Paper Ed2 12

8/9/2019 Paper Ed2 12

1/15

8/9/2019 Paper Ed2 12

2/15

Journal of Babylon University/Pure and Applied Sciences/ No.(1)/ Vol.(1)! "#11

Fig (1) Designed and manufactured wind tunnel

YXY

8/9/2019 Paper Ed2 12

3/15

8/9/2019 Paper Ed2 12

4/15


The design conforms to the recommendations given in the references arlow et

al,)444 , oo& , )47*, 8ong + 1ar&inson,)449", given here in , on the state : of art to

find optimum construction ratio, wall profile , wire screen and honeycomb selection

for the settling chamber to have good flow qualities in the test section. The flow

qualities are initiated in the large settling chamber where very low velocities permit

use of multiple wire screens honeycomb without incurring much losses to ma&e flowuniform and reduced turbulence level. The settling chamber cross section is made to

contract considerably to reach test section at a short test possible distance.

Interpretation of results is sensitive to the test section airflow quality, (ousif et

al,)44), design the wind tunnel with crosssection 0.7; m" square and ).;m" long

and contraction ratio ustable test section could used either as the environment,

subsonic or climatic wind tunnel. 2iterature survey indicates acceptable value of mean

velocity variation of about ?0.*= to 0.;=, flow angularity of about ?0.;, turbulence

intensity of 0.;=and power factor for this type of tunnel of about *.; .

@$act solutions have not proved sufficient to predict adverse velocity gradient near

the wall trial and error methods and empirical relationships were deployed wherevernecessary. Also A'#(#)* software is used to chec& the distribution of velocity +

pressure.

esign, construction and testing of a subsonic wind tunnel that the wind tunnel

simulates a high speed stream of air to flow past a model of the ob>ect being

testedB5ig.3"C

ind tunnel is a chamber though which air is forced at controlled velocities in order

to study the effects of aerodynamic flow around airfoils or scale models.

The parts of wind tunnel areD

). A settling chamber to straighten the flow and control turbulence intensity.

*. A contraction !one to ta&es a large volume of low :velocity air and reduces it

to a small volume of high velocity air without creating turbulence.

3. A test section where the test article and sensors are installed.

8/9/2019 Paper Ed2 12

5/15

$eoretical Analysis of %ind $unnel

&ontraction DesignA$isymmetric steady ideal irrotational flow in cylindrical coordinate system is

considered sufficient for the appro$imation the design Aldruby ,)477".

ontinuity equationD

Irrotationality conditionD

ombining it reduces to sto&e eltrame equationD

ith particular solution of the form

And general solution of the fromD

En substituting general solution equation3" in continuity equation )"

The velocity components becomes (ousif et al ,)44)" D

Indication at $ a$is, r F0,urF0F*f$",vrF0F0 GGGGGGGGGGGG..G.;"

As $H, the boundary condition at the end of contraction should be D

-H*f$"

5or all r GGGGGGGGGGGGGGGG.G.. 6"

/H0

This is possible when f$" H0 as $ HJ suggesting formation of f$" to satisfy the

following two conditions hyperbolic in nature.

hange in a$ial velocity pattern should be in continuity from inlet to outlet ofcontraction.

Airflow at the inlet and outlet of contraction to be uniform therefore, the simplest

form of f$" which satisfies the .. isD

here & ) is constant representing controlling parameter for the accuracy of

contraction, the two constants A and ) can be determine form the ..

Kowever equation 9" has unsatisfaction wall velocity gradients resulting in delayed

wall infle$ion for contraction.

ohen suggested modification as followers (ousif et al, )44)"D

YX^

8/9/2019 Paper Ed2 12

6/15


here constants & ), A and ) are as in eq. 9" and 8 * is a constant of small value

while * can be determine from the ..at $F0.

#ubstituting @q.4" in @q.3", the stream function becomesD

efining the profile of wall contraction in terms of only two variable $ and r.

A, ) and * can be obtained by specifying velocities at the inlet and outlet at $F0. & ) and & * are the geometric aspect ratios of contraction affecting slenderer and the

position of point of contraction.

$est SectionA closed rectangular test section with filleted corners considerably less power than the

open type recommended length of test section should be less than twice the ma$imum

width (ousif et al ,)44) and #hapiro ,)4;3" gives the drop in static pressure asD

here &Fconstant D 0.0)6 to 0.0

8/9/2019 Paper Ed2 12

7/15

This givers the requires divergence and the test section outlet width.

DiffuserAssuming constant velocity distribution, gives diffuser efficiency asD

here, p) and p* are static pressures at the cross sections at A)diffuser inlet" and

A* diffuser outlet", -) is the mean velocity at A). The low efficiency arises out of

large loss moreover of energy during pressure recovery from flow &inetic energy.

Adverse pressure gradient associated with velocity reduction favoures undesirable

flow separation from the walls. 1arameters efficiency diffuser performance are rate of

e$pansion cross section shape , turbulence level .2. effect and effect of wall shape,

%ecommended Aldruby, )477" values of diffuser divergence is ;N to )0N beyond

which efficiency drops rapid as it depends on the turbulence level and empirical

relationship derived from e$perimental results.

The ratio *O.2 thic&nessPma$. diffuser width" Q0.00; give 40= and deterioratesrapidly as the ratio e$ceeds 0.00;" .2arger diffuser angle can be reduced by .2.

control such as use of deflector vanes and Por introduce spiral flow to energies by

mi$ing high energy core flow with .2. growth improving up to 70= for larger

diffuser angle up to ;3N.

Fan

5an differs from airplane propellers by usage of multi blades to avoid pulsation and

usage of thin blades with pre rotating vanesPaft straightened vans to avoid

rotationality. In open circuit wind tunnel aft fans have little effect on test section

airflow quality and therefore vans may not be required. The parameters affecting fan

efficiency is tip clearance, component of radial flow, blade liftPdrag ratio, number of blade, blades %e. 'o. www.ARA.org ", blade angle and tip compressibility effect

Flow 'uantities

hile uniform and steady airflow is desirable at the test section, in practice

associates with low frequency fluctuations and high frequency turbulence. The state

of art of design to reduce the turbulence level. %eduction of turbulence is achieved

through identifying the sources improve construction the usage of damping screen and

honey comb in the setting chamber considerably improves flow qualities, if the

turbulence intensity is above 0.9=". #creens reduce turbulence level but it has little

effect on removing swirl and lateral mean velocity variation. Koneycomb with cell

length S7O cell diameter, proved to be suitable Abbaspour and #hv>ee,*004" smaller

si!e honey comb diameter could help abatement of turbulence and in some cause may

dispense with the use of screen. #quare and he$agons shaped honey comb give better

result than circular section.

*erimental *art+anufacturing and &alibration te Subsonic %ind tunnel

The specification of test sectionD

ross section F 30cmO30cm

2ength F 60 cm

/elocity≈ 3; mPsec

YX_

http://www.amca.org/http://www.amca.org/

8/9/2019 Paper Ed2 12

8/15


Fig!(,) te used Fan

Area ratio-.

Fan-The used fan is shown in 5ig

8/9/2019 Paper Ed2 12

9/15

$est Section DesignTest suction walls must have minimal .2." growth to avoid static pressure drop on

account of energy losses i.e. avoid a hori!ontal buoyancy,phenomena.

In this study the wor&ing section is construction of clear 1erspe$ with crosssection of

300mmO300mm" and length of 600 mm. 5ig 7" represent schematic + 5ig.9"represented photo of the test section.

Fig() *oto of test section wit s*ecimen Fig () scematic diagram of test section

Diffuser DesignThe iffuser slope angle is )93 degree" to avoid separation of shown in 5ig.4 a,b"

Fig!(3!a) Diffuser Design Fig!(3!b) Diffuser *oto

$e &onstructionThe bell mouth settling chamber, test section, and diffuser were separately constructed

as an unit >oint together by bolt and isolated by cascade which can decreased the

vibration between the element of the device at less the device is hung on a load table

and isolated by cascade too, the part of the fan is support in the diffuser and isolated

from it. 5ig)0" shows the >oining of the parts of wind tunnel .

Fig! (14) 5oining te wind tunnel

YXa

8/9/2019 Paper Ed2 12

10/15


The fan caused vibration which is sha&e the device tightly . uring final assembly, care was

ta&en to align center lines of each section to be in one straight hori!ontal line. The >oints were

sealed to give smoothness of flow and oil was coated to reduce .2.

The final step is to commission the new tunnel into operation and to measure flow quantities .

+odel of $urbine 6lade

The models mounting device is built and support in the test section by /" shapeshown in 5igs. ))"+)*".

Fig(11) $e mounted of te s*ecimen

Fig (12) Scematic diagram of blade turbine

Final AssemblyThe wind tunnel is assembly from the mentioned parts, and the final apparatus is

shown in 5igs)3"+)

8/9/2019 Paper Ed2 12

11/15

Fig (1") Scematic of wind tunnel sown te *itot tube *osition

Fig (1,)$e final assembly of wind tunnel

7umerical SimulationA'#(#)* is a finite element analysis A'#(#)* help, *004" , have the capability to

analy!e a wide range of different problems. There are two ways to use the A'#(#)*

interactively through the graphical userD

). interface -I".

*. used batch file and A'#(# commands.

The main complete flowchart and A12 Ansys 1arametric esign 2anguage" for the

numerical simulation 'a&asone ,*006" is shown in 5ig.);".

YYX

Pprep9

@t,),fluid)

8/9/2019 Paper Ed2 12

12/15


Fig! (1/)- Flow cart and A#D8 for numerical Simulation

The used element in this analysis is 52-I)

8/9/2019 Paper Ed2 12

13/15

flow show good agreement with the A'#(# results of blade geometry. 5ig.)4"

#hown the shadow capture of the turbine blade at Rach no. F 0.097.

Fig!(1) Static *ressure distribution on te u**er surface

Fig! (13) Sadow ca*ture of turbine blade at +ac no!4!4

5igs.*0" and *)" are shown the velocity distribution and velocity vector, and 5ig.

**" shows the pressure distribution and 5ig.*3" shows the stream line in thedesigns wind tunnel with turbine blade by A'#(#)* software.

YY)

8/9/2019 Paper Ed2 12

14/15


Fig!(24) 9elocity distribution

Fig!(21) 9elocity 9ector

Fig!(22) #ressure distribution

Fig!(2") Stream line distribution

&onclusions

)In our research , the wind tunnel is manufactured and calibrated for subsonic flow .* ood agreement is observed for testing turbine blade specimen with comple$

geometry.

3 The present e$perimental results show good agreement with A'#(#)* results .

8/9/2019 Paper Ed2 12

15/15

Aldruby R. I. , )477 Y1rinciple of a!dynamicY, *nd edition.

A'#(#)* online help, A'#(# Ranuals, *004.

Asher K. #hapiro /ol. )+ /ol. *, )4;3Y The ynamics and Thermodynamics of

ompressible 5luid 5lowY, %eference Kand boo&".

arlow X. ., %ea .K and 1ope A. )444. Y2ow: #peed ind Tunnel TestingZ , Xohn

iley Inc.ien&iewic! . )446 ['ew Tools in ind @ngineeringY X. ind. @ng. Aerod., 6;3

)", *94300.

oo& .'.X,)47* [#imulation Technique for #hort Test : #ection ind TunnelM

%oughness , arrier , and Ri$ing evice RethodY , ambridge -niversity

1ress,)*6)36.

8ong 2. + 1ar&inson ./. , )449. [A 3 Tolerant ind Tunnel for eneral ind

@ngineering TestY , 1roceeding of the 3rd International olloquium on luff ody

Aerodynamics and Applications , X. wind @ng. Ind. Aerod. , 649),49;47;.

'a&asone (. and (oshimoto #. and #tolars&i T.A. , *006 Y@ngineering Analysis with

A'#(# #oftwareY, )st published.

(ousif A. K. ,Tlib. #. Xawad + X. A. 8adihm , )44)Yesign #tudy of a 2ow #peedind TunnelY , Xournal of Rilitary @ng. ollage, 'o.6, p.)7

Paper Ed2 12

Documents

Transcript of Paper Ed2 12