The boundaries of the domains of existence of flow regimes past single dimples made as spherical segments on a flat plate are determined with the use of available experimental results. Regimes of a diffuser-confuser flow, a horseshoe vortex, and a tornado-like vortex in the dimple are considered. Neither a horseshoe vortex nor a tornado-like vortex is observed in dimples with the relative depth smaller than 0.1. Transformations from the diffuser-confuser flow regime to the horseshoe vortex regime and from the horseshoe vortex flow to the tornado-like vortex flow are found to depend not only on the Reynolds number, but also on the relative depth of the spherical segment. Dependences for determining the boundaries of the regime existence domains are proposed, and parameters at which the experimental results can be generalized are given.Key words: dimple, vortex structures, horseshoe and tornado-like vortices, enhancement of heat transfer. Introduction.A large amount of experimental data on specific features of the flow structure and heat transfer in the flow past surfaces with spherical cavities (dimples) has been accumulated. There are some summary papers where numerous aspects of vortex formation in dimples under the action of a large number of factors were considered [1-6]. It was found in the first experiments that the increase in heat transfer in the flow past a surface with dimples made as spherical segments is accompanied by an insignificant relative increase in hydraulic resistance [7]. Large-scale dynamic vortex structures observed in a wide range of flow regimes are formed inside the dimples and in their vicinity. Despite a large number of publications dealing with using surfaces with dimples, however, some disputable issues have not been resolved (hypothesis of the tornado-like mechanism of heat-transfer enhancement [7,8], role of dimples in flows with high turbulence levels [9, 10], and determining the boundary between the domains of existence of separated and non-separated flows past spherical cavities [11,12]). This is explained by the complexity of the phenomenon, which is affected not only by the flow velocity, but also by the dimple size, shapes of its edges, character and thickness of the boundary layer at the dimple entrance, and channel height.Here, we make an attempt to determine the boundaries of the domains of existence of flow regimes in the dimple, depending on the external flow velocity and on the geometric characteristics of the dimple, by means of analyzing available experimental data. Similar studies were performed previously (see, e.g., [11,13]), but the previous results can now be refined because a large amount of new data appeared.Regimes of the Flow Past Dimples. The following flow regimes past a surface with a dimple can be identified. As low velocities, the flow has a diffuser-confuser character. The streamlines passing close to the dimple become curved. As the flow velocity is increased, the flow becomes separated near the leading edge of the dimple, forming a recirculation zone,...

This paper provides the primary results of an experimental study into the fluid flow and heat transfer features at a cross-flow of a dimpled tube in a rectangular-shaped duct between two adjacent dimpled tubes. The cylindrical dimples were engraved on each tube surface both in the staggered and in-line mode; altogether nine dimpled tubes were tested in the range of the Reynolds number Re from 8,000 to 115,000. The first group (four samples) represents tubes structured with symmetrical dimples drilled in the radial direction, while the second group (five samples) is tubes with asymmetrical dimples. In the latter case each dimple was made in such a way that its axis is parallel to the tube diameter with a certain clearance between axes. For comparisons a row of smooth tubes of the same configuration was tested under identical fluid boundary conditions. Three factors primarily influencing heat transfer are under consideration in this paper: a) increase in a heat exchange surface due to a tube dimpling, b) variations in the flow pattern, c) interaction between boundary layer and main flow. Behind a smooth tube in confined space the reverse flow zone grows initially to Re = 37,000 however decreases at larger Reynolds numbers. Unlike this, behind a dimpled tube in confined space the reverse flow zone reduces at low Reynolds numbers to reach minimum magnitude at Re = 10,000–28,000, and increases afterwards to become approximately constant at Reynolds numbers over 45,000. It has been found, the reverse flow length depends on the Reynolds number, dimple parameters and configuration. The frequency spectrum of the dimpled tube is different from that occurring for a smooth tube. A few frequency ‘picks’ with corresponding the Strouhal numbers were registered including those typical to a single dimple on a flat plate. The heat transfer enhancement rates of around 45%–55% compared with a smooth tube in confined space were obtained depending on dimple parameters and flow regimes. Increase in the heat transfer enhancement rate for tubes with shallow dimples exceeds growth of heat exchange surface due to a dimpling. Increases in a pressure drop at the tube bundle caused by dimpling do not exceed 14%.

No abstract

No abstract