Jet Noise Reduction of Double-Mixing Exhaust System

Shao Wanren，Wu Fei，He Jingyu，Li Xiaodong

1.Shenyang Engine Design and Research Institute，Aviation Industry Corporation of China，Shenyang 110015，P.R.China；2.China Academy of Aerospace Aerodynamics，Beijing 100074，P.R.China；3.School of Energy and Power Engineering，Beihang University，Beijing 100191，P.R.China（Received 16Apri 2015；revised 3March 2016；accepted 21March 2016）



Jet Noise Reduction of Double-Mixing Exhaust System

Shao Wanren1＊，Wu Fei1，He Jingyu2，Li Xiaodong3

1.Shenyang Engine Design and Research Institute，Aviation Industry Corporation of China，Shenyang 110015，P.R.China；
2.China Academy of Aerospace Aerodynamics，Beijing 100074，P.R.China；
3.School of Energy and Power Engineering，Beihang University，Beijing 100191，P.R.China
（Received 16Apri 2015；revised 3March 2016；accepted 21March 2016）

A jet noise reduction technique by using the external chevron nozzle with lobed mixer in the double-mix-ing exhaust system is investigated under cold conditions.The computations of jet field and the experiments of noise field are conducted with scaled model of high-bypass-ratio turbofan engine mixing exhaust system composed of external chevron nozzle with lobed mixer.The computational results indicate that comparing with the baseline nozzle with lobed mixer，the external chevron nozzle with lobed mixer increases mixing of jet and ambient air near the nozzle exit.The experimental results show that the external chevron nozzle with lobed mixer has better jet noise reduction at low frequencies，and this reduction rises with the increase of chevron bend angle.The experimental results also show that the external chevron nozzle with lobed mixer has sound pressure level（SPL）increase which is not obvious at high frequencies.With chevron bend angle increasing，SPL has relatively marked increase at 60°（directivity angle measured from upstream jet axis）and little fluctuations at 90°and 150°.The external chevron nozzle with lobed mixer has overall sound pressure level（OASPL）reduction in varying degrees at 60°and 150°，but it has little OASPL increase at 90°.

aerospace propulsion system；high-bypass-ratio turbofan engine；mixing exhaust system；lobed mixer；chevron nozzle；jet noise reduction

0 Introduction

Jet noise is one of the main noise sources of turbofan engine.To meet increasingly stringent noise standards published by International Civil Aviation Organization（ICAO）and related countries，the jet noise reduction technique is one of the important issues in aviation industry［1］.

The total acoustic power emitted from a jet is nearly proportional to the eighth power of the jet velocity on the whole.Spectra of jet noise have a direct bearing on the jet field structures and they have clear directivities［2］.For this reason，jet noise reduction could be realized by reducing jet peak velocity，changing jet field structures and controlling radiation way of jet noise.With regards to the high-bypass-ratio turbofan engines，the highly efficient mixing between core flow and fan flow，as well as the enhanced mixing of jet and ambient air are two key ways for jet noise reduction.

Since the mid-1980s，an advanced forced mixer named as lobed mixer has been employed in turbofan engine mixing exhaust systems for enhancing the mixing between core flow and fan flow.The technique can help to reduce the noise. For example，CFM56-5Cengine by using lobed mixer produces 12.8dB reduction compared with CFM56-3engine.Another technique for enhancing the mixing of jet and ambient air by using chevron nozzle is also employed in turbofan engines with separated exhaust system.The basic mechanism of the chevron nozzle is to create array stream-wise vortexes which enhance the mixing of jet and ambient air［3-6］，thus reducing jet noise atlow frequencies.

In recent years，the studies on the aerodynamic performance of forced mixing exhaust system with lobed mixer［7-12］and the jet noise reduction technique of separated exhaust system and mixing exhaust system［13-16］in turbofan engines had been successively conducted.But the comprehensive study on the combination of forced mixing between core flow and fan flow with lobed mixer and enhanced mixing between jet and ambient air with chevron nozzle had not been nearly conducted at home and abroad.In this paper，an experimental study on jet noise reduction by using the lobed mixer and chevron nozzles is conducted on the jet noise test rig under cold conditions with scaled model.The far field spectra of jet noise and the effect of chevron bend angles on noise reduction are discussed.Besides，numerical simulations of flow fields are performed to reveal the mechanism of double-mixing on jet noise reduction.

1　Nozzle Models

For subsonic jet noise，the nozzle exit structure has a remarkable influence on the jet field.It is expected that different nozzle exit structures have different effects on jet noise reduction. Chevron nozzle is regarded as an effective way to reduce jet noise with minimal performance or weight impacts in high-bypass-ratio turbofan engine.Based on Refs.［10—15］，three types of scale model nozzles were designed to study the effects of different chevron nozzles on jet noise reduction.These nozzles include a nozzle with conventional annular mixer（Nozzle-A），a nozzle with 16-lobes scalloped lobed mixer（Nozzle-B），and three nozzles coupled with 16-lobes scalloped lobed mixer and chevron outlets（Nozzle-C）. Three types of scale model nozzles and their parameters are shown in Fig.1and Table 1.Five different nozzles have the same core flow exit area，fan flow exit area，and nozzle exit area.

Fig.1　Schematic diagram of scale model nozzles

Table1　Parameters of external chevron nozzles

2　Flow Field Simulation

2.1 Computational procedures

Numerical simulation is performed at a core flow exit with Ma0.8and a fan flow exit with Ma0.5.Combinations of tetrahedral grid and hexahedral grid are used in the solution domain. Computational grid for Nozzle-B is shown in Fig.2.

The CFD analysis of the steady-state jet plume flow fields is performed using commercial CFD software，which solves the Navier-Stokes（N-S）equations using a standard k-εturbulence model.The compressible turbulence flow and the no-slid surface condition are used in the flow field computations and wall function is used in the wall treatment options.

Fig.2　Computational grid for Nozzle-B

2.2 Computational results

Fig.3presents the comparisons of velocity contours on five cross sections of Nozzle-A，Nozzle-B and Nozzle-C1.Compared with Nozzle-A，it is seen that Nozzle-B has a tremendous flow field changes.The lobed mixer strongly distorts the mixing layer to form a lobe-shaped velocity distribution inside the nozzle（As seen at X/Dj=-1 and-0.6，where Djis nozzle exit diameter）. The continuing mixing causes the flow to form a round mixing layer downstream the nozzle exit at X/Dj=2—10.The enhanced mixing significantly reduces the potential core length and the jet velocity of Nozzle-B，thus contributing to a noise reduction.

Compared with Nozzle-B，it is also seen that Nozzle-C1has a slight flow field changes inside the nozzle.Chevron nozzle changes the cross-sectional shape of the jet and increases mixing of jet and ambient air near the nozzle exit，thus leading to a rapid increase in the width of the mixing layer.The core jet length and the jet velocity of Nozzle-C1are further reduced.This can result in more reduction of jet noise.

3　Experiment

3.1 Test rig and measuring implements

The experiments are performed in the full anechoic chamber of Fluid and Acoustic Laboratory in Beihang University which meets the condition of free noise field.With acoustic wedges placed on the walls，the ceiling and the floor，minimizing sound reflections，the full anechoic chamber has a size of 8.9m×6.8m×4.65m（Excepting wedges）and a cut-off frequency of 250Hz.The dual-flow cold jet noise test rig（See Fig.4）is composed of test section，flow control section and gas source.The test section is located in the anechoic chamber and is used for the installation of model experiment nozzles and acoustic measuring implements.Flow control section is placed in the control room.The total pressure of the airflow can be adjusted by the pneumatic pressure valve，intelligent pressure transmitter or manual pressure regulator.There is a gas source outside the laboratory，which is used for supplying dry compressed air.Acoustic measuring implements are quarter inch condenser microphones，which have a dynamic range from 56dB to 170dB and a sensitivity of 1.33mV/Pa.Condenser microphone calibration is conducted before every experiment.Atmospheric pressure and temperature are recorded before and after every experiment.

3.2 Measurement scheme

In order to study the amplitude，the frequency characteristics of each component of jet noise and noise directivities，according to the laboratory measurement conditions and the propagation characteristics of each jet noise components，the far field acoustic measurement scheme is obtained by the microphone phased array consisting ofeight condenser microphones arranged on a circular arc at the center of the nozzle exit plane，whose radius is 60times the diameter of nozzle exit.Eight microphones are arranged in the far field every 15°from 45°（Directivity angle measured from upstream jet axis）to 150°，as shown in Fig.5.

Fig.3　Velocity contours on five cross sections

Fig.4　Dual-flow cold jet noise test rig

Fig.5　Vertical view of measurement scheme

4　Results and Discussion

4.1 Far field spectra of jet noise

The experiments are performed at a core flow exit with Ma0.8and a fan flow exit with Ma0. 5.For limited space，we only give the comparisons of noise frequency spectra from Nozzle-C0，Nozzle-C1and Nozzle-C2with Nozzle-A and Nozzle-B（See Fig.6）at upward direction（Directivity angle=60°），sideline direction（Directivity angle =90°）and downward direction（Directivity angle =150°）.In consideration of jet noise directivities，the ordinate datum points at 60°and 90° have the same value of sound pressure level（SPL）and the ordinate datum point at 150°increases 10dB than the ordinate datum points at 60°and 90°.

Note that all the frequency spectra in Fig.6is normalized as a Strouhal number（St）calculated by

where fis the observation frequency，Djthe nozzle exit diameter，and Ujthe nozzle exit jet velocity.SPL is calculated by

where pis the sound pressure，p0=2×10-5Pa，which is the standard sound pressure.

As it is seen from Fig.6，similar to Nozzle-A and Nozzle-B，Nozzle-C0，Nozzle-C1and Nozzle-C2have far field frequency spectra which have parabolic distribution with low on both sides and high in the middle at 60°，90°and 150°，which meets similitude law of the far field frequency spectra.In addition，alike to Nozzle-A and Nozzle-B，the comparisons of Figs.6（a）—（c）show that Nozzle-C0，Nozzle-C1and Nozzle-C2successively increase SPL values at low frequencies as directivity angle increases from 60°to 150°.Because the jet at 150°is mainly composed of large scale vortexes which mainly radiate low frequency noise［17-19］.Fig.7presents the computational instantaneous density contours for Nozzle-A，and shows jet noise sources of high-bypass-ratio turbofan engine.

Fig.6　Comparisons of noise frequency spectra from Nozzle-C0，Nozzle-C1and Nozzle-C2with Nozzle-A and-B

Fig.7　Jet noise sources of high-bypass-ratio turbofan engine

4.2 Effect of chevron bend angle

At a core flow exit with Ma0.8and a fan flow exit with Ma0.5，the comparisons of Figs.6（a）—（c）show that Nozzle-C2has a best noise reduction at low frequencies among three different external chevron nozzles with 16-lobes scalloped lobed mixer，produces a 3—4dB reduction compared with Nozzle-A and a 0.7—1.7dB reduction compared with Nozzle-B at 60°，90°and 150°compared with Nozzle-B，the results of noise reduction of Nozzle-C0，Nozzle-C1and Nozzle-C2are specifically discussed as follows：

（1）At 60°（See Fig.6（a）），Nozzle-C0，Nozzle-C1and Nozzle-C2have a slight noise reduction at low frequencies，and produce about 0.1—1.3dB reduction as a whole，but have a clear noise increase at high frequencies compared with noise reduction at low frequencies.Because the double mixing of the external chevron nozzle with lobed mixer further enhances the mixing of jet and ambient air，and makes more large scale vortexes transfer to small ones.Low frequency noise is radiated mainly by large scale vortexes，while high frequency noise is radiated mainly by small scale vortexes.For Nozzle-C0，Nozzle-C1and Nozzle-C2，when the bend angle increases from 11°to 22°，noise reduction at low frequencies increases slightly，correspondingly，noise increase at high frequencies increases obviously.For Nozzle-C0，Nozzle-C1and Nozzle-C2，noise reduction at low frequencies is about 0.1—0.4dB，0.7—0.9dB and 0.7—1.3dB，respectively.This main reason is that the chevrons enhance the mixing of jet and ambient air and make large scale vortexes transfer to small ones with the increase of chevron bend angle from 11°to 22°.

（2）At 90°（See Fig.6（b））and 150°（See Fig.6（c）），SPL variation of Nozzle-C0，Nozzle-C1and Nozzle-C2at low frequencies is similar to that at 60°，and produce about 0.1—1.0dB reduction at 90°and about 0.1—1.7dB reduction at 150°as a whole.About 0.1—0.3dB reduction of Nozzle-C0，about 0.1—0.5dB reduction of Nozzle-C1，and about 0.7—1.0dB reduction of Nozzle-C2are produced at 90°.About 0.1—0.7dB reduction of Nozzle-C0，about 0.7dB reduction of Nozzle-C1，and about 1.5—1.7dB reduction of Nozzle-C2are produced at 150°.As the chevron bend angle increases from 11°to 22°，SPL variation of Nozzle-C0，Nozzle-C1and Nozzle-C2at high frequencies is different to that at 60°，which is fluctuant and is not obvious compared with that at 60°.High frequency SPL of Nozzle-C0，Nozzle-C1and Nozzle-C2slightly increases compared with that of Nozzle-B at 90°，and is equal to that of Nozzle-B at 150°.

4.3 Noise directivities

Fig.8is the comparisons of noise directivity from Nozzle-C1with Nozzle-A and Nozzle-B at a core flow exit with Ma0.8and a fan flow exit with Ma0.5.In Fig.8，over all sound pressure level（OASPL）is the sum of the different SPLs，which have a range from 200Hz to 50 000Hz. OASPL is calculated by

OASPL=

10lg（100.1SPL1+100.1SPL2+…+100.1SPLn）（3）

As it is seen from Fig.8，the double-mixing of Nozzle-C1further enhances the mixing of jet and ambient air.Nozzle-C1has a OASPL reduction in varying degrees at 60°and 150°，and the reduction is about 0.2dB at 60°and about 1.1dB at 150°.But Nozzle-C1has a slightly OASPL increase of about 0.5dB at 90°.In addition，compared with Nozzle-A，Nozzle-C1has a clear OASPL reduction of about 4.6dB at 150°，but has some OASPL increase of about 1.7dB at 90°and about 0.9dB at 60°.

Fig.8　Comparisons of noise directivity from Nozzle-C1 with Nozzle-A and Nozzle-B

5　Conclusions

In this paper，the investigation shows that compared with the nozzle with lobed mixer，the optimum external chevron nozzle with lobed mixer increases mixing of jet and ambient air near the nozzle exit and has better noise reduction at low frequencies.The main results of this present are described as follows：

（1）Compared with the nozzle with lobed mixer，the external chevron nozzle with lobed mixer largely reduces jet noise at low frequencies. The jet noise reduction at low frequencies increa-ses a little with the increase of chevron bend angle.Nozzle-C2（Chevron bend angle of 22°）produces a 0.7—1.7dB reduction compared with the nozzle with lobed mixer and a 3—4dB reduction compared with the nozzle with annular mixer at low frequencies.

（2）Compared with the nozzle with lobed mixer，the external chevron nozzle with lobed mixer has a SPL increase which is not obvious at high frequencies.SPL has relatively clear increase at 60°and little fluctuation at 90°and 150°with chevron bend angle increasing.

（3）Compared with the nozzle with lobed mixer，Nozzle-C1with a chevron bend angle of 18°has OASPL reduction in varying degrees at 60°and 150°.It produces about 0.2dB reduction at 60°and about 1.1dB reduction at 150°，but has a slightly OASPL increase of about 0.5dB at 90°.

References：

［1］ U.S.Environment Protection Agency.Noise standards for aircraft type certification（Modification to Far Part 36）：EPA 550/9-76-013［R］.Washington DC：［s.n.］，1994.

［2］ CSANADY G T.The effect of mean velocity variations on jet noise［J］.Journal of Fluid Mechanics，1966，26（1）：186-197.

［3］ VINOD G M，WILLIAM N D.Lobed mixer design for noise suppression，acoustic and aerodynamic test data analysis：NASA-CR-2002-210823［R］.2002.

［4］ TESTER B J，FISHE M J A.Contribution to the understanding and prediction of jet noise generation in forced mixers：AIAA 2004-2897［R］.2004.

［5］ ZAMAN K B M Q.Supersonic jet mixing enhancement by″Delta-Tabs″：AIAA 92-3548［R］.1992.

［6］ PAN Chengxiong，ZHANG Jingzhou，SHAO Wanren，et al.Numerical research on flow dynamics and infrared radiation characteristics for scaled chevron convergent nozzle［J］.Journal of Aerospace Power，2010，25（7）：1518-1523.（in Chinese）

［7］ LIU Youhong，FAN Chao，XIE Yi，et al.Effect of lobe numbers on the performance of a lobed mixer in the forced mixing exhaust system［J］.Journal of Aerospace Power，2010，25（8）：1683-1689.（in Chinese）

［8］ LI Teng，LIU Youhong，XIE Yi，et al.Effect of ration of height to width of lobe on performance offorced mixing exhaust system［J］.Journal of Aerospace Power，2013，28（8）：1736-1743.（in Chinese）

［9］ DING Yulin，LIU Youhong，XIE Yi，et al.Effect of serrated cut at trailing edge on performance of lobed mixer in forced mixing exhaust system［J］.Journal of Aerospace Power，2012，27（10）：2236-2242.（in Chinese）

［10］LIU Youhong，ZHANG Shaopeng，YANG Xu，et al. Effects of cut angles at trailing edge of lobe on aerothermodynamic performance of forced mixing exhaust system in turbofan engine［J］.Journal of Prolulsion Technology，2013，34（6）：734-740.（in Chinese）

［11］LIU Youhong，DU Liwei，XIE Yi，et al.Effect of elliptical cut at trailing edge of lobed mixer on aerodynamic and thermal performance of a turbofan forced mixing exhaust system［J］.Science Technology and Engineering，2013，13（18）：5226-5233.（in Chinese）

［12］SHAN Yong，ZHANG Jingzhou，XU Liang.Numerical investigation of aerodynamic and mixing characteristics of scarfed lobed mixer［J］.Transactions of Nanjing University of Aeronautics and Astronautics，2009，26（2）：130-136.（in Chinese）

［13］ZHANG Zhengwei，ZHANG Jingzhou，SHAO Wanren，et al.Numerical caculation of flow and acoustics fields for separated exhaust nozzle with bypass offset［J］.Journal of Aerospace Power，2012，27（5）：1139-1145.（in Chinese）

［14］SHAO Wanren，XU Xihai，HE Jingyu，et al.Experimental and numerical study of chevron nozzle for jet noise reduction［C］∥Proceedings of Asian Joint Conference on Propulsion and Power（AJCPP2012-140）.Xi′an，China：JCPP，2012.

［15］SHAN Yong，ZHANG Jingzhou，SHAO Wanren，et al.Experimental and numerical research on jet noise suppression with chevron nozzle for turbofan engines［J］.Acta Aeronautica et Astronautica Sinica，2013，34（5）：1046-1055.（in Chinese）

［16］SHAN Yong，ZHANG Jingzhou.The effects of turbofan exhaust system geometry parameters on sound pressure level［J］.Journal of Prolulsion Technology，2010，31（3）：309-313.（in Chinese）

［17］TAM C K.Jet noise：Since 1952［J］.Journal of Fluid Dynamics，1998，10（4）：393-405.

［18］CROW S C，CHAMPAGNE F H.Orderly structure in jet turbulence［J］.Journal of Fluid Mechanics，1971，48（3）：547-591.

［19］BROWN G L，ROSHKO A.On density effects and large structure in turbulent mixing layers［J］.Journal of Fluid Mechanics，1974，64（4）：775-816.

Dr.SHAO Wanren is currently a researcher at Shenyang Engine Design and Research Institute，Aviation Industry Corporation of China.His major research interests are aeroengine exhaust nozzle design and exhaust noise research.

Mr.WU Fei is currently an engineer at Shenyang Engine Design and Research Institute，Aviation Industry Corporation of China.His major research interests are aeroengine exhaust nozzle design and exhaust noise research.

Dr.HE Jingyu is currently an engineer at China Academy of Aerospace Aerodynamics.His major research interests are aerospace aerodynamics research and exhaust noise research.

Dr.LI Xiaodong is a currently professor at School of Energy and Power Engineering，Beihang University.His major research interests are aeroacoustics research and unsteady aerodynamics research.

（Executive Editor：Xu Chengting）

TB52 Document code：A Article ID：1005-1120（2016）02-0129-08

＊Corresponding author，E-mail address：shaowanren135@163.com.

How to cite this article：Shao Wanren，Wu Fei，He Jingyu，et al.Jet noise reduction of double-mixing exhaust system［J］. Trans.Nanjing Univ.Aero.Astro.，2016，33（2）：129-136.

http：//dx.doi.org/10.16356/j.1005-1120.2016.02.129