T. Gunnar Johansson

Influence of external conditions on transitionally rough favorable pressure gradient turbulent boundary layers

Laser Doppler anemometry measurements are carried out in order to investigate the influences of the external conditions on a transitionally rough favorable pressure gradient turbulent boundary layer. The acquired data is normalized using the scalings obtained by the means of equilibrium similarity of the outer flow. The point at hand is to not only understand the interaction between the rough surface and the outer flow but also to include the external pressure gradient as the flow evolves in the streamwise direction. It is found that the velocity profiles show the effects of the upstream conditions imposed on the flow when normalized with the free-stream velocity. However, the profiles do collapse when normalized with U ∞ δ*/δ, demonstrating that this scaling absorbs the roughness effects and upstream conditions. An augmentation in the Reynolds stresses occurs with an increase in the roughness parameter and a decrease due to the external favorable pressure gradient. However, close to the wall, there is an increase due to the favorable pressure gradient while on the outer part of the boundary layer there is a decrease in magnitude due to this imposed effect. The near-wall peak of the ⟨ u 2 ⟩ component is dampened by the surface roughness condition due to the destruction of the viscous sublayer. In addition, the shape of the profile in the inner region tends to flatten due to the surface roughness. The upstream wind-tunnel speed also plays an important role thus creating a Reynolds number dependence on the outer flow of the Reynolds stress components. Furthermore, through 11 consecutive downstream locations, the skin friction coefficient is obtained for smooth and rough favorable pressure gradient data. The skin friction shows dependencies on the Reynolds number, the roughness parameter, and the favorable pressure gradient condition in the transitionally rough regime; while for the fully rough regime, it becomes form drag and the dependencies are on the favorable pressure gradient and the Reynolds shear stress. The external condition effects are isolated with a fixed parameter comparison. Favorable pressure gradient effects slow down the growth of the boundary layer while the surface roughness promotes its growth.

The effects of the upstream conditions on a low Reynolds number turbulent boundary layer with zero pressure gradient

An experiment on turbulent boundary layers with low local Reynolds numbers in a zero pressure gradient has been carried out using the laser-Doppler anemometry technique and a similarity analysis of the Reynolds averaged Navier-Stokes equations. This experiment seeks to investigate the effect of the local Reynolds number and the upstream conditions on the downstream development of the mean flow and the turbulent quantities. It was found that by fixing the upstream conditions (i.e. such as the wind-tunnel speed and trip wire size), the mean velocity profiles and Reynolds stresses tend to collapse in the outer flow even though the Reynolds number, R θ, varied from approximately 700 up to 5321. However, when the upstream conditions were changed and the local Reynolds number was held constant, neither the profiles of the wall-normal Reynolds stress nor those of the Reynolds shear stress collapsed, thus showing the effects of the upstream conditions. Moreover, no effect on the mean velocity profiles or the long...

Inner and outer scalings in rough surface zero pressure gradient turbulent boundary layers

A new set of experiments have been performed in order to study the effects of surface roughness and Reynolds number on a zero pressure gradient turbulent boundary layer. In order to properly capture the x dependence of the single point statistics, consecutive measurements of 11 streamwise locations were performed which enabled the use of the full boundary layer equations to calculate the skin friction. This quantity was obtained within 3% and 5% accuracy for smooth and rough surfaces, respectively. For the sand grain type roughnesses used, only the Zagarola and Smits scaling, U∞δ*∕δ, was able to remove the effects of roughness and Reynolds number from the velocity profiles in outer variables. However, each scaling used for the velocity deficit profiles resulted in self-similar solutions for fixed experimental conditions. When examining the Reynolds stresses in the inner region [i.e., 0<(y+ϵ)+<0.1δ+], the ⟨u2⟩ component showed the largest influence of roughness, where the high peak near the wall was decrea...

Near-Wall Measurements in Turbulent Boundary Layers Using Laser Doppler Anemometry

Near wall measurements have been performed in a zero pressure gradient turbulent boundary layer at low to moderate local Reynolds numbers using Laser-Doppler Anemometry in order to investigate how accurately the wall shear stress can be determined. Also, scaling problems are particularly difficult at low Reynolds numbers since they involve simultaneous influences of both inner and outer scales and this is most clearly observed in the near-wall region. In order to fully describe the zero pressure gradient turbulent boundary layer at low to moderate local Reynolds numbers it is necessary to accurately measure a number of quantities. These include the mean velocity and Reynolds stresses, and their spatial derivatives all the way down to the wall (y+ ∼1). Integral parameters that need to be measured are the wall shear stress and boundary layer thickness, particularly the momentum thickness. Problems with the measurement of field properties get worse close to a wall, and they get worse for increasing local Reynolds number. Three different approaches to measure the wall shear stress were examined. It was found that small measurement errors in the mean velocity close to the wall significantly reduced the accuracy in determining the wall shear stress by measuring the velocity gradient at the wall. The constant stress layer was found to be affected by the advection terms. However, it was found that taking the small pressure gradient into account and improving on the spatial resolution in the outer part of the boundary layer made the momentum integral method reliable.Copyright

An Experimental Study of Surface Temperature Distribution on Effusion-Cooled Plates

A parametric study of temperature distribution on effusion-cooled plates under conditions typical for combustion chambers was performed using infrared thermography. In this investigation, the effects of different temperature ratios, velocity ratios of the two air streams, the injection hole spacing, inclination angle of the injection holes, and the thermal heat conductivity of the plates were studied. For a given amount of cooling air, the cooling efficiency was found to increase markedly with a reduction in hole spacing, i.e., when the number of holes was increased. Reducing the injection angle results in more attached jets, especially for small amounts of cooling air, and marginally lowers the wall temperature. A high thermal conductivity of the plate was found to decrease its surface temperature in front of the first row of holes but not the mean temperature in downstream positions. The most important operational parameters were the temperature ratio and the velocity ratio of the hot and cold air streams. An almost linear relation was found between the temperature ratio and the surface temperature when the jet velocity was large compared to the crossflow velocity. For plates with sparse hole spacing, a change in the velocity ratio had a small effect on the surface temperature, whereas the effect was large for dense hole spacings and the same amount of cooling air.

Upstream Condition Effects on Turbulent Boundary Layers Subject to Favorable Pressure Gradients

The effects of the upstream conditions in favorable pressure gradient boundary layers are studied by carrying out an experiment using laser-doppler anemometry over multiple traverses along a smooth plate. A set of upstream conditions composed of upstream wind tunnel speed, tripwire location, and the strength of the pressure gradient is analyzed. The similarity analysis of the equations of motion for pressure gradient flows is used to obtain the scales for the outer flow. For this study, the mean deficit profiles show a small dependence on the strength of the pressure gradient when scaled with the freestream velocity U ∞ . Second, the upstream conditions effects are removed from the velocity deficit profiles when normalized by the U ∞ δ*/δ scaling. However, the Reynolds stress profiles show the effects of upstream conditions and the strength of the pressure gradient. Finally, these favorable pressure gradient flows are found to be nonequilibrium flows because the pressure parameter A is not constant. In addition, three quadrants are found to describe all pressure gradient flows: one for adverse pressure gradient, one for favorable pressure gradient, and one for quasi-laminar flows, where different values are obtained and these are dependent on the experimental conditions. The quadrants are obtained by plotting log (U ∞ /U ∞i ;) versus log(δ/δ i ).

Design, Performance Evaluation and Endwall Flow Structure Investigation of an S-Shaped Intermediate Turbine Duct

Annular S-shaped intermediate turbine ducts are used in modern multi-spool jet engines to connect the high pressure turbine with the low-pressure turbine. The trend towards engines with larger by-pass ratios requires the future intermediate turbine ducts to be shorter and have larger radial off-set. This paper deals with the design and performance evaluation of a state-of-the-art annular S-shaped intermediate turbine duct. The details of the design of the intermediate turbine duct are presented together with static pressure measurements and oil film flow visualization along the endwalls, and area traverses at the inlet and outlet planes using a 5-hole probe. The measurements were done for three operating points of the turbine. From the flow visualization no separation could be detected at design point conditions, but for off-design conditions regions of separation were detected on the guide vanes located within the inter-turbine duct. The pressure loss coefficient was shown to be comparable for the two cases with lowest swirl angle, but the design point showed a slightly lower pressure loss. For the case with the largest flow angle the pressure loss coefficient was clearly larger than for the other two cases, which can be associated with the separation found on the guide vanes.Copyright

Experimental and Numerical Investigation of an Aggressive Intermediate Turbine Duct: Part 1 - Flowfield at Design Inlet Conditions

Demands on lower emissions and reduced noise levels drive the design of modern turbofan engines toward high by-pass ratios. The design of the intermediate turbine ducts, connecting the high- and low-pressure turbines, will become more important as the turbofan engine by-pass ratios are increased. In order to introduce more aggressive designs there is a need to understand the flow features of high aspect ratio and high diffusion ducts. This is part one of a two-part paper, presenting a comparison between an experimental study and a CFD analysis of the flowfield of an aggressive intermediate turbine duct for design inlet conditions. Part one focuses on the on-design conditions and the second part will focus on off-design conditions. The experimental study was performed in a large-scale, low-speed turbine facility. The work presented highlights some of the challenges associated with more aggressive intermediate ducts for the next generation of turbofan engines. The main flow features are successfully reproduced by the CFD, but there are discrepancies found in the predicted local velocity and loss levels. An explanation of the discrepancies between the experimental data and the CFD results is provided and an attempt to track the origin of these differences is made.

Experimental investigation of the time-averaged flow in an intermediate turbine duct

Intermediate turbine ducts are used in modern multi-spool jet engines to connect the high pressure turbine with the lowpressure turbine. The trend towards turbofan engines with larger by-pass ratios requires the radial off-set between the high-pressure and low-pressure turbines to increase with a corresponding increase in radial off-set for the intermediate turbine ducts. Other improvements of the ducts is to make them shorter and more diffusing but this strive towards more aggressive design increases the risk for separation. This paper deals with an experimental investigation of the time-averaged mean flow field and turbulence development in an aggressive intermediate turbine duct (downstream a rotating turbine stage) using a 5-hole probe and 2-component hot-wire anemometry. In addition the duct endwall static pressure distribution is discussed. The investigation revealed the complex flow structure development within the duct, where corotating vortices emanating from the break-up of the tip gap shear-layer dominates the flow pattern.

Numerical simulation of effusion cooling with comparisons to experimental data

CFD simulations of a slanted jet in crossflow were made and compared to detailed 3D-LDA measurements. Three turbulence models were studied using Fluent v6.1: realisable k-e, SST-k-ω and Reynolds stress model (RSM). The SST-k-ω model and the RSM were both able to capture the recirculation zone in the jet wake, which was not the case with the realisable k-e model. The turbulent kinetic energy field was not well predicted by the k-e or the SST-k-ω model. However, the Reynolds stress field was captured qualitatively correct by the RSM.