-
ria
hi
Volute casing
conuded and, raditirs s
2011 Elsevier Ltd. All rights reserved.
rbomae is reqw, forolute cly use
to restrict the ow against the given head without incurring
signif-icant losses.
The total ow of the impeller discharge passes through the
vo-lute throat area and only a part of the impeller capacity
passesthrough any other section. There were ve geometrical
parametersin volute design. The most important parameter was the
volutethroat area [6]. If the volute throat area is too small in
comparison
The design objective is to achieve a uniform ow at volute
inlet.This is usually attained at the design ow rate only; at off
designconditions the volute is either too small or too large and a
pressuredistortion develops circumferentially around the volute
passage.The pressure distortions reduce the stage performance and
havea direct impact on diffuser and impeller ow stability. The
stableoperating range of the compressor is also decreased [1]. The
pres-sure distribution in any cross section is mostly determined by
thevertical nature of the ow. The strong vortex ow induces
acentrifugal effect, which results in an increasing pressure
gradienttowards radially outward direction [3].
Corresponding and presenting author. Tel.: +91 261 2201994.
Applied Energy 90 (2012) 335343
Contents lists availab
lseE-mail address: [email protected] (B.D. Baloni).the
centrifugal blower. The blower of this type is used to
provideforced draft to boiler and general industry due to their
simplestructure, small size and low production cost. In this type
ofblower, an impeller forces the gas from the impeller and leads
itaway at a high pressure. Before the gas leaves the stator i.e.
volute,its kinetic energy is reduced and partially converted into
pressureby diffusing action. The stator takes no part in the head
generationhence consequently all theoretical consideration in
casing designdeals with losses. But the casing plays an important
part in locatingbest efciency point by virtue of its hydraulic
losses and its ability
volute having an equal width [15,17]. To avoid shock and
separa-tion loss at the volute tongue, the volute angle av is made
to corre-spond to the direction of the absolute velocity vector at
theimpeller discharge ow angle. At higher specic speeds, besidesthe
volute angle, the length and shape of the tongue become
moreimportant [4,15,16] Removal of a portion of the tongue reduces
theefciency from 85% to 81% while restoring the tongue to the
origi-nal shape the efciency can be brought back to 85% [15].
Depend-ing upon casing, the tongue may be sharp or blunt but it
must bealong the streamline [16].Backward curved blade
impellerThree-dimensional probeFull throttle
1. Introduction
To improve the design of any Tustanding of the ow of such
machinwith an experimental study of the ovolute of a centrifugal
blower. The vthan the impeller exit width is wide0306-2619/$ - see
front matter 2011 Elsevier Ltd.
Adoi:10.1016/j.apenergy.2011.02.016chine, a better under-uired.
This paper dealsthe different designs ofasing with width largerd
without a diffuser in
with the optimum value, the peak efciency will decrease
slightlyand move towards a lower capacity [15]. If volute area is
too large,the peak efciency may increase but will move towards a
highercapacity, efciency at partial capacity will be lower in this
case[15]. The volute exit to inlet area ratio has a signicant
effect onthe diffusion and losses into the volute ow processes [2].
A volutewith width exceeding the impeller width is more efcient
than aKeywords:Turbomachine
the experimental data, analyzed results were presented for the
pressure recovery coefcient and losscoefcient.Pressure recovery and
loss coefcient vacentrifugal blower volute designs
Beena D. Baloni a,, S.A. Channiwala a, V.K. MayavansaDepartment
of Mechanical Engineering, SVNIT Surat, IndiabDepartment of
Mechanical Engineering, MSU -Vadodara, India
a r t i c l e i n f o
Article history:Received 8 July 2010Received in revised form 28
January 2011Accepted 12 February 2011Available online 5 March
2011
a b s t r a c t
An experimental study waswith backward blade shrostant angular
momentumstudied at various angularwith full throttle open conwhich
gave ow paramete
Applied
journal homepage: www.ell rights reserved.tions in the two
different
b
ducted using two different designs of volute casing of a
centrifugal blowerimpeller. The volute casing designs were based on
the principle of con-constant mean velocity. For both design of
volute the ow elds were
dial and axial locations in the volutes. The experiments were
conductedon at inlet. Analysis was done with the help of
three-dimensional probe,uch as stagnation pressure, static pressure
and ow directions. Based on
le at ScienceDirect
Energy
vier .com/locate /apenergy
-
The outline of this paper is as follows. First, the
experimentalset-up and instrumentation is described, followed by
details oftwo different designs of volute casing. Finally results
of both thecasing, based on analysis of experimental data at
various angular,radial and axial locations is discussed for the
pressure recoverycoefcient and loss coefcient. Also the ow
condition aroundthe impeller is analyzed at immediate exit.
2. Experimental set-up and instrumentation
Fig. 1 shows existing blower experimental set-up. The
singlestage blower has backward straight blade type impeller
with
Nomenclature
Cp pressure recovery coefcientP local static pressure, mm of
waterPave average static pressure within the Volute, mm of waterPo
local stagnation pressure, mm of waterPoav average stagnation
pressure within the volute, mm of
waterr radius, mm
336 B.D. Baloni et al. / Applied Eneshroud, having decreasing
width from inlet to outlet. The detailsof impeller data are
provided in Table 1. To rotate the impeller,the motor of 5.5 KW
(7.5 HP) at speed of 2900 r.p.m. was used.
The volute casing, stator of blower was design based on two
dif-ferent approaches namely constant angular momentum andconstant
mean velocity. To visualize the probe in the casing, itwas made of
acrylic 8 mm thick sheet and FRP on the side of casing.To analyze
the pressure, tapping were made at azimuth angle from0 to 360 at an
interval of 30 as shown in Fig. 2.
Five-hole probe was used for three-dimensional ow survey,which
was calibrated in a suction subsonic wind tunnel havingworking
section of 50 cm 60 cm. It was calibrated for zero set-ting,
dynamic pressure and pitch angle. Probe tubes were con-nected to
the multi tube inclined manometer having water asFig. 1.
Experimental set-up of Blower unit.
Table 1Data specication of impeller.
Inlet Outlet
Diameter (cm) 30 42.5 Number of Blade 12Blade width (cm) 17.5 13
Blade thickness (cm) 0.5Blade angle, b (deg.) 20 483.1. Constant
angular momentum design
In this design concept the angular momentum of the ow
isconsidered constant through out the volute casing. It is also
knownas Free vortex design. The ow for a free vortex design
followsthe equation,
r Vu constant 1In this design, if the ow through volute is
incompressible, and
volute has constant width then the direction of streamline
remainsconstant [17].uid for measurement of pressure. Flow eld was
measured at fourangular positions at 180, 240, 300 and 360. At each
angularlocation ve radial positions were selected for the purpose
of owsurvey along each axial distance. A traversing mechanism was
de-signed to provide movement to the probe in all
three-directionsalong with simultaneous yawing in ow passage as
shown inFig. 3. Probe traversing device controlled the position of
the proberadially and circumferentially with the help of vertical
lead screwwith machining nut and a guided rod assembly. The axial
andyaw movement of the probe is controlled by probe holder withthe
help of threaded rod with lock nut and disk as shown in Fig. 4.
3. Design aspects of volute casing
Many researches had gone into more and more systematic de-sign
of centrifugal blower. Different authors have suggested differ-ent
procedure although each has a slightly different method
ofcalculation. The volute design process is often subject to many
con-straints such as maximum casing size, location of piping to
whichthe volute connect, avoiding necessary auxiliary equipment
etc.,often limit the designers option [7]. Usually design methods
forvolutes are based either on the assumption of constant
angularmomentum [12] or on the assumption of constant mean
velocity[14]. Osborne [11], Church [8], Aungier [7], and Yahya [17]
amongothers have incorporated the detail design of
one-dimensionalvolute.
Rc3 velocity distribution factorVm mean velocity of ow within
Volute, m/sV 02 actual absolute velocity at Impeller exit, degreeVu
tangential velocity of ow within volute, m/sW load coefcientav
volute angle, degree
rgy 90 (2012) 3353433.2. Constant mean velocity design
The volute is designed on the basis of constant mean
velocitythrough out the volute. Constant mean velocity Vm can be
obtainedby multiplication of volute velocity distribution factor
(Rc3) withactual absolute velocity at impeller exit (V 02).
Rc3 V 02 Vm constant 2To obtain high efciency, it is necessary
to maintain constant
velocity of the uid in the volute passage at design point [17].
Thiswould give uniform static pressure distribution around
theimpeller.
For both the design concepts, main parameters of volute
casingare shown in Table 2.
-
Motor
Fig. 2. Line diagram o
Fig. 3. Probe traversing mechanism.
Fig. 4. Probe holder.
B.D. Baloni et al. / Applied EneAnalysis of ow at different
angular, radial and axial positionsin both volute casing is
revealed that maximum 15% variations oc-cur with the average value
of r Vu in constant angular momen-tum concept whereas in constant
mean velocity concept,
f existing set-up.
Table 2Main parameters of volute casings.
Parameters Constant angularmomentum
Constant meanvelocity
Volute base circle diameter (cm) 46.75 46.75Volute exit diameter
(cm) 100.57 125.26Volute width (cm) 26 26Throat length (cm) 26.91
39.26Volute tongue angle (deg) 15 21.84Tongue radius (cm) 24.35
25.51
rgy 90 (2012) 335343 337analyzed value of mean velocity was
found 26.789 m/s instead ofdesign mean velocity 26.7387 m/s.
4. Experimental results and discussion
The ow parameters analysis was carried out after measure-ment of
stagnation pressure, static pressure, Flow velocity and
owdirections at 180, 240, 300 and 360 through out the radial
dis-tance at various axial positions. Also from suction to exit
i.e. A 7(Fig. 2) for the holes near impeller. For presentation of
the owconditions, the angular locations at 180, 240, 300 and
360weredesignated as Sections 1, 3, 5 and 7 respectively. The
radial loca-tions were non-dimensionalised by the radial width at a
given sec-tion. They were expressed in terms of percentages i.e.
from 0 to100. The ow was surveyed at ve radial positions at each
section.The axial distance were non-dimensionalised by the
impellerwidth at exit and expressed in terms of percentages, 0%
being onhub side, 100% being on tip (shroud) side, 23% to 0%
representsoverhung portion on hub side and 100188% represents
overhungportion on shroud side. Thus the ow elds were presented
interms of sections, non-dimensional radial and non-dimensional
ax-ial distances.
The average values of stagnation pressure, static pressure
andvelocity were determined based on area average at different
angu-lar positions for the holes near the impeller as shown in Fig.
57.
The plot of average stagnation pressure reveals that the
owdelivered by the impeller is of higher stagnation pressure on
exitside compared to near tongue. The variation is almost nil
towardsexit of the volute casing (210360). But two designs produce
owwith different stagnation pressure. Volute designed with
constantangular momentum principle delivered ow with higher
stagna-tion pressure which leads to higher head.
-
120
140
160
180
200
0 30 60 90 120 150 180 210 240 270 300 330 360
Angular position (Degree)
Stag
natio
n Pr
essu
re(m
m o
f wat
er)
Volute based on "constant angular momentum" Volute Based on
"Constant Velocity Method"
Fig. 5. Variations of average stagnation pressure of ow at the
exit of impeller at various angular positions.
20
24
28
32
36
40
0 30 60 90 120 150 180 210 240 270 300 330 360 390
Angular position (degree)
Velo
city
(m/s
)
Volute basde on "Constant Angular momentum" methodVolute based
on "Constant mean velocity" method
Fig. 6. Variations of average velocity of ow at the exit of
impeller at various angular positions.
Average static pressure of flow at exit of impeller at various
angular positions.
40
60
80
100
120
140
160
0 30 60 90 120 150 180 210 240 270 300 330 360 390
angular positions (degree)
stat
ic p
ress
ure
(mm
of w
ater
)
Volute based on "Constant angular momentum" methodVolute based
on "Constant mean velocity" method
Fig. 7. Variations of average static pressure of ow, at the exit
of impeller at various angular positions.
338 B.D. Baloni et al. / Applied Energy 90 (2012) 335343
-
recat s
EneVariations in PressureDistance
-0.2
0
0.2
0.4
0.6
0.8
Rec
over
y co
effic
ient
(Cp)
B.D. Baloni et al. / AppliedThe ow conditions delivered by
impeller with both designsof volute were also compared by plotting
average values ofvelocity and static pressure (Figs. 6 and 7). With
both the de-signs, ow delivered by impeller near tongue region
(30150)is to be of higher velocity and lower static pressure. The
velocitydecreases and static pressure increases as ow moves
fromtongue to exit section of the volute (towards 360), this
justiesthe theory of one-dimensional isentropic ow with variable
area.Like stagnation pressure, the values of average static
pressureand velocity were lower with volute casing designed
withconstant mean velocity principle. Thus the analysis shows
thatdesign of volute inuences the ow conditions delivered by
theimpeller.
-0.8
-0.6
-0.4
% Radi
Pres
sure
Variations of Pressure Distance a
-0.8
-0.6
-0.4
-0.2
0
0.2
0.4
0.6
0.8
% Radial
Pres
sure
reco
very
coe
ffici
ent
(Cp)
Variation of Pressure recDistance at end
-0.8
-0.6
-0.4
-0.2
0
0.2
0.4
0.6
0.8
% Radia
Pres
sure
Rec
over
y C
oeffi
cien
t(C
p)
Section 1 Sectio
Fig. 8. Pressure recovery coefcient variations near Impeller
widthovery coefficient along Radial tarting of Impeller
rgy 90 (2012) 335343 339The rise in pressure on account of
conversion of kinetic energyinto enthalpy is expressed in the forms
of pressure recovery coef-cient (Cp) dened as,
Cp P Pave=Poav Pave 3While loss is expressed as loss coefcient
(W) and dened as,
W Poav Po=Poav Pave 4where (Poav Pave) is equal to the dynamic
head and also expressedas qV2. Since the values of stagnation
pressure and static pressurevary along axial & angular
positions of volutes, their average valueswere obtained by area
average technique at various angularpositions. These average values
at angular positions were used to
al Distance
recovery coefficient along Radial t mid of the Impeller
Distance
overy coefficient along Radial of the Impeller Width
l Distance
n 3 Section 5 Section 7
of volute designed on constant angular momentum concept.
-
recoting
EneVariation of Pressure at star
0.00
0.20
0.40
0.60
0.80
very
coe
ffici
ent
Cp)
340 B.D. Baloni et al. / Applieddetermine average value for the
whole impeller. Po and P were thelocal values at a particular
location of volute.
The values of pressure recovery coefcient were
calculatedthroughout the volute casings. Figs. 8 and 9 represent
the varia-tions in pressure recovery coefcient at starting, middle
and endof Impeller width for different angular positions. It is
observed thatfor both the design concept the pressure recovery
coefcient islowest at Section 1 and highest at Section 7. This
shows that aswe move from suction to exit the value of local static
pressure in-creases due to conversion of kinetic energy into
pressure energy.Which indicates that according to [9,10,13,14]
casings serves its
-0.80
-0.60
-0.40
-0.20
% Rad
Pres
sure
reco (
Variations in Pressure recodistance at mid o
-0.80
-0.60
-0.40
-0.20
0.00
0.20
0.40
0.60
0.80
% Radi
Pres
sure
reco
very
coe
ffici
ent
(Cp)
Variations in Pressure recodiatance at en
-0.80
-0.60
-0.40
-0.20
0.00
0.20
0.40
0.60
0.80
% Radia
Pres
sure
reco
very
coe
ffici
ent
(Cp)
Section 1 Sectio
Fig. 9. Pressure recovery coefcient variations near Impeller
wivery along radial distance of Impeller
rgy 90 (2012) 335343function well. Also as we move radially
outward the Cp value in-creases at all angular positions except at
Sections 3 and 5 in caseof Constant angular momentum design
concept. At Sections 1and 3 value of Cp comes negative at hole near
impeller, whichshows that the ow is delivered by the impeller at
lower staticpressure than the average static pressure of the
volute. It also indi-cates a pressure drop from impeller outlet to
the volute.
The pressure recovery coefcient variations were also pre-sented
for both design concepts in the form of contours made byMatlab
software in Figs. 10 and 11. As we move from suction toexit of the
volute i.e. from Sections 1 to 7, regions covered by high
ial Distance
very coefficient along radial f the impelller width
al Distance
very coefficient along radial d of the impeller
l Distance
n 3 Section 5 Section 7
dth of volute designed on constant mean velocity concept.
-
EneB.D. Baloni et al. / Appliedvalue of Cp, increases at
radially outward position from about 10%to 60% of radial distance
in case of the constant angular momen-tum design concept. And
following the same about 3080% of ra-dial distance in case of the
constant mean velocity designconcept. These results indicate that
as we move from suction to
Fig. 10. Pressure recovery coefcient variations in volute
designed on constantangular momentum concept.rgy 90 (2012) 335343
341exit of volute the conversion of kinetic energy into pressure
energytakes place. In case of volute designed by constant
angularmomentum, it is observed that near impeller region the value
of
Fig. 11. Pressure recovery coefcient variations in volute
designed on constantmean velocity concept.
-
Ene342 B.D. Baloni et al. / AppliedCp came negative. For another
designed volute these regions arelimited up to near hub portion of
the impeller region. As we movefrom suction to exit these regions
of negative Cp value decreases.Compared to other, high pressure
recovery regions are more incase of constant mean velocity designed
volute. The gradientsare also less in case of constant mean
velocity designed volute
Fig. 12. Loss coefcient variations in volute designed on
constant angularmomentum concept.rgy 90 (2012) 335343which shows
better uniformity of the ow compared to constantangular momentum
designed volute [17].
The variations of Loss coefcient for both the designed
volutecasings were represented in the form of contours made by
Matlabsoftware in Figs. 12 and 13. It is clear from the gures that
com-pared to other regions of volute casings the losses are more
near
Fig. 13. Loss coefcient variations in volute designed on
constant mean velocityconcept.
-
the tip region of the impeller. While near the hub region of
theimpeller and in Section 1 at overhung portion of casing wall
sidethe values of loss coefcient came negative because, the loss
coef-cient compares the local stagnation pressure loss to average
inletstagnation pressure. The local pressure depends upon the
stagna-tion pressure on the same streamline at the impeller exit
[3].Due to the circumferential pressure distortion at the volute
off-design operation, stagnation pressure delivered by impeller
washigher than the average stagnation pressure. As we move from
suc-tion to exit of the volute casings, the higher value of colour
bar isdecreased from 1 to 0.3, which indicates value of loss
coefcientdecreased towards the exit of volute. In volute designed
based onconstant mean velocity, the gradients were less compared
toother design concept. It indicates that ow was more uniform
inthis volute casing then the constant angular momentum de-signed
volute [17]. The results justify the known fact that volutegeometry
affects the ow delivered by the impeller [5].
5. Conclusions
An experimental analysis has been undertaken to elucidate
theeffect of volute casing designs on the ow from the impeller of
acentrifugal blower. The pressure recovery coefcient as well as
losscoefcient variations in the casing were investigated. The
follow-
Acknowledgement
The experimental work presented in this paper has been
carriedout at Faculty of Technology & Engineering, The Maharaja
SayajiraoUniversity. All support from faculties and the staff
members isgratefully acknowledged.
References
[1] Ayder E, Van den Braembussche R, Brasz JJ. Experimental and
theoreticalanalysis of the ow in a centrifugal compressor volute.
ASME J Turbomach1993;115:5829.
[2] Hassan AS. Inuence of the volute design parameter on the
performance of acentrifugal compressor of an aircraft turbocharger.
In: Proc IMechE, vol. 221;2007. p. 695704.
[3] Hagelstein D et al. Experimental and numerical investigation
of the ow in acentrifugal compressor volute. Trans ASME
2000;122:2231.
[4] Ji Chunjun, Wang Yajun, Yao Lixin. Numerical analysis and
optimization of thevolute in a centrifugal compressor. ICPE-2007,
October 2327, 2007,Hangzhou, China; 2007.
[5] Rutschi K. The effect of the guide apparatus on the output
and efciency ofcentrifugal pumps. Schwiez Bauz 1961;79:23340.
[6] Worster RC. The ow in volutes and its effect on centrifugal
pumpperformance. Porc Inst Mech Eng 1963;177:84375.
[7] Aungier RH. Centrifugal compressors: a strategy for
aerodynamic design andanalysis. New York: ASME Press; 2000.
[8] Church AN. Centrifugal pumps and blowers. Joftr.witey and
Sons, HemispherePublishing Corporation; 1989.
[9] Eck B. Design and operation of centrifugal, axial-ow and
cross-owfans. Oxford: Pergamon Press; 1975.
[10] Karassik, Roy Carter. Centrifugal pumps. McGraw Hill Inc.;
1960.
B.D. Baloni et al. / Applied Energy 90 (2012) 335343 343ing
conclusions deduced from the experimental results are:
(1) The ow within the volute casing based on constant
meanvelocity design concept gives better ow conditions thanthat
based on the constant angular momentum. The gradi-ent of both the
ow parameters were less in case of constantmean velocity design,
suggests more ow uniformity com-pared to constant angular momentum
design concept.
(2) Variation in the pressure recovery is larger up to 50%
ofradial distance from the impeller towards radially
outwarddirection.
(3) Value of loss coefcient was decreased as ow move fromsuction
to exit of volute.[11] Osborne WC. Fans. Pergomon Press; 1961.[12]
Peiderer C. Centrifugal pumps for liquids and gases. Berlin,
Germany: Springer; 1948.[13] Shepherd DG. Principle of
turbomachinery. New York: Macmillan Publishing
Co., Inc.; 1956.[14] Stepanoff AJ. Centrifugal & axial ow
pumps: theory, design and
application. New York: Wiley; 1948.[15] Stepanoff AJ. Turbo
blowers: theory, design and application of centrifugal and
axial ow compressors and fans. New York: John Wiley & Sons,
Inc.; 1955.[16] Turton RK. Principles of turbomachinery. New York,
London: E. & F. N. Spon
Ltd.; 1984.[17] Yahya SM. Turbine compressors and fans. New
Delhi: Tata Mcgraw Hill
Publishing Company Limited; 2002.
Pressure recovery and loss coefficient variations in the two
different centrifugal blower volute designsIntroductionExperimental
set-up and instrumentationDesign aspects of volute casingConstant
angular momentum designConstant mean velocity design
Experimental results and
discussionConclusionsAcknowledgementReferences