The nanofluid literature contains many claims of anomalous convective heat transfer enhancement in both turbulent and laminar flow. To put such claims to the test, we have performed a critical detailed analysis of the database reported in 12 nanofluid papers (8 on laminar flow and 4 on turbulent flow). The methodology accounted for both modeling and experimental uncertainties in the following way. The heat transfer coefficient for any given data set was calculated according to the established correlations (Dittus-Boelter’s for turbulent flow and Shah’s for laminar flow). The uncertainty in the correlation input parameters (i.e. nanofluid thermo-physical properties and flow rate) was propagated to get the uncertainty on the predicted heat transfer coefficient. The predicted and measured heat transfer coefficient values were then compared to each other. If they differed by more than their respective uncertainties, we judged the deviation anomalous. According to this methodology, it was found that in nanofluid laminar flow in fact there seems to be anomalous heat transfer enhancement in the entrance region, while the data are in agreement (within uncertainties) with the Shah’s correlation in the fully developed region. On the other hand, the turbulent flow data could be reconciled (within uncertainties) with the Dittus-Boelter’s correlation, once the temperature dependence of viscosity was included in the prediction of the Reynolds number. While this finding is plausible, it could not be conclusively confirmed, because most papers do not report information about the temperature dependence of the viscosity for their nanofluids.

To investigate the relationship between local burning velocity and flame displacement speed with different density ratio, a numerical analysis was performed using DNS databases of statistically steady and fully developed turbulent premixed flames. The local burning velocity based on fuel consumption rate is considered to be most appropriate as the definition of burning rate because combustion is a kind of chemical reactions. Since it is impossible, however, that the local burning velocity is obtained experimentally by using any present measurement technology, the numerical results using the local burning velocity cannot be compared with any experimental data. Hence the flame displacement speed, which can be obtained and compared easily with experiments, has previously been used for numerical analyses. Thus, to realize comparison of numerical results using the local burning velocity with some experimental ones, it is necessary to reveal the relationship between the local burning velocity and the flame displacement speed.

Heat transfer and pressure drop measurements for horizontal macro-tubes under uniform wall heat flux boundary condition have been conducted by various researchers in recent years. From their studies, it was shown that good agreements were observed in the laminar and turbulent regions. However, for the transition region, the heat transfer and pressure drop characteristics depended on various factors, such as inlet configuration, buoyancy effect, and surface roughness. In a recent study by Tam et al. (2010), they measured the heat transfer and pressure drop simultaneously for a horizontal macro-tube with and without internally micro-fins and concluded that under the heating condition, the transition Reynolds number range for heat transfer and pressure drop were completely different. The transition Reynolds number range was documented in their research in great detail. However, for horizontal micro-tubes, there is no information in the literature on the simultaneous behavior of the heat transfer and pressure drop, especially in the transition region. In order to fill in this gap, an experimental setup was built to measure the heat transfer and pressure drop simultaneously for a horizontal micro-tube under uniform wall heat flux boundary condition. Water was used as the test fluid and the test section was a stainless steel micro-tube with 1000μm diameter. For heat transfer, the results indicated that the micro-tube had an earlier start and end of transition compared to the macro-tube and, in the turbulent region, an increase in heat transfer due to the surface roughness was observed. For friction factor under isothermal condition, the micro-tube had a narrower transition range due to the roughness compared to the macro-tube. For friction factor under heating condition, the laminar data and the start of transition were different from the isothermal case, and the effect of heating was not seen on the end of transition.

Transportation accidents frequently involve liquids dispersing in the atmosphere. An example is that of aircraft impacts, which often result in spreading fuel and a subsequent fire. Predicting the resulting environment is of interest for design, safety, and forensic applications. This environment is challenging for many reasons, one among them being the disparate time and length scales that must be resolved for an accurate physical representation of the problem. A recent computational method appropriate for this class of problems has been developed for modeling the impact and subsequent liquid spread. This involves coupling a structural dynamics code to a turbulent computational fluid mechanics reacting flow code. Because the environment intended to be simulated with this capability is difficult to instrument and costly to test, the existing validation data are of limited scope, relevance, and quality. A rocket sled test is being performed where a scoop moving through a water channel is being used to brake a pusher sled. We plan to instrument this test to provide appropriate scale data for validating the new modeling capability. The intent is to get high fidelity data on the break-up and evaporation of the water that is ejected from the channel as the sled is braking. These two elements are critical to fireball formation for this type of event involving fuel in the place of water. We demonstrate our capability in this paper by describing the pre-test predictions which are used to locate instrumentation for the actual test. We also present a sensitivity analysis to understand the implications of length scale assumptions on the prediction results.

Cooling of supercritical CH 4 /N 2 mixture is the most important heat transfer process during coalbed methane (CBM) liquefaction. In this paper, numerical studies of the turbulent convective heat transfer of supercritical CH 4 /N 2 flowing inside a vertical circular tube has been conducted with Lam-Bremhorst low Reynolds turbulence model. The present numerical investigations focus on the effects of the nitrogen content, heat flux and flow orientation. Results indicate that as nitrogen content increases, the maximum heat transfer coefficient gradually decreases and corresponds to lower temperature. Heat transfer coefficient is slightly affected by heat flux in the liquid-like region, and increases with increasing heat flux in the gas-like region. Buoyancy effect gradually increases with decreasing bulk temperature, and reaches its maximum at the pseudo-critical point, and then drops as bulk temperature further decreases. It is significant in the liquid-like region, and negligible in the gas-like region. At the same time, buoyancy effect enhances heat transfer in the upward flow and impairs it in the downward flow.

A lean premixed CH 4 air flame (LPF) impinges with a CH 4 diluted with N 2 diffusion flame (DF) having different turbulence conditions to create a lean heterogeneous combustion model such as a stratified combustion. The local quenching recovery processes of LPF and DF interacting with the turbulence in an opposed flow have been investigated experimentally using a Particle Image Velocimetry movie. The local quenching phenomena can be observed frequently with approaching the global extinction condition. The local quenching may trigger to global extinction. However, in many cases, the flame can recover from the local quenching phenomena and create the stable flame. There are three distinct local quenching recovery mechanisms namely a passive mode, an active mode, and an eddy transportation mode. These three modes depend on the local flame propagation mechanism, the bulk flow motion, and the eddy motion by turbulence. In the passive mode, the bulk flow plays an important role on the recovery process. The local quenching area is drifting outward from the stabilization point by the bulk flow and then, it is displaced by the stable flamelets. In the active mode, the local quenching area is recovered by the self-propagating wrinkled LPF from somewhere in the active zone. The active mode is observed only when the turbulence is added to the premixed flame side. In the eddy motion mode, the local quenching area is recovered by the eddy transportation. That is, the flamelet is transport by the eddy motion and the local quenching area is replaced. The wrinkled flamelet having self-propagation plays a very important role for the local quenching recovery mechanism. The turbulence on the premixed flame not only induces high possibility for the local quenching but also helps to recover from the local quenching.

Three-dimensional direct numerical simulation (DNS) with a detailed kinetic mechanism has been conducted for statistically-planar turbulent flame and turbulent V-flame of hydrogen–air mixture to clarify the effects of mean flow velocity on principal strain rates at flame front and on flame geometry. Reynolds numbers based on Taylor micro scale and turbulent intensity are selected to 60.8 and 97.1, and mean flow velocities for V-flame are 10 and 20 times laminar burning velocity. From results of DNS, eigenvalues and eigenvectors of strain tensor are evaluated to investigate characteristics of strain field near flame and flame normal alignments with the principal axes of strain in detail. It has been revealed that Reynolds number affects both magnitude of strain rates and alignment between flame normal and principal axis of strain, and that the magnitude of mean flow velocity affects flame normal alignments in turbulent V-flame.

Flow and heat transfer characteristics of TBAB hydrate slurry were investigated experimentally. The Reynolds number, diameter of the tubes and solid fraction were varied as experimental parameters. For laminar flow condition, it was found that the ratio of the coefficients of pipe friction and Nusselt numbers increases with solid fraction, and the rate of increase is high in the case of a low Reynolds number. For turbulent flow condition, the ratio of the coefficients of pipe friction and Nusselt numbers was 1 for each condition in the case of a low solid fraction. On the other hand, the ratio of the Nusselt numbers increased with the solid fraction in the high-solid fraction region. Moreover, it was found that the effects of the difference of the size and shape of hydrate particles on the coefficients of pipe friction are large. On the other hand, the effects of the difference of the hydrate particles on the Nusselt number are small.

Three-dimensional direct numerical simulation (DNS) of turbulent hydrogen-air premixed flames in a constant volume vessel at relatively high Reynolds numbers have been conducted considering detailed kinetic mechanism and temperature dependence of the transport and thermal properties. The flame behavior and heat transfer characteristics are investigated in the vessel. The flame is strongly affected by the growth of the internal pressure which is caused by the temperature rise in the vessel. Since the pressure increase makes the flame thickness thin, the heat release rate of each flame element is augmented. The local pressure rise due to the dilatation also enhances turbulence and finer scale vortices appear, which make the flame surface more complicated and result in an increase of the flame surface area. Due to the increase of the mean pressure in the vessel, the maximum wall heat flux induced by the flame front is enhanced during the combustion.

Circulating water systems (CW) and safety water systems (SW) in various power plants use vertical pumps to pump water from pump intakes. A properly designed pump intake structure prevents the occurrence of strong surface vortices, which might inhibit the proper functioning of the pump. Although several standards for experimental testing of pump intake structure suitability exist, our goal is to find a way to predict such vortices numerically, from a single-phase simulation. In such a process, we had already eliminated some of the turbulence models. In the current paper we confirm that Scale Adaptive Simulation (SAS) turbulence model with the curvature correction (CC) factor applied is well suited for such flows. By using a methodology for determining the vortex air core length, the SAS-CC turbulence model results were compared to the experimental data for two selected temperatures. The results show better agreement than the laminar simulations in terms of higher mean value accuracy and lower scattering.

A two-dimensional numerical analysis is made on the effect of aspect ratio on flow and heat transfer characteristics of natural convection in a slender fluid layer enclosed between two vertical plates of different temperatures. It is shown that the boundary layer instability induces the hook-shaped flows and that its growth disrupts the large-scale circulatory flow into cellular flows of smaller scales, which in part include the renewed boundary layers. Then it is suggested that there exists a transition region where the average heat transfer coefficient increases with the increase of the height of a vertical fluid layer, as is seen in the transition region from laminar to turbulent natural convection along a single vertical plate. Discussion is also made on the nature of instability to yield cellular flows of smaller scales.

The accuracy and computational expense of various radiation models in the simulation of turbulent jet flames are compared. Both nonluminous and luminous methane-air non-premixed turbulent jet flames are simulated using a comprehensive combustion solver. The combustion solver consists of a finite-volume/probability density function-based flow–chemistry solver interfaced with a high-accuracy spectral radiation solver. Flame simulations were performed using various k -distribution-based spectral models and radiative transfer equation (RTE) solvers, such as P-1 , P-3 , finite volume/discrete ordinates method (FVM/DOM), and Photon Monte Carlo (PMC) methods, with/without the consideration of turbulence-radiation interaction (TRI). TRI is found to drop the peak temperature by close to 150 K for a luminous flame (optically thicker) and 25–100 K for a nonluminous flame (optically thinner). RTE solvers are observed to have stronger effects on peak flame temperature, total radiant heat source and NO emission than the spectral models. P-1 is found to be the computationally least expensive RTE solver and the FVM the most expensive for any spectral model. For optically thinner flames all radiation models yield excellent accuracy. For optically thicker flames P-3 and FVM predict radiation more accurately than the P-1 method when compared to the benchmark line-by-line (LBL) PMC.

With increasing requirements for model validation when comparing computational and experimental results, there is a need to incorporate detailed representations of measurement devices within the computational simulations. Thermocouples are the most common temperature measurement transducers in flames and fire environments. Even for the relatively simple thermocouple transducer, the coupling of heat transfer mechanisms particularly under unsteady flow conditions leads to interesting dynamics. As experimentalists are well aware, the experimentally determined thermocouple values are not the same as the local gas temperatures and corrections are often required. From the computational perspective, it is improper then to assume that the predicted gas temperatures should be the same as the temperatures that an experimentalist might measure since the thermal characteristics of the thermocouple influence the indicated temperature. In this study we investigate the thermal characteristics of simulated thermocouples in unsteady flame conditions. Validation exercises are presented to test the underlying thermocouple model. Differences are noted between the predicted thermocouple response and expected response. These differences are interpreted from the perspective of what modeling artifacts might drive the differences.

Detached turbulent flows are difficult to predict numerically and often serve as benchmark cases for developing new numerical schemes and new turbulent models. Turbulent flow over periodic hills is one such examples, since the flow exhibits separation and reattachment on a smoothly and/or sharp curved geometry, strong pressure gradients and fluctuation of the separation point in time. These cases have been chosen by many authors for testing different turbulence simulation approaches. When the bottom wall is heated, the complexity of the problem increased, since convective heat transfer is defined by small scale turbulent structures close to the wall. We developed a Reynolds-Averaged Navier-Stokes and Large Eddy Simulation solver based on the velocity-vorticity formulation of Navier Stokes equations. RANS equations are coupled by a low-Reynolds number turbulent model, while Smagorinsky subgrid model is used for LES. The governing equations are solved with a numerical solution algorithm, which is based on the boundary element method. The pressure field is computed in a post processing step by solving a Poisson equation. The single domain as well as domain decomposition approaches are applied. The developed method was validated using flow over periodic hills test case.

Fast evolving techniques for macroscale graphene and ultrathin graphite material production are promising for applications of graphene-based materials in thermal management. A numerical comparison of aluminium and graphene-based plate-fin heat exchangers is conducted. Anisotropic thermal conductivity of graphene-based solution shows an improvement of up to twenty percent in heat rejection over the aluminium design. Thermal and hydraulic performance is characterized for both designs over a range of air flow rates in both laminar and turbulent regimes. Steady and unsteady 3-D conjugate simulations reveal a faster equilibration rate for the graphene-based solution, minimizing thermal lag that must be accounted for in on-demand electronics cooling. The combination of improved heat rejection, rapid response rate, and low material density make a graphene-based solution uniquely suited to aerospace thermal management.

The horizontal frame members that often protrude from the inner surface of a window can significantly effect the convective heat transfer rate from this inner surface to the room. The purpose of the present numerical study was to determine how the size of a pair of horizontal frame members effect this heat transfer rate. The flow has been assumed to be steady and conditions under which laminar, transitional, and turbulent flows occur are considered. Fluid properties have been assumed constant except for the density change with temperature that gives rise to the buoyancy forces, this being dealt with using the Boussinesq approach. The governing equations have been solved using the FLUENT commercial CFD code. The k-epsilon turbulence model with standard wall functions and with buoyancy force effects fully accounted for has been used. The solution has the following parameters: the Rayleigh number, the Prandtl number, the dimensionless window recess depth, and the dimensionless width and depth of the frame members. Results have been obtained for a Prandtl number of 0.74.

Importance of turbulence and radiation interaction (TRI) has been investigated in a turbulent channel flow by using direct numerical simulation (DNS) to clarify detailed turbulent flow structure and heat transfer mechanisms. To investigate the effect of correlation functions between gas absorption and temperature fluctuation, the two cases of correlation are tested. Consequently, the TRI effect can be clearly observed when the correlation is positive. This fact provides the evidence that radiative intensity is enhanced by the turbulent fluctuation. The DNS results suggest the significance in the fundamental aspect of TRI. Furthermore, effects of frictional Reynolds number, Re τ , are investigated. Comparing with the case of Re τ = 150, the location of the enhancement peaks of Re τ = 300 shifts toward the walls. It is found that the relative importance of the TRI correspond to the structure of temperature fluctuation intensity originated from the differences of the Re τ .

Evaporation heat transfer from the hot water flow to the cold air flow in a horizontal and rectangular flow channel was examined. The water temperature was 35°C ∼ 65°C. The air velocity was 0.02 m/s ∼2.57 m/s. The heat transfer rate from the water flow to the air flow became large with an increase in the air velocity and the water temperature. The evaporation heat transfer was much larger than the convection heat transfer and dominant in the heat transfer. The ratio of the evaporation heat transfer rate to the total heat transfer rate was approximately 0.9 ∼ 0.7 in the present experimental conditions. It showed the slightly decreasing tendency for the air velocity. The evaporation heat transfer coefficient showed strong dependency on the air velocity in both the laminar and the turbulent flow region of the air flow. The convection heat transfer coefficient showed the same tendency for the Reynolds number of the air flow as that for the air single-phase flow in the turbulent flow region although the value was much larger than that of the single-phase flow. In the laminar flow region, the convection heat transfer coefficient was constant as in the single-phase flow when the water temperature was low, although the value itself was much larger than that of the single-phase flow. As the water temperature became high, the convection heat transfer coefficient became large and showing dependency on the Reynolds number of the air flow. As the Reynolds number of the air flow became further small, the convection heart transfer coefficient greatly decreased irrespective of the water temperature.

For the investigation of turbulent structure in drag reducing flow with polymer solution blown from the channel wall (wall blowing), instantaneous velocity field has been precisely measured in the x-z plane at different locations along the wall-normal direction via Particle Image Velocimetry (PIV). Polymer solutions with 25 ppm and 100 ppm of weight concentration were tested at a blowing ratio of 1.2×10 −4 and at 20000 of Reynolds number. About 5% and 11% of drag reduction (DR) rate was obtained, respectively. As a result of this experiment, turbulent statistic data showed that the Root Mean Square (RMS) of streamwise velocity fluctuation increased and RMS of spanwise velocity fluctuation was suppressed comparing with water flow. We found that these low-speed streaks became relatively regular in the buffer layer, including an increase of both length and width, which indicated a depression of turbulence by polymer diffused in the buffer layer.

Simultaneous Particle Image Velocimetry (PIV) measurement and Planar Laser Induced Fluorescence (PLIF) measurement at the same position were performed to clarify the relationship between spatial structure and mass transfer in the drag reducing surfactant flow. In the drag reducing flow, mass flux is largely suppressed in the near-wall region with increasing drag reduction rate. To discuss the relationship between coherent motion and drag reduction more detail, weighted probability density function was also calculated. As a result of simultaneous measurement, diffusion of wall-normal direction is largely suppressed and this indicated that turbulent coherent structure changes and sweep and ejection which produce the skin frictional drag are suppressed.