Effects of Tip Clearance On Rotor/Stator In-Teraction Tonal Noise of Axial Fan

Download as pdf or txt
Download as pdf or txt
You are on page 1of 8

The 21st International Congress on Sound and Vibration

13-17 July, 2014, Beijing/China

EFFECTS OF TIP CLEARANCE ON ROTOR/STATOR IN-


TERACTION TONAL NOISE OF AXIAL FAN
Liangfeng Wang, Weiyang Qiao, Weijie Chen, Liang Ji
School of Power and Energy, Northwestern Polytechnical University, Xi’an 710072, China
e-mail: wlfzxlgnl@126.com

The tip leakage flow is crucial importance for the turbomachinery design and operation for its
contribution in loss and noise production. This study is focused on the effects of tip clear-
ance(TC) on tonal noise behavior of a single stage axial fan. The flow-field/acoustic-field
hybrid model is used to evaluate the tonal noise level from a time-accurate CFD result. The
hybrid model is based on the three-dimensional unsteady Reynolds-averaged Navier-Stokes
(URANS) equations and the ducted blade-rows aerodynamics and aeroacoustics(DBAA) the-
ory. The results indicate that the high amplitude areas of unsteady loading fluctuation are
mainly around the leading edge of stator vane, and the effect of TC on the noise source dis-
tribution is mostly in the tip region. The varies of total acoustic power level is less than 1dB
as the TC is changed from 0mm to 1mm. The effect of TC on the total acoustic power level is
not same for different frequency or different propagated direction.

1. Introduction
In recent years, commercial aircraft noise has become a major concern for society, and the
aircraft engine noise is a main noise source, which includes the tonal noise and the broadband noise.
For the subsonic turbomachinery, the tonal noise is an important component. The design of effec-
tive new technology to reduce aircraft noise is dependent on identifying and understanding the noise
source and noise generation mechanisms. In the design of modern axial fan for commercial high-
bypass turbo-fan engine, aeroacoustics signatures has become a main factor that are considered. The
tip leakage flow is of crucial importance for the turbomachinery design and operation for its contri-
bution in loss and noise production1, 2. This study is focused on the effects of tip clearance(TC) on
tonal noise behavior of a single stage axial fan.
The nature of a tip leakage vortex observed in a rotor tip region has been studied by many re-
searchers because of its important role on the flow field; for example, Inoue and Kuroumaru3,
StorerandCumpsty4, and Lakshminarayana et al.5 They showed that the tip leakage flow has a three-
dimensional and an unsteady nature, and effect on a loss production and a noise generation. How-
ever, most studies are mainly focused on the vortical flow in a blade passage with relation to aero-
dynamic performance without considering the sound generation due to the unsteady behavior of the
vortical flow.
Only a few studies are concerned with the noise due to rotor tip leakage vortex, which was
thought broadband naturally. Kameier and Neise6 showed that the reverse flow of the rotor tip in
turbomachines was the important source of the tip clearance noise. The noise due to tip clearance
flow on low speed axial flow fans was analyzed by an experimental measurement using hot-wire
probes rotating with the fan blades by T. Fukano et al.7, 8 The results show that the spectral peaks

ICSV21, Beijing, China, 13-17 July 2014 1


21st International Congress on Sound and Vibration (ICSV21), Beijing, China, 13-17 July 2014

appeared in sound spectra although a tip clearance noise is broadband naturally and the peak fre-
quency is proportionally increased by the increase of the rotational speed and also the flow rate at
constant rotational speed. Julien Grilliat et al.9, 10 carried out a combined near and far field aero-
acoustic measurement during a low Mach number experiment about the tip leakage flow past a sin-
gle non-rotating airfoil. They concluded that the clearance flow generate sound when leaving the
clearance region, and the tip leakage vortex feeds unsteady perturbations into the blade wall pres-
sure field that become sound sources as they are scattered by the tip edge or the trailing edge corner.
Alessandro Corsini et al.11 tested blade-tip end-plates configurations designed as a passive noise
control in a duct fan. The beneficial effect on the fan aeroacousitc signature was obtained as a result
of the changes they induced in tip-leakage-flow behavior.
In the said literatures, the tip-clearance flow noise is recognized as a broadband noise, which
is generated as turbulent boundary–layer noise from the blade surfaces and end-walls. In tur-
bomachinery, the tip clearance also affects the tonal noise since it produces vortexes that rotate with
rotor blade with blade passing frequency. Zhao L12 assessed the effects of tip leakage vortex on the
stator/rotor interaction tonal noise and its unsteady behavior by 3D unsteady numerical simulations
and the Triple-Plane Pressure matching strategy. His results showed that the tonal noise level of
turbine stage is decreased with the enlargement of tip clearance. However only tip clearance of
0mm, 1mm and 2mm were investigated because the amount of the computer efforts is required so
much for Triple-Plane Pressure matching strategy, moreover, he didn’t research the effects of tip
clearance on rotor/stator interaction tonal noise and on the noise level in upstream direction.
The present paper investigated effects of tip clearance on rotor/stator interaction tonal noise of
a single stage fan, the flow-field/acoustic-field hybrid model is used to evaluate the tonal noise level
from a time-accurate CFD result. The hybrid model is based on the three-dimensional unsteady
Reynolds-averaged Navier-Stokes (URANS) equations and the ducted blade-rows aerodynamics
and aeroacoustics(DBAA) theory. In the URANS/DBAA hybrid model, the detailed design parame-
ters of the turbomachinery blades can be related to the noise level, and the amount of the computer
efforts is not required so much because the Lighthill’s acoustic analogy theory was used.

2. Calculation models and numerical scheme


2.1 Calculation model and grid
In present paper, a single stage fan was chosen for research. Major parameters of geometry
and operating condition are shown in Table 1.
Table 1. Major parameters of geometry and operating condition of the fan
Number of Rotor Blade 24 Axial Flow Mach Number 0.325
Number of Stator Vane 36 RPM of Operating Point 7000
Hub to Tip Ratio 0.3063 Total pressure ratio 1.21

24 fan rotor blades and 36 stator vanes allow a combination of 2-rotor/3-stator airfoil to be
simulated. A computational domain is shown in Figure 1, and a meridional plane of a computational
domain is shown in Figure 2.
Figure 3 indicates the efficiency of the fan model varies with number of grid. The results
came from steady simulations with single rotor and vane passage, based on which 60 layers grid
along span direction (without tip clearance) was chosen. Four cases are modeled in present work,
the TC height are 0mm, 0.25mm, 0.5mm, 1mm (0.35% blade height). The value of dimensionless
wall distance (y+) is below 3.5 in tip region. Figure 4 gives the velocity distribution at TC for dif-
ferent number of grid layers in tip region. To capture flow field at TC precisely, there are at least 15

ICSV21, Beijing, China, 13-17 July 2014 2


21st International Congress on Sound and Vibration (ICSV21), Beijing, China, 13-17 July 2014

cells set in tip region. In our cases, we set 15 cells for 0.25mm, 0.5mm TC height and 20 cells for
1.0mm TC height.

Figure 1. A computational domain divided into 5 Figure 2. Meridional plane of a computational do-
sub-domains main

Figure 4. Velocity profile inside the tip clearance of


Figure 3. Fan Efficiency vs Grid Number rotor outlet (1.0mm height) for different number of
cells

2.2 Numerical method and boundary condition


The simulation was carried out by commercial CFD (Computational Fluid Dynamics) solver
ANSYS CFX. It solves Reynolds Average N-S equation under Generalized Coordinates, by finite
volume method for space discretization and implicit dual-time stepping method for time discretiza-
tion. High resolution scheme for space and second order backward Euler scheme for the time were
used to discretize the equation.
At the inlet, total pressure and total temperature were specified, the flow direction is along the
rotational axial. At the exit, average static pressure was specified. No slip adiabatic wall was im-
posed on shroud, hub, and the surface of rotor blades and stator vanes.

3. UNRANS/DBAA hybrid model


UNRANS/DBAA hybrid model, which is based on the Goldstein duct aeroacoustics equation,
viz. acoustic analogy equation, simulates the turbomachinery tonal noise by coupled the results of
turbomachinery three-dimensional unsteady numeric simulation to the Goldstein equation.
For the subsonic turbomachinery, we can neglect the contributions of the volume quadrupole
term and the volume displacement effects, the acoustic pressure in the duct can be written as13:
p  x,     pmn ( ) m,n  m,n r  exp(im  i n,m x1 ) (1)
m n

Where p is acoustic pressure, pmn ( ) is duct mode amplitude,  m,n is eigenfunction of duct,
 m,n is eigenvalues of annular duct modes, m and n are circumferential mode order and radial
mode order, respectively.

ICSV21, Beijing, China, 13-17 July 2014 3


21st International Congress on Sound and Vibration (ICSV21), Beijing, China, 13-17 July 2014

Unsteady loading on the stator surface, which is regarded as a fan tone noise source in this
study, is directly calculated by CFD. Then, pmn ( ) can be calculated based on the duct acoustics
theory. The sound power spectral density is calculated by [14]:
 (rT2  rH2 ) M 2 (1  M 2 )2 ( / U )kmn ( )
Wmn ( )    pmn ( ) (2)
0U [ / c0  Mkmn ( )]2

Where kn,m  ( / c0 )2  (1  M 2 ) m2 ,n , M is axial Mach number, U is axial fluid velocity,


rT and rH are duct radius and hub radius, respectively.
The effects of rotor-stator spacing and rotating speed on turbomachinery tonal noise have
been studied using the model. Table 2 shows the effect of rotor-stator spacing on tonal noise, which
were obtained by UNRANS/DBAA model and empirical model. The empirical model used here
was 10log10(C/S), which was use in Heidmann fan noise prediction model15, where C is chord of
blade in upstream and S is rotor/stator spacing. “PWL_1s” means the sound power level (PWL) at
the normal rotor-stator spacing, “PWL_2s” means the PWL when the rotor-stator spacing is two
times of normal value.
Table 2. Major parameters of geometry and operating condition of the fan
URANS/DBAA Empirical
1BPF 2BPF 3BPF Model
PWL_1s /dB 102.0 97.5 90.6
PWL_2s /dB 98.7 89.4 78.3
ΔPWL /dB -3.3 -8.1 -12.3 -3

Figure 5 gives the comparison of URANS/DBAA with Lowson and TPP prediction about the
variation of PWL with rotating speed of a turbine stage. The detailed parameters of the turbine stage
are in chapter 3 of reference [16].

Figure 5. Comparison of URANS/DBAA with Lowson and TPP prediction about the variation of PWL with
rotating speed

Form Table 2 and Figure 5, we can see that the hybrid model can be used to study the effects
of detailed design parameters of turbomachinery blades on tonal noise.

4. Results and discussions


4.1 Characteristics of noise Source contribution
Figure 6 shows unsteady loading distribution on the stator surface, which is calculated by
CFD analysis. A component of BPF, 2BPF and 3BPF for different tip clearances are shown in Fig-
ures. The left figures show the amplitude of unsteady loading, and the right figures show real parts

ICSV21, Beijing, China, 13-17 July 2014 4


21st International Congress on Sound and Vibration (ICSV21), Beijing, China, 13-17 July 2014

of instantaneous unsteady loading. Each figure has its own color range, so that it is easy to under-
stand characteristics of a noise source. Unsteady loading does not contain a component of steady
loading.

TC=0mm TC=0.25mm TC=0.5mm TC=1mm TC=0mm TC=0.25mm TC=0.5mm TC=1mm


Amplitude Real Part
(a) BPF Component

TC=0mm TC=0.25mm TC=0.5mm TC=1mm TC=0mm TC=0.25mm TC=0.5mm TC=1mm


Amplitude Real Part
(b) 2BPF Component

TC=0mm TC=0.25mm TC=0.5mm TC=1mm TC=0mm TC=0.25mm TC=0.5mm TC=1mm


Amplitude Real Part
(c) 3BPF Component
Figure 6. Unsteady loading distribution on the stator surface directly calculated by CFD

From Figure 6, we can see that a high amplitude area can be observed near the leading edge in
all BPF harmonics and for all tip clearances. Focused on a BPF component of Figure 6, high ampli-
tude areas were observed not only near the leading edge but also around the trailing edge of the tip
region, which means that this region is also the main noise source. Positive and negative values
around the leading edge in the right figures come from the phases of rotor-wake and stator interac-
tion. Existence of one and a half pairs of positive and negative values means that the stator leading
edge intersects the rotor-wake one and a half on average. As the TC increases, the amplitude and its

ICSV21, Beijing, China, 13-17 July 2014 5


21st International Congress on Sound and Vibration (ICSV21), Beijing, China, 13-17 July 2014

phase distribution (see Real Part distribution) of the unsteady loading are very similar to each other
for all TC height. The differences are mainly in the tip region, which means that for the fan model,
the effects of rotor blade TC height on the flow-field are very small, the changes of flow-field
around the stator vane are mainly in the tip region.

4.2 Acoustic Results


Acoustic waves generated from unsteady loading on the stator surface and propagated in a fan
duct are calculated. Figure 7 shows the acoustic power level (upstream direction) of each acoustic
mode of 0mm and 1mm tip clearance. Acoustic modes consist of circumferential modes m and radi-
al modes n . Only cut-on modes are shown in the Figure. Radial modes more than 8th order are not
shown here. One circumferential mode is cut-on in BPF, and two/three circumferential modes are
cut-on in 2BPF/3BPF respectively. It can be found that the effects of TC height on acoustic power
level of each acoustic mode are very little.

1BPF 2BPF 3BPF


(a) TC=0mm

1BPF 2BPF 3BPF


(b) TC=1mm
Figure 7. Comparison of acoustic power level of each acoustic mode between TC=0mm and TC=1mm

Total acoustic power level of different tip clearance is predicted. Total acoustic power level is
calculated by means of summing up acoustic power level of all the modes. Table 3 shows the acous-
tic power level of different tip clearance, where “Fwd” and “Aft” are the upstream and downstream
direction of fan aerodynamic flow, respectively.
Table 3. Acoustic power level of different tip clearance
Fwd Aft
TC= TC= TC= TC= TC= TC= TC= TC=
0mm 0.25mm 0.5mm 1mm 0mm 0.25mm 0.5mm 1mm
PWL_1BPF 132.04 131.71 131.57 131.85 138.74 138.53 138.31 138.39
PWL_2BPF 138.90 138.77 138.91 139.09 138.60 138.49 138.43 138.45
PWL_3BPF 126.33 126.40 126.58 126.60 130.29 130.33 130.43 130.42

ICSV21, Beijing, China, 13-17 July 2014 6


21st International Congress on Sound and Vibration (ICSV21), Beijing, China, 13-17 July 2014

From Table 3, we can see that, for the fan model, as the TC of fan blades changed from 0mm
to 1mm, the varies of total acoustic power level is less than 1dB. For the noise of 1BPF, the acous-
tic power level of both “Fwd” and “Aft” decreased as the TC changed from 0mm to 0.5mm, but the
acoustic power level of TC=1mm is higher than that of TC=0.5mm. For the noise of 2BPF, the
acoustic power level of “Fwd” decreased as the TC changed from 0mm to 0.25mm, but increased as
the TC changed from 0.25mm to 1mm, and the acoustic power level of “Aft” decreased as the TC
changed from 0mm to 0.5mm, but the acoustic power level of TC=1mm is higher than that of
TC=0.5mm. For the noise of 3BPF, the acoustic power level of both “Fwd” and “Aft” increased as
the TC increased. The results indicated that the effect of TC on the total acoustic power level is not
same for different frequency or different propagated direction. The changed trend of acoustic power
level when the TC height is larger than 1mm will be studied in the future work. Compared with the
experiment results of the effect of TC on fan noise17, it can be found that the effect of TC on the fan
tonal noise is very complicated. For different fan models, different frequency, different operating
points, and different propagated direction, the results will be different.

5. Conclusion
The effect of tip clearance on tonal noise of a single stage fan is studied in the paper. The re-
sults are summarized as follows:
1. The high amplitude areas of unsteady loading fluctuation are mainly around the leading
edge of stator vane, which means that the region is the main source of the tonal noise.
2. The effect of rotor blade TC height on the noise source distribution of the stator vane sur-
face and on the acoustic power level of each acoustic mode are very small, the changes of noise source
distribution are mainly in the tip region.
3. The varies of total acoustic power level is less than 1dB as the TC is changed from 0mm to
1mm. The effect of TC on the total acoustic power level is not same for different frequency or dif-
ferent propagated direction.
The changed trend of acoustic power level for larger TC, for different operating points will be
studied in the future work.

Acknowledgments
This work was supported by NSFC under grant 51276149.

REFERENCES
1
Denton J D. Loss mechanisms in turbomachinery. ASME Paper 93-GT-435.
2
Fukano, T., Takamatsu, Y. The effects of tip clearance on the noise of low-pressure axial and
mixed flow fans, Journal of Sound and Vibration, 105, 291-308, (1986)
3
M. Inoue, M. Kuroumaru. Structure of tip clearance flow in an isolated axial compressor rotor,
American Society of Mechanical Engineers, Journal of Turbomachinery,111 (3) , 250–256,
(1989)
4
J.A. Storer, N.A. Cumpsty. Tip leakage flow in axial compressors, American Society of Me-
chanical Engineers, Journal of Turbomachinery , 113, 252–259, (1991).
5
B. Lakshminarayana, M. Zaccaria, B. Marathe. The structure of tip clearance flow in axial
flow compressors, American Society of Mechanical Engineers, Journal of Turbomachinery,
117, 336–347, (1995).
6
F. kameier, W. Neise. Experimental study of tip clearance losses and noise in axial tur-
bomachines and their reduction, ASME, Journal of Turbomachinery, 119, 460-471, (1997).

ICSV21, Beijing, China, 13-17 July 2014 7


21st International Congress on Sound and Vibration (ICSV21), Beijing, China, 13-17 July 2014

7
Choon-Man Jang, Tohru Fukano and Masato Furkawa. Effects of the Tip Clearance on Vorti-
cal Flow and Its Relation to Noise in an Axial Flow Fan, JSME International Journal, Series
B, 11(3), 356-356, (2003).
8
T. Fukano, C.-M. Jang. Tip clearance noise of axial flow fans operating at design and off-
design condition. Journal of sound and vibration, 275, 1027-1050, (2004).
9
Julien Grilliat, Mare C. Jacob, Roberto Camussi and Giovanni Caputi-Gennaro. Tip Leakage
Experiment–Part One: Aerodynamic and Acoustic Measurements, 13th AIAA/CEAS Aeroa-
coutics Conference (28th AIAA Aeroacoustics Conference), AIAA 2007-3684.
10
M. C. Jacob, J. Grilliat, R. Camussi and G. Caputi Gennaro. Aeroacoustic investigation of a
single airfoil tip leakage flow, Aeroacoustics, 9(3), 253-272, (2010).
11
Alessandro Corsini, Franco Rispoli and A. G. Sheard. Aerodynamic Performance of Blade
Tip End-Plates Designed for Low-Noise Operation in Axial Flow Fans. Journal of Fluids
Engineering, 131, 081101, (2009).
12
Zhao L, Qiao W.Y, etc. Tip Clearance effects on Unsteady Behavior of the Tip Leakage Vor-
tex and Stator/Rotor Interaction Tonal Noise. Proceedings of ASME Turbo Expo 2012.
GT2012-68549.
13
Munjal M L. Acoustics of ducts and mufflers with application to exhaust and ventilation sys-
tem design. John Wiley & Sons, 1-15, (1987).
14
Ventres C S. Turbofan Noise Generation. Volume 1: analysis . NASA CR-167952, (1983).
15
Heidmann M F. Interim prediction method for fan and compressor source noise. NASA TM X-
71763, (1975).
16
Zhao L. Theory and Method Investigation of the Aerodynamic-Acoustics Integration Design
in Turbine. Doctor of Science Thesis, Northwestern Polytechnical University, 59-79, (2012).
17
Christopher E Hughes, Richard P Woodward, Gary G Podboy. Effect of Tip Clearance on Fan
Noise and Aerodynamic Performance, 11th AIAA/CEAS Aeroacoustics Conference (26th AI-
AA Aeroacoustics Conference), AIAA 2005-2875.

ICSV21, Beijing, China, 13-17 July 2014 8

You might also like