Experimental development of a measurement technique to resolve the
radial distribution of fan aero-acoustic emissions
S. Bianchi
a)
, A. G. Sheard
b)
, I. R. Kinghorn
b)
, A. Corsini
c)
and F. Rispoli
c)
(Received: 24 January 2008; Revised: 29 September 2008; Accepted: 29 September 2008)
This paper presents an experimental technique developed to assess the
performance of a family of axial fans for compact cooling units. It is used to
study fan rotor noise sources to enable the effectiveness of design modifications
to be assessed. By varying the radial positions of a near field microphone,
correlations with a far field microphone are used to study noise source
signatures along the blade span. The correlations reveal distinctive acoustic
signatures that are described in detail. The methodology has been shown to be
effective in identifying: (i) the blade tip feature with the best acoustic
performance; and (ii) other significant noise sources along the blade span.
© 2009 Institute of Noise Control Engineering.
Primary subject classification: 11.4.1; Secondary subject classification: 72
1 INTRODUCTION
Fans that operate in the cooling systems of industrial
heat exchangers and in building heating and ventilation
are a source of noise pollution. The link between acous-
tic emissions and the aerodynamic features of the rotor
flow has been studied extensively; in particular, the
work of Wright
1
, Cumpsty
2
and Holste and Neise
3
has
enhanced understanding of the aero-acoustics of turbo-
machines.
According to the works of Cumpsty
2
and Holste and
Neise
3
the amount of power radiated into noise by a
rotor is generally small compared with the aerody-
namic power released to the air flow. Noise is radiated
by forces, volume displacements and nonlinearities
which are either unsteady or periodic in their effects.
According to the origin of the noise produced, the
source mechanisms considered in this study can be
divided into three categories: inflow turbulence noise,
impeller noise and tip flow noise (Fukano et al.
4,5
and
Sharland
6
).
The first source originates from the unsteady loading
due to the ingested turbulence fluctuations. Inflow
turbulence was seen to be an important noise source
over a range of frequencies in the work of Wright
1
for
low speed fans and Cumpsty
2
for high speed turboma-
chinery. Their analysis also focused on the humped
nature of the low frequency spectrum, below the BPF,
as being due to the large scale components of incoming
turbulence. These large scale components give nearly
periodic disturbances as they are swept through the
rotor plane.
As far as the self-generated noise from the impeller,
as recognized as far back as 1976
1
, it is produced by the
pressure fluctuations from the turbulent boundary layer
on the blade and by the shedding of vortices from the
trailing edge. This confirms earlier observations by
Sharland
6
, stating that the local forces set up from the
random fluctuation of a moving flow on a surface, may
well act as acoustic sources and so, at least in the first
instance, it is reasonable to consider the turbulent
boundary layer inside a fan as a possible source of
noise. Moreover, the work of Cumpsty
2
argued that the
shear layer, interacting with the rotating blades should
lead to additional noise in some manner. The same
lifting surfaces that create the shear layer produce a
dipole-like source that rotates with the blade. When the
dipole rotates with the machine the repetitive pattern of
the fluctuating pressure passing a given point (as a
stationary strut) produces the narrowband tone known
as the BPF. This tone and its harmonics are the
hallmark of this kind of rotating machinery.
When looking at the tip region of a fan blade, it is
recognized as the most significant noise source, as
stated in dedicated studies on the far field emission of
this source (Longhouse
7
, and Fukano and Jang
8
).
Marcinowski
9
, who was the first to present a study of
noise associated with tip dynamics, demonstrated that
increases in broadband noise levels occur with increas-
ing tip clearance, with the largest changes being appar-
a)
Dipartimento di Meccanica e Aeronautica, Sapienza Uni-
versity of Rome, Via Eudossiana 18, Roma, 00184,
ITALY; email: bianchi@dma.ing.uniromal.it
b)
Fan Technology, Fläkt Woods Ltd, Axial Way, Colchester,
Essex, C04 5ZD, UK.
c)
Dipartimento di Meccanica e Aeronautica, Sapienza Uni-
versity of Rome, Via Eudossiana 18, Roma, 00184, ITALY
360 Noise Control Eng. J. 57 (4), July-Aug 2009