The explosive growth in recent years, in the publication of both technical articles on enhanced heat and mass transfer as well as new journals in which they appear, has posed a pertinacious challenge to the task of compiling and reviewing the relevant literature. Manual searches by culling contents of journals, archival proceedings, and reports require an exorbitant amount of time and are becoming near impossible. Computerized search engines, while useful and evolving in sophistication, also have limitations based on their algorithms for targeted information filtering. To highlight some of these obstacles that require wider attention, the current status of the literature growth is evaluated, along with a discussion of the associated issues in collecting and categorizing them, and future needs for a dedicated digital library.
Very high heat transfer enhancement can be achieved in single-phase flows by using twisted-tape inserts in circular tubes. The primary convection mechanism is the generation of helical swirl or secondary fluid motion that is induced by the helical curvature of the tape insert. This promotes cross-stream mixing and sharper wall gradients, which are further aided by the increased flow velocity due to the tube partitioning and blockage along with an effectively longer helical flow length. These phenomena are scaled for both laminar and turbulent flow regimes, and an evaluation of the transition is also given to highlight the damping effects of tape-generated swirl. The nature of swirl and its dimensionless scaling, and concomitant development of predictive correlations for heat transfer coefficients and friction factors are discussed. Finally, a brief discussion of the quantification of heat transfer enhancement by means of twisted tapes is given so as to extend their application in heat exchanger design.
The present investigation deals with a differential formulation to estimate the eddy diffusivity together with the universal velocity for fully developed turbulent flows in a tube. The subsequent theoretical predictions of wall friction coefficients and Nusselt numbers are in reasonable agreement with the classical solution of Blasius wall friction coefficient and Dittus and Boelter correlation for heat transfer, Nu=0.023 Re0.8 Pr1/3. The differential form of the eddy diffusivity equation is rendered into an explicit form by regression of the computer runs for wide range of 200<R+<3000 and 1<Pr<1000: εmν=0.0195(y+u+)1-exp-y+u+1002. The prediction of Nusselt numbers from the eddy diffusivity expression satisfactorily agree with the Dittus and Boelter heat transfer correlation. In addition, the analysis can be further modified by altering the correction factor to agree with the predictions from Petukhov–Gnielinski correlation. Thus the efficacy of the model is established for different correlations with a modification in the correction factor for mixing length.
Condensation of vapours on a vertical fin in the presence of non-condensable gas is formulated and solved for a wide range of system parameters, making use of the analogy between heat and mass transfer. The formulation enables the derivation of special cases such as condensation of vapours on a vertical isothermal surface with and without non-condensable gas in the vapour. Besides, the analysis of condensation of vapour on a fin has revealed non-isothermal conditions of the thin fin with the temperature of the liquid–vapour interface also changing along the length of the fin. The numerical results from the analysis are rendered into dimensionless equations that can be used to predict the mean condensation heat transfer coefficients.
The turbulent film boiling on a horizontal cylinder is formulated and solved giving due importance to thermal radiation. In the analysis the test surface is maintained under isothermal conditions and the thermophysical property variation of the vapor due to the steep temperature gradients in the vapor boundary layer is included in the formulation. The analysis is checked with some experimental data relevant to turbulent film boiling. The agreement between the two is found to be very satisfactory.
Energy and materials saving considerations, as well as economic incentives, have led to efforts to produce more efficient heat exchange equipment. Common thermal-hydraulic goals are to reduce the size of a heat exchanger required for a specified heat duty, to upgrade the capacity of an existing heat exchanger, to reduce the approach temperature difference for the process streams, or to reduce the pumping power. The first two objectives translate to an increase in the average heat flux of the heat exchanger, or the encouragement of high heat fluxes. In the case of systems with a specified heat dissipation, the goal is to cool the device, or accommodate a high heat flux, at moderate temperature difference. Implicit in these objectives, energy reduction (improvement of first law efficiency) and temperature difference reduction (improvement of second law efficiency) are important to global environmental protection.
Based on the study of gross flow maldistribution in an experimental electrical heater, this paper presents the effect of flow nonuniformity on the performance of heat exchangers. First, it is shown that it is much more important to understand maldistributions for electrical heaters than for two-fluid heat exchangers. The study of the flow distribution in a particular heater shows that reverse flows may occur for poor inlet header design. Suggested here is a simple way to homogenize the flow distribution and a simple law to calculate, with good accuracy, the velocity ratio (ratio of the highest velocity in the tubes to the lowest velocity). The original fluid distribution is applied to heat exchangers (condensers, counterflow and crossflow heat exchangers), and it is shown that gross flow maldistribution leads to a loss of effectiveness of about 7% for condensers and counterflow heat exchangers, and up to 25% for crossflow exchangers, for velocity ratios up to 15.
This paper presents an experimental study dealing with the basic nucleate boiling concerning two finned surfaces placed in a narrow channel. The influence of both the channel width and the orientation of the base surface (horizontal or vertical) are discussed. The experiments were performed in a saturated pool of FC-72 while the channel widths investigated were 2.0 mm and 0.5 mm. The experimental data are compared with those obtained in the case of the unconfined situation of the extended surfaces. Channel width reduction does not affect the heat transferred to the liquid in the case of vertical orientation of the base surface, while it causes a drastic reduction in the heat transfer behavior in the case of a horizontal base surface. For the latter situation, vapor stagnation in the gap was observed after the maximum heat flux had been reached.
The thermal behavior of three vertical inline heaters placed in narrow channels and cooled by one of two dielectric fluids (FC-72 or Galden HT-55) was experimentally studied. The influence of the channel width on natural convection and boiling heat transfer is discussed. The experiments were carried out at saturation or under subcooled conditions, and the channel widths investigated were 33 mm, 2 mm, and 0.5 mm. The reduction of channel width decreases the heat transfer coefficient for natural convection, but it causes an increase in heat transfer performance for nucleate boiling. In particular, for the boiling data taken in saturation conditions, the better thermal behavior was noted for the 0.5-mm channel width; whereas, in the case of a subcooling of 10 K, the better thermal behavior was noted for the 33-mm channel width. In the latter test condition, on the heat transfer surface, nucleate boiling and natural convection probably occur at the same time. The data obtained are useful for the design of compact microelectronic packages.
The turbulent film boiling from a horizontal cylinder is theoretically investigated considering the properties of the vapor as temperature dependent. In addition the effect of the thermal radiation from the surface of the cylinder on the film boiling heat transfer coefficients is studied. Some of the two-phase boundary layer characteristics are established. Comparison of the results from the present theory with relevant experimental data revealed very satisfactory agreement. An equation is proposed from the computational results with the aid of regression.
General correlations are developed and verified for friction factor and heat-transfer coefficient for single-phase turbulent flow in internally augmented tubes. Data from frequently referred investigations were gathered for a wide range of tube parameters with ed, 0.01 to 0.2; pd, 0.1 to 7.0; α90, 0.3 to 1.0, and flow parameters Re, 5000 to 250,000 and Pr, 0.66 to 37.6. The data were applied to a linear model to obtain normalized correlations that were then modified to approach smooth tube correlation as the roughness variables became very small. A shape function was included in the friction correlation to account for different rib profiles. These correlations were then verified by independent experiments. Different types of commercial ribbed tubes were tested with water as working fluid under heating conditions. Five wire coil inserts were tested in another apparatus using air under cooling conditions. The parameters tested were Re = 5000–60,000, ed = 0.023–0.127, pd = 0.12–1.17, and α90 = 0.34–0.91. The results proved that the correlations can be successfully applied to a wide range of roughness types and Prandtl number. The present correlations are superior, for this large database, to those presented by other investigators. The friction factor correlation predicts 96% of the database to within ±50% and 77% of the database to within ±20%. Corresponding prediction figures for the heat-transfer correlation are 99 and 69%.
Introducing roughness in tubes increases the turbulent heat-transfer coefficient but there is usually a much larger increase in friction factor for constant geometry and flow conditions. In this study, generalized rib-tube correlations are used in optimizing the roughness geometry for two different objective functions or performance evaluation criteria (PEC). Considered are the maximum heat transfer for given pumping power and surface area (R-3) and the minimum surface area for constant heat load and pumping power (R-5). Only tube-side flow was considered. The PEC, with their associated constraint equations, were evaluated by using a simple iterative procedure over a wide range of parameters (Re = 10,000–100,000, Pr = 0.7–35, ed = 0.01–0.1, pd = 0.1–1.0, and α90 = 0–1.0). The optimum rib geometry had ed = 0.02, pd = 0.1, and α90 = 1.0. The Reynolds and Prandtl numbers had little influence on heat-transfer augmentation. The improvement in the effectiveness due to the ribbed tubes has been illustrated for different tube and flow variables to aid in the preliminary design of enhanced heat exchangers.
The paper presents the results of an experimental study that was carried out to determine the Prandtl number influence on the thermal-hydraulic performance of enhanced tubes. The enhanced tubes were identical in all aspects, except the roughness height. Emphasis was placed on turbulent flow, with water as the working fluid. Experiments were performed for Prandtl numbers from 10.2 to 5.8. Data for friction factor were obtained under both isothermal and heating conditions. The study confirmed the strong influence of rib roughness on heat transfer and pressure drop increase. In addition, the results of the investigation strongly suggest that the optimum roughness is application specific, because it is dependent on Reynolds and Prandtl numbers. The results show that the presence of entrance effects and the influence of temperature on the friction factor of enhanced tubes may be significant. The experiments show that a rib height-to-diameter ratio of 0.02 provides the best overall performance for water at Pr = 10; while at Pr = 6.0, the optimum e/d depends on the Reynolds number.
The development of thermal-hydraulic design tools for rectangular offset strip fin compact heat exchangers and the associated convection process are delineated. On the basis of current understanding of the physical phenomena and enhancement mechanisms, existing empirical f and j data for actual cores are reanalyzed. The asymptotic behavior of the data in the deep laminar and fully turbulent flow regimes is identified. The respective asymptotes for f and j are shown to be correlated by power law expressions in terms of Re and the dimensionless geometric parameters α, δ, and γ. Finally, rational design equations for f and j are presented in the form of single continuous expressions covering the laminar, transition, and turbulent flow regimes.
Single-tube pool boiling tests were performed with saturated pure refrigerants and binary mixtures of refrigerants. Generally, with pure refrigerants, the High Flux surface performed better at the higher heat fluxes compared to the Turbo-B tube, and both enhanced surfaces performed significantly better than smooth surface. In tests of R-11/R-113 mixtures, the enhanced surfaces had much less degradation in heat transfer coefficient due to mixture effects compared to smooth tubes; the largest degradation occurred at a mixture of 25% R-11/75% R-113. Under boiling in saturated aqueous solution of calcium sulfate, with a single tube, effects of fouling were more pronounced at the higher heat fluxes for all surfaces. Two staggered tube bundles were tested with tube pitch-diameter ratios of 1.17 and 1.50. For the pure refrigerant, tests on the smooth-tube bundle indicated that the effects on the heat transfer coefficient of varying mass flux, quality, and tube-bundle geometry were small, except at low heat fluxes. Neither enhanced surface showed any effect with changing mass flux or quality. The binary mixture bundle-boiling tests had results that were very similar to those obtained with the pure refrigerants. When boiling a refrigerant-oil mixture, all three surfaces (smooth, High Flux, and Turbo-B) experienced a degradation in its heat transfer coefficient; no surface studied was found to be immune or vulnerable to the presence of oil than another surface.
Two-phase flow boiling from bundles of horizontal tubes with smooth and enhanced surfaces has been investigated. Experiments were conducted in pure refrigerant R-113, pure R-11, and mixtures of R-11 and R-113 of approximately 25, 50, and 75% of R-113 by mass. Tests were conducted in two staggered tube bundles consisting of fifteen rows and five columns laid out in equilateral triangular arrays with pitch-to-diameter ratios of 1.17 and 1.5. The enhanced surfaces tested included a knurled surface (Wolverine`s Turbo-B) and a porous surface (Linde`s High Flux). Pool boiling tests were conducted for each surface so that reference values of the heat transfer coefficient could be obtained. Boiling heat transfer experiments in the tube bundles were conducted at pressures of 2 and 6 bar, heat flux values from 5 to 80 kW/m{sup 2}s, and qualities from 0% to 80%, Values of the heat transfer coefficients for the enhanced surfaces were significantly larger than for the smooth tubes and were comparable to the values obtained in pool boiling. It was found that the performance of the enhanced tubes could be predicted using the pool boiling results. The degradation in the smooth tube heat transfer coefficients obtained in fluid mixtures was found to depend on the difference between the molar concentration in the liquid and vapor.
Passive augmentation techniques such as surface disruptions are being increasingly used in heat exchangers. Although many working correlations have been suggested to predict their thermal-hydraulic characteristics, the physical phenomena governing the heat transfer enhancement have not been clearly understood. The paper describes a qualitative study on the flow phenomena near an enchanced surface. Water was used as the working fluid. Experiments were conducted for different coil wire diameters and for a Reynolds number of 150-2600. The results show the simultaneous existence of different flow patterns in enhanced flow. Also, the study confirmed that the developing length is very much smaller than that of a smooth tube, even for laminar flow.