0
Research Papers

# Modeling Nonuniform Bleed in Axial Compressors

[+] Author and Article Information
S. D. Grimshaw

Whittle Laboratory,
University of Cambridge,
1 JJ Thomson Avenue,
Cambridge CB3 0DY, UK
e-mail: sdg33@cam.ac.uk

G. Pullan, T. P. Hynes

Whittle Laboratory,
University of Cambridge,
1 JJ Thomson Avenue,
Cambridge CB3 0DY, UK

Contributed by the International Gas Turbine Institute (IGTI) of ASME for publication in the JOURNAL OF TURBOMACHINERY. Manuscript received January 29, 2016; final manuscript received February 15, 2016; published online April 12, 2016. Editor: Kenneth C. Hall.

J. Turbomach 138(9), 091010 (Apr 12, 2016) (11 pages) Paper No: TURBO-16-1024; doi: 10.1115/1.4032845 History: Received January 29, 2016; Revised February 15, 2016

## Abstract

The coupling between the bleed system and the flowfield of a downstream compressor stage is studied using two approaches. In the first approach, three-dimensional, full annulus, unsteady computations simulate the flow in a low-speed research compressor with nonuniform bleed extraction. Comparisons with experimental data show that the flow prediction in the main annulus is accurate to within 0.005 of flow coefficient and $0.5deg$ of flow angle. The computational fluid dynamics (CFD) is then used to provide a description of flow within the bleed system itself. In the second approach, a two-dimensional mean radius model, similar to that adopted by Hynes and Greitzer in the previous work on compressor stability, is used to simulate the response of the compressor to nonuniform bleed. This model is validated against experimental data for a single-stage compressor, and despite the inherent assumptions (two-dimensional flow and simplified compressor response), provides a satisfactory prediction of the flow for preliminary design purposes with orders of magnitude less computational cost than full 3D CFD. The model is then used to investigate the effect of different levels of bleed nonuniformity and of varying the axial distance between the bleed and the downstream stage. Reducing bleed nonuniformity and moving the stage away from the bleed slot are predicted to reduce the circumferential nonuniformity of the flow entering the stage.

<>

## Figures

Fig. 1

Meridional view of test compressor showing measurement planes

Fig. 2

Comparison of passage-averaged experimental and CFD data, ϕ¯stage=0.43. (a) Rig inlet, (b) upstream of bleed slot, (c) stage inlet, (d) downstream of rotor, and (e) downstream of stator.

Fig. 3

Rig inlet, CFD calculated flow field, ϕ¯stage=0.43 and bleed rate = 4.2%

Fig. 4

Upstream of slot, CFD calculated flow field, ϕ¯stage=0.43 and bleed rate = 4.2%

Fig. 5

Stage inlet, CFD calculated flow field, ϕ¯stage=0.43 and bleed rate = 4.2%

Fig. 6

Downstream of rotor row, CFD calculated flow field, ϕ¯stage=0.43 and bleed rate = 4.2%

Fig. 7

Passage-averaged distribution of stagnation pressure coefficient at stage inlet and downstream of rotor, ϕ¯stage=0.43 and bleed rate = 4.2%

Fig. 8

Downstream of stator row, CFD calculated flow field, ϕ¯stage=0.43 and bleed rate = 4.2%

Fig. 9

CFD calculated local bleed rate at 80% of slot height compared with experimental measurements, ϕ¯stage=0.43 and bleed rate = 4.2%. The CFD local bleed rate is averaged over 6 deg sections.

Fig. 10

CFD calculated contours of radial velocity coefficient at different circumferential locations. Overlaid are projected streamlines in the same plane as the contours, ϕ¯stage=0.43 and bleed rate = 4.2%. (a) −90 deg from center of duct, (b) −10 deg from center of duct, (c) 0 deg from center of duct, and (d) 10 deg from center of duct.

Fig. 11

CFD calculated streamlines in bleed system. Streamlines are seeded in the off-take duct and traced backward toward the plenum chamber and bleed slot, ϕ¯stage=0.43 and bleed rate = 4.2%. (a) Front-on view and (b) top-down view.

Fig. 12

Flow coefficient upstream and downstream of the row of sinks evaluated from the potential flow model

Fig. 13

Flow angle upstream and downstream of the row of sinks evaluated from the potential flow model

Fig. 14

Comparison of passage-averaged experimental data and model output upstream of bleed slot, ϕ¯stage=0.43. (a) Distribution of flow coefficient, ϕ, (b) distribution of absolute flow angle, α, and (c) distribution of relative flow angle, αrel.

Fig. 15

Comparison of passage-averaged experimental data and model output upstream of bleed slot, ϕ¯stage=0.38. (a) Distribution of flow coefficient, ϕ, (b) distribution of absolute flow angle, α, and (c) distribution of relative flow angle, αrel.

Fig. 16

Comparison of passage-averaged experimental data and model output downstream of bleed slot, ϕ¯stage=0.43. (a) Distribution of flow coefficient, ϕ, (b) distribution of absolute flow angle, α, and (c) distribution of relative flow angle, αrel.

Fig. 17

Comparison of passage-averaged experimental data and model output downstream of bleed slot, ϕ¯stage=0.38. (a) Distribution of flow coefficient, ϕ, (b) distribution of absolute flow angle, α, and (c) distribution of relative flow angle, αrel.

Fig. 18

Measured spanwise distributions of flow angle at different circumferential locations, ϕ¯stage=0.43. (a) Upstream of bleed slot and (b) downstream of bleed slot.

Fig. 19

Distribution of flow coefficient at stage inlet with different models for λ (a) ϕ¯stage=0.43 and (b) ϕ¯stage=0.38

Fig. 20

Distribution of flow coefficient at stage inlet with modifications to ψ (a) ϕ¯stage=0.43 and (b) ϕ¯stage=0.38

Fig. 21

Relative flow angle distributions at stage inlet for different levels of bleed nonuniformity, ϕ¯stage=0.38. (a) Local bleed extraction rate, % and (b) distribution of relative flow angle, αrel.

Fig. 22

Relative flow angle distributions at stage inlet with different axial spacing between bleed and stage, ϕ¯stage=0.38

## Discussions

Some tools below are only available to our subscribers or users with an online account.

### Related Content

Customize your page view by dragging and repositioning the boxes below.

Related Journal Articles
Related Proceedings Articles
Related eBook Content
Topic Collections