Measurement and Prediction of Flow in Annular S-shaped Ducts

Britchford, K. M.; Manners, A. P.; McGuirk, James J.; Stevens, S. J.

doi:10.1016/b978-0-444-89802-9.50077-0

Cited by 19 publications

(30 citation statements)

References 2 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Over-prediction and local maxima in the profiles that were otherwise not registered by the experiment are seen in all three computational models. This is consistent with the results of Britchford et al 2 and attributed to the inaccuracy of the turbulence models. Notwithstanding the small differences seen between the measurements and simulations in Fig.…”

Section: B Validation Of Predicted Resultssupporting

confidence: 93%

“…The LUT cases feature detailed descriptions of the facility, duct geometry, test conditions and experimental measurements in a number of papers and two Ph.D. theses. [2][3][4][5][6] Without an axial compressor stage upstream of transition duct, air was drawn from the atmosphere into a plenum by compressors located within the exhaust. The flow adjacent to the inner and outer casings was artificially tripped before entering the inlet section, which was seven hydraulic diameters long.…”

Section: A Experimental Data Setmentioning

confidence: 99%

“…1 A number of investigations performed with the Loughborough University of Technology (LUT) low-speed onespool compressor rig have examined unstrutted and strutted annular transition ducts with area ratio (AR=A ex /A in ) of one and low and moderate loading. The early work by Britchford et al [2][3][4] provided experimental data on the mean flow field and turbulence within the duct using five-hole probes and Laser Doppler anemometry (LDA). Two inlet cases were considered, 'clean' inlet, in which boundary layers were developed along an upstream entry length, and a single stage axial compressor upstream of the duct.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

RANS Simulations for Sensitivity Analysis of Compressor Transition Duct

Milanović¹,

Whiton²,

Florea³

et al. 2014

50th AIAA/ASME/SAE/ASEE Joint Propulsion Conference

View full text Add to dashboard Cite

This paper presents a numerical investigation of a strutted annular transition duct used to connect the low pressure and high pressure compressors of aircraft gas turbine engine. Data was generated for one-and six-strut configurations with the United Technologies (UTC) flow solver. For reference, some cases were also run with the ANSYS FLUENT software package. The validity of the computed velocity and pressure loading distributions was established with the existing experimental data available in the open literature. Mean streamwise velocity profiles are in good agreement with data along the entire length of the duct with some discrepancies in the boundary layers adjacent to each casing. Axial velocity distributions at all locations show a good match in the width of the wake. However, the wake strength is not always well matched. Axial velocity contours illustrate a well-defined wake at the strut trailing edge with no significant regions of flow separations, similar to the experimental data set. Obtained axial static pressure loading distributions along the midline of the duct, about the strut and duct-strut combination indicate good match with measurements. The static pressure loading distributions along the strut at different heights are presented and compare very well with experiment. Additionally, at 10 percent strut height the superposition of strut and duct static pressure distributions shows good agreement between simulation and theoretical data up to 70 percent length. The loss distributions across the duct at four axial locations and integrated losses indicate increasing development of loss near the endwalls through the duct. Finally, sensitivity studies examining the influence of duct length and Mach number on pressure loading were also performed for the case of the one-strut duct. Nomenclature A = area C p = pressure coefficient L = transition duct axial length M = Mach number P = stagnation pressure R = radius of flow curvature U = mean streamwise velocity component U mean = area-weighted spatial mean value 1 2 h = passage height k = turbulent kinetic energy m  = mass flow rate r = radius x = axial distance  = stagnation pressure loss coefficient  = density Superscripts  = mass-weighed spatial mean value Subscripts a = upstream b = downstream ex = exit in = inlet

show abstract

Section: B Validation Of Predicted Resultssupporting

confidence: 93%

Section: A Experimental Data Setmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

RANS Simulations for Sensitivity Analysis of Compressor Transition Duct

Milanović¹,

Whiton²,

Florea³

et al. 2014

50th AIAA/ASME/SAE/ASEE Joint Propulsion Conference

View full text Add to dashboard Cite

show abstract

“…Little et al [19] predicted the flow in a dump diffuser combustion system and suggested that the latter failing leads to an underprediction of system total pressure loss for eddy-viscosity models, although this can be masked by the overprediction of impingement loss. Britchford et al [20] demonstrated the superiority of the Reynolds stress model over a k-" model in capturing the combined effects of strong pressure gradients and streamline curvature in an annular S-shaped duct. Similarly, Walker [21] also showed that a Reynolds stress model gave better agreement with measured data in a dump diffuser combustion system.…”

Section: Numerical Methodologymentioning

confidence: 99%

“…This is caused by the outboard movement of fluid, within the diffuser, which is due to the influence of the radial pressure gradient associated with the downstream flame-tube blockage. However, this movement is restricted by the outer casing of the datum diffuser and this is further suggested by the pitch-angle profile, which reaches a maximum of 20 . In contrast, the integrated diffuser tries to accommodate the impact of the downstream blockage on the flowfield as it tries to diffuse the flow.…”

Section: B Prediffuser Exitmentioning

confidence: 99%

Annular Diffusers with Large Downstream Blockage Effects for Gas Turbine Combustion Applications

Walker

Carrotte

Denman

2011

Journal of Propulsion and Power

View full text Add to dashboard Cite

., 2011.Annular diffusers with large downstream blockage effects for gas turbine combustion applications. Journal of Propulsion and Power, 27(6), pp. 1218-1230. Additional Information:• This paper was published in the Journal of Propulsion and In engineering applications, diffuser performance is significantly affected by its boundary conditions. In a gas turbine combustion system, the space envelope is limited, the inlet conditions are generated by upstream turbomachinery, and the downstream geometry is complex and in close proximity. Published work discusses the impact of compressor-generated inlet conditions, but little work has been undertaken on designing diffusers to accommodate a complex downstream geometry. This paper considers the design of an annular diffuser in the presence of a large downstream blockage. This is most applicable in the combustion system of a low-emission landbased aero-derivative gas turbine, where immediately downstream of the diffuser approximately 85% of the flow moves outboard and 15% moves inboard to supply the various flame-tube and turbine-cooling features. Several diffuser concepts are numerically developed and demonstrate 1) the interaction between the diffuser and downstream geometry and 2) how this varies with changes in diffuser geometry. A preferred concept is experimentally evaluated on a low-speed facility that simulates the combustion system and provides compressorgenerated inlet conditions. A conventionally designed aero-derivative diffuser system is also evaluated and, with reference to this datum, the system total pressure losses are reduced by between 20 and 35%.= mass flow rate N = rotor speed P, p = local total or static pressure R, r = radius T, t = local total or static temperature To 1 = total temperature in rig inlet U = velocity U blade = rotor midpassage blade speed Va = midpassage axial velocity y = dimensionless wall distance = kinetic energy coefficient = diffuser cant angle R = change in radius = total pressure loss coefficient = density = area-weighted spatially averaged mean valuẽ = mass-weighted spatially averaged mean value

show abstract