Abstract

In the past few decades, “buzz-saw” noise was mostly measured and predicted along the shroud wall. Uniform or nonuniform axial flows were applied to the predictions. Besides, the strength of the “buzz-saw” noise was widely assumed to be identical along the radius. However, nonuniform background flows and distinct radial distributions of shock strength are observed in almost all transonic fans. A possible way to solve these problems is to couple the shock trajectory with the evolution of the shock wavefront. In case the state of background flow field varies slowly, the shock trajectory is depicted by geometric acoustics. Meanwhile, the evolution of the wavefront can be solved by the governing equation of the weak-shock. Under this framework, a method is proposed to tackle the prediction of the “buzz-saw” noise under nonuniform axial and radial flows. This method is first validated by the test data from the literature. Then, it is applied to predict the near-field noise generated by the ideal and four modified versions of NASA rotor 67. The results indicate that the nonuniform radial and axial flows introduced by the wall boundary have strong effects on the distribution of the “buzz-saw” noise. Additionally, the eccentric-force problem is revealed as a side effect of blade sorting, which is an efficient method to suppress the “buzz-saw” noise. A bi-pyramid blade sorting strategy is proposed to suppress the eccentric force introduced by other blade sorting strategies.

Highlights

  • The one-step prediction of the “buzz-saw” noise by computational aeroacoustics (CAA) is very expensive and lacks numerical stability due to the existence of shockwaves

  • Engine nacelles tend to be shortened for drag reduction, which significantly enhances the non-uniformity of inflow under operation conditions with attack angles

  • Hawkings23 discussed the influence of nonuniform axial flow on the propagation of the “buzz-saw” noise by geometric acoustics; the proposed method by Hawkings is not practical for prediction

Read more

Summary

INTRODUCTION

“Buzz-saw” noise is a series of irregular shockwaves generated by transonic fan or compressor blades. McAlpine et al. expanded the prediction of the “buzz-saw” noise to the lined duct by numerically solving the modified weak-shock relation Their method is validated by the test data of the transonic research fan from Rolls–Royce.. The methods mentioned above are based on the assumptions of the uniform axial flow and uniform radial distribution of the “buzzsaw” noise This assumption is violated by complex blade geometry and in the presence of a spinner, nacelle lip, and axisymmetric duct.. In order to predict the “buzz-saw” noise with radial distribution, Mathews et al. proposed a stepwise and stripwise version of the Morfey and Fisher’s method under the assumption that the acoustic energy is bounded in stream tubes. Adetifa investigated the “buzz-saw” noise propagation in axisymmetric ducts with a similar stepwise method to account for the effect of the nonuniform axial flow.

Framework of the propagation method
Propagation method
Pre-process the “real” source to sawtooth waves
VALIDATION OF THE METHOD
APPLICATION IN NASA ROTOR 67
Source from rotor with random stagger variations
Sources from blade-sorted rotors
Propagation of the sources
Findings
CONCLUSIONS
Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call