Introduction

A variety of techniques are used for enhancing convective heat transfer rates in gas turbine engine passages used for internal cooling of turbine airfoils and combustion chamber liners. These include rib turbulators, pin fins, jet impingement cooling, dimpled surfaces, surface roughness, surfaces with protrusions or other types of turbulence promoters, and swirl chambers. All of these devices act to increase secondary flows and turbulence levels to enhance mixing, in some cases, to form coherent fluid motions in the form of streamwise oriented vortices. Such vortices and secondary flows not only act to increase secondary advection of heat away from surfaces, but also to increase three-dimensional turbulence production by increasing shear and creating gradients of velocity over significant flow volumes. These then give larger magnitudes of turbulence transport over larger portions of the flow fields. The overall objective of each device is then significant enhancement of turbulence transport and convective heat transfer coefficients with minimal increases in pressure penalties.

To determine the capabilities of these devices in accomplishing these tasks, spatially-resolved, spatially-averaged, and globally-averaged heat transfer coefficient data are needed, along with friction factors. Also helpful are details of the flow field structure because this aids the development of numerical models and prediction schemes, and because this provides important insight into the flow structural characteristics responsible for local heat transfer coefficient augmentations. However, in spite of the value of heat transfer data, obtained together with friction factor and flow structure data, very few papers present such data together for the internal cooling augmentation devices mentioned. The present paper is aimed at remedying this deficiency for internal passages with rib turbulators.

A significant number of experimental and numerical studies address the effects of rib turbulators on heat transfer in internal channels. Considered are single pass and multi-pass channels, square and rectangular channels, channels with and without rotation, and rotating channels with different orientations with respect to the axis of rotation.

The earliest experimental studies consider single pass, stationary channels with no rotation. Of these studies, Han et al. [1] address the effects of rib shape, angle of attack, and pitch-to-height ratio. According to the investigators, ribs with 45 ° inclinations produce better heat transfer performance than ribs with 90 ° orientations, when compared at the same friction power. Han and Park [2] vary the channel aspect ratio, and conclude that the best heat transfer performance is obtained using a square channel with a rib turbulator angle of attack from 30o to 45o. This range of angles of attack also yield the best heat transfer performance for rectangular channels. Han et al. [3] indicate that best heat transfer enhancements in square channels are produced by V-shaped ribs with 45o and 60o arrangements, followed by 45o and 60o parallel ribs, which are followed by 45o and 60o crossed ribs. Han et al. [4] investigate wedge-shaped and delta-shaped turbulence promoters in square channels, and compare their performance with existing data for different types of rib turbulators. Delta-shaped ribs generally perform better than the wedge-shaped ribs, especially when the delta-shaped ribs on opposite walls are aligned, and arranged with a backward flow direction. The investigators also indicate that broken configurations of delta-shaped ribs and wedge-shaped ribs both give better performance than full length configurations. Taslim et al. [5] provide additional evidence that 45o ribs produce higher thermal performance factors than 90o ribs. The authors also indicate that, of the configurations examined, the highest heat transfer enhancements and highest friction factors are produced by low-blockage ratio V-shaped ribs. In a later study, Taslim et al. [6] study twelve different geometries of ribs that are placed on all four walls of channels with both square and trapezoidal cross sections. Compared to channels with ribs on two walls, heat transfer coefficients and thermal performance factors are enhanced.

More recent studies of stationary channels with no rotation consider single pass and multipass channels. Wang et al. [7] present heat transfer results from square ducts with 45 ° ribs. Thurman and Poinsatte [8] measure heat transfer and bulk air temperature in a three-pass duct with orthogonal ribs and bleed holes both located on one wall. According to the investigators, changing the locations of the ribs relative to the holes produces large changes to surface heat transfer coefficient distributions. Cho et al. [9] employ continuous and discrete, parallel and cross arrays of ribs in a single-pass square duct. Discrete ribs with gaps in between are found to produce more uniform heat transfer coefficient distributions than continuous ribs.

Other recent experimental studies investigate heat transfer in channels with rotation and square cross-section. Of these studies, Johnson et al. [10] and Fan et al. [11] examine 4-pass serpentine channels and indicate that the best performance is produced by ribs with 45o arrangements. Parsons et al. [12], Johnson et al. [13], Zhang et al. [14], Parsons et al. [15], and Dutta and Han [16] consider 2-pass channels with different rib turbulator configurations. Three of these studies [13,15,16] also consider the influences of changing the channel orientation with respect to the axis of rotation. Heat transfer data are presented which show that the influences of Coriolis forces and cross-stream flows decrease as channel orientation changes from normal to angled. Another pair of studies by Park et al. [17,18] consider the effects of rib size on local heat transfer coefficients with radially outward flow and transverse ribs on the trailing and leading walls of an internal passage.

Experimental heat transfer studies using channels with rotation and rectangular cross-section are less numerous. Of studies in this area, Taslim et al. [19,20] employ single pass channels with orthogonal rotation, and either staggered transverse ribs [19], or 45 ° ribs arranged with perpendicular orientations on opposite channel walls [20]. In both cases, more pronounced rotation effects are evidenced as the channel aspect ratio increases, or as the rib blockage ratio decreases. Azad et al. [21] employ a two-pass channel with rib turbulators on leading and trailing sides at angles of 45 ° with respect to the mainstream flow. Rotating rib wall heat transfer coefficients are 2 to 3 times values measured on rotating smooth walls. Heat transfer coefficients on the first pass trailing side and second pass leading side are enhanced by rotation, whereas the first pass leading side and second pass trailing side values are diminished by rotation. In addition, 45 ° ribs, which a"e parallel on opposite channel walls, produce higher augmentations than 45 ° ribs, which are perpendicular on opposite channel walls.

An experimental and numerical investigation by Dutta et al. [22] examines the effects of rib turbulators on heat transfer behavior in a rotating, two-pass channel with triangular cross-section. Two channel orientations with respect to the axis of rotation are considered, along with the effects of channel orientation on secondary flows.

A number of other experimental investigations address flow behavior (without heat transfer) in channels with rib turbulators. Of these, Bonhoff et al. [23] and Schabacker et al. [24] consider non-rotating channels, and Tse and Steuber [25], and Prabhu and Vedula [26] consider rotating channels. In one case, different velocity components are measured in a serpentine channel with 45 ° ribs [25], and in another, surface static pressure variations are measured with different channel aspect ratios, and different rotational speeds in a rectangular channel with tranverse ribs on one wall [26].

Computational studies of flows and heat transfer in ducts with rib turbulators consider straight single pass ducts [27,28,30,33,35], two pass ducts [31,32,36], two pass ducts with U shaped channels in between [29,34], 90 ° orthogonal ribs [27,30,33,34], 45 ° angled ribs [28,29,31, 32, 35,36], and rotation [27,29,31-36]. In one case [28], the angled ribs placed on two opposite walls of the channel are rounded. The results from the most notable of these investigations with rotation show that the secondary flows induced by angled ribs, rotating buoyancy, and Coriolis forces produce strong non-isotropic turbulence stresses and heat fluxes, along with important alterations to flow fields and surface heat transfer coefficients.

The present experimental study is conducted using a large-scale test section, without rotation, so that detailed, spatially-resolved surface heat transfer coefficients, spatially-resolved flow structure (instantaneous and time-averaged), and friction factors can be measured. A single-pass channel with aspect ratio of 4 is employed, which models internal cooling passages employed near the mid-chord and trailing edge regions of turbine airfoils used in gas turbine engines for utility power generation. The ribs are placed so that they are perpendicular to each other on the two widest, opposite walls of the channel with 45 ° angles with respect to the streamwise flow direction. Reynolds numbers, based on channel height, range from 270 to 90,000. The one other study which uses a stationary, non-rotating test section, similar rib geometry, and same channel aspect ratio [2], does not present spatially-resolved surface heat transfer data or any flow structure data. Two other studies with similar (but not exactly matching) rib turbulator geometry and non-rotating test sections, either use a square channel [5] or a channel with an aspect ratio of 20 [7].

The results given in the present paper are thus new and unique because of these differences, and because new spatially-resolved heat transfer and flow structure data are presented and inter related to each other, something which is impossible for experimental rotating ribbed channel studies [10-22,25-26]. Included in the present study for different channel Reynolds numbers are: (i) time-sequences of flow visualization images, illustrating instantaneous flow structure, (ii) time-averaged distributions of local streamwise velocity, local streamwise vorticity, local total pressure, and local static pressure, (iii) spatially-resolved, spatially-averaged, and globally-averaged surface Nusselt number data, and (iv) friction factor data. Also discussed are the effects of thermal boundary layer development on local Nusselt number distributions. These results are valuable because the spatial resolution of the heat transfer and flow measurements is much better than provided by rib turbulator results from other sources.

0 0

Post a comment