Effect of Tip-to-Shroud Clearance on Turbulent Heat Transfer From a Shrouded, Longitudinal Fin Array

1986 ◽  
Vol 108 (3) ◽  
pp. 519-524 ◽  
Author(s):  
E. M. Sparrow ◽  
D. S. Kadle

Experiments were performed to determine the response of the heat transfer from a longitudinal fin array to the presence of clearance between the fin tips and an adjacent shroud. During the course of the experiments, the clearance was varied parametrically, starting with the no-clearance case; parametric variations of the fin height and of the rate of fluid flow through the array were also carried out. Air was the working fluid, and the flow was turbulent. The fully developed heat transfer coefficients corresponding to the presence and to the absence of clearance were compared under the condition of equal air flowrate, and substantial clearance-related reductions were found to exist. For clearances equal to 10, 20, and 30 percent of the fin height, the heat transfer coefficients were 85, 74, and 64 percent of those for the no-clearance case. The ratio of the with-clearance and no-clearance heat transfer coefficients was a function only of the clearance-to-fin-height ratio, independent of the air flowrate, the fin height, and the fin efficiency model used to evaluate the heat transfer coefficients. The presence of clearance slowed the rate of thermal development in the entrance region.

1967 ◽  
Vol 89 (3) ◽  
pp. 258-268 ◽  
Author(s):  
A. W. Black ◽  
E. M. Sparrow

An experimental investigation, supported by analysis, was performed to determine the heat transfer characteristics for turbulent flow in a circular tube with circumferentially varying wall temperature and wall heat flux. Air was the working fluid. The desired boundary conditions were achieved by electric heating within the wall of a tube whose thickness varied circumferentially. In this way, ratios of maximum-to-minimum wall heat flux as large as two were attained. Local heat transfer coefficients, deduced from the experimental data, display a circumferential variation that is substantially smaller than the heat flux variation. In general, lower heat transfer coefficients correspond to circumferential locations of greater heating, while higher coefficients correspond to locations of lesser heating. The predictions of prior analyses appear to overestimate the circumferential variation of the heat transfer coefficient. A specially designed probe was employed to measure the radial and circumferential temperature distributions within the flowing airstream. On the basis of these measurements, as well as from the heat transfer results, it is concluded that, in the neighborhood of the wall, the tangential turbulent diffusivity is greater than the radial turbulent diffusivity. The axial thermal development was found to be more rapid on the lesser-heated side of the tube than on the greater-heated side. Experimentally determined circumferential-average heat transfer coefficients agreed well with the predictions of analysis.


1984 ◽  
Vol 106 (1) ◽  
pp. 55-63 ◽  
Author(s):  
P. Souza Mendes ◽  
E. M. Sparrow

A comprehensive experimental study was performed to determine entrance region and fully developed heat transfer coefficients, pressure distributions and friction factors, and patterns of fluid flow in periodically converging and diverging tubes. The investigated tubes consisted of a succession of alternately converging and diverging conical sections (i.e., modules) placed end to end. Systematic variations were made in the Reynolds number, the taper angle of the converging and diverging modules, and the module aspect ratio. Flow visualizations were performed using the oil-lampblack technique. A performance analysis comparing periodic tubes and conventional straight tubes was made using the experimentally determined heat transfer coefficients and friction factors as input. For equal mass flow rate and equal transfer surface area, there are large enhancements of the heat transfer coefficient for periodic tubes, with accompanying large pressure drops. For equal pumping power and equal transfer surface area, enhancements in the 30–60 percent range were encountered. These findings indicate that periodic converging-diverging tubes possess favorable enhancement characteristics.


1995 ◽  
Vol 117 (2) ◽  
pp. 316-322 ◽  
Author(s):  
K. Ichimiya

Experiments were conducted to determine the turbulent heat transfer and flow characteristics of an oblique impinging circular jet within closely confined walls using air as a working fluid. The local temperature distribution on the impingement surface was obtained in detail by a thermocamera using a liquid crystal sheet. A correction to the heat flux was evaluated by using the detailed temperature distribution and solving numerically the three-dimensional equation of heat conduction in the heated section. Two-dimensional profiles of the local Nusselt numbers and temperatures changed with jet angle and Reynolds number. These showed a peak shift toward the minor flow region and a plateau of the local heat transfer coefficients in the major flow region. The local velocity and turbulent intensity in the gap between the confined insulated wall and impingement surface were also obtained in detail by a thermal anemometer.


1983 ◽  
Vol 105 (3) ◽  
pp. 527-535 ◽  
Author(s):  
E. M. Sparrow ◽  
N. Cur

The effects of flow maldistribution caused by partial blockage of the inlet of a flat rectangular duct were studied experimentally. Local heat transfer coefficients were measured on the principal walls of the duct for two blockages and for Reynolds numbers spanning the range between 6000 and 30,000. Measurements were also made of the pressure distribution along the duct, and the fluid flow pattern was visualized by the oil-lampblack technique. Large spanwise nonuniformities of the local heat transfer coefficient were induced by the maldistributed flow. These nonuniformities persisted to far downstream locations, especially in the presence of severe inlet flow maldistributions. Spanwise-average heat transfer coefficients, evaluated from the local data, were found to be enhanced in the downstream portion of the duct due to the flow maldistribution. However, at more upstream locations, where the entering flow reattached to the duct wall following its separation at the sharp-edged inlet, the average coefficients were reduced by the presence of the maldistribution.


1978 ◽  
Vol 100 (4) ◽  
pp. 588-594 ◽  
Author(s):  
K. K. Koram ◽  
E. M. Sparrow

Pipe flow experiments were performed to study the heat transfer in the separation, reattachment, and redevelopment regions downstream of a wall-attached blockage in the form of a segmental orifice plate. Water was the working fluid, and the Reynolds number encompassed the range from about 10,000–60,000. The extent of the flow blockage was varied from one-fourth to three-fourths of the tube cross section. Heat transfer coefficients were determined both around the circumference of the uniformly heated tube and along its length. The axial distributions of the circumferential average Nusselt numbers show an initial increase, then attain a maximum, and subsequently decrease toward the fully developed regime. These Nusselt numbers are much higher than those for a conventional thermal entrance region. The unsymmetric blockage induces variations of the Nusselt number around the circumference of the tube. Axial distributions of the Nusselt number at various fixed angular positions reveal the presence of two types of maxima. One of these is associated with the reattachment of the flow and the other occurs due to the impingement of flow deflected by the blockage onto the tube wall. The circumferential variations decay with increasing downstream distance, but the rate of decay for the case of the smallest blockage is remarkably slow. Although most of the tests were performed for Pr = 4, supplementary experiments for Pr = 8 showed that the results are valid for a range of Prandtl numbers.


1981 ◽  
Vol 103 (3) ◽  
pp. 415-422 ◽  
Author(s):  
S. C. Lau ◽  
E. M. Sparrow ◽  
J. W. Ramsey

A systematic experimental study was carried out to determine how the heat transfer characteristics of a turbulent tube flow are affected by the length and diameter of a cylindrical plenum chamber which delivers fluid to the tube. The net pressure loss due to the presence of the plenum was also measured. The experimental arrangement was such that the fluid experiences a consecutive expansion and contraction in the plenum before entering the electrically heated test section. Air was the working fluid, and the Reynolds number was varied over the range from 5,000 to 60,000. It was found that at axial stations in the upstream portion of the tube, there are substantially higher heat transfer coefficients in the presence of longer plenums. Thus, a longer plenum functions as an enhancement device. On the other hand, the plenum diameter appears to have only a minor influence in the range investigated (i.e., plenum diameters equal to three and six times the tube diameter). The fully developed Nusselt numbers are independent of the plenum length and diameter. With longer plenums in place, the thermal entrance length showed increased sensitivity to Reynolds number in the fully turbulent regime. The pressure loss coefficient, which compares the plenum-related pressure loss with the velocity head in the tube, increases more or less linearly with the plenum length. With regard to experimental technique, it was demonstrated that guard heating/cooling of the electrical bus adjacent to the tube inlet is necessary for accurate heat transfer results at low Reynolds numbers but, although desirable, is less necessary at higher Reynolds numbers.


1966 ◽  
Vol 88 (2) ◽  
pp. 170-174 ◽  
Author(s):  
E. M. Sparrow ◽  
J. R. Lloyd ◽  
C. W. Hixon

An experimental investigation of the effect of asymmetrical heating on fully developed turbulent heat transfer has been carried out. The test apparatus was a rectangular duct of aspect ratio 5:1. The duct was constructed so that the two long sides of the rectangular cross section could be heated at different preselected rates, while the two short sides were unheated. Two cases of asymmetrical heating were studied: (a) One of the two long sides was heated, while the second was unheated; (b) both of the long sides were heated, with the heating rate at one side being twice that of the other. For the first case, the heat transfer coefficients are lower than those for the symmetrically heated duct. For the second case, the coefficients for the more strongly heated wall are also below the values for symmetrical heating, while the coefficients for the lesser-heated wall are greater than the symmetric heating results. These findings are in qualitative agreement with analytical predictions for the parallel-plate channel. Furthermore, by applying an analytically motivated correlation procedure (reference [10]), it was shown that overall Nusselt number results for asymmetric heating could be brought into virtual coincidence with those for symmetric heating.


1979 ◽  
Vol 101 (1) ◽  
pp. 23-28 ◽  
Author(s):  
E. M. Sparrow ◽  
R. G. Kemink

Experiments have been performed to study how fluid withdrawal at a branch point in a tube affects the turbulent heat transfer characteristics of the main line flow downstream of the branch. Air was the working fluid. The experiments were carried out for several fixed test section Reynolds numbers and at each Reynolds number the ratio of the withdrawn flow to the test section flow (hereafter designated as the flow split number) was varied systematically. Local heat transfer coefficients were determined both around circumference and along the length of the tube, and circumferential average coefficients were also evaluated. The circumferential average Nusselt numbers in the thermal entrance region are much higher than those for a conventional turbulent pipe flow having the same Reynolds number, and the differences are accentuated at higher values of the flow split number. When normalized by the corresponding fully developed value, the axial distribution of the circumferential average Nusselt number is relatively insensitive to the Reynolds number for a fixed flow split. The thermal entrance lengths, based on a five percent approach to fully developed conditions, are in the 20 to 30 diameter range, which is substantially greater than that for conventional turbulent air flows. Circumferential variations on the order of ±20 percent are induced by the fluid withdrawal process. For the most part, these variations are dissipated upstream of x/D = 10.


Sign in / Sign up

Export Citation Format

Share Document