Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                

Flow Investigation Inside A C Shaped Curved Diffuser: Presented by

Download as pptx, pdf, or txt
Download as pptx, pdf, or txt
You are on page 1of 44

FLOW INVESTIGATION INSIDE A C SHAPED CURVED

DIFFUSER
Presented by
KONETI LEELASAGAR
Roll Number – 18ME4307
Under the Guidance of

Dr. Arup Kumar Biswas Dr. Prasanta K. Sinha


Associate Professor Professor and Principal
Dept. of Mechanical Engineering DIATM
NIT Durgapur G.T. Road, Rajbandh
Durgapur – 713209 Durgapur – 713212
West Bengal, India West Bengal, India

National Institute of Technology Durgapur


Department of Mechanical Engineering
Durgapur – 713209, West Bengal, India
CONTENTS

• Introduction
• The problem under consideration
• Literature review
• Experimental facility and methodology
• Results and discussion
• Conclusions
• References

2
INTRODUCTION

• A diffuser is a mechanical device which recovers the pressure energy from the
flowing fluid at the expense of its kinetic energy.
• Diffusers find very wide applications in the field of turbo-machinery and aerospace.
These are designed in different shapes and sizes to meet the specific application.
• Curved diffusers are finding huge applications in the field of aircraft applications
due to space restrictions and design compatibility.
• Study of flow parameters within the curved diffusers has been a research area for
the last century to many researchers.
• Pioneer in this was Eustine (1910) who has investigated the nature of flow within
the curved diffusers.
• In curved channels the radial pressure gradient, resulting from the centrifugal force
produce significant secondary flows.

Contd.
3
• A secondary flow is a relatively minor flow superimposed on the primary flow.
This flow takes place in a plane normal to the main flow. It has been verified that
the secondary flow can influence the flow parameters.
• The presence of curvature generates centrifugal force. The centrifugal force on the
flowing fluid results in the development of a secondary motion, which is
manifested in the form of a pair of counter-rotating vortices.

4
THE PROBLEM UNDER CONSIDERATION
• The geometry of the test curved diffuser as shown in the fig.3(a) is C shape of
90° of dimension 100 mm X 100 mm at inlet and at outlet its dimension is 150
mm X 100 mm with a centerline length of 750 mm.
• It is constituted of four segments of 22.5° each. The entire test piece is made of
Perspex sheet.
• Two straight constant area ducts of cross-sectional area 100 mm x 100 mm and
of 150 mm X 100 mm are connected as extension pieces at the inlet and exit of
the test piece respectively.
• Middle section of all the six segments are considered as six sections and they are
named as Inlet Section, Section-A, Section-B, Section-C, Section-D and Outlet
Section.
• The radius of centerline curvature of the test diffuser is fixed at 477.5 mm.
• Height of the diffuser remains constant i.e. 100 mm throughout from inlet to
outlet but its width changes from 100 mm to 150 mm as shown in fig.3(a)

5
Fig 3(a). Geometry of C Shape curved diffuser
6
B-SEC C-SEC
D-SEC

A-SEC
INLET SEC OUTLET-SEC

Fig 3(b) C Shape curved diffuser


7
Measurement of Flow Parameters
 A 10 mm width is cut on the top face of each section so that three hole probe can easily
travel through the whole section.
 At inlet section, the distance of the hole from the wall-2 is 5, 10, 15, 25, 35, 45, 55, 65,
75, 85, 90, 95 mm respectively, as shown in Fig.3(b)
 Readings are taken at different depth starting from the height of (1, 3, 5, 7, 9, 11, 13, 15,
20, 25, 30, 40, 50, 60, 70, 75, 80, 85, 87, 89, 91, 93, 95, 97, 99) mm, total 25 measuring
locations for one hole then at a distance of 10 mm from the wall again repeat the same
process.
 For taking the measurements, the particular hole was connected to the pre-calibrated
Three hole Pressure Probe then to inclined tube manometer through flexible tube.
 At Inlet section (12 x 25) 300 locations, Section-A (14 x 25) 350 locations, Section-B
(14 x 25) 350 locations, Section-C (16 x 25) 400 locations, Section-D (16 x 25) 400
locations and at Outlet section (18 x 25) 450 point locations measurements were taken.
 Flow parameters Such as velocity, total pressure and static pressure were calculated.

8
Fig-3(b): Measuring location of inlet section
9
LITERATURE REVIEW
• Sullerey [1] in 1984 performed a comparison between straight diffuser and curved
diffuser. He concluded that the straight diffuser performance is higher than that of
curved diffuser.
• Sahin in 1995, [2] used gauze screens to control the flow inside their wide-angle
diffusers in which they achieved improved flow uniformity and reasonable pressure
recovery.
• Agrawal and Singh, 2003[3] by carrying out flow visualisation and wall static
pressure measurements on an elliptical centre line 90º curved diffuser of large area
ratio (AR = 3.4), detected large separation pockets on the convex wall. The
separated flow affected the diffuser performance severely and the performance
evaluation showed poor pressure recovery and high losses.
• Mullick and Majumdar [4] in 2006 studied the performance of a 22.5° / 22.5°
circular cross section S-shaped diffuser. The experimental results indicated the
generation of secondary flow in the form of a pair of counter rotating vortices in
the first half, which changes its senses of rotation in the second half.

10
• Sedlar, and Prihoda, 2007, [5], modelled numerically the flow phenomena
occurring in turbulent flows through rectangular curved diffusers of AR = 1.5,
Δβ = 900, with straight inlet and exit parts. Regions of flow separation were
marked and were found to be dependent both on diffuser geometry and
Reynolds number. Secondary flows of different structures of multiple vortices
were found to form which were influenced mainly by the nature of flow
separation.
• Sumida [6] in 2009 carried out preliminary experiments on a pulsating turbulent
flow in a conical diffuser with a divergence angle (2θ) of 12°. It was found that
the distributions of the pressure and velocity exhibit complicated behaviours,
which are different from those in a steady flow.
• P. K. Sinha [7] in 2011 investigated an experiment on 42° C-shape diffusing
duct. They measured the mean velocity, static pressure and total pressure along
the flow passage of the diffuser. They observed that with the increase in area
ratio pressure recovery increases up to certain point then with further increase in
area ratio Pressure recovery decreases.

11
EXPERIMENTAL FACILITY AND
METHODOLOGY
Experiment is carried out using the facility of wind tunnel as shown in fig. 3(a) at
the aerodynamics laboratory of NIT Durgapur.

Settling chamber Conical diffuser


Flexible
coupling

Conical diffuser

Flow
Contraction
Contraction cone controller Air supply
Nylon screen Flexible
Settling coupling Air supply
Air
cone unit
chamber unit
supply
unit
Flow control
Contraction cone Flow controller
arrangement

Fig. 3(a)
Contd. 12
Air Supply Unit

For present investigation, a


centrifugal blower is directly
coupled to a 5.5 KW motor (ABB
make) of 2870 rpm. The mean
flow rate was controlled by flow
controller. Centrifugal blower is
shown in the Fig. 3(b).

Fig. 3(b)

Contd.

13
Flow Controller

A metal gate valve is fitted


at blower intake to control
the flow rate of blower as
shown in the Fig. 3(c). By
rotating it clockwise it’s
opening increases and for
anticlockwise rotation it
decreases.

Fig. 3(c)

Contd.

14
Flexible Coupling

•A flexible extension made of


canvas of dimension 0.155m x
0.310m as shown in the Fig.3(d) is
fitted between the blower outlet and
remaining portion of the wind
tunnel set up.
•This piece minimizes the vibration
transmitted by the air supply unit to
the test piece.

Fig. 3(d)

Contd.

15
Conical Diffuser

•The blower is followed by


conical diffuser as shown
in the fig 3(e). It is made
of G.I. sheet and of length
1.38m having rectangular
inlet of dimension 0.155m
x 0.310m and outlet
diameters of 0.600m
respectively with a
diverging angle of 6.66.
•It smoothly slows the air
velocity and reduces
turbulence level. Fig. 3(e)

Contd.

16
Settling Chamber

The settling chamber as shown in the Fig.3(f) is a straight cylindrical shell


through which the discharge from the conical diffuser flows. It is of uniform
diameter of 0.600m and length 2.88m. Nylon screens are provided at three
locations in the settling chamber in the transverse direction to flow for
straightening the flow and reducing the turbulence level of the flow.

Settling chamber

Nylon screen

Fig. 3(f)
Contd.

17
Contraction Cone
•The contraction cone as
shown in the Fig. 3(g) is a
pyramidal structure made of
wood and of square section
at both of its inlet and outlet.
The inlet dimensions of the
contraction cone are 0.30m x
0.30m and outlet dimension
of 0.10m x 0.10m.
•This is used to achieve
uniform velocity profile at
the inlet of the diffuser as
well as reduce the turbulence
level at its exit. Fig. 3(g)
Contd.

18
Multitube Inclined Manometer

In the present experimentation a multi tube


inclined manometer (kerosene (S.G.= .81))as
manometric fluid) as shown in the Fig.3(h)
has been used to measure the pressure at
different locations . Ten number of glass
tubes are fitted side by side over a wooden
board and one end of all the tubes were
connected to a common manifold. The other
ends were connected with flexible tubes, one
of those tubes was connected to the point
where the pressure was to be measured. The
inclined manometer was fixed at an angle of
35°.

Fig. 3(h)
Contd.

19
Three-hole pressure probe

• The three-hole pressure probe is a combination of three tubes connected side by


side. The measuring tip of it is either L-shaped or Cobra shape.
• For the present study L-shaped pressure probe was used and the geometry of the
probe is shown in the Fig3(i) & Fig.3(j).The two side tubes of the three hole
pressure probe are chamfered at approximately 45° and the centre tube is cut at 90°
with the flow direction and it is placed at opposite direction of flow.
• Three-hole pressure probes, using the principle of Prandtl's tube, can measure flow
direction in the yaw plane with respect to a reference plane and hence measurement
can be done in aligned with the direction of flow in yaw plane.

Fig.3(j)
Fig.3(i) 20
 
Traversing mechanism

• The traversing mechanism used with the probe,


requires an arrangement for rotating the probe about
its longitudinal axis in addition to its longitudinal
motion. Longitudinal motion is achieved by rotating a
lead screw by hand in the clockwise-direction fitted
with traversing mechanism.

• The movement of the probe in z-direction from the


bottom surface of the test diffuser was measured using
a scale having least count of 1mm. The scale was
fixed to a support parallel to the movement of the
probe in z-direction. Fig.3(k)

21
RESULTS AND DISCUSSIONS
• The flow development in the C-shaped diffuser was investigated through
the mean velocity distribution, total pressure distribution and static
pressure distribution.
• In the present study the total and static pressure is normalized with inlet
averaged dynamic pressure (0.5ρ Uavi2) and the velocity is normalized with
inlet average velocity(Uavi).
• The contours have been drawn using ‘SURFER’ graphics software
package. In the following sections the experimentally obtained results are
analyzed for the curved diffuser at different sections.
• Experiment for the curved C shape diffuser was carried out at inlet
averaged velocity of 40 m/s.

Contd.

22
Mean velocity contours
Fig4.1(a) referred to inlet
section at average inlet velocity
of 40 m/s, illustrates an uniform
distribution of flow through out
the entire cross sectional area
except at close neighbourhood of
top and bottom walls, this is due
to
 i)accumulation of low
momentum fluid on the walls.
 ii)corner effect
iii)the effect of downstream
curvature

Contd.
Fig 4.1(a) 23
Fig4.1(b) refer to section
A(11.250) at the inlet
velocity of 40 m/s, depicts
that flow with maximum
velocity is occupying major
portion of the cross section
adjoining to wall-1 while the
velocity on wall -2 is
relatively less.
This is due to the higher
pressure on the concave
wall-2 compared to convex
wall-1 as a result of
centrifugal force.

Contd.
Fig 4.1(b) 24
Fig 4.1(c) refers to section
B(33.750), the flow suffers
successive turning and
diffusion.
Low momentum fluid is seen
to accumulate on wall-1 due to
the generation of secondary
motion.
Migration of this low
momentum fluid on to wall-1
affects the uniformity of flow
along this wall. The core flow
with maximum velocity is
pushed from wall-1 towards
the mid vertical plane.
The flow developed at this
section is symmetric about the
mid horizontal plane.

Contd.
Fig 4.1(c)
25
Fig4.1(d) refers to section
C(56.250) at the inlet
velocity of 40 m/s, exhibits
fluid with further decreasing
momentum to accumulate
on the convex wall-1 and
disturb the uniformity of
flow over a significant area.
Core flow with maximum
velocity is seen to be shifted
further away from wall-
1,compared to that at the
previous section.

Fig 4.1(d) Contd.


26
Fig 4.1(e) refers to section
D(78.750),Low momentum
fluid is seen to accumulate
at an increasing rate on wall-
1 side as compared to that at
section-C.
The core flow with
maximum velocity is
marginally shifted towards
wall 2 as compared to that
of previous section.

Contd.
Fig 4.1(e)
27
At last stage the flow moves
through the straight outlet
duct.
Fig4.1(f) refers to outlet
section at the inlet velocity of
40 m/s, exhibit more area on
wall-1 side being occupied by
low momentum fluid.
The pattern of velocity
contours indicate a significant
improvement in quality of
flow compared to that of
previous section and justifies
the installation of straight
duct.

Fig 4.1(f)
28
Mean total pressure contours
Fig4.2(a) refers to inlet section
,total pressure distributions are
uniform over the entire cross
section, except over small areas
close to the top and bottom
walls, due to accumulation of
low momentum fluid over small
areas close to the top and bottom
walls.

Contd.
Fig 4.2(a) 29
The contours at section-
A, refer to fig4.2(b) depict
a decrease in total pressure
towards wall-2,this is an
indication of decelerated
flow along wall-2.

Contd.
Fig 4.2(b)
30
Fig4.2(c) shows, as the
flow moves downwards,
total pressure is seen to
decrease along wall-1 due to
accumulation of low
momentum fluid.
This establishes a
continuous decelerating
flow with accumulation of
low momentum fluid on
wall-1.

Fig 4.2(c) Contd.


31
Fig4.2(d)shows, total pressure
is seen to decrease further along
wall-1.
This establishes a continuous
decelerating flow with
accumulation of low momentum
fluid on wall-1.

Fig 4.2(d) Contd.


32
Fig4.2(e) shows, total pressure
is seen to decrease further along
wall-1.
This establishes a continuous
decelerating flow with
accumulation of low momentum
fluid on wall-1.

Contd.
Fig 4.2(e)
33
Fig4.2(f) shows, total
pressure is seen to decrease
more along wall-1.
This indicates a continuous
decelerating flow with
accumulation of low
momentum fluid on wall-1.

Fig 4.2(f)
34
Mean static pressure contours
Fig4.3(a) referred to inlet section
,exhibit uniform distribution of
static pressure over the entire
section, except over small areas
close to the top and bottom walls
due to accumulation of low
momentum fluid.

Contd.
Fig 4.3(a) 35
Fig4.3(b) referred to static
pressure distribution of Section-
A, Centrifugal force acts on the
fluid towards wall-2 due to
turning, As a result flow gains
higher velocity along WL-
1,compared to WL-2.
Compared to previous section
there is increase in static pressure
on the top and bottom walls

Fig 4.3(b) Contd.


36
Fig 4.3(c) shows that growth
in static pressure is seen faster
along WL-1 compared to WL-
2,Indicating a relatively stronger
retardation of flow on wall-1 and
helping low momentum fluid to
accumulate over there.

Fig 4.3(c) Contd.


37
As the flow moves
downstream, static pressure
increases more rapidly along
wall-1 compared to that
along wall-2.
High static pressure along
wall-1 indicates the shifting
of the core flow ,with
maximum velocity towards
wall-2.

Fig 4.3(d) Contd.


38
Fig4.3(e) shows that the static
pressure further increases along
wall- 1.indicating the shifting
of the core flow with maximum
velocity towards wall-2.

Fig 4.3(e) Contd.


39
Fig4.3(f) shows that
as the flow moves
downstream at the outlet
section the static
pressure increases more
rapidly along WL-1
compared to that of
WL-2.

Fig 4.3(f)
40
CONCLUSIONS

From the present investigation the following conclusions have been drawn:
1.The mean velocity distribution at the inlet of the C-shape diffuser is uniform and
symmetric about the mid-horizontal plane except at the vicinity of the top and bottom
walls. This behaviour may be attributed to the accumulation of low momentum fluid over
the walls.
2.In C-Shape diffuser, the core flow with maximum velocity in the downstream beyond
the angle of turn of approximately 22.5° is gradually pushed towards WL-2 by the
continuously accumulating low momentum fluid at WL-1,Indicating the transportation of
low momentum fluid by the generated secondary motion.
3.It is observed that the distribution of total pressure contours for all the measuring
sections has got a similarity with the corresponding mean velocity contours. As flow
moves downstream the total pressure is seen to decrease continuously along WL-1.
4.Due to diffusion, the static pressure on the four walls increases continuously along the
flow direction, with similar trend at top and bottom walls at corresponding angle of turn.

41
REFERENCES

1. Mizuno, A. and Ohashi, H. 1984 “A study of flow in a two-dimensional diffuser


with an oscillating plate”, Transactions of the Japan Society of Mechanical
Engineers, Series B, Vol. 50, No. 453, pp. 1223-1230 (in Japanese).
2. J. Eustice, Flow of Water in Curved Pipes, Proc. Royal Society London, 1910,
Series A84, pp. 107-118.
3. J. Eustice, Experiments of Streamline Motion in Curved Pipes, Proc. Royal Society
London, 1911, Series A85, pp.119-131.
4. R. W. Fox, and S. J. Kline, Flow Regimes in Curved Subsonic Diffusers, Trans. of
ASME, Journal of Basic Engineering, Sept. 1962, pp. 303-316.
5. M. B. Sajanikar, S Kar, and U. S. Powle, Experimental Investigation of
Incompressible Flow in Curved Diffusers, Proc. 11th National Conference of
FMFP, Hyderabad, India, 1982.

Contd.
42
6. Sumida, M. 2009 “Experimental study of pulsating turbulent flow through a
divergent tube”, in Matos, D. and Valerio, C. ed., Fluid Mechanics and Pipe
Flow, Chapter 11, Nova Science Pub.
7. King, C. V. and Smith, B. L. 2011 “Oscillating flow in a 2-D diffuser”,
Experiments in Fluids, Vol. 51, pp. 1577-1590.
8. Bounous, O., 2008 “Studies of the ERCOFTAC Conical Diffuser with Open
FOAM” Research Report 2008:05 (ISSN: 1652-8549). Department of Applied
Mechanics, Chalmers University of Technology, Sweden.

43
THANK YOU

44

You might also like