxygen (LOX) stream in the central jet of the coaxial injector is surrounded by a fast, light, gaseous hydrogen annular stream. At the root of the interpenetration process between the two phases is a strong shear destabilizing the central liquid jet, which further fragments into a more or less uniform spray. This process, which is contrasted with the case of a simple jet issuing in a quiescent environment, 2 is known as airblast atomization. 3

Effect of Angle of Attack on Hypersonic

An experimental investigation of the effects of angle of attack on hypersonic boundary-layer stability on a ared-cone model was conducted in the low-disturbance Mach-6 Nozzle-Test-Chamber Facility at NASA Langley Research Center. Hot-wire anemometry diagnostics were applied to identify the boundary-layer instability mechanisms that lead to transition. The present results show that the boundary layer becomes more stable on the windward ray and less stable on the leeward ray relative to the zero-degree angle-of-attackcase. The second-mode instability dominates the transition process at a 0-deg angle of attack; however, on the windward meridian at an angle of attack this mode was stabilized. On the leeward meridian the frequency of the dominant instability was higher than the estimated frequency of the second-mode disturbance; thus the dominant transition mechanism may be other than a second-modedisturbance. Nonlinear effects, such as growth saturation, harmonic generation, and spectral broadening, were observed in the transitional and turbulent ow regimes.

Mixed Convection in the Laminar Flow of Viscoelastic Liquids Through Rectangular Ducts

Flow patterns and heat transfer coefe cients are obtained for the fully developed e ow of viscoelastic liquid e owing laminarly inside ducts of rectangular cross sections. The effect of natural convection is analyzed. The results are obtained by integrating the conservation equations together with the Criminale‐ Ericksen‐ Filbey constitutive model, via a control-volume technique. Wide ranges of the relevant parameters are investigated. Different e ow regimes are observed, depending upon the relative importance of viscoelastic and buoyancy-driven forces. These changes in e ow regime are responsible for dramatic changes in the heat transfer behavior, which have been previously observed experimentally, as reported in the literature.