Designing and Construction of Low Speed Wind Tunnel WiWu To Investigate The Robustness of Small Model Aircrafts and Launcher Controllers

(1)

MASTER'S THESIS

Designing and Construction of Low Speed Wind Tunnel WiWu To Investigate The Robustness of Small Model Aircrafts and

Launcher Controllers

Vikas Dalal

Master of Science (120 credits) Space Engineering - Space Master

Luleå University of Technology

Department of Computer Science, Electrical and Space Engineering

(2)

Julius Maximilian University of Würzburg Faculty of Computer Science VIII

Institute of Aerospace Information Technology Luleå University of Technology

Department of Space Science Space Campus Kiruna

Designing and Construction of Low Speed Wind Tunnel WiWu

To investigate the robustness of small model aircrafts and launcher controllers

Master Thesis in the subject

Space Science and Technology

presented by

Vikas Dalal

(3)

Julius Maximilian University of Würzburg Faculty of Computer Science VIII

Institute of Aerospace Information Technology

Luleå University of Technology Department of Space Science

Space Campus Kiruna

Designing and Construction of Low Speed Wind Tunnel WiWu

To investigate the robustness of small model aircrafts and launcher controllers

Master Thesis in the subject

Space Science and Technology

presented by

Vikas Dalal

Completed at University of Würzburg Faculty of Computer Science VIII Institute of Aerospace Information Technology

Supervisors:

Prof. Dr.-Ing. Sergio Montenegro Prof. Dr. Lars-Göran Westerberg

M.Sc. Eng. Atheel Redah

(4)

(5)

Declaration

I hereby declare that this submission is my own work and that, to the best of my knowledge and belief, it contains no material previously published or written by another person nor material which to a substantial extent has been accepted for the award of any other degree or diploma of the university or other institute of higher learning, except where due acknowledgment has been made in the text.

Wurzburg, Vikas Dalal

(6)

Acknowledgments

I especially would like to thank Prof. Dr. Ing. Sergio Montenegro for entrusting me with this project, as it did not only allow me to improve my skills in the eld of Aerospace Manufacturing and Designing, but it was also a great opportunity and chance to gain practical experience as a part of my studies at the University of Würzburg. It has been a great pleasure and honour. Additionally, I would like to give my thanks to his wife Magda and their sons Felix, Nils and Markus, who supported the project in the phase of construction.

For my supervisor Atheel Redah's critical input and encouragement throughout the design and construction process of the wind tunnel WiWU, I feel very grateful.

Furthermore, I am pleased to have Prof. Dr. Lars-Göran Westerberg as my supervisor from the Lulea University of Technology.

I extend my deepest gratitude to my competent and driven team of fellow students making this project possible - namely Richard Pradeep, Ravi Agarwal, Dinesh Kumar Babu, Hamza Hassan and Salman Khan, who were willing to encounter plenty of chal- lenges with me from the very beginning until the end of the project. Moreover, I would like to thank Muhammad Faisal for his wonderful guidance.

Finally, I feel deeply indebted towards my friend Martina for her invaluable support during this project.

(7)

List of Figures

1.1 Benjamin Robins Whirling Arm[20] . . . 4

1.2 Bicycle used by Wright brothers in 1901, photo by NASA(http://www.grc.nasa.gov/WWW/k- 12/airplane/wrights/biketest.html) . . . 5

1.3 Wind Tunnel in 1901, photo by NASA(http://wright.nasa.gov/airplane/test1901.html) 6 2.1 Boundary Layer representation with laminar ow and turbulant ow. . 9

3.1 Open loop Suckdown Wind Tunnel . . . 13

3.2 Closed Loop Wind Tunnnel with closed test section[51] . . . 14

3.3 Basic 3D design of WiWU . . . 15

3.4 Fan (EL 710 EC 01 | 119359), purchased from Ruck Ventilation GmbH (Boxberg, Germany) . . . 17

3.5 Electronics design to operate the fans. . . 17

3.6 3D design of WiWU Test Section . . . 19

3.7 3D design of the WiWU Diuser . . . 21

3.8 3D view of Contraction section . . . 23

4.1 Arrangements of two Fans in Diuser's end (Front view) . . . 25

4.2 WiWU controller board . . . 26

4.3 Wooden Model of Diuser and Test Section . . . 27

4.4 Attachment of two polycarbonate plates . . . 27

4.5 Finished Diuser section outdoor . . . 28

4.6 Completed test section with Anemometer . . . 29

4.7 Compleated Test section of WiWU with Airplane Model . . . 30

4.8 Basic frame of Contraction cone . . . 31

4.9 Completed Contraction cone of WiWU . . . 32

4.10 WiWU with air feedback channel . . . 34

(10)

5.1 Open Loop test condition . . . 36

5.2 With Air feedback channel test condition . . . 36

5.3 Turbulent ow . . . 37

5.4 Operating mode of Hot Wire Anemometer Probe . . . 39

5.5 Schematic of Sampleing array at test section of Wind Tunnel VIII . . . 41

5.6 Data Acquistion Software of Hot Wire Anemometer PCE-423 . . . 42

5.7 Hot Wire Anemometer with probe and USB cable . . . 43

5.8 Example of Test results in text le . . . 44

5.9 Plot of Air velocity in test section against time of Sample number10 in (a)Open Loop and with (b)Air feedback tunnel . . . 45

5.10 Color map of Mean Velovity at 40% of frequency . . . 47

5.13 Placement of WiWU in the room . . . 50

5.14 Color map of Turbulence Intensity at 40% of frequency . . . 52

5.17 Involight FM1500 DMX fog machine . . . 56

6.1 Complete WiWU Close loop with settling chamber . . . 58

(11)

Chapter 1 Introduction

The testing of small model aircrafts and launcher controllers at the Institute of Aerospace Information Technology of the University of Wuerzburg had indicated the need for a low speed wind tunnel, which could be fullled in form of the WiWU(Wind Tunnel Uni- versity of Wuerzburg). The main requirements for this wind tunnel were to have a test section, allowing to test an aircraft model of a 1m wing span and to provide a wind speed of > 15m/s. This precondition resulted in a cross section area of 1500mm × 700mm and length to 1500mm. The primary intention of building this wind tunnel was to investigate the robustness of small model aircrafts and launcher controllers and to give a rsthand experience to students in the eld of Aerospace manufacturing.

1.1 Background and Motivation

The invention of the wind tunnel can be without a doubt considered as a milestone in the history of Aviation and Space Technology. Throughout the world scientists have been doing research on how uid tends to react on shapes of dierent bodies. Leonardo da Vinci (1452-1519) was the rst to suggest the attachment of a model of a physical body in a certain position in order to be able to test its reaction on pressurised air owing over it. According to him, forces are equal, no matter if air is moved over a body or vice versa. His theory, however, was nevertheless exposed to vehement criticism until it gained validity[10, 18]. An experiment in which dierent models at some speed were moved in still air was easier and cheaper to conduct. This idea was rst implemented by the English mathematician Benjamin Robins (1707-1751) as the Whirling Arm. He created a machine with a1.26m- feet long arm which could be spun by a falling weight

(12)

attached to a wheel. The tip of this arm could eventually reach the velocity of a few feet per second. With this machine he had proved that air resistance is an inuential factor for ying objects. Consisting of a whirling arm his device lead to a mechanism in which the weight M spun a drum as well as rotated the test object P [19].

Figure 1.1: Benjamin Robins Whirling Arm[20]

The English scientist Sir George Cayley (1773-1857) became famous for his concept of a modern airplane¹ in early 1799. He was the rst who built a whirling arm dedicated to the study of ight. He had used the whirling arm for his basic aeronautical experiments to predict the drag (resistance) which the moving body experienced in air. His work displayed the rst research done on airplane aerodynamic. It was eventually presented as evidence for the circumstance that the higher the aspect ratio of the wing is, the higher the lift over the drag[21, 22]. Later it was then recognized by Samuel Langley while he was using the rotating arm inside a room that it actually leads the air to produce eddies. Due to this he decided to do his experiments outdoors. Under high turbulence, however, one could not examine the true relative velocity between the test model and air. Apart from that it was even more dicult to determine the small forces aecting the test model under high speed[19, 21].

Francis Wenham, a member of the Aeronautical Society of England, contributed to the development of the wind tunnel as a marine engineer. He is credited to have constructed and successfully used the very rst wind tunnel in 1871. In the publication

1A machine with xed wings, a fuselage, horizontal and vertical structures

(13)

Wind Tunnels of NASA ² Wenham remarks that it "had a trunk 12 feet long and 18 inches square, to direct the current horizontally, and in parallel course."Air was blown through a box with the help of a fan³. It was used to measure the lift and drag of the air on various models in the wind tunnel. His research helped to understand the eect of the aspect ratio, concerning the fact that long, narrow wings are able to maintain a higher lift than stubby wings by indistinguishable wing areas[23].

Otto Lilienthal (1848-1896), the world's rst true aviator, used a rotating arm for his experiments with a velocity of 10m/s. Using the data on aerofoil collected by Lilienthal, the brothers Orville and Wilbur Wright performed a test in 1901 for which they utilized a bicycle (see Figure 1.2).

Figure 1.2: Bicycle used by Wright brothers in 1901, photo by NASA(http://www.grc.nasa.gov/WWW/k-12/airplane/wrights/biketest.html)

For that they installed two test objects on a wheel with the bicycle running at a speed of 15miles/h. The up and down streets of Dayton were helpful to reach the speed.

At the end the results did not match their expectations. In 1901, the test done by the Wright brothers at Kitty Hawk resulted in a huge failure. Wilbur Wright nally explained that in one glide the machine rose higher and higher till it lost headway.

Later he commented on the experiment with the words: We cast it all aside, and

2Donald D. Baals, William R. Corliss(1981),Wind Tunnels of Nasa:Volume 440 of NASA SP, Wash- ington,USA

3Steam powered fan

(14)

decided to rely entirely upon our own investigations[27]. Consequently, the Wright brothers began to conduct new tests for their aspired ight. They gured out that they needed a complete wind tunnel. Within the same year they built one with a test section of area 0.4m × 0.4m (See g.), allowing a maximum wind speed of 60km/h.

With the new data the Wright brothers eventually planned to write a new handbook for aerodynamics. The day came on December 17th in 1903 when the Wright brothers mangaged to y at Kitty Hawk for exactly 59 seconds, covering a distance of 862 feet on the ground[24, 19].

Figure 1.3: Wind Tunnel in 1901, photo by NASA(http://wright.nasa.gov/airplane/test1901.html) The problem of scaling models in an ideal relation to real aircrafts in wind tunnels

was solved by Samuel Reynolds (1842-1912) from the University of Manchester. He revealed in one of his experiments that the airow eect is the same on the scale model as on the full scale object. It was based on the condition that some ow parameters are equal. Reynolds combined the measure of an aerodynamic object's size and its ow

eld's density speed, as well as its viscosity into a simple ratio. All these measures were expressed by him in only one factor, called Reynolds number (Re)[25, 26]. Ludwig Prandtl (1875-1953), a member of the Motorluftschistudiengesellschaft (MStG), Goet- tingen in Germany, was in charge of working on a wind tunnel project to measure the drag and pressure distribution in airship hulls ⁴. What distinguished Prandtl's wind

4For more please read Airship Technology edited by Gabriel Alexander Khoury

(15)

tunnel, built in 1908, from former devices was the fact that it displayed a closed circuit wind tunnel with an air duct of 2m × 2m throughout. A honeycomb structure served to reduce turbulence in ow[28, 24]. Many test facilities were built during World War I and II in order to gain the maximum aerodynamic and control abilities in aircrafts.

New aircrafts were eventually faster and more maneuverable. Later the scientic com- petition for space technology was initiated. The wind tunnel was now used for testing spacecrafts, as well as high speed aircrafts. For these purposes the worldwide largest wind tunnel was constructed by the NASA in Ames Research Center, California. It is capable of testing a 30m wing span and 426m long aircraft[34]. In previous times wind tunnels had been mainly built for military purposes. Nowadays many universities use wind tunnels to educate students in the eld of Aerospace. One example is the low speed wind tunnel MUB, located in the Department of Mechanical Engineering of the Technical University of Braunschweig. This wind tunnel includes two test sections of 1300×1300 and 800×800 mm2, where the maximum air speed is 60m/s in the big test section[35].

(16)

Chapter 2 Fundamental of uid mechanics for Low Speed Wind Tunnels

2.1 Boundary Layer

In wind tunnels the Boundary Layer plays an essential role in experiments. It is for this reason that in this section a brief overview is given on that part of the subject matter.

Two-dimensional boundary layers are recommended to decrease the disparity between theory and tests [38].Whenever any uid ows over an object's surface the molecules of the last layer¹ are likely to be attached to the surface of the object. Consequently, the velocity of this layer is identical with the object's velocity. In terms of the wind tunnel walls, this layer velocity will be zero; due to the wall shear stress² this specic condition is known as Slip Condition. The velocity of uid varies from zero to maximum in upright layers. It is this type of layer, formed near the wall of the wind tunnel, known as Boundary Layer, where viscosity plays an important role.

It leads to a laminar form at low(Re), whereas the ow converts to turbulent ow as Reincrease[39, 40]. According to the British physicist and engineer Osborne Reynolds

the general character of motion of uids in contact with solid surfaces depends on the relation between a physical constant of uid, and the product of the linear dimensions of space occupied by the uid, and the velocity[12]. If Lts complies with the length of the test section, Uts complies the velocity of air within the test section, then Reynolds number is shown by Re. Therefore one can rewrite all these parameters in the following

1Which is near to the object's surface

2Shear stress between wind tunnel's wall or object surface and closet layer of uid to them

(17)

Figure 2.1: Boundary Layer representation with laminar ow and turbulant ow.

equation:

Re = U_tsL_ts

ν (2.1)

where υ is kinetic viscosity³[13, 14].

Figure 2.1 (y) shows the height of free stream velocity U from the wall of the wind tunnel. Delta shows the Boundary layer thickness. UL is the wall velocity. Figure 2.1(a) shows the laminar ow in the boundary layer and gure 2.1(b) shows the turbulent

ow.There are many denitions for Laminar ow. According to Smith uid can ow in one of two ways. One is in smooth, layered fashion, in which the streamlines all remain in the same relative position with respect to each other. This type of ow is referred to as laminar ow. At high Reynold's numbers the layer of air ow nearest to the wall surface acts like the wall surface. Due to many swirls being formed in this layer, all molecules become amalgamate, moving in an irregular fashion[10, 11].

2.2 The Continuity Equation

The mathematical equation that represents the conservation of mass of moving uid is known as the Continuity Equation. Suppose that a uid is in motion with the speedv, distance s moves as uid in a time interval of ∆t then s can be calculated as below:

3It can be dened as inherent friction of adjoining layers in uid moving at dierent velocities.

(18)

s = v∆t (2.2) Presumed that the uid is in motion in a tube of a cross sectional area ofA, the volume V of the uid can be express at a point as:

V = v∆tA (2.3)

The mass ow rate of this uid in the tube can be calculated by the following equation:

∆m

∆t = ρAV (2.4)

where ρis uid density. If the mass of the uid is constant between two points of the tube⁴, this type of ow is called Steady Flow⁵. As a result the mass ow rate will be constant at both points[37]. This can be expressed in form of the following equation:

ρ₁A₁V₁ = ρ₂A₂V₂ (2.5)

In the case that the uid within the tube is incompressible and at low speed, its densities at both points of the tube should be the same. Thus the equation can be rewritten as:

A₁V₁ = A₂V₂ (2.6)

2.3 Bernoulli Equation

Bernoulli's Equation basically represents the relation between velocity, density, and pressure. Since density is a constant, as explained in the previous section, the following equation expresses the relation of pressure and velocity betweeen P2 and the conditions at P1[15]:

P₁+ 1 2

m

V v²₁+ m

V gh₁ = P₂+ 1 2

m

V v²₂+ m

V gh₂ (2.7)

In the following equation, presented by Nicholas in College Physics to understand more deeply the relation of presure and velocity,^m_V is the density of the uid which can be written as ρ. Therefore one could rewrite it in the form below, known as Bernoulli's

4in other words without additional uid between these points

5independent on time

(19)

Equation :

P₁+1

2ρv²₁+ ρgh₁ = P₂+ 1

2ρv²₂+ ρgh₂ (2.8)

P₁, P₂- static pressure at point 1 and point 2 v₁, v₂- ow speed at point 1 and point 2

h1, h2- height of two ends of the tube at point 1 and point 2

In the case that v = o the pressure at two points is equal. Hence it only appears when the uid is in motion. If the Bernoulli Equation is expressed in terms of the work energy theorem, then the total mechanical energy of the uid is conserved when moving from one place to the other. Still, a part of the energy is likely to be transfered from kinetic to potential or vice versa.

If the air within the wind tunnel is a incompressible uid, then the Continuity Law is valid for all the section of wind tunnel:

A1v1 = A2v2 (2.9)

Apart from that A1,A2 display the area at two points, whereas v1,v2are velocities at point 1 and point 2. The air ow through the wind tunnel will get pressure losses, which can be compensated by a raised pressure of the fans. As a result, the ratio between the lost pressure in a particular section and the dynamic pressure at the entrance of the wind tunnel can be written in the following form:

K₁ = M H

1 2ρ1v1

(2.10) Where K is the loss coecient without dimension,M H can be dened as pressure loss at the section of measurement of the loss coecient [17].

(20)

Chapter 3 Designing the Low-Speed Wind Tunnel WiWU:

3.1 Wind Tunnel classication

Wind tunnels can be classied in two categories on the basis of geometry, the rst being Open Loop Wind Tunnels and the second Closed Loop Wind Tunnels. Further predetermined requirements in terms of performance and dimensions of the WiWU, as well as technical means are listed in the table below:

Type Subsonic model

Max. speed >15m/s

Min. speed 2m/s

Test Section 1500 × 1500 × 700 mm³ Visualization with smoke Table 3.1: Requirements for the WiWU

3.1.1 Open Loop Wind tunnel

This kind of wind tunnel features no loop feedback to its contraction. Its specic design was devoloped by the French engineer Gustave Eiel in 1909, thus it is also called Eiel Wind Tunnel [21]. The main sections of this type are (1) contraction, (2) test section, (3) diuser and (4) fan as shown in gure 3.1 .This category of wind tunnels is usually used in spacious, closed rooms in order to have the benets of natural feedback air.

The generic term Open Loop Wind Tunnel can be divided into two further types:

(21)

The Suckdown Wind Tunnel in which air is sucked down by a fan through the areas of contraction, test section and diuser.

The Blower Wind Tunnel in which air is blown through the areas of diuser, test section and contraction.

Figure 3.1: Open loop Suckdown Wind Tunnel

3.1.2 Closed Loop Wind tunnel

In contrast to the previously depicted type, this category of wind tunnels, which was devised originally in Göttingen/Germany (Cf. previous section), is characterised by a loop feedback to its contraction. Besides the sections of contraction, test section, diuser and fan it consists also of the additional sections of turning vanes and the loop. Denite advantages of a Closed Loop Wind Tunnel are a high quality ow, as well as a decrease of pressure loss. The category can be divided in the following two subcategories:

The Open Test Section in which air is blown from the contraction cone to an open space between the contraction area and the diuser.The test model is usually placed in this space.This type of test section creates much higher pressure loss.

The Closed Test Section in which air is blown from the contraction cone to a closed wall test section. The test model is usually located in this closed wall section. The walls of the test section cause Wall friction[44].

(22)

Figure 3.2: Closed Loop Wind Tunnnel with closed test section[51]

3.2 General Design of the WiWU

The rst challenge in the designing process of the WiWU was the low budget assigned for this project, resulting in a very restricted choice of material and equipment. Apart from that, the rather small room¹ provided for the construction of the wind tunnel, was another important factor which had to be considered in the designing phase.

With regard to the low budget, the concept of a Suckdown Open Loop Wind Tunnel turned out to be most suitable and promising for the basic design of the WiWU, since it allowed to reduce the costs for constructing a close loop for the wind tunnel. In addition to that, parts of the budget could be saved by selecting a rectangular shape instead of a round or semi-rectangular shape of the wind tunnel. The Continuity equation for wind tunnel can be rewritten for the WiWU as following:

ρ₁A₁V₁ = ρ₂A₂V₂ = ρ₃A₃V₃ (3.1)

ρ1A1V1 = constant (3.2)

121m×5m

(23)

Figure 3.3: Basic 3D design of WiWU

This is valid for all sections of the WiWU. On the basis of the given requirements the

rst basic 3D-model of the WiWU was designed as illustrated in the gure 3.3.

3.3 Drive Section of WiWU

3.3.1 Selection of suitable fans for the WiWU

Fans (or propellers) make up the most indispensable part of a wind tunnel, as they accelerate the air ow within a given area and ensure that it is kept at a high speed.

It is therefore not surprising that companies providing them demand high prices. Nev- ertheless, in order to guarantee a high performance of the WiWU, the fans selected for this wind tunnel make up the part of the purchased equipment and material, where

(24)

a compromise in favour of a cost reduction has been made least. Important predetermined criteria like had to be kept in mind for the selection of suitable fans for the WiWU, the wind speed in the test section and the loss of static pressure in the wind tunnel.

As previously mentioned, the required area of the test section had been dened with a dimension of 1500×700 mm sq, whereas the intended ow speed in the test section was supposed to be >15 m/s. Based on the required speed of the ow speed at the diuser can be determined by applying Bernoulli's equation. To ensure an equal air ow in the rectangular test section area, two parallel fans (instead of one) were included in the design concept. This in particular is also an advantage in terms of cost reduction, as a single fan comparatively requires a bigger motor which is quite expensive to operate.

Apart from that, it was expected that in the constructing phase two fans of medium size would be easier to handle compared to a single massive fan. Speaking of practicable methods, it seems important to point out, that the WiWU was mainly built by a small team of students (5-6) within a short time period of 3-4 months. Consequently, the technical skills and working experience of the team members had to be set within the design concept of the wind tunnel, in order to avoid complications throughout the construction process.

In the case that the diuser has same the dimensions as the test section, Bernoulli's equation can be applied as follows:

U_testA_test= U_{f an}A_{f an} (3.3) where Utestand Uf an are the velocities in test section and fan respectively. Atestand Af an represent the cross sectional area of test section and fan. In order to ensure that the velocity in the test section will be 15m/s , the performance of the fans have to ensure at least the same velocity. Since fans of this performace category would have exceeded the given budget of our project, an alternative was seeked. After consulting with a great number of dierent companies, providing fans suitable for the purpose of the project, an axial fan with duct conguration was purchased. With regard to the amount which had to be invested, a good cost-performance ratio has been achieved.

The typical air ow given in the provided data sheet for this type of fan was indicated with 20,200 m3/h or 5.61 mqube/s. The gure below shows one of the purchased fans (see also appendix):

(25)

Figure 3.4: Fan (EL 710 EC 01 | 119359), purchased from Ruck Ventilation GmbH (Boxberg, Germany)

3.3.2 Electronics Design

Twin-fans with two separate variable frequency drive controllers can be of an advantage in many ways. Electronic controllers make it not only possible to reach a maximum eciency of the fan system, but also allow ideal test conditions for various airplane models and test objects to be created, as one can set precisely a certain wind speed within a few seconds.

Figure 3.5: Electronics design to operate the fans.

It is important to have a complete control over the electronics system control with

(26)

regard to emergency cases which might need to stop the fans immediately or slow down the speed. A basic electronics design,which guarantees an uncomplicated and safe handling of the fan system for students working with the WiWU, is illustrated in the gure 3.5. Due to the possibility of controlling the air ow speed, it is easy to simulate dierent levels and structures of air turbulence, resulting in resistance which could similarly be caused by buildings or launchers in a horizontal ying state. Since fans of this size and power, running at a high speed, can be a risk for the user's safety, an electronic system which can be operated via a computer from some distance, serves here best.

3.4 Test Section

The test section is located between the diuser and the area of contraction. The open and closed type are the two most common categories of test sections, each having its own advantages and disadvantages. In constrast to the closed type, open test sections need to be adjusted from time to time, since it is the part of a wind tunnel which encounters the highest forces exerted by the air ow [4, 5]. In case of a closed type, signicant pressure losses have to be attributed to the walls. Furthermore, an airplane model used in the test section also tends to create disturbance in the ow, resulting in additional pressure losses [4, 31]. Open test sections feature even higher pressure losses than the closed type, but have the advantage that one has an easy and fast access to the respective test model.

For the project of the WiWU an Open Loop Wind Tunnel was considered to be best, as it has a lower energy ratio compared to a Closed Loop Wind Tunnel. Consequently, an open test section was not suitable for the WiWU. One of the requirements for this project was to have the possibility of using an airplane model (or a launcher model),² of 1mwing span for wind tunnel tests; therefore the dimensions of the test section had to be laid out for this purpose, having at least a width to height ratio of1.5m, whereas the length of the test section was based on the length of the aircraft model. Moreover, the size of a wind tunnel test section is related to the ow speed and the size of the fan(s) The larger the test section, the bigger the fan(s) will be. For a bigger test section more powerful fans are required to produce the same ow of air. Correspondingly it can be said, that the features of a wind tunnel's test section determine signicantly the design

2This model will be in a static position for the rst test, but in future if possible a small model in

ying conditions can be used for further tests.

(27)

for the rest of the construction ³. The size and conditions of the room where the wind tunnel will be located also have to be considered[17]. It is important to keep in mind that the shape of the test section also aects the force being exerted on the model.

In the case of the WiWU a rectangular shape will be used, mainly for the reason that it is relatively simple to manufacture. In order to have access to the respective airplane model, a window in one of the test section's walls was included. Hermann Glauert, a British aerodynamicist, who investigated how the walls of a wind tunnel induce the eect on the wings of test model, proposed the following:

4α = δ_wA_wing A_{T S}

c_L (3.4)

δ_wrepresents the boundary correction factor.cL shows non-dimensional lift coecient, wind axis AT S equates the area of the test section. Awing stands for the wing area. δw

depends on the shape of the test section[4].This equation can be taken into account as a wall correction method. Figure (3.6) illustrates the design of the test section⁴

Figure 3.6: 3D design of WiWU Test Section

3Size of diuser and contraction.

4All designations are in mm and refer to the inner dimensions of the test section.

(28)

3.5 Diuser

The diuser is the last section of a wind tunnel, located before the fans. In a Suckdown Wind Tunnel the fans must be positioned at the end of the wind tunnel rather than at the front, to produce a less turbulent air ow in the test section, since the fans are signicantly involved in the creation of turbulence. The diuser allows the air ,coming from the test section, to expand and leads to a decrease of velocity as the cross section area of the diuser is larger. A challenge given in the designing of the diuser, is the necessity to decrease the speed of air ow within a short distance, without causing a separation of the boundary laysers.The process of decreasing dynamic pressure and increasing static pressure in the diuser, supports the fans to achieve a high performance[33]. Consequently, a Suckdown Wind Tunnel can be considered as advantageous here, since it creates a vacuum in the test section rather than blowing turbulent air there.

The air ow in the diuser depends on its shape and dimensions. The boundary layers are thick at the entry of the diuser. It is important that the angle of the diuser is kept as small as possible so that the boundary layers will not separate. [17].If the skin friction and density of the uid are constant, then the loss coecient can be determined as follows:

K_d = K_f + K_ex (3.5)

The geometry of the diuser is given by the area ratio and the diuser angle. To ensure a uniform ow, the diuser angle should not be more than 5⁰. The area ratio of the diuser should be limited to 2.5. A conical shape of the diuser is useful to increase the air ow speed in the test section[1, 2].

In case of the WiWU, the diameter of the fans which will be used to achieve the commensurate ow is 718 mm⁵ minimum. Two fans will be in operation. In order to guarantee sucient space for both fans, the area at one end of the diuser should be bigger than 810000 mm². The other end should be expedient to the test section. The determined area for the WiWU's test section is 1500 × 700mm². For the design of the diuser a small angle is acceptable. In this case the angle for the diuser should not exceed 7⁰[3] . Corresponding with these parameters a 3D-design of the diuser is given in form of the gure (3.7). For the fans an area of 1610700 mm²has been determined.

5The diameter of the fan is 717±2 mm, as illustrated in the Fan Drawing provided by www.ruck.eu

(29)

Thus, the fans need to run at 17m/s to reach the ow velocity of 25 m/s in the test section⁶. The diuser angles are delimitated to 6⁰.

Figure 3.7: 3D design of the WiWU Diuser

3.6 Contraction Cone

The contraction cone is the biggest section of a wind tunnel, which is located after the test section at the entry of air, leading to a compression of the air ow. As a result, the mean ow velocity for the test section of the wind tunnel will be increased.

It can be explained with the help of Bernoulli's equation, that a big cross sectional area of a contraction cone helps to reach the maximum ow speed in the wind tunnel's test section. Another eect of the contraction cone is that it decisively controls the turbulence decay rate. The ow through the contraction cone declines the turbulence levels and modies the turbulence by diminishing the uctuating velocity deviations during the ow through the contraction cone.The boundary layer separation begins at a laminar state and remains as it is. The most important parameter when building the contraction cone is its ratio⁷. A contraction ratio of 6 to 10 is common in most of the small subsonic wind tunnels[8, 7, 9].

6These calculations are based on Bernoulli's equations, with regard to the loss coecient

7Area ratio between test section and area at contraction entry.

(30)

The settling chamber⁸ is normally placed at the very beginning of the wind tunnel.

Hence, it is the section where the air ow usually enters the wind tunnel. When having a contraction cone, it is possible to locate the settling chamber at its beginning, where the ow speed is low. This is of an advantage as it leads to a reduction of pressure losses. Further important parameters when designing the contraction cone are its wall shape and length.[9]

The size of the room provided for the construction of the WiWU did not allow it to include a contraction cone with a ratio of 6-10, as it is recommended by experts (cf.

Mehta, 1989). Therefore some adjustments in terms of the dimensions and shape of the contraction cone were done retrospectively. Eventually, for the WiWU a contraction ratio of 5.6 was applied. One end of the contraction cone was determined to have a dimension of 1500 × 700mm², whereas for the area of the second end a dimension of 3450 × 1610 mm²was suggested .

A very small length of the contraction cone can cause a boundary layer separation at the inlet. The larger the length of the contraction cone, the more probable it is to have a thick boundary layer, but this time it could be at way-out of air ow towards test section. In case of the WiWU the length of the contraction cone does not exceed 2200mm. The pressure loss throughout the contraction can be determined by applying the equation given below[17] :

K_n = (0.32)(f_av)(L_n

D_ts) (3.6)

D_tsrepresents the test section diameter,fav stands for the average friction factor,Lnis the length of contraction, whereas Kn equates the contraction loss coecient. For two dimensional incompressible ows, the boundaries of contraction channel walls are quite

ush until the end of the contraction cone. The shape of the contraction's boundary can be determined by enumerating the constant velocity beside the bowed portion and angle of the inclination⁹. For the WiWU the contraction cone shape was then designed in the modeling software Catia 3D. The gure below (3.8) shows the contraction shape in 3D. All dimensions in this gure are given in millimeter (mm).

8Combination of honeycomb and screens in series.

9Angle of straight portion to the wind tunnel axis

(31)

Figure 3.8: 3D view of Contraction section

(32)

Chapter 4 Construction

4.1 Drive Section

4.1.1 EL 710 Duct Fan with EC motor

For this project two duct fans with a diameter 714mm of EL 710 D4 01 series were selected, as their high eciency allows it to save a lot of energy. According to the manufacturer the eciency of these types of fans is 50% higher than in the case of con- ventional tube fans. This can be mainly attributed to very complex three dimensional blades used for the rotor and the stator, which reduces the loss of energy during the process of energy transformation of the fan. These fans are manufactured as mixed fans with guided vanes. Concerning the aspect of maintenance, free long existence bearings were installed to the fans of the WiWU. The motor of the fan is protected by a special hub, so that it will not cause an interruption of the air ow. Moreover, this hub is advantageous as it increases the air ow quality and eciency of the fan, as well as serves the safety of users.

(33)

Figure 4.1: Arrangements of two Fans in Diuser's end (Front view)

The duct of the fan increases the safety level, as it protects individuals from hazards which could be caused by the blades of the fan and motor. The casing and blades of the fan are made of aluminum material, which signicantly reduces its total weight to only 49.0 kg. These fans provide the possibility of single step mains operations of 400V/50Hz. Each Fan features a 3-phase motor with a maximum operating current of 7.7 A. The utmost pressure from the fan is 1000 pa, at maximum 1450 1/min rpm of the motor. The allowed voltage for the fan is 400 V. For this purpose for this purpose special electrical connections with a high voltage supply were therefore required. Two fans are installed as shown in gure 4.1. To assure the safety of fans two screens for fans are installed.

4.1.2 Controller Unit

In order to be able to operate these fans properly, two Typ ER23K inverters from

BLEMO Frequenzumrichter (Frankfurt a. M./Germany) has been used. Each of the fans is connected to a single inverter, making it possible to control them separately from each other. Both inverters were installed on a controller board, together with an On/o switch, an emergency switch and a socket to connect the controllers to electricity (See gure 4.2). In the case of emergencies both fans can be stopped immediately by pressing the emergency button.

(34)

Figure 4.2: WiWU controller board

4.2 Diuser

Initially, it was decided to delegate the construction of the wind tunnel to a manufacturing company, but after investing in fans and controllers there was a very limited amount left within the budget and in addition to this the time remaining for the project was also restricted. Since it would have been too expensive to commission professionals with the construction of the WiWU, the project was eventually completely conducted by students who were supervised, as well as practically supported by external members of the department.

The rst step was to choose suitable material, being easy to handle and inexpensive at the same time. On the basis of the completed WiWU design, several materials for the construction were chosen. As metal is not only expensive, but also not very easy to convert, wood was considered the best choice. For the recent design of the diuser material like plywood, plexiglass, polycarbonate plates, thin exible wooden plates, and various kinds of mounting hardware were purchased. Parts like the fans and controllers had already been ordered and many of the wooden beams, provided by local hardware stores, were ideal for the intended construction. Eventually, some small corrections in the design of the wind tunnel, with regard to the choice of material, were made before the nal construction was approached. In order to be able to gain more knowledge of and experience in the manufacturing of big sections several small models were made

(35)

with the help of cardboard and small wooden beams.

Figure 4.3: Wooden Model of Diuser and Test Section

For building the diuser section wood and polycarbonate plates (16mm thick and 1200mm wide) were used. These polycarbonate plates, called Guttagliss dual Hohlka- mmerplatten, have the advantage of being especially light and are available in a length of 2000-6000 mm. The inner structure contributes to the stability of the material (see

gure 4.4).

Figure 4.4: Attachment of two polycarbonate plates

Wooden beams (40 × 60mm²) and steel clamps were used to support the structure.

To increase the stability, two side walls of the diuser were manufactured with wooden plates, as they were cheaper and easier to cut into the angle when compared to plexi glass. A wooden sheet of 3mm was installed at one end of the diuser with two round

(36)

cut of Fan diameter in it, being used for an precise attachment of the fan system to the diuser.Without this wooden sheet the loss of air was high. Initially, the fans were attached directly to the diuser.

Figure 4.5: Finished Diuser section outdoor

4.3 Test section

The wind speed at the entry of the test section at maximum speed of the fans was at some points16 − 18m/s under the condition of a very high turbulence. Until this point of the construction phase several diculties had come up. Consequently, the following improvements were suggested and targets set for the next steps:

1. Get more space (or even a separate lab) for the wind tunnel, to ensure a better air ow to the test section and to guarantee easy handling.

2. Increase visibility of the test section, as well as a reducing of weight.

(37)

3. Create easy acess to the inner area of the test section, in order to allow dierent kinds of measurements from the side of the wind tunnel as well as the ability to reach airplane models (or other objects) used in experiments.

4. Increase the velocity under the condition of less turbulence.

A new room could be provided for the further construction of the WiWU, so that the next section, namely the contraction cone, could be manufactured indoors (For more details see next chapter).

Figure 4.6: Completed test section with Anemometer

A new makrolon type of glass was ordered with 98% visibility and 10mm thickness for the side walls of the test section. One side of the test section was reserved for a window. After several designs had been considered, the window was nally realized with a dimension of1000mm×500mm as it is shown in the gure below. Eventually, a handle and ve locks were installed on the window, in order to ensure easy and comfortable access to the test section and to contribute to the safety of users. In addition to that, two holders were attached, which can be used to hold the window in an opened position.

(38)

Figure 4.7: Compleated Test section of WiWU with Airplane Model

4.4 Contraction Cone

The contraction cone - the biggest part of the wind tunnel - turned out to be the most challenging section which had to be built for the WiWU, since its shape was largeand complex, and a great deal of time had to be invested into manufacturing it. Originally, it had been planned to commission a company with the construction of a contraction cone having a cubic curve cross section . But due to budget issues and in order to improve the performance of the contraction cone by changing from rectangular cone to cubic cone with smooth curve. The construction work for the contraction cone was done by team of students.

The newly xed conditions in terms of material needs for the contraction cone were the following:

1. It should be wood, as this material can be more easily converted than metal.

2. It should be in planks with a dimension of 3500mm × 2100mm. Apart from that, supporting beams (min.3500mm long, in one piece) which would not bend, were needed.

(39)

3. The material should be light, since the contraction cone is very big and a heavy construction would make it complicated to adjust with the rest of the wind tunnel.

4. The wooden plates should be very exible with form the required shape of the contraction cone.

After a long period of research and many inquires directed to companies, a suitable material could be found, provided by Klöpfer Holzhandel (Höchberg/Würzburg) ,which oers a very good selection of wooden material in above-average dimensions.

Figure 4.8: Basic frame of Contraction cone

The required wooden planks could be ordered in a size of 5000mm × 2300mm (500cm long,3mm thick ), whereas the wooden beams were purchased from there in a size of 5000mm × 60mm × 80mm. The planks were easy to bend and cut to build the curved shape of the contraction cone.The construction was started with the bottom layer of the contraction. All boundary points for the walls of the contraction cone were determined with the help of a 3D- model designed in Catia (software). Later these points

(40)

Figure 4.9: Completed Contraction cone of WiWU

were transferred on the wooden planks, which were cut with the help of an electrical jigsaw.The wooden beams were directly attached to the lower layer as bearings for this section, so that in the next step also the upper layer could be installed. Steel angel brackets and screws were used to attach the vertical beams with the horizontal ones and the respective wooden plank also assist the permanent attachment of the bearings to the lower side of the contraction. For connecting the planks with the beams at headed screws, bolts and a special type of glue were utilized. Additionally, the beams on each side (including the upper and lower parts) were stabilized by aluminum strips.

In the last step the wooden planks could be installed at the beams. The completed construction cone eventually was at one end (near the test section) 810mm high and at the other end 1730mm high¹. Figure 4.9 illustrates the completed contraction cone of the WiWU.

4.5 Assembly of wind tunnel WiWU

After completion of the construction of all sections of the wind tunnel, the next step was to join them together and positioned them at the correct hight so that the centre line of all sections was even. Each section was very big and heavy. The biggest problem was to move each section with its table what make task more dicult. The fan was

1The given dimensions all refer to the outer part of the contraction cone; the inner dimensions are the same as given in the last chapter

(41)

permanently xed to the table by nuts and bolts to make sure a very stable basis was created which helps decrease vibrations. A sum of three tables served the purpose of supporting the entire construction of the WiWU, except the contraction cone. The fans were on one table, the diuser section was on second table and the test section on the third table seperately. The purpose of using the disconnected tables for the WiWU, was to reduce the vibrations transfer from the drive section to the test section.

A special type of tape and thick foam was used as insulation in the wind tunnel.

Long nuts with bolts were used to tighten all the sections together. After nishing all construction all of the corners were sealed with special tape as well as extra glue normally uses in for the sealing of ventilation systems. It helps to minimize pressure losses and and irregularities in the ow withn the wind tunnel. The lowest part of wind tunnel lowest point of contraction was 24cm high from oor. The lowest part of the wind tunnel² was 24cm high from oor.

An air feedback tunnel of dimension 1500mm × 700mm was bought from a ventilation company. This feedback channel made a partial loop for air from the fan to the contraction. The complete tunnel was made of metal and consequently very heavy. The company provided only sections of the channel, later which were assembaled within the WiWU. The most dangerous moment was to build this channel over the contraction because of very less space between roof and upper part of contraction. This chanel is supported with help of very strong metal legs. Figure 4.10 shows the assembling of the feedback channel.

2Lowest point of contraction

(42)

Figure 4.10: WiWU with air feedback channel

(43)

Chapter 5 Experimental Composition and Hot Wire Anemometry

More details on that will be given in the next section of this chapter (cf. 5.2). Before- hand, the most signicant parameters and formulas should be exposed at this point.

5.1 Introduction

It is important to be aware of its performance and parameters before the application of the newly constructed WiWU for experiments. Thus, the turbulence intensity and mean wind velocity in the test section, under the conditions of an open loop and air feedback channel, have been investigated extensively¹. For this purpose the method of multi-point anemometry, using a hot wire anemometer- proved in our case to be most eective.

When the wind ow in the wind tunnel test section is not completely laminar but with ξ of uctuations, then in a particular time t during the wind ow, a eeting value for the wind velocity can be calculated as follows[42]:

U = U_mean+ ξ (5.1)

The turbulence intensity Tin given in the wind tunnel can be determined by taking the ratio of standard deviation of the wind speed into account , as well as the mean wind

1See also last chapter

(44)

Figure 5.1: Open Loop test condition

Figure 5.2: With Air feedback channel test condition

(45)

speed, resulting in the following equation[29, 30]:

T_in = U_rms

U_mean (5.2)

U_mean and Urms(shown in gure )can be calculated as hereinafter[41] :

Umean= 1 N

N

X

i

Ui (5.3)

Urms = v u u t

( 1

N − 1

N

X

i

(Ui− Umean) )

(5.4) whereN is the number of samples.

Figure 5.3: Turbulent ow

The average of these measurements at each checkpoint provides useful information about the conditions under the respective wind speed. The average speed Umean when

(46)

multiplied with a cross sectional area Λof the test section, equates to the volumetric airowQ in m³/sec as displayed below[29]:

Q = ΛU_mean (5.5)

5.2 The Method of Hot Wire Anemometery

The most common instruments to measure the wind velocity in a wind tunnel are the Constant Temperature Anemometer (also called Hot Wire Anemometer) and the Pitot Tube Anemometer.

A Constant Temperature Anemometer's (CTA) functional principle lies in the process of a varying heat transfer in the probe with the convective heat transfer coecient h, since the wind ow rate varies.

In contrast to that, a Pitot Tube Anemometer (PTA) operates according to the con- version process of kinetic energy of ow to pressure energy, where the impact pressure² hence give the velocity of ow.

For testing the wind ow quality in WiWU 's test section, a Hot Wire Anemometer was selected for the following reasons:

The air ow in its test section is low, resulting in a low dierential velocity pressure. This can cause uncertainty during measurements taken in a PTA. Despite this, with a CTA still air (low up to 0.2 m/s) can be measured with high accuracy.

The signal to noise ratio is indicative in a PTA for inadequate uctuations in ow.

When using a CTA it is possible to handle these type of uctuations properly.

A PTA needs to be a long, straight duct for excellent performance, whereas a CTA probe is much smaller.

Higher frequency response of CTA than PTA , which make the CTA more eective in uctuation measurement[45].

As mentioned in the last section, the main intention of this experiment is to determine the turbulence intensity given in the test section. In addition to the explanation above, the illustration below should be useful in order to comprehend the working procedure of a CTA:

2The dierential pressure of the measured static pressure by the tube and its pressure buildup.

(47)

Figure 5.4: Operating mode of Hot Wire Anemometer Probe

A feedback amplier is used in the CTA circuit to conserve the constant mean temperature of the wire. A heat balance equation of for the heated wire of resistance R, length L and diameter dw with curent I can be formed as follows:

dc_w

dt T_w = P − Q (5.6)

dc_w

dt T_w = I²R_w− πLd_wh(T_w − T_adw) (5.7) where cw is specic heat of wire, h shows heat coecient and Twis temperature of wire.

For CTA dTw/dt = 0 in equation (23) in limitation of amplier used in anemometer.

Or more precisely, the heat stored in the wire can be described as the dierence of electrical power in wire P and aerodynamic heat transfer out Q from that. King's Law³ specied by P.C. Stainback and K.A. Nagabushana with constant T0 and densityρas follows

E² = A + B√

u (5.8)

whereE represents mean voltage across probe wire.In the following equationu = f(E)

3Prof. L.V. King gave a formula to correlate air ow velocity component to axis of linear wire and heat transfer coecient of linear wire[49].

(48)

results in

u = A₁− A₂E²+ A₂E⁴ (5.9)

Later it was discoverd that A1− A₃ are the functions of temperature T [43, 41, 45, 47].

It is possible with a CTA to receive instant measurements of the wind velocity and temperature at a specic checkpoint in the test section. Measurements of the air ow rate when using a CTA are quite similar in Pitot tubes. The wind speed is measured at every transverse point of the test section. During all measurements the time of response of the respective CTA should be kept in mind [?]. In general anemometers are able to record velocity uctuations at frequencies of up to 100kHz[32].

5.3 Experiment Composition

A hot wire anemometer attached to a computer has been used for the measurements.

The anemometer probe was placed at more than 56 checkpoints to collect velocity samples under dierent conditions⁴. In order to take measurements at various points with the help of a probe, seven holes were bored in the lower part of the wind tunnel's test section. These holes were placed at a distance of 21.42cm to each other in a x-axis transverse direction, in order to ensure that the probe of the hot wire anemometer can reach all points of the test section. Hot wire anemometer could transform in vertical direction. The distance between two checkpoints is 17.5cm in y-direction longitudinal as shown in gure 5.5. The measurement was done at total 28 dierent points of test section.

Initially, it was decided upon that the total number of samples to be taken at each checkpoint was1000 at 5 dierent positions in the x-direction decided. The anemometer could take measurements of the instantaneous velocity(Uτ) over a time interval (τ) of one second. Since the anemometer could provide one sample per second, it was later decided to reduce the number of samples per checkpoint but increase the check points.

The number of total samples to take at each checkpoints was taken at a certain fan speed was 600 samples. The samples taken at each checkpoint, wind speed was respectively set up to 40% ,70 %, and 100% of the total frequency. Data from the anemometer was recorded continuouswithly with the help of a computer. Before measurements could be taken at a new checkpoint, the fans had to run for at least 10min (or more) in advance,

4The tests were conducted with an open loop, an air feedback as well as with some temporary screens.

(49)

in order to have a stable ow.

Figure 5.5: Schematic of Sampleing array at test section of Wind Tunnel VIII

5.4 Gradual Data Logging process with CTA PCE- 423

Using a CTA PCE- 423, measurements with a minimum air speed of 0.1m/s and a maximum air speed of 25.0 m/s can be taken. The resolution of this anemometer lies in a range up to 0.01 with +/-(5%+1d) accuracy. It can measure temperatures from 0-50°C with of up to +/-1°C accuracy. As the probe that is connected to the CTA PCE-423 has a diameter of 12 mm at maximum, it does not eect the wind ow remarkably. Consequently, the features of this anemometer make it possible to capture

uctuations in the wind ow precisely in the respective time domain in order to get correct measurements from the CTA the follwing steps should be considered:

1. Placeing the probe at a certain checkpoint in the test section, and adjust its height and direction, if necessary.

2. Connecting the probe's plug to its input socket.

3. Attaching the anemometer to the computer using USB port after having installed the data acquisition software provided by the CTA PCE-423.

(50)

Figure 5.6: Data Acquistion Software of Hot Wire Anemometer PCE-423

4. Making sure that the Anemometer head cover is closed before switch on the anemometer, using the Power On/O Button and wait for 5-10 seconds for cali- bration.

5. Adjusting the direction of the probe by using the arrow sign situated at the head of the probe.

6. As soon as the protecting cover of the probe is removed, the sensor will come in contact with air. Then the air ow variations passing over it dissipate the heat, so that the anemometer will respond by measuring the air velocity.

(51)

Figure 5.7: Hot Wire Anemometer with probe and USB cable

7. Before start gathering information, let the fans should run at the desired speed atleast for 5 min.

Follow all steps again for the next sample. ( For more details see appendix)

5.5 Data Analysis

In this section, the data was analyzed with the help of the data acquisition software Matlab. An example of the logging text le of data gethered by the software is shown

in gure (5.8). Data collected at one checkpoint is given in the appendix. Plots of velocity in relation to time cannot be saved simultaneously. To understand the data with regard to the plots, a Matlab code was written for each sample was transformed

into a matrix.

(52)

Figure 5.8: Example of Test results in text le

The velocity (m/s) in relation to time was plotted for each sample as in gure 5.9.

5.5.1 Ploting Velocity vs Time

Here is the example of plots from the measurement at sample M10 (g.) point in test section using an open loop, air feedback channel and both fans running at full power.With the help of the equation 19 and 18, the mean velocity and turbulence intensity were calculated for each check point and plotted for the test section as shown in the gure 5.9. The heading of the plot indicates the mean velocity of the respective sample and turbulence intensity at the particular sample point. AFC stands for Air Feedback Channel.

(53)

(a)

(b)

Figure 5.9: Plot of Air velocity in test section against time of Sample number10 in (a)Open Loop and with (b)Air feedback tunnel

(54)

Plot (a) on the one hand shows a turbulence intensity of 1.5% and a mean velocity of 11.96 m/s under an open loop test condition. On the other hand plot (b) indicatesplot (b) indicates a turbulence intensity of 2.7% , as well as a mean velocity of 11.36 m/s, using an air feedback channel. High uctuations under AFC- conditions cause a higher turbulence than in the case of an open loop.

5.5.2 Mean Velocity Colormap

To understand the behavior of airow velocities at dierent points of the cross section, Matlab colormaps have been created.Under both conditions with and without air feedback channel and three dierent fan speeds respectively, in total 84,000 samples have been taken. Figure 5.10 shows the color map of the mean velocities at sample point M14 while both fans were running at 40% of the total frequency.The color bar in gure 5.10 indicates the velocity in m/s. The heading of the graph refers to the mean velocity in the test section at a particular speed of the fans and test conditions where OL represents an open loop and AFC represents the air feedback channel. In the plots the x-axis and y-axis show the position of the sample points in the test section. To read these color maps the schematic of sample array in gure 5.8 will be helpful.

(55)

(a)

(b)

Figure 5.10: Color map of Mean Velovity at 40% of frequency

(56)

(a)

(b)

(57)

(a)

(b)

(58)

These color maps provide clear results for the velocities and the mean velocities in the test section. When the fans are running at a low speed, the dierence between the test conditions of mean velocities under an OL and AFC is slow as 0.4 -0.6 m/s, whereas at a high speed of the fans it increases up to 1m/s. In an open loop the velocity has a more uniform pattern than under AFC- conditions at a particular speed of the fans.

Despite this, in all plots the velocity is lower on the right than on the left side.

(a)

(b)

Figure 5.13: Placement of WiWU in the room

The color bar in gure 5.12 shows that at some points in the test section the velocities with a AFC were very high (up to 16 m/s), but at the same speed of the fans under OL- conditions not more than 12 m/s were achieved. An explanation for this phenomena is

(59)

that the contraction cone has a larger area for the air ow on the left side as on the right (side). This restriction is given by the uneven height of roof, as shown in gure (5.3 b), the wind tunnel was not exactly in the middle of wind tunnel but near to the right wall and far to the left wall. This wall also could be a reason for turbulence in the test section.

5.5.3 Turbulence Intensity Colormap

The following color maps indicate the turbulence intensity at dierent checkpoints in the test section at dierent speeds of the fans. Here the color bar shows the turbulence intensity. The heading of the graph indicates the turbulence intensity in the test section at a particular speed of the fans. As previously shown, OL represents an open loop, whereas AFC represents an air feedback channel. As mentioned above, the x-axis and y-axis indicate the respective positions of the checkpoints in the test section. In order to read the color map schematic of the sample array in gure (5.8) will be helpful.

(60)

(a)

(b)

Figure 5.14: Color map of Turbulence Intensity at 40% of frequency

(61)

(a)

(b)

(62)

(a)

(b)

Designing and Construction of Low Speed Wind Tunnel WiWu To Investigate The Robustness of Small Model Aircrafts and Launcher Controllers

MASTER'S THESIS

Designing and Construction of Low Speed Wind Tunnel WiWu To Investigate The Robustness of Small Model Aircrafts and

Launcher Controllers

Vikas Dalal

Designing and Construction of Low Speed Wind Tunnel WiWu

Space Science and Technology

Vikas Dalal

Designing and Construction of Low Speed Wind Tunnel WiWu

Space Science and Technology

Vikas Dalal

Contents

List of Figures

Chapter 1 Introduction

1.1 Background and Motivation

Chapter 2

Fundamental of uid mechanics for Low Speed Wind Tunnels

2.1 Boundary Layer

2.2 The Continuity Equation

2.3 Bernoulli Equation

Chapter 3

Designing the Low-Speed Wind Tunnel WiWU:

3.1 Wind Tunnel classication

3.1.1 Open Loop Wind tunnel

3.1.2 Closed Loop Wind tunnel

3.2 General Design of the WiWU

3.3 Drive Section of WiWU

3.3.1 Selection of suitable fans for the WiWU

3.3.2 Electronics Design

3.4 Test Section

3.5 Diuser

3.6 Contraction Cone

Chapter 4 Construction

4.1 Drive Section

4.1.1 EL 710 Duct Fan with EC motor

4.1.2 Controller Unit

4.2 Diuser

4.3 Test section

4.4 Contraction Cone

4.5 Assembly of wind tunnel WiWU

Chapter 5

Experimental Composition and Hot Wire Anemometry

5.1 Introduction

5.2 The Method of Hot Wire Anemometery

5.3 Experiment Composition

5.4 Gradual Data Logging process with CTA PCE- 423

5.5 Data Analysis

5.5.1 Ploting Velocity vs Time

5.5.2 Mean Velocity Colormap

5.5.3 Turbulence Intensity Colormap

Fundamental of uid mechanics for Low Speed Wind Tunnels

3.1 Wind Tunnel classication

3.5 Diuser

4.2 Diuser