Effect of melt superheat on heat transfer coefficient for aluminium solidifying against copper chill

Solidification of metal castings inside moulds is mainly dependent on the heat flow from the metal to the mould which is in turn proportional to an overall heat transfer coefficient h which includes all resistances to heat flow such as the presence of an air gap. In the present work the heat transfer coefficient is determined using a directional solidification set-up with end chill for solidifying commercial-purity aluminium with different superheats (40 K and 115 K) against copper chill. A computer program solving the heat conduction and convection in the solidifying metal is used together with the experimental temperature history in order to determine the heat transfer coefficient at the interface. The variation of h as a function of time, surface temperature and gap temperature for each melt superheat is found. The results indicate that h reaches a maximum value for surface temperature close to the liquidus. The analysis of heat flux from the metal to the mould indicates that it is mainly by conduction. The air gap size is evaluated with time, surface temperature and with melt superheat. It is found that higher h values and smaller gap sizes are obtained with higher superheats.     

Experimental study of laminar natural convection heat transfer from a vertical cylinder to slush nitrogen under constant heat flux conditions

Natural convection heat transfer from a vertical cylinder immersed in slush and subcooled liquid nitrogen and subjected to constant heat fluxes was investigated in order to determine the relative merits of slush nitrogen (SlN₂) for immersion cooling. A glass dewar was used as a test vessel in which a cylindrical heater was mounted vertically, and heat transfer measurements were carried out for SlN₂ and subcooled liquid nitrogen (LN₂) in the laminar flow range. The results revealed advantages of SlN₂ over subcooled LN₂ in natural convection cooling. The local temperatures of the heated surface surrounded by solid nitrogen particles are measured to increase at much slower rates than in subcooled LN₂, which is due to the latent heat of fusion of solid nitrogen. Even after the solid nitrogen particles surrounding the heater are apparently depleted, the average heat transfer coefficients for SlN₂ are still found to be greater than those for LN₂ with the improvement in heat transfer being larger for lower Grashof number regime. Our analysis also indicates that solid nitrogen particles in close proximity to heated surface do not discourage local convection due to the porous nature of SlN₂, making the heat transfer in SlN₂ more effective than in the case of solid–liquid phase change of nitrogen involving melting and conduction processes.     

Effect of bath and specimen temperature on the thermal stress resistance of brittle ceramics subjected to thermal quenching

The effect of specimen and bath temperature on the failure of brittle ceramics in a thermal quench experiment was studied by quenching glass and alumina rods in water and silicon oil baths at different temperatures. The results were discussed in terms of the variation of heat transfer coefficient of the quenching media and the change in material properties as a function of temperature. It was found that the usual assumption of constant heat transfer coefficient and material properties may lead to considerable errors in the quantitative interpretation of the results of thermal quench experiments. Effective values for the film coefficient of heat transfer for water and oil baths were estimated as a function of film temperature from thermal quench data. Recommendations were made for the selection of quenching media and for the procedure to be followed in reporting the results.     

Estimation of the Heat Transfer Coefficient in a Liquid–Solid Fluidized Bed Using an Artificial Neural Network

A non-iterative procedure has been developed using an artificial neural network (ANN) for estimating the fluid–particle heat transfer coefficient, h_fp, in a liquid–solid fluidized system. It is assumed that in a liquid–solid system, the liquid temperature is time dependent, and the input parameters and output parameters for the ANN are considered on a linear scale. The output configuration yields an optimal ANN model with 10 neurons in each of the three hidden layers. This configuration is capable of predicting the value of Bi in the range of 0.1–10 with an error of less than 3%. The heat transfer coefficient estimated using the ANN has been compared with the data reported in the literature and found to match satisfactorily.© Koninklijke Brill NV, Leiden and Society of Powder Technology, Japan, 2008     

Heat transfer characteristics in centrifuge melt-spinning

Centrifuge melt-spinning (CMS) is a new technique for the production of rapidly solidified metallic ribbons. In CMS, centrifugal forces are used twice: to eject the liquid melt on to the quenching substrate (a copper rim) by rotation of the casting crucible, and to ensure prolonged contact of the solidifying ribbon with the heat extraction sink by making the quenching rim rotate too, in the opposite direction. The heat transport in CMS has a Newtonian nature, as it can be considered as a constant-resistance heat transfer process. Calculated heat transfer coefficients h range between (1.55 to 4.30)×10^–6 W m^–2 sec^–1, a half to one order of magnitude higher than for conventional melt-spinning. Increasing the ejection pressure from 1.8 to 269 kPa causes the apparent heat transfer coefficient to increase by a factor of three. Conversely to conventional melt-spinning, two additional phenomena contribute to the heat transfer characteristics in CMS at high extraction velocities: forced convection and mechanical dragging of the melt. The overall effect is a net improvement of the heat transfer ability in CMS as compared to conventional melt-spinning.     

Experimental research and numerical simulation on cryogenic line chill-down process

The empirical heat transfer correlations are suggested for the fast cool down process of the cryogenic transfer line from room temperature to cryogenic temperature. The correlations include the heat transfer coefficient (HTC) correlations for single-phase gas convection and film boiling regimes, minimum heat flux (MHF) temperature, critical heat flux (CHF) temperature and CHF. The correlations are obtained from the experimental measurements. The experiments are conducted on a 12.7 mm outer diameter (OD), 1.25 mm wall thickness and 7 m long stainless steel horizontal pipe with liquid nitrogen (LN₂). The effect of the lengthwise position is verified by measuring the temperature profiles in near the inlet and the outlet of the transfer line. The newly suggested heat transfer correlations are applied to the one-dimensional homogeneous transient model to simulate the cryogenic line chill-down process, and the chill-down time and the cryogen consumption are well predicted in the mass flux range from 26.0 kg/m² s to 73.6 kg/m² s through the correlations.     

Study on the effect of refrigerant distributing nonuniformity on the performance of falling-film evaporator with liquid recirculation system

To investigate the effect of liquid distributing nonuniformity on the heat transfer of horizontal-tube falling film, a horizontal-tube falling-film air-cooled water chiller with an ejector liquid recirculation system (LRS) is presented, and then the finite difference model of the horizontal-tube falling-film evaporator is given. Also a new method of dividing the liquid refrigerant distributor into several subzones is proposed as well as the model of liquid refrigerant distributing nonuniformity is established. The analysis results show that increasing the liquid spraying flowrate in a certain range is obviously valid for enhancing the evaporator capacity, and the suitable R_l is about 1.2, the corresponding value of φ is about 0.85. The analysis results also suggest that the higher the liquid recirculating ratio, the more the turning point of nonuniformity coefficient, and the turning point of nonuniformity coefficient is 0.1483 when the liquid recirculating ratio is equal to 1.21. The performance of falling-film heat transfer is more sensitive to the number of subzones under the condition of higher nonuniformity coefficient, and a suggested optimal number of partitioned subzones is 4.     

Complex Model of the Efficiency of Rectification Plates. 3. Countercurrent Motion of Phases in Partial Mixing of a Liquid

A model of mass transfer in the countercurrent motion and partial mixing of a liquid on a plate is analyzed. The liquid is considered to be consisting of two parts, one of which () is totally mixed and the other of which (1 – ) moves in the regime of ideal displacement. The intensity of the mixing is determined by the relation of these parts. In the model, the compositions of the flows on ideal and actual plates at a certain distance h for the vapor and h ₁ for the liquid from the site of injection of the phases coincide; the vapor leaving an ideal plate and the liquid arriving at it are in equilibrium. The dependences of the difference of the concentrations of a highly volatile component in the liquid before and after the plate on the efficiency, the flow rates of the vapor and the liquid, and the coefficient of equilibrium are derived. Particular cases where h = h ₁ and h = h ₁ = 0.5 and where they are equal to zero or to unity are analyzed. Corresponding relations of the efficiencies are derived. The complex countercurrent model is analyzed with allowance for the mixing and for the fields of its application.     

Heat transfer enhancement in distribution transformers using TiO2 nanoparticles

A transformer is an electrical device that transfers electrical energy between windings by electromagnetic induction while producing a considerable amount of heat in circuits. The heat produced in windings is removed by an appropriate heat transfer fluid such as liquid dielectrics. The cooling and insulating of a liquid dielectric depend on the properties of the oil filling the transformer. One of the approaches to enhance the thermal and dielectric properties of transformer oil is employing an appropriate nanoparticle in a transformer.In this paper, a three-phase distribution transformer is simulated three-dimensionally in order to study the heat transfer efficiency for pure oil (single phase) and nanofluid (TiO₂ nanoparticles- transformer oil). For both models, the electromagnetic field in solid sections and heat transfer in fluid and solid sections of the transformer are simultaneously investigated. The simulation results show that the presence of TiO₂ nanoparticles in the transformer oil increases the heat transfer coefficient, i.e. adding 1% (vol/vol) of TiO₂ nanoparticles to the transformer oil increases the Nusselt number from 2.17 to 2.49, while the maximum temperature of transformer components decreases from 47.20?°C to 43.05?°C.     

Thiobacillus denitrificans immobilized biotrickling filter for NO2 removal

Nitrogen dioxide (NO₂) removal efficiency of a biotrikling filter was evaluated under different operating conditions. Activated alumina (AA) was used as the immobilization matrix for Thiobacillus denitrificans (T. denitrificans) in the biotrickling filter. Batch studies were conducted to find out the degradation kinetics of nitrate and nitrite for a concentration range of 600–10,000 mg/L expressed as nitrogen. Nitrite exhibited maximum degradation rate followed by nitrate. Electron acceptor in the form of NO₂ gas showed least removal efficiency. Bio-kinetic parameters for T. denitrificans, by utilizing nitrate and nitrite as electron acceptors, were also evaluated. The μ_max (Maximum specific growth rate) and Y_T (Yield coefficient) values for T. denitrificans in the presence of nitrate and nitrite were 1.03 h⁻¹, 0.275 and 0.63 h⁻¹, 0.1316 respectively. Column study was conducted to find the adsorption and desorption potential of activated alumina. The adsorbed NO₂ from AA could easily be desorbed using distilled water with an efficiency of 76±0.8%. Once fed batch studies were conducted to evaluate the NO₂ removal efficiency by a biotrickling filter. With an influent NO₂ gas concentration of 2,735 ppm, the reactor could achieve a removal efficiency of 99% within 2 min from gas phase and within 96 h from the liquid phase, with an average biomass concentration of 200 mg/g of AA. The mechanism of NO₂ gas removal in the biotrickling filter seems to be the dissolution of NO₂ in water to form NO₃⁻, conversion of NO₃⁻ to NO₂ ⁻, and finally to N₂ gas.     

Structure and properties of highly crystallized graphite films based on polyimide Kapton

Highly crystallized graphite films were prepared by heat treatment of carbonized polyimide films (Kapton) at temperatures of 2700 and 3050° C. Interlayer spacing d ₀₀₂ at room temperature, and electrical resistivity, magnetoresistance and Hall coefficient at room and liquid nitrogen temperatures were measured. All of these data indicate high crystallinity of the graphitized Kapton films obtained. For the graphite films heat treated at 3050° C mean-square mobilities were estimated from the magnetoresistance data at 1 T to be 0.91 m² V^–1 sec^–1 at room temperature and 2.3 m² V^–1 sec^–1 at liquid nitrogen temperature; the value at liquid nitrogen temperature corresponds to that for a pyrolytic graphite heat treated at 3200° C for 1 h (PG 3200). Magnetic field dependence of Hall coefficient at liquid nitrogen temperature for this sample also agrees well with that for PG 3200. Scanning electron micrographs on the surfaces show that the present graphite films consist of grains of large crystallites, and grain size increases as the crystallinity of the material improves.     

Transverse heat transfer coefficient in the dual channel ITER TF CICCs. Part III: Direct method of assessment

The data resulting from the thermal-hydraulic test of the ITER TF CICC are used to determine the flow partition and the overall effective heat transfer coefficient (h_BC) between bundle and central channel in a direct way, i.e. by analysis of the heat transfer between both flow channels, based on the mass and energy balance equations and the readings of thermometers located inside the cable. In cases without a local heat source in the considered cable segment the obtained h_BC values were consistent with those obtained in earlier studies by analysis of experimental data using indirect methods. It was also observed that the transverse heat transfer was strongly enhanced in a cable segment heated from outside. This phenomenon results from the mass transfer from the bundle region to the central channel. The experimental h_BC data obtained for the case without a heat source in the considered segment were also compared with those calculated using various heat transfer correlations.     

The cryopumping of water vapour in the continuum pressure region

Pumping speed measurements in the continuum pressure region, P>10^?3 torr, have been made for water vapour impinging on copper spheres and coils cooled to liquid nitrogen temperatures. Water vapour flow rates between 0.06 mg s^?1 and 420 mg s^?1 were used. The volumetric pumping speed was constant over the pressure range 2 × 10^?3 torr to 2 × 10^?2 torr and was, as expected, higher than that obtained in the free molecular flow region. Above 2 × 10^?2 torr the pumping speed decreased and possible reasons for this were investigated and are discussed. These included the effects of inadequate heat transfer from the liquid nitrogen refrigerant to the cryopump, a poor thermal conductivity of the cryodeposit, and an impurity, nitrogen gas, in the water vapour.     

Influences of solute content, melt superheat and growth direction on the transient metal/mold interfacial heat transfer coefficient during solidification of Sn–Pb alloys

Several factors such as alloy composition, melt superheat, mold material, roughness of inner mold surface, mold coating layer, etc., can affect the transient metal/mold heat transfer coefficient, h_i. An accurate casting solidification model should be able to unequivocally consider these effects on h_i determination. After this previous knowledge on interfacial heat transfer, such models might be used to control the process based on thermal and operational parameters and to predict microstructure which affects casting final properties. In the present work, three different directional solidification systems were designed in such a way that thermal data could be monitored no matter what configuration was tested with respect to the gravity vector: vertical upward and downward or horizontal. Experiments were carried-out with Sn–Pb hypoeutectic alloys (5 wt.% Pb, 10 wt.% Pb, 15 wt.% Pb and 30 wt.% Pb) for investigating the influence of solute content, growth direction and melt superheat on h_i values. The experimentally obtained temperatures were used by a numerical technique in order to determine time-varying h_i values. It was found that h_i rises with decreasing lead content of the alloy, and that h_i profiles can be affected by the initial melt temperature distribution.     

An inverse problem of simultaneously estimating contact conductance and heat transfer coefficient of exhaust gases between engine's exhaust valve and seat

The conjugate gradient method using two search step sizes is used to solve the inverse problem of simultaneously estimating the periodic thermal contact conductance, h_c(t), and the heat transfer coefficient of the exhaust gases, h_g(t), between the exhaust valve and seat in an internal combustion engine. The importance of the determination of h_c(t) and h_g(t) lie in that they are the critical factors for designing the cooling system and the insulation of the exhaust valve. The inverse analysis is based on the temperature measurements taken from the sensors placed in both the valve and seat regions during the transient process of operation. In this study two unknown timewise-varying functions h_c(t) and h_g(t) are to be estimated at the same time, thus two search step sizes with each one corresponding to each unknown function are derived. The results show that the CPU time for the inverse solutions using two search step sizes are greatly reduced than using just one search step size¹ for the determination of two unknowns, besides, it also shows that the inverse solutions are reliable even when the measurement errors are considered. The advantage of the conjugate gradient method is that no a priori information is needed on the variation of the unknown quantities, since the solution automatically determines the functional form over the domain specified. The successful development of the present technique can be applied to any kind of two-dimensional periodic contact problems, such as the determination of a two-dimensional contact conductance problem² and the temperature or heat flux behaviour on the inside wall of internal combustion engines³.     

Effect of lubricating oil on cooling heat transfer of supercritical carbon dioxide

In this research, the cooling heat transfer coefficient and pressure drop of supercritical CO₂ with PAG-type lubricating oil entrained were experimentally investigated. The inner diameter of the test tubes ranged from 1 to 6 mm. The experiments were conducted at lubricating oil concentrations from 0 to 5%, pressures from 8 to 10 MPa, mass fluxes from 200 to 1200 kg m⁻² s⁻¹, and heat fluxes from 12 to 24 kW m⁻².In comparison to the oil-free condition, when lubricating oil entrainment occurred, the heat transfer coefficient decreased and the pressure drop increased. The maximum reduction in the heat transfer coefficients—about 75%—occurred in the vicinity of the pseudocritical temperature. The influence of oil was significant for a small tube diameter and a large oil concentration. From visual observation, it was confirmed that this degradation in the heat transfer was due to the formation of an oil-rich layer along the inner wall of the test tube. However, when the oil concentration exceeded 3%, no further degradation in the heat transfer coefficient could be confirmed, which implies that the oil flowing along with CO₂ in the bulk region does not influence the heat transfer coefficient and the pressure drops significantly. For a large tube at a lower mass flux, no significant degradation in the heat transfer coefficient was observed until the oil concentration reached 1%. This is due to the transition of the flow pattern from an annular-dispersed flow to a wavy flow for a large tube, with CO₂ flowing on the upper side and the oil-rich layer on the lower side of the test section.     

Complex Model of the Efficiency of Rectification Plates. 4. Cross Motion of Phases

A model of mass transfer in the cross motion of phases is proposed in which the compositions of the flows on ideal and real plates at a certain distance h for a vapor and h ₁ for a liquid from the site of their injection are equal. The composition of the vapor after a plate is equal to the average value of the compositions of the vapor after the initial and final portions in the direction of liquid motion that, for an ideal plate, are in equilibrium with, respectively, the arriving and discharging liquid. Dependences of the efficiency in the vapor and liquid phases on the parameters of a real plate that are identical and confirm the equality of these efficiencies are derived. An analysis of particular cases of the model for h = h ₁ and h = h ₁ = 0.5 and for their boundary values when they are equal to zero or to unity is made. Corresponding relations of the efficiencies are obtained.     

Heat transfer from metallic filaments (whiskers) into helium under overcritical pressure for temperatures between 3.7 and 7.2 K

The heat transfer coefficient α, near the critical temperature, T_cO, was determined for several whiskers from the In-Pb alloy system. For this purpose the hysteresis of the voltage-temperature (V-T_B) transition curves at fixed currents, I, and of the V-I characteristics at fixed helium bath temperature, T_B, was determined. The advantage of using measurements made with whiskers is that there is no heat transfer to a substrate and negligible heat transfer to the contracts. The only heat transfer is that to the surrounding helium.     

Heat Transfer in a Packed Bed at Moderate Values of the Reynolds Number

Results are given of an experimental investigation of the coefficient of wall heat transfer of a round tube filled with a packed medium formed by monodisperse glass spheres of different diameters (d _p = 0.9, 3.2, 8.9 mm) in a stabilized region of heat transfer under conditions of filtering of water and aqueous solution of glycerin. A two-layer model of heat transfer is used to calculate the contribution made by the heat resistance of the flow core and of the wall zone using the measured coefficients of heat transfer and temperature profiles across the packed bed. The form of dependence for the effective coefficient of thermal conductivity is determined. Data are given of the measurement of the coefficient of wall heat transfer of annular channel filled both with a single layer of spheres with packing of two types (cubic and rhombohedral) and with several layers of spheres with random packing in a stabilized region of heat transfer under conditions of filtering of water. It is demonstrated that, in the case of inertial mode of filtering of liquid through the packed bed, the values of the Nusselt number both in the tube and in the annular channel correspond to the relation Nu _e (d _e/D)Pe^1/2 _e . A semi-empirical correlation is suggested, which generalizes well our experimental data (and the data of other authors) on heat transfer in the tube and in the annular channel. A theoretical model is suggested, according to which the variation of heat transfer is defined by the behavior of the effective coefficient of thermal conductivity _ef/ _f Pe^1/2 _d associated with the predominant contribution made to convective heat transfer by the transport processes in vortex cells with closed lines of flow.     

Heat transfer from Ni–W tapes in liquid nitrogen at different orientations in the field of gravity

Ni–W tapes of the micrometric thickness are considered as the basis for the cost-effective manufacturing of coated conductors – the 2nd generation of high-temperature superconductor (HTS). Many HTS applications involve widely-available and inexpensive liquid nitrogen. The transition from superconducting to normal state may however occurs due to unexpected temperature fluctuations. In this case Ni–W tape is significantly heated by electrical current propagating through it. The amount of heat transferred from the tape to coolant is defined by heat transfer from the surface of tape to liquid nitrogen. The heat transfer, in turn, is strongly dependent on the tape orientation in the field of gravity. The present paper reports the experimental results on the heat transfer from Ni–W tape to a pool of liquid nitrogen. The heat transfer coefficients are quantified for three subsequent heat transfer regimes: natural convection of liquid nitrogen, nucleate boiling regime and film boiling. The dependence of heat transfer coefficient on inclination angle of the tape from vertical are experimentally clarified for each regime. The expression for the heat transfer coefficient at different inclination angles is derived for the case of nucleate boiling.