David L. Rigby

Effect of Squealer Tip on Rotor Heat Transfer and Efficiency

Calculations were performed to simulate the tip flow and heat transfer on the GE-E 3 first-stage turbine, which represents a modern gas turbine blade geometry. Cases considered were a smooth tip, 2 percent recess, and 3 percent recess. In addition, a two-dimensional cavity problem was calculated. Good agreement with experimental results was obtained for the cavity calculations, demonstrating that the k-ω turbulence model used is capable of representing flows of the present type. In the rotor calculations, two dominant flow structures were shown to exist within the recess. Also areas of large heat transfer rate were identified on the blade tip and the mechanisms of heat transfer enhancement were discussed. No significant difference in adiabatic efficiency was observed for the three tip treatments investigated.

A three-dimensional coupled internal/external simulation of a film-cooled turbine vane

A three-dimensional Navier-Stokes simulation has been performed for a realistic film-cooled turbine vane using the LeRC-HT code. The simulation includes the flow regions inside the coolant plena and film cooling holes in addition to the external flow. The vane is the subject of an upcoming NASA Lewis Research Center experiment and has both circular cross-sectional and shaped film cooling holes. This complex geometry is modeled using a multiblock grid, which accurately discretizes the actual vane geometry including shaped holes. The simulation matches operating conditions for the planned experiment and assumes periodicity in the spanwise direction on the scale of one pitch of the film cooling hole pattern. Two computations were performed for different isothermal wall temperatures, allowing independent determination of heat transfer coefficients and film effectiveness values. The results indicate separate localized regions of high heat flux in the showerhead region due to low film effectiveness and high heat transfer coefficient values, while the shaped holes provide a reduction in heat flux through both parameters. Hole exit data indicate rather simple skewed profiles for the round holes, but complex profiles for the shaped holes with mass fluxes skewed strongly toward their leading edges.

Internal Passage Heat Transfer Prediction Using Multiblock Grids and a k-ω Turbulence Model

Numerical simulations of the three-dimensional flow and heat transfer in a rectangular duct with a 180° bend were performed. Results are presented for Reynolds numbers of 17,000 and 37,000 and for aspect ratios of 0.5 and 1.0. A k-ω turbulence model with no reference to distance to a wall is used. Direct comparison between single block and multiblock grid calculations are made. Heat transfer and velocity distributions are compared to available literature with good agreement. The multi-block grid system is seen to produce more accurate results compared to a single-block grid with the same number of cells.Copyright

TopMaker: A Technique for Automatic Multi-Block Topology Generation Using the Medial Axis

ABSTRACT A two-dimensional multi-block topology generation technique has been developed. Very general configurations are addressable by the technique. A configuration is defined by a collection of non-intersecting closed curves, which will be referred to as loops. More than a single loop implies that holes exist in the domain, which poses no problem. This technique requires only the medial vertices and the touch points that define each vertex. From the information about the medial vertices, the connectivity between medial vertices is generated. The physical shape of the medial edge is not required. By applying a few simple rules to each medial edge, the multi-block topology is generated with no user intervention required. The resulting topologies contain only the level of complexity dictated by the configurations. Grid lines remain attached to the boundary except at sharp concave turns where a change in index family is introduced as would be desired. Keeping grid lines attached to the boundary is especially important in the area of computational fluid dynamics where highly clustered grids are used near no-slip boundaries. This technique is simple and robust and can easily be incorporated into the overall grid generation process. mesh generation by Price and Armstrong [5,6] using Midpoint

Improved Film Cooling Effectiveness by Placing a Vortex Generator Downstream of Each Hole

Calculations are presented demonstrating the effect of placing a delta vortex generator downstream of a film cooling hole. The effects of blowing ratio, density ratio, and spanwise pitch are included in the study. Flow over a flat plate with film cooling holes oriented at a 30 degree angle was investigated. The Reynolds numbers based on the freestream velocity and the hole diameter was 11,300. The simulation was performed using the Glenn-HT code, a full three-dimensional Navier-Stokes solver using the Wilcox k-ω turbulence model. A structured multi-block grid was used with approximately one million cells, and average y+ values on the order of unity. Local and span averaged effectiveness are presented. Analysis and visualization of the flow are presented as well as a discussion on the mechanisms which contribute to the dramatic improvement in effectiveness. The results demonstrate that the delta vortex generator was able to annihilate the up-wash vortex pair produced by the film hole and produce a down-wash pair downstream.

Unsteady Analysis of Blade and Tip Heat Transfer as Influenced by the Upstream Momentum and Thermal Wakes

The effect of the upstream wake on the blade heat transfer has been numerically examined. The geometry and the flow conditions of the first stage turbine blade of GE s E3 engine with a tip clearance equal to 2 percent of the span was utilized. Based on numerical calculations of the vane, a set of wake boundary conditions were approximated, which were subsequently imposed upon the downstream blade. This set consisted of the momentum and thermal wakes as well as the variation in modeled turbulence quantities of turbulence intensity and the length scale. Using a one-blade periodic domain, the distributions of unsteady heat transfer rate on the turbine blade and its tip, as affected by the wake, were determined. Such heat transfer coefficient distribution was computed using the wall heat flux and the adiabatic wall temperature to desensitize the heat transfer coefficient to the wall temperature. For the determination of the wall heat flux and the adiabatic wall temperatures, two sets of computations were required. The results were used in a phase-locked manner to compute the unsteady or steady heat transfer coefficients. It has been found that the unsteady wake has some effect on the distribution of the time averaged heat transfer coefficient on the blade and that this distribution is different from the distribution that is obtainable from a steady computation. This difference was found to be as large as 20 percent of the average heat transfer on the blade surface. On the tip surface, this difference is comparatively smaller and can be as large as four percent of the average.

Automatic block merging methodology using the method of weakest descent

A methodology for automatic block merging is developed to the point where a complicated multi-block grid system is supplied and a merged system with connectivity information is returned. The objective of the method is to produce the minimum number of blocks. The recently developed Method of Weakest Descent (MWD) is described and its application to representative test cases is presented. The MWD is based on the premise that the more internal faces of a multi-block grid that can be removed, the lower the final number of blocks. With each internal face that is removed, other faces are disqualified from being removed as a result of the merging of two blocks. The MWD chooses which internal face to remove so as to minimize the number of disqualified faces, thus diminishing the set of available faces at the slowest rate (i.e. weakest descent). Reducing the available faces at the slowest rate allows more blocks to be merged before all valid internal faces are removed. When more than one candidate disqualify the same number of other faces, the choice among those candidates is made randomly. Because of this randomness, each application of the MWD can produce different results. Many tests have been done, with the number of initial blocks ranging from 12 to 7936. The tests have shown that, for most cases, every application of the MWD produces a result near the expected minimum. Repeated application increases the likelihood of realizing the actual minimum.

Heat Transfer in a Complex Trailing Edge Passage for a High Pressure Turbine Blade: Part 1 — Experimental Measurements

A combined experimental and numerical study to investigate the heat transfer distribution in a complex blade trailing edge passage was conducted. The geometry consists of a two pass serpentine passage with taper toward the trailing edge, as well as from hub to tip. The upflow channel has an average aspect ratio of roughly 14:1, while the exit passage aspect ratio is about 5:1. The upflow channel is split in an interrupted way and is smooth on the trailing edge side of the split and turbulated on the other side. A turning vane is placed near the tip of the upflow channel. Reynolds numbers in the range of 31,000 to 61,000, based on inlet conditions, were simulated numerically. The simulation was performed using the Glenn-HT code, a full three-dimensional Navier-Stokes solver using the Wilcox k-omega turbulence model. A structured multi-block grid is used with approximately 4.5 million cells and average y+ values on the order of unity. Pressure and heat transfer distributions are presented with comparison to the experimental data. While there are some regions with discrepancies, in general the agreement is very good for both pressure and heat transfer.

Heat Transfer in a Complex Trailing Edge Passage for a High Pressure Turbine Blade: Part 2 — Simulation Results

A combined experimental and numerical study to investigate the heat transfer distribution in a complex blade trailing edge passage was conducted. The geometry consists of a two pass serpentine passage with taper toward the trailing edge, as well as from hub to tip. The upflow channel has an average aspect ratio of roughly 14:1, while the exit passage aspect ratio is about 5:1. The upflow channel is split in an interrupted way and is smooth on the trailing edge side of the split and turbulated on the other side. A turning vane is placed near the tip of the upflow channel. Reynolds numbers in the range of 31,000 to 61,000, based on inlet conditions were simulated numerically. The simulation was performed using the Glenn-HT code, a full three-dimensional Navier-Stokes solver using the Wilcox k-ω turbulence model. A structured multi-block grid is used with approximately 4.5 million cells, and average y+ values on the order of unity. Pressure and heat transfer distributions are presented with comparison to the experimental data. While there are some regions with discrepancies, in general the agreement is very good for both pressure and heat transfer.Copyright

TRAF3D.MM - A multi-block flow solver for turbomachinery flows

An overview of a methodology for simulating steady flow and heat transfer in turbomachinery is presented. The methodology is the basis for a computer code called TRAF3D.MB. The objective behind the development of the methodology and the computer code is to improve the capability to predict heat transfer in turbomachinery flows. The computer code is used to study turbomachinery flows and to test turbulence models in the prediction of turbomachinery flows. Key aspects of the methodology are (a) multi block grid systems for complicated geometries, (b) finite volume discretization and (c) multigrid convergence acceleration. A target has also been to make the computer code modular, for example to allow flexibility in implementing and testing turbulence models. Currently two turbulence model have been implemented, an algebraic turbulence model and a two equation (k-co) turbulence model. Sample results are presented.