See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/328382128

Vertical Axis Wind Turbine

Thesis · July 2018

DOI: 10.13140/RG.2.2.12592.81927

CITATIONS READS
0 1,253

2 authors:

Ahmed El.Attar Ahmed Shahin

Alexandria University Alexandria University
2 PUBLICATIONS   0 CITATIONS    2 PUBLICATIONS   0 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

numerical simulation of the j-2x rocket engine nozzle during the start up condition View project

Vertical Axis Wind Turbine View project

All content following this page was uploaded by Ahmed El.Attar on 19 October 2018.

The user has requested enhancement of the downloaded file.

PERFORMANCE ANALYSIS, DESIGN AND MANUFACTURING OF

VERTICAL AXIS WIND TURBINE

A B.SC. SENIOR PROJECT

Authors
Ibrahim Gaber Abd El.maksod

Ahmed Awni Hafez Abd El.Aal

Ahmed Mohamed Elsayed Hussien

Ahmed Mohamed Hassan Hafez

Ahmed Moustafa Ahmed Hassan El.attar

Ahmed Walid Mohamed Shahin

Ali farag Abd El.naser Ahmed El.kilany

Ghada Mohamed Ahmed Emara

Mohamed Zakaria Faried Mohamed kassim

Nehal Ashraf Saeed Ibraheem Mowafy

Noura Ayman Salah Abd El.maasoud

Under supervision of

Dr. Yahia Al.driny

 Acknowledgments

Special thanks are extended to

Faculty of Engineering, Alexandria University, Egypt ─ Dean.Dr. Saied Allam, Dr. Osama
El.samni, Dr. Mahmoud Badawy, Dr. Abdulsalam Olleak, Dr. Hussein Metwally, Dr. Ibrahim
El.faham.

SRTA City at Bourg El.arab ─ Dean Dr. Wlaa Shitaa, Dr. Mohamed el-Sayed, Dr. Gasser el-
Hosenny, Eng. Omar Hassan, Eng. Alaa Salah.

Egyptalum - Aluminum company ─ chairman Eng. Abd el.Zaher Abd El-satar

2
 Abstract

With the recent surge in fossil fuels prices, demands for cleaner energy sources, and government
funding incentives, wind turbines have become a viable technology for power generation. Currently,
horizontal axis wind turbines (HAWT) dominate the wind energy market due to their large size and high
power generation characteristics. However, vertical axis wind turbines (VAWT) are capable of producing
a lot of power, and offer many advantages over (HAWT). The main objective of this project is to design
and build a self-starting vertical axis wind turbine to operate in low wind speed condition. Parametric
studies were performed using momentum based models in QBlade v0.963 software and computational fluid
dynamics (CFD) in ANSYS fluent 17.2 software to get the optimum design that achieve the project
objectives. Stress analysis and vibration analysis were performed using ANSYS mechanical 17.2 and
ANSYS modal 17.2 respectively for checking both the proper material selection and safe operating. We
found from the research that the mass moment of inertia is the most dominant design parameter when you
design wind turbine to work in low wind speed conditions. Self-starting ability depending on the airfoil
type, turbine solidity, fixed pitch angle and the mass moment of inertia. Wood is a suitable material for
blade construction in small-scale wind turbines.

3
 Contents

Page

Chapter 1: Introduction................................................................................................................................. 12

1.1 Introduction................................................................................................................................... 12
1.2 Wind power............................................................................................................................. ...... 12
1.2.1 Power coefficient................................................................................................................ . 13
1.3 Wind Turbines............................................................................................................................... 14
1.3.1 Horizontal Axis and Vertical Axis Wind Turbines ........................................................... ... 14
1.3.2 Vertical axis wind turbines – VAWTs.............................................................................. ... 15
1.3.2.1 VAWT types......................................................................................................... . 16
1.3.2.1.1 Savonius turbines................................................................................. 16
1.3.2.1.2 Darrieus turbines.................................................................................. 16
1.3.2.2 Straight-blade VAWTs........................................................................................ .. 17
1.4 Project objectives.......................................................................................................................... . 17

Chapter 2: Litarature review.......................................................................................................................... 18

2.1 History of VAWTs........................................................................................................................ . 18

2.2 Development of Darrieus type of VAWTs .................................................................................... . 19
2.3 Analysis of VAWT performance.................................................................................................. . 20
2.4 Design analysis of VAWT............................................................................................................ . 21

Chapter 3: Methodology................................................................................................................................ 23

3.1 Introduction................................................................................................................................... 23
3.2 VAWT design parameters ............................................................................................................. 23
3.2.1 Turbine swept area............................................................................................................ ... 24
3.2.2 Tip speed ratio..................................................................................................................... 24
3.2.3 Number of blades.............................................................................................................. .. . 25
3.2.4 Turbine aspect ratio............................................................................................................. 25
3.2.5 Turbine solidity................................................................................................................... 25
3.2.6 Airfoil type.......................................................................................................................... 25
3.2.7 Pitch angle........................................................................................................................... 25
3.2.8 Mass moment of inertia.................................................................................................... . ... 26
3.3 Aerodynamics of VAWT.............................................................................................................. . 26
3.3.1 The actuator disk theory and Betz limit............................................................................. ... 26
3.3.2 Aerodynamics of Straight Blade Darrieus Type VAWT................................................... ... 30

4
 3.4 VAWT Performance analysis methodology.................................................................................. 32
3.4.1 Momentum based models ................................................................................................. ... 33
3.4.1.1 Single Stream-tube Model..................................................................................... 33
3.4.1.2 Multiple Stream-tube Model................................................................................ .. 33
3.4.1.3 Double Actuator Disk Theory.............................................................................. .. 34
3.4.1.4 Double Multiple Stream-tube Model (DMST)..................................................... .. 35
3.4.1.4.1 DMST model validation using QBlade software.................................. .. 37
3.4.3 Viscous models – Computational Fluid Dynamics (CFD)................................................ ... 38
3.4.3.1 CFD turbulence model validation........................................................................ .. 39
3.4.3.1.1 Experimental rotors features .................................................................. 39
3.4.3.1.2 Computational domain generation and optimization............................ .. 39
3.4.3.1.3 Transition Turbulence model optimization............................................ 40
3.4.3.1.4 Grid Generation....................................................................................... 41
3.4.3.1.5 Boundary Conditions ............................................................................... 42
3.4.3.1.6 Solver settings and calibration................................................................ . 43
3.4.3.1.7 Coefficient of Power Calculations .......................................................... . 44
3.4.3.1.8 Results..................................................................................................... 44
3.4.3.2 Flow driven rotor simulation................................................................................... 45
3.4.3.2.1 6-DOF solver Coupled with CFD............................................................ . 45
3.4.3.2.2 6-DOF method validation........................................................................ 46
3.4.3.2.2.1 6-DOF parameters................................................................ 46
3.4.3.2.2.2 Time step calculations.......................................................... 46
3.4.3.2.2.3 Coefficient of power calculations ........................................ . 46
3.4.3.2.2.4 Validation results and discussion......................................... . 47
3.5 VAWT Design analysis methodology.......................................................................................... . 48
3.5.1 Forces acting on VAWT blades and rotor........................................................................... . 48
3.5.2 Wind turbine construction................................................................................................... 49
3.5.2.1 Blade construction................................................................................................. 49
3.5.2.2 Turbine Arms........................................................................................................ . 50
3.5.2.3 Turbine hub........................................................................................................... . 50
3.5.2.4 Driven shaft........................................................................................................... 50
3.5.2.5 Bearings and bearing housing............................................................................... . 50
3.5.2.6 Tower.................................................................................................................... . 51
3.5.3 Material selection................................................................................................................ 52
3.5.4 Stress analysis...................................................................................................... ............... . 52
3.5.4.1 Geometry............................................................................................................... 52
3.5.4.2 Material insertion.................................................................................................. . 53
3.5.4.3 Grid generation...................................................................................................... 53
3.5.4.4 Loads modeling..................................................................................................... 53
3.5.4.5 Specify the supports.............................................................................................. . 53
3.5.4.6 Post processing...................................................................................................... 53
3.5.5 Vibration analysis............................................................................................................... . 54
3.5.6 Cost analysis....................................................................................................................... . 54

Chapter 4: Results & Discussion.............................................................................. ................................... ... 55

4.1 Introduction................................................................................................................................... 55
4.2 VAWT Parametric study............................................................................................................... 55
4.2.1 Turbine swept area.............................................................................................................. . 55
4.2.2 The aspect ratio effect on performance using DMST QBlade............................................. . 56
4.2.3 Solidity effect on performance using DMST QBlade......................................................... . 57
4.2.4 Airfoil selection using DMST QBlade................................................................................ . 58
4.2.4.1 Optimum (AoA) for every airfoil.......................................................................... . 58
4.2.4.2 Airfoil type effect on turbine performance............................................................ . 59
4.2.5 Optimum tip speed ratio using CFD Mesh motion method ................................................. . 60
4.2.6 Pitch angle Effect on performance using CFD 6-DOF method........................................... . 61

5
 4.2.7 Turbine characteristics data sheet....................................................................................... . 63
4.3 Design case using CFD mesh motion method............................................................................... . 64
4.4 Material selection.......................................................................................................................... 65
4.5 FEA stress and vibration analysis results ...................................................................................... . 67
4.6 Cost analysis.................................................................................................................... 69

Chapter 5: Conclusion & Recommendations.............................................................................................. ... 70

5.1 Conclusions................................................................................................................................... 70
5.2 Recommendations......................................................................................................................... 71

Appendix A: Hand calculations..................................................................................................................... 72

Bibliography................................................................................................................. ............................ 78

6
 List of figures

Page

Figure 1.1 Types of vertical axis wind turbines................................................................................. 14

Figure 1.2 Darrieus turbines different configurations....................................................................... 16
Figure 1.3 Egypt average wind speed Atlas on Dec 2005.................................................................. 17
Figure 2.1 an example of VAWTs in the Sistan Basin in the border region of Iran
and Afghanistan................................................................................................................................ 19
Figure 2.2 Possible variants of Darrieus VAWT: H-type (a), V-type (b), Troposkien type (c), and
Gorlov (helical type, d)..................................................................................................................... 20
Figure 3.1 H- darriues VAWT schematic.......................................................................................... 24
Figure 3.2 H- Pitch angle of the turbine blade................................................................................... 26
Figure 3.3 Air flow through wind turbine......................................................................................... 27
Figure 3.4 Velocity and pressure distribution along.......................................................................... 28
Figure 3.5 VAWT flow velocities and blade stream......................................................................... 30
Figure 3.6 Overview of the development of the stream tube............................................................. 33
Figure 3.7 2D Schematic of the stream-tube Model.......................................................................... 34
Figure 3.8 Schematic of the two actuator disk behind each............................................................... 35
Figure 3.9 Schematic of the double multiple stream-tube................................................................. 35
Figure 3.10 Iterative Procedure used to calculate the flow velocity in DMST................................... 36
Figure 3.11 QBlade DMST model power coefficient comparison measured and simulated data...... 37
Figure 3.12 Computational domain and Boundary Conditions......................................................... 40
Figure 3.13 Mesh around the rotor and blade for transition SST model simulation........................... 41
Figure 3.14 Mesh around the rotor and blade for K-epsilon realizable model simulation................. 42
Figure 3.15 Fluid domain Boundary conditions................................................................................ 43
Figure 3.16 CFD numerical and experimental data........................................................................... 44
Figure 3.17 CFD 6DOF method numerical results and experimental data........................................ 47
Figure 3.18 Aerodynamics forces acting on rotor blade.................................................................... 48
Figure 3.19 Blade main components................................................................................................. 49
Figure 3.20 VAWT arms.................................................................................................................. 50
Figure 3.21 VAWT hub.................................................................................................................... 50
Figure 3.22 VAWT Shaft with bearings and housing........................................................................ 50
Figure 3.23 VAWT construction on SolidWorks 2016..................................................................... 51
Figure 3.24 VAWT FEA geometry groups....................................................................................... 52
Figure 4.1 Aspect ratio effect on turbine performance...................................................................... 56

7
Figure 4.2 Solidity effect on turbine performance............................................................................. 57
Figure 4.3 airfoil type effect on lift, drag and moment coefficients................................................... 58
Figure 4.4 Airfoil type effect on performance................................................................................... 59
Figure 4.5 our turbine performance curve......................................................................................... 60
Figure 4.6 different blade pitch angle performance curves............................................................... 62
Figure 4.7 VAWT starting time from rest......................................................................................... 62
Figure 4.8 Blade loading aerodynamic forces................................................................................... 64
Figure 4.9 Turbine rotor loading aerodynamic torque....................................................................... 64
Figure 4.10 FEA stress analysis using ANSYS mechanical.............................................................. 67
Figure 4.11 Total deformation in case of resonance.......................................................................... 68

8
 List of tables

Page

Table 1.1 Typical Cp values for various wind turbines .......................................................................... . 13

Table 3.1 Momentum model and viscous model advantages & disadvantages. ................................. . 32
Table 3.2 URANS Turbulence models comparison............................................................................... . 45
Table 4.1 aspect ratio simulation input values ........................................................................................ . 56
Table 4.2 Solidity and chord length.......................................................................................................... . 57
Table 4.3 pitch angle effect on performance ........................................................................................... . 61
Table 4.4 Turbine main design data .......................................................................................................... . 63
Table 4.5 Martials & properties ................................................................................................................ . 65
Table 4.6 Vibration analysis results .......................................................................................................... . 68
Table 4.7 project cost analysis................................................................................................................... . 69

9
 Nomenclatures

K.E Kinetic Energy

m Air mass
u∞ Free Stream wind velocity
P Theoretical power of the turbine
𝑚̇ Air mass flow rate
ρ Air density
A Projected frontal area of turbine
CP Coefficient of power
Pmech,out Turbine Mechanical output power
I The wind turbulence intensity
σu Air standard deviation
R Turbine rotor radius
H Height of turbine \ Blade length
λ Turbine tip speed ratio
ω turbine rotational speed
σ Turbine solidity
c chord length
uR Wind velocity at the rotor
T Thrust force
uw Wake velocity
a induction factor
B Number of blades
CD Blade drag coefficient
CL Blade lift coefficient
Cn Normal force coefficient
Ct Tangential force coefficient
CT Thrust coefficient
CQ Torque coefficient
FN Normal force
FT Tangential force
Nθ Number of stream tubes

10
pd Downwind pressure of the rotor
pu Upwind pressure of the rotor
po Pressure of undisturbed air
P’ pressure drop across rotor plane
Qi Instantaneous torque
Qavg Average torque
Tavg Average thrust force
Ti Instantaneous thrust force
ue Equilibrium value of wind velocity
ui Induced wind velocity
w Relative wind speed

11
 Chapter

Introduction
1
1.1 Introduction

The rising concerns over global warming, environmental pollution, and energy security have increased
interest in developing renewable and environmentally friendly energy sources such as wind, solar,
hydropower, geothermal, hydrogen, and biomass as the replacements for fossil fuels. Wind energy can
provide suitable solutions to the global climate change and energy crisis. The utilization of wind power
essentially eliminates emissions of CO 2 , SO2 , NOx and other harmful wastes as in traditional coal-fuel
power plants or radioactive wastes in nuclear power plants. By further diversifying the energy supply, wind
energy dramatically reduces the dependence on fossil fuels that are subject to price and supply instability,
thus strengthening global energy security. During the recent three decades, tremendous growth in wind
power has been seen all over the world. In 2009, the global annual installed wind generation capacity
reached a record-breaking 37 GW, bringing the world total wind capacity to 158 GW. As the most
promising renewable, clean, and reliable energy source, wind power is highly expected to take a much
higher portion in power generation in the coming decades (Tong, 2010).

Wind energy has a number of benefits and advantages. Wind power is a clean and environmentally
friendly energy source. As an inexhaustible and free energy source, it is available and plentiful in most
regions of the earth. In addition, more extensive use of wind power would help reduce the demands for
fossil fuels, which may run out sometime in this century, according to their present consumptions.
Furthermore, the cost per kWh of wind power is much lower than that of solar power (El.Ali, 2007).

1.2 Wind power

Kinetic energy exists whenever an object of a given mass is in motion with a translational or rotational
speed. When air is in motion, the kinetic energy in moving air can be determined as
1
𝐾. 𝐸 = 𝑚𝑢∞ 2 (1.1)
2

12
 Where m is the air mass and 𝑢∞ is the mean wind speed over a suitable time period. The wind power
can be obtained by differentiating the kinetic energy in wind with respect to time, i.e.:
𝑑𝐾.𝐸 1
𝑃= = 𝑚̇𝑢∞ 2 (1.2)
𝑑𝑡 2

However, only a small portion of wind power can be converted into electrical power. When wind passes
through a wind turbine and drives blades to rotate, the corresponding wind mass flow rate is

𝑚̇ = 𝜌𝐴𝑢∞ (1.3)

Where ρ is the air density and A is the swept area of blades. Substituting (1.3) into (1.2), the available
power in wind P can be expressed as
1
𝑃 = 𝜌𝐴𝑢∞ 3 (1.4)
2

An examination of equation (1.4) reveals that in order to obtain a higher wind power, it requires a higher
wind speed, a larger swept area, and a higher air density. Because the wind power output is proportional to
the cubic of the mean wind speed, a small variation in wind speed can result in a large change in wind
power (Tong, 2010).

1.2.1 Power coefficient

The conversion of wind energy to electrical energy involves primarily two stages: in the first stage,
kinetic energy in wind is converted into mechanical energy to drive the shaft of a wind generator. The
critical converting devices in this stage are wind blades. For maximizing the capture of wind energy, wind
blades need to be carefully designed (Tong, 2010).

The power coefficient Cp deals with the converting efficiency in the first stage, defined as the ratio of
the actually captured mechanical power by blades to the available power in wind:
𝑃𝑚𝑒𝑐ℎ,𝑜𝑢𝑡
𝐶𝑝 = (1.5)
𝑃

Because there are various aerodynamic losses in wind turbine systems, for instance, blade-tip, blade-
root, profile, and wake rotation losses, etc., the real power coefficient Cp is much lower than its theoretical
limit, usually ranging from 30 to 45%.

Table 1.1 Typical Cp values for various wind turbines (Chang, 2005).

13
 There is a theoretical limit in the efficiency of a wind turbine determined by the deceleration the wind
suffers when going across the turbine. For HAWT, the limit is 19/27 (59.3%) and is called Lanchester-Betz
limit (Tong, 2010).

For VAWT, the limit is 16/25 (64%) (Paraschivoiu, 2002). These limits come from the actuator disk
momentum theory which assumes steady, inviscid and without swirl flow.

1.3 Wind Turbines

A modern wind turbine is an energy-converting machine to convert the kinetic energy of wind into
mechanical energy and in turn into electrical energy. Various wind turbine concepts have been developed
and built for maximizing the wind energy output, minimizing the turbine cost, and increasing the turbine
efficiency and reliability. Wind turbines can be classified according to the turbine generator configuration,
airflow path relatively to the turbine rotor, turbine capacity, the generator-driving pattern, the power supply
mode, and the location of turbine installation (Tong, 2010).

1.3.1 Horizontal Axis and Vertical Axis Wind Turbines

The HAWT is the most common turbine configuration. The propellers and turbine mechanisms are
mounted high above the ground on a huge pedestal. It is a matter of taste as to whether they enhance the
landscape. However, there is no denying that the height at which their mechanisms are located is a
disadvantage when servicing is required. Also, they require a mechanical yaw system to orient them such
that their horizontal axis is perpendicular to and facing the wind. As potential power generation is related
to the swept area (diameter) of the rotor, more power requires a larger diameter. The blades experience
large thrust and torque forces, so size is limited by blade strength (MECH 4010 Design Project Vertical
Axis Wind Turbine, 2005).

Several typical vertical-axis wind turbines are shown in Figure 1.1. The blades of the vertical-axis wind
turbines rotate with respect to their vertical axes that are perpendicular to the ground. A significant
advantage of vertical-axis wind turbine is that the turbine can accept wind from any direction and thus no
yaw control is needed. Since the wind generator, gearbox, and other main turbine components can be set
up on the ground, it greatly simplifies the wind tower design and construction, and consequently reduces
the turbine cost. However, the vertical-axis wind turbines must use an external energy source to rotate the
blades during initialization. Because the axis of the wind turbine is supported only on one end at the ground,
its maximum practical height is thus limited. Due to the lower wind power efficiency, vertical-axis wind
turbines today make up only a small percentage of wind turbines (Tong, 2010).

Figure 1.1 Types of vertical axis wind turbines

(Schaffarczyk, 2014)

14
1.3.2 Vertical axis wind turbines - VAWTs

Vertical-axis wind turbines (VAWTs) come in a wide and interesting variety of physical configurations
and they involve a range of complex aerodynamic characteristics. Not only were VAWTs the first wind
turbines to be developed but they have also been built and operated at a scale matching some of the biggest
wind turbines ever made. VAWTs in principle can attain coefficients of performance, C p max, that are
comparable to those for horizontal-axis wind turbines (HAWTs) and they have several potentially
significant advantages over the HAWTs.

These advantages include the fact that VAWTs are cross-flow devices and therefore accept wind from
any direction. Thus, in principle, they do not need a yaw mechanism to ensure that they are aligned to the
wind as is the case with all horizontal axis machines. Another key advantage is that the mechanical load
may be connected directly to the VAWT rotor shaft and located at ground level. This removes the need for
a substantial tower to support the weight of equipment such as the gearbox, generator and yaw mechanism.
There is also no need for slip rings or flexible cables to connect the generator to the load, which can be an
important point for small-scale turbines.

From the 1970s to the 1990s a number of research groups and companies developed and built hundreds
of VAWTs and a great deal has been learnt from that experience. But despite the inherent advantages of
VAWTs they have fallen significantly behind HAWTs in recent years in terms of technical development
and in the size and number of units manufactured. This has occurred for a number of reasons, not least
because of some inherent disadvantages of VAWTs.

As the VAWT blades rotate about the main rotor shaft the velocity of the air relative to the blade is
constantly changing in respect of both magnitude and direction. In addition, each blade will interact with
the wakes of other blades, and possibly its own wake, when it passes through the downstream half of its
path about the turbine axis. Both these effects result in fluctuating aerodynamic forces on the blades, which
in turn lead to a potentially significant fatigue issue for the design of the blades and overall turbine structure.
The fluctuating blade loads also lead to a varying torque transferred to the mechanical load.

Many designs of VAWTs produce very low torque when they are stationary and may produce negative
torque at low tip speed ratios, so they must be powered up to a speed at which the aerodynamic torque is
sufficient to accelerate the rotor to normal operational speeds. A further disadvantage is that parasitic drag
losses may be high for a given VAWT design. This situation can arise when the VAWT blades need to be
mounted on structures (spars, beams, cables, etc.) that rotate with the blades or are located upstream of the
blades. The drag forces on these passive components can lead to significant parasitic losses in respect to
rotor torque and power output. This has inhibited the successful development of a number of VAWT
designs (Tong, 2010).

Nevertheless there continues to be widespread interest in VAWTs as a means of generating electrical

and mechanical energy from the wind. Novel VAWT turbine designs appear relatively frequently at the
time of writing and a number of small companies appear to be undertaking development of VAWTs for
small-scale application, particularly in respect to domestic dwellings.

15
1.3.2.1 VAWT types

A wide variety of VAWTs have been proposed over the past few decades and a number of excellent
bibliographies on VAWTs have been published that summarize research and development of these devices,
including the survey by Abramovich (Abramovic, 1987). Some of the more important types of rotor design
are highlighted in the following sections.

1.3.2.1.1 Savonius turbines

The Savonius turbine has been popular with both professional and amateur wind turbine developers
over the years, not least because of its simple and robust construction.

Many variations of the Savonius rotor have been developed and tested. However, because of the
inherently high solidity and hence high mass of the Savonius turbine it has not been used for large-scale
electricity production.

The Savonius rotor is primarily a drag device with some inherent augmentation of the rotor performance
available due to the air flow across each vane and mutual coupling of the two halves of the rotor. Like all
drag machines it has a low operating tip speed ratio. This makes it less suitable for electricity generation
than devices with higher tip speeds, since a high shaft speed is generally preferred to minimize the step-up
ratio requirement of the gearbox coupling a rotor to a conventional electrical generator.

1.3.2.1.2 Darrieus turbines

In 1931 the invention by Darrieus (Darrieus, 1931)of his rotor with a high tip speed ratio opened up
new opportunities for VAWTs in regards to electricity generation. The fundamental step forward made by
Darrieus was to provide a means of raising the velocity of the VAWT blades significantly above the free-
stream wind velocity so that lift forces could be used to significantly improve the coefficient of performance
of VAWTs over previous designs based primarily on drag. Darrieus also foresaw a number of embodiments
of his fundamental idea that would be trialed at large scale many decades later. These included use of both
curved-blade shown in figure 1.2 and straight blade versions of his rotor. He also proposed options for
active control of the pitch of the blades relative to the rotor as a whole, so as to optimize the angle of attack
of the wind on each blade throughout its travel around the rotor circumference .

Figure 1.2 Darrieus turbines different configurations

16
1.3.2.2 Straight-blade VAWTs

The name Darrieus is usually associated with the curved-blade version of Darrieus’ patent. However, a
great deal of work over the past three decades has gone into the development and analysis of the straight-
blade version of his original invention, which is sometimes known as the H-VAWT from the shape of the
blades and supporting spars.

1.4 Project objectives

The objective of this project is design and manufacturing H-Darrius Vertical axis wind turbine to
produce clean energy suitable for low average wind velocities regions. The wind velocities in Alexandria,
Egypt where we will install our turbine have an average wind speed of five meter per second as shown in
figure 1.3.

Figure 1.3 Egypt average wind speed Atlas

on Dec 2005 (Wind Atlas, 2005)

Another objective of this project is that the turbine having self-starting ability which is a major problem
for H-Darrius vertical axis wind turbine.

Our target is reaching the most highly possible performance of a vertical axis wind turbine, we made
that by studying the different variables that have effects on the performance. We followed different methods
to study this variables to make our turbine more efficient and more reliable, after that we moved to another
step where we studied the stress and vibrations to make our design safer to operate at extreme conditions
of high wind velocities. In our manufacturing process we took in our consideration the overall cost of the
turbine to be suitable for developing remote regions, as well as the quality and availability of the material
in our local market in Alexandria, Egypt.

17
 Chapter

Literature review
2
2.1 History of VAWTs

VAWTs appear to have been developed long before their horizontal axis cousins. One of the reasons
for this is that the VAWT has a number of inherent advantages including the fact that a drive shaft may be
connected directly from the rotor to a mechanical load at ground level, eliminating the need for a gearbox.
The early pioneers involved in the development of wind turbines many centuries ago applied VAWTs to
the milling of grain, an application where the vertical axis of the millstone could be easily connected to the
VAWT rotor. Quite a number of excellent review articles have been published in the past detailing the
historical development of wind turbines of all types (Golding, 1976) (Shepherd, 1994). Virtually all of these
reviews suggest that the very earliest wind turbines were indeed VAWTs and it is thought that these were
first used in Persia for milling grain more than 2000 years ago. These early wind turbines were essentially
drag devices with a rotor comprising a number of bundles of reeds, or other simple blades, on a timber
framework. The rotor was housed within a walled enclosure that channeled the flow of wind preferentially
to one side of the rotor thereby generating the torque necessary to rotate the millstone. This type of device
was still in use during the latter half of the 20th century and an example located in the border region of
Afghanistan and Iran see figure 2.1 (Muller, 2009).

The Persian and Sistan VAWTs had rigid vanes to generate torque whereas other designs have used
sails that can effectively pitch with respect to their alignment on the rotor and thus can potentially increase
efficiency. An example of a Chinese VAWT of the type used for many years for pumping applications, and
which was described by King (King, 2004) for pumping brine for salt production.

18
 Figure 2.1 an example of VAWTs in the Sistan Basin in the border region of Iran
and Afghanistan. Note in the right hand image how the upstream wall is
used to expose only one half of the rotor to the.

2.2 Development of Darrieus type of VAWTs

The Darrieus wind turbine was patented by (Darrieus, 1931) with Troposkien, H and V-shaped rotors,
but other blade geometries have since been developed (modified Troposkien, Sandia type, etc.), with the
aim of optimizing both aerodynamic and structural performances. For this reason, the curved blade shape
was mainly used in large turbines. An important aerodynamic disadvantage of this architecture is the
decreasing radius close to the top and the bottom of the rotor, where blades experience low peripheral
speeds and therefore generate very little torque Figure 2.2 c. In order to overcome this limit, a different
concept was introduced: in 1986 the first H-Darrieus, or Musgrove VAWT Figure 2.2 a, was prototyped.
In this case, the blades are straight and therefore the radius is the same over the whole length of the blade.
On the other hand, in contrast to the Troposkien blade, extra strength is necessary to cope with the
centrifugal force causing bending moment, blade support struts have to be considered and should be
designed in order to limit the aerodynamic drag loss.

19
 Figure 2.2 Possible variants of Darrieus VAWT: H-type (a), V-type (b), Troposkien type (c), and Gorlov
(helical type, d) (Battisti, 2012).

The Darrieus turbine is usually characterized by a periodic variation of the aerodynamic forces during
every rotation cycle. A straightforward method to get rid of the resulting pulsating loads is to incline the
blade, as proposed by A. Gorlov (Gorlov, 1998), which lead to a helical shaped Darrieus turbine design
Figure 2.1 d, patented in the mid-1990s for hydro applications and adapted for wind energy exploitation
since the early to mid-2000s.

One of the characteristics of the Darrieus family of turbines is that they have a limited self-starting
capacity because there is often insufficient torque to overcome friction at start-up. This is largely because
lift forces on the blades are small at low rotational speeds and for two-bladed machines in particular the
torque generated is virtually the same for each of the stationary blades at start-up, irrespective of the rotor
azimuth angle relative to the incident wind direction. Moreover, the blades of a Darrieus rotor are stalled
for most azimuth angles at low tip speed ratios. As a result large commercial machines generally need to
be run up to a sufficiently high tip speed for the rotor to accelerate in a given wind velocity (Tong, 2010)

Self-starting capability may be enhanced through a number of strategies including: increasing solidity;
using an odd number of blades; providing a form of blade pitch mechanism; and using blades that are
skewed so that the blade azimuth angle is a function of axial distance along the rotor. A recent study of the
self-starting characteristics of small Darrieus machines has been reported by Hill (Hill, 2009).

2.3 Analysis of VAWT performance

As in the case of HAWTs, there are a number of levels of complexity with which one might analyze
the performance of the VAWTs as outlined by authors including (Touryan, 1987), (Strickland, A review of
aerodynamics analysis methods for vertical axis wind turbines, 1986) and Wilson (Wilson, 1994) .Allet et
al. (Allet, 1997) classified the four main approaches to modelling of VAWTs as: (i) momentum models;
(ii) vortex models; (iii) local circulation models; and (iv) viscous models, where the latter would include
full viscous flow computational fluid dynamics (CFD) models. Other bibliographic sources of information
on analysis methods include those of Abramovich (Abramovic, 1987) and Islam et al. (Islam, 2008). An
extremely useful and relatively simple momentum model uses the actuator disc/blade element momentum
theory (or strip theory) in a similar manner to the momentum model used for analysis of HAWTs. This type
of analysis can be implemented relatively easily by anyone with a basic knowledge of fluid mechanics.

20
 Computational fluid dynamics - CFD – turbulence models were developed and modified to predict the
turbine performance and simulate these complex aerodynamics found in the turbine air flow. Authors
(Lanzafame, 2014) (Castelli M. , 2011) they were developing turbulence models and chose a set of solver
sittings to model the VAWT using sliding mesh method in CFD by ANSYS Fluent software then compare
the numerical results to actual experimental results and both were in a good match . They recommended in
their papers (i) transition SST (ii) SST K-omega (iii) K- epsilon turbulence models for modeling the air
flow around the turbine.

Furth more some researchers developed what is called flow driven rotor simulation in CFD described
by (Wang, 2011). In his paper he explain 6DOF method to model the movement of the turbine and make
the turbine rotational speed depending on the wind speed. This method was very useful in predicting the
overall output power and the rotational speed that the turbine will work on.

The development in design of Darriues turbine continued and many researchers studied the different
variables and parameters that affect the performance of the turbine using numerical and analytical ways to
get the optimum design. Aerodynamics investigation and performance analysis were performed by
(Vaishnav, 2007) using CFD models and he found that the tip speed ratio play a major rule in turbine
performance and the range of operating range of tip speed ratio is depending on airfoil type. The CFD
modeling 3D results and 2D results of VAWT were compared by (Alaimo, 2015) in his article shows that
the generation of the tip vortices in 3D simulation greatly reduces the aerodynamic performance.

The aspect ratio and solidity parameters were studied by (Brusca, 2014) in his paper By analyzing the
factors which influence the Reynolds number, it was found that the ratio between blade height and rotor
radius (aspect ratio) influences the Reynolds number and as a consequence the power coefficient. It has
been highlighted that a turbine with a lower aspect ratio has several advantages over one with a higher
value. The advantages of a turbine with a lower aspect ratio are: higher power coefficients, a structural
advantage by having a thicker blade (less height and greater chord), and greater in-service stability from
the greater inertia moment of the turbine rotor. Brusca also recommended the optimum range of solidity
that guarantee self-starting ability of the turbine.

Despite the recommendations of (Brusca, 2014) the grater the radius and thicker the chord that will
increase the mass moment of inertia which it was not recommended for self-starting by (Yang, 2013) Yang
conclude that the lesser moment of inertia’s VAWT own biggish torque coefficient ,power coefficient
before get the max power coefficient, so it has trend of easy start.

2.4 Design analysis of VAWT

Several publications were published about this topic. (Castelli, 2013) In his paper he studied the blade
loading numerically using CFD and describe the main three forces (i) Normal force, (ii) tangential force
and (iii) Centrifugal force, then conclude that the dominant force was the centrifugal force coming from
rotation and not the aerodynamic force coming from air flow. (Brown, 2010) Where he modeled and
simulate the blade loading using finite element analysis and compared to experimental results. In his paper
he was recommending using FEA due to the complexity of the turbine design, also done study on blade
vibration to ensure there is no resonance and his turbine is safe to operate. Other researchers go to study the
materials which the turbine should be made from (Islam M. , 2008) he made comparisons between the
available materials based on their mechanical properties and costs.

21
 The work published was great and very useful for us as an establishment we build our project on. Our
target was to use this information we collected to achieve our objectives. Optimum designing of H- Darrieus
VAWT which operate in low wind speeds and has the ability to self-starting, and manufacturing this turbine
from available material having good mechanical properties and low cost.

22
 Chapter

Methodology
3
3.1 Introduction

This chapter includes the different methods we followed in designing our turbine to get the optimum
design which achieved the project objectives.

There were two scopes we followed to achieving our objectives:

1. Performance analysis: where we study the different parameters that effect the performance.
Parametric studies were performed using different methods, were we get the overall dimensions
and different characteristics of the turbine.
2. Design analysis: where we select the materials which the turbine is made of. Stress analysis
and vibration analysis were performed to ensure safe operation of the turbine under extreme
conditions.

3.2 VAWT design parameters

The wind turbine parameters considered in the design process are:

1. Turbine swept area

2. Tip speed ratio
3. Number of blades
4. Turbine Aspect ratio
5. Turbine solidity
6. Airfoil type
7. Pitch angle
8. Mass moment of inertia

23
3.2.1 Turbine swept area

The swept area is the section of air that encloses the turbine in its movement, the shape of the swept
area depends on the rotor configuration, this way the swept area of an HAWT is circular shaped while for
a straight-bladed vertical axis wind turbine the swept area has a rectangular shape and is calculated using:

𝐴 = 2𝑅𝐻 (3.1)
R

Figure 3.1 H- darriues VAWT schematic

Where A is the swept area [m2 ], R is the rotor radius [m], and H is the blade length [m]. The swept area
limits the volume of air passing by the turbine. The rotor converts the energy contained in the wind in
rotational movement so as bigger the area, bigger power output in the same wind conditions.

3.2.2 Tip speed ratio

The tip speed ratio is an extremely important factor in wind turbine design, which is defined as the ratio
of the tangential speed at the blade tip to the actual wind speed, i.e.:
𝜔𝑅
𝑇𝑆𝑅 = (3.2)
𝑢∞

Where R is the radius of the hub, and ω is the angular speed of blades.

If the blade angular speed ω is too small, most of the wind may pass undisturbed though the blade
swept area making little useful work on the blades. On the contrary, if ω is too large, the fast rotating blades
may block the wind flow reducing the power extraction. Each rotor design has an optimal tip speed ratio at
which the maximum power extraction is achieved (Castillo, 2011)

24
3.2.3 Number of blades

The number of blades has a direct effect in the smoothness of rotor operation as they can compensate
cycled aerodynamic loads. For even number of blades as two blades the fluctuation in rotor torque was
higher than odd number of blades as three blades turbine. Such behavior could represent a great advantage
for three-bladed small VAWT architectures, for which the increased manufacturer and installation costs are
not as important as for large rotors (Battisti L. , 2016).

3.2.4 Turbine aspect ratio

The ratio between blade height to rotor radius (aspect ratio) is play an important role in design process
of VAWT, Basically high aspect ratio means that we get high rotational speed than rotor torque for the
same power and vice versa. Aspect ratio influences the Reynolds number and as a consequence the power
coefficient (Brusca, 2014).

3.2.5 Turbine solidity

The solidity σ is defined as the ratio between the total blade area and the projected turbine area (Tullis,
2008). It is an important non-dimensional parameter which affects self-starting capabilities and for straight
bladed VAWTs is calculated with:
𝐵∗𝑐
𝜎= (3.3)
𝑅

Where B is the number of blades, c is the blade chord; it is considered that each blade sweeps the area
twice. This formula is not applicable for HAWT as they have different shape of swept area.

Solidity determines when the assumptions of the momentum models are applicable, and only when
using high σ ≥ 0.4 a self-starting turbine is achieved (Tong, 2010).

3.2.6 Airfoil type

Airfoil proper selection for your VAWT is important in many ways. The airfoil should be selected based
on many factors:

1. Chambered or symmetrical
2. Lift/drag ratio
3. Aerodynamic performance to increase output power
4. Airfoil thickness

3.2.7 Pitch angle

It was clearly seen that proper airfoil selection make great contributions to the improvement of
aerodynamic performance, but it was difficult to give consideration to self-staring capability, higher power
coefficient and wider operational region. Fortunately, it was found out that pitch angle had significant
impacts on overall performance of wind turbine see figure 3.2 (Klimas, 1981) (Lazauskas, 1992).

25
 Figure 3.2 H- Pitch angle of the turbine blade

In-offset blade pitch angle enters the effective region with high power coefficient in advance, while the
out-offset blade pitch angle can delay to exist the effective region (Zhang, 2014). From that we investigate
the effect of different pitch angles to increase the coefficient of power.

3.2.8 Mass moment of inertia

The biggish moment of inertia is disadvantage of VAWT’s start, effecting reasonable design ( rate of
high and radius )and result of brake Therefore , it is necessary that study the influence of moment of inertia
of VAWT.

Angular acceleration = Aerodynamic moment / Mass moment of inertia

The VAWT of the lesser moment of inertia has good start performance, but the steady of performance
is bad with wind velocity increased; it’s because that the process is unsteady when wind turbine is rotated,
that is to say, the angular velocity is different at the different azimuth angle, the angular velocity has obvious
wave with azimuth angle’s difference for the lesser moment of inertia’s wind turbine (Yang, 2013).

3.3 Aerodynamics of VAWT

3.3.1 The actuator disk theory and Betz limit

A simple model, generally attributed to Betz (1926) can be used to determine the power, the thrust of
the wind on the ideal rotor and the effect of the rotor operation on the local wind field.

The simplest aerodynamic model of a wind turbine is known as ‘actuator disk model’ in which the rotor
is assumed as a homogenous disk that extracts energy from the wind.

The theory of the ideal actuator disk is based on the following assumptions (Duncan, 1962):
 Homogenous, incompressible, steady state fluid flow, no frictional drag;
 The pressure increment or thrust per unit area is constant over the disk;
 The rotational component of the velocity in the slipstream is zero;
 There is continuity of velocity through the disk;
 An infinite number of blades.

26
The analysis of the actuator disk theory assumes a control volume as shown in Figure 3.3.

Figure 3.3 Air flow through wind turbine (M ulugeta, 2015).

In the above control volume, the only flow is across the ends of the stream tube. The turbine is
represented by a uniform actuator disk which creates a discontinuity of pressure in the stream tube of air
flowing through it. Note also that this analysis is not limited to any particular type of wind turbine.

From the assumption that the continuity of velocity through the disk exists, the velocities at section 2
and 3 are equal to the velocity at the rotor:

𝑢2 = 𝑢3 = 𝑢𝑅 (3.4)

For steady state flow, air mass flow rate through the disk can be written as:

𝑚̇ = 𝜌𝐴𝑢𝑅 (3.5)

Applying the conservation law of linear momentum to the control volume enclosing the whole system,
the net force can be found on the contents of the control volume. That force is equal and opposite to the
thrust, T which is the force of the wind on the wind turbine. Hence, from the conservation of linear
momentum for a one dimensional, incompressible, steady flow the thrust is equal and opposite to the change
in momentum of air stream:

𝑇 = −𝑚̇ (𝑢 ∞ − 𝑢 𝑤 ) (3.6)

Work is not done on either side of the wind turbine rotor. Thus the Bernoulli equation can be used in
the two control volumes on either side of the actuator disk. Between the free-stream and upwind side of the
rotor (from section 1 to 2 in Figure 3.3) and between the downwind side of the rotor and far wake (from
section 3 to 4 in Figure 3.3) respectively:
1 1
𝑝𝑜 + 2 𝜌𝑢 2∞ = 𝑝𝑢 + 2 𝜌𝑢 2𝑅 (3.7)

1 1
𝑝𝑑 + 2 𝜌𝑢 2𝑅 = 𝑝𝑜 + 2 𝜌𝑢 2𝑤 (3.8)

27
The thrust can also be expressed as the net sum forces on each side of the actuator disk:

𝑇 = 𝐴𝑝 ′ (3.9)

Where,

𝑝 ′ = (𝑝𝑢 − 𝑝𝑑 ) (3.10)

By using equations (3.7) and (3.8), the pressure decrease, 𝑝 ′can be found as:
1
𝑝 ′ = 2 𝜌(𝑢 2∞ − 𝑢 2𝑤 ) (3.11)

And by substituting equation (3.11) in to equation (3.9)

1
𝑇 = 2 𝜌𝐴(𝑢 2∞ − 𝑢 2𝑤 ) (3.12)

By equating the thrust values from Equation (3.6) in which substituting Equation (3.5) in place of 𝑚̇ and
Equation (3.12), the velocity at the rotor plane can be found as:
𝑢 ∞+𝑢 𝑤
𝑢𝑅 = (3.13)
2

If an axial induction factor, a is defined as the fractional decrease in the wind velocity between the free
stream and the rotor plane, then
𝑢 ∞−𝑢 𝑅
𝑎= (3.14)
𝑢∞

𝑢 𝑅 = 𝑢 ∞ (1 − 𝑎) (3.15)

Using equation (3.13) and (3.15) one can get:

𝑢 𝑤 = 𝑢 ∞ (1 − 2𝑎) (3.16)

The velocity and pressure distribution are illustrated in Figure 3.4. Because of continuity, the diameter of flow
field must increase as its velocity decreases and there occurs sudden pressure drop at rotor plane which contributes to
the torque of rotating turbine blades.

Figure 3.4 Velocity and pressure distribution along

the

stream tube

28
 The power output, P is equal to the thrust times the velocity at the rotor plane:

𝑃 = 𝑇𝑢 𝑅 (3.17)

Using equation (3.12)

1
𝑃 = 𝜌𝐴(𝑢2∞ − 𝑢2𝑤 )𝑢𝑅 (3.18)
2

And substituting for uR and uw from Equations (3.15) and (3.16) to Equations (3.18)

𝑃 = 2𝜌𝐴𝑎(1 − 𝑎)2 𝑢 3∞ (3.19)

The power performance parameters of a wind turbine can be expressed in dimensionless form, in which the
power coefficient, CP is given in the following equation:
𝑃
𝐶𝑝 = (3.20)
0.5𝜌𝐴𝑢 3∞

Using Equation (3.19) and (3.20), the power coefficient CP becomes:

𝐶𝑝 = 4𝑎(1 − 𝑎)2 (3.21)

The maximu m CP is determined by taking the derivative of Equation (3.21) with respect to (a) and setting it equal
to zero which yields:

16
(𝐶𝑝 )𝑚𝑎𝑥 = = 0.5926 (3.22)
27

When a = 1/3

This result indicates that if an ideal rotor were designed and operated so that the wind speeds at the
rotor were 2/3 of the free stream wind speed, then it would be operating at the point of maximum power
production. This is known as the Betz limit. From Equations (3.12) and (3.16) the axial thrust on the disk can be
written in the following form:

𝑇 = 2𝜌𝐴𝑎(1 − 𝑎)𝑢 2∞ (3.23)

Similar to the power coefficient, thrust coefficient can be defined by the ratio of thrust force to dynamic
pressure as shown in the following equation:
𝑇
𝐶𝑇 = 0.5𝜌𝐴𝑢 2 (3.24)
∞

Using Equation (3.23) and (3.24), the thrust coefficient CT becomes:

𝐶𝑇 = 4𝑎(1 − 𝑎) (3.25)

Note that CT has a maximum of 1.0 when a = 0.5 and the downstream velocity is zero. At maximum power
output (a =1/3), CT has a value of 8/9.

29
3.3.2 Aerodynamics of Straight Blade Darrieus Type VAWT

As the VAWT have a rotational axis perpendicular to the oncoming airflow, the aerodynamics involved are
more complicated than of the more conventional HAWT. The main disadvantages of VAWT are the high local
angles of attack involved and the wake coming from the blades in the upwind part and from the axis.

Compared to Savonius rotor, Darrieus rotor usually works at relatively high tip speed ratio which makes it
attractive for wind electric generators. However, they are not self-starting and require external ‘excitation’ to
cut-in (Sathyajith, 2006) (Claessens, 2006). If the straight blade Darrieus type VAWT is represented in a two
dimensional way Figure 3.5 the aerodynamic characteristics are more obvious.

Figure 3.5 VAWT flow velocities and blade stream

tube
As can be seen from the Figure 3.5 the relative velocity, W can be obtained from the cordial velocity
component and normal velocity component:

𝑤 = √(𝑢 𝑖 sin(𝜃) )2 + (𝑢 𝑖 cos (𝜃) + 𝜔𝑅 )2 (3.26)

Where ui is the induced velocity through the rotor,  is the rotational velocity, R is the radius of the turbine, and
 is the azimuth angle. The relative velocity can be written in non-dimensional form using free stream velocity:

𝑤 𝑢𝑖 𝑢𝑖 𝜔𝑅 2
= √( sin(𝜃))2 + ( cos(𝜃) + ) (3.27)
𝑢∞ 𝑢∞ 𝑢∞ 𝑢∞

From Equation (3.15) substituting the induced velocity ui in place of uR results:

𝑢 𝑖 = 𝑢 ∞ (1 − 𝑎) (3.28)

From equation (3.2) and using equation (3.28), equation (3.27)

𝑤
= √((1 − 𝑎)sin⁡(𝜃)2 + ((1 − 𝑎)𝑐𝑜𝑠𝜃 + 𝜆)2 (3.29)
𝑢∞

30
 From the geometry of Figure 3.5, the local angle of attack can be expressed as:
𝑢 𝑖sin⁡(𝜃)
tan(𝛼) = 𝑢 (3.30)
𝑖 cos(𝜃)+𝜔𝑅

Similarly to relative velocity the induced velocity can be put in non-dimensional form and using equation
(3.28) the above equation yields:
(1−𝑎)sin⁡(𝜃)
𝛼 = tan−1 (( ) (3.31)
1−𝑎) cos(𝜃) +𝜆

The normal and tangential force coefficient can be expressed as:

𝐶𝑛 = 𝐶𝐿 cos (𝛼) + 𝐶𝐷 sin⁡(𝛼) (3.32)

𝐶𝑡 = 𝐶𝐿 sin⁡(α) + 𝐶𝐷 cos⁡(𝛼) (3.33)

Where CL is the lift coefficient and CD is the drag coefficient for angle of attack  .

Then the normal and tangential forces for single blade at a single azimuthal location are:

1
𝐹𝑁 = 𝜌𝑤 2 (𝐻𝑐)𝐶𝑛 (3.34)
2

1
𝐹𝑇 = 𝜌𝑤 2 (𝐻𝑐)𝐶𝑡 (3.35)
2

Where ‘H’ is the blade height and ‘c’ is the blade chord length.

Using Figure 3.4, the instantaneous thrust force which is the force of the wind on the turbine experienced
by one blade element in the direction of the airflow, is:

𝑇𝑖 = 𝐹𝑇 cos⁡(𝜃) − 𝐹𝑁 sin⁡(𝜃) (3.36)

Using equations (3.32), (3.33), (3.34), (3.35) and (3.36) the instantaneous thrust force can be expressed as:
1
𝑇𝑖 = 𝜌𝑤 2 (𝐻𝑐)(𝐶𝑛 sin(𝜃) − 𝐶𝑡 cos⁡(𝜃) (3.37)
2

Since it is tangential force component that drives the rotation of the wind turbine and produces the torque
necessary to generate electricity the instantaneous torque or the torque by a single blade at a single azimuthal
location is:

𝑄𝑖 = 𝐹𝑇 𝑅 (3.38)

Substituting equation (3.35) in equation (3.38) yields:

1
𝑄𝑖 = 2 𝜌𝑤 2 (𝐻𝑐)𝐶𝑡 𝑅 (3.39)

31
3.4 VAWT Performance analysis methodology

In order to model the performance of a vertical-axis wind turbine there are four main approaches (Tong,
2010):

 Momentum models
 Vortex models
 Local circulation models
 Viscous models – Computational Fluid Dynamics (CFD)

Each model has its advantages and drawbacks, the main advantage of momentum models is that their
computer time needed is said to be much less than for any other approach.

For this project we used both momentum models and viscous models (CFD) to increase the reliability of
our design. Each method of the two have its advantages and its disadvantages see table 3.1

Table 3.1 M omentum model and viscous model advantages & disadvantages.

Momentum based models Viscous based models

 The main advantage of momentum
models is that their computer time
needed much less than for any other
approach.
 Useful tool in the design process as it
gives quicker and qualitative results
Advantages than other methods.
 Due to its flexibility, CFD has been gaining
popularity for analyzing the complex,
unsteady aerodynamics involved in the
study of wind turbines (Ferreira, 2007)
(Vassberg, 2005) and has demonstrated an
ability to generate results that compare
favorably with experimental data (Edwards,
2008) (Torresi, 2008).
 CFD has shown no problems predicting the
performance of either high- or low-solidity
wind turbines or for various tip speed ratios.

 Neglect the effect of dynamic stall.  Predicting the performance of a wind

 Neglect the effect of wake interaction. turbine using CFD typically requires large
 The 1D simplification introduces some computational domains with sliding
error by neglecting span-wise velocity interfaces and additional turbulence
and 3D effects on the blade modeling to capture unsteady affects;
(Vallverdu, 2014). therefore, CFD can be computationally
 The algorithm is not accurate for small expensive (Carrigan, 2012).
wind turbines at small tip speed ratios
Disadvantages (Claessens, 2006)
 Limited to solidities below 0.2
 Consider the flow as (homogeneous,
incompressible, steady state flow; no
frictional drag or non-rotating wake).

32
3.4.1 Momentum based models

Momentum based models are based on the actuator disk theory, which is generally used for rotor
aerodynamics, adjusted for the VAWT. The basic model is called the single stream-tube model. This model was
developed in two directions figure 3.6.

Figure 3.6 Overview of the development of the stream tube

models the

3.4.1.1 Single Stream-tube Model

stream tube

This model was first developed by Templin for the VAWT (Templin, 1974). It is based on the actuator
disk theories applicable for propellers as discussed above in section 3.3.1 and is the most basic model based on
the momentum theory. The flow through the turbine is assumed to have one constant velocity.

3.4.1.2 Multiple Stream-tube Model

Multiple stream-tube model is developed by Strickland (Strickland, The darrieus turbine: a performance
prediction model using Multiplestream tubes, 1975) and is also based on the momentum theory. The main
improvement with respect to the single stream-tube model is that more stream-tubes make different induced
velocities possible Figure 3.7. Each stream tube has its own velocity, allowing a change in velocity depending
on the direction perpendicular to the free stream flow. The accuracy is dependent on the number of stream-tubes
used. The momentum balance is carried out separately for each stream tube.

For each of these stream-tubes the momentum equations and blade elements have to be calculated, resulting
in “N” interference (induction) factors. The total span of the single stream-tube is divided in multiple stream
tubes using a fixed angle:

2𝜋
∆𝜃 = 𝑁 (3.40)
𝜃

33
 Here, a single blade passes each stream-tube twice per revolution in the upstream and downstream.

Figure 3.7 2D Schematic of the stream-tube M odel

models the
The average aerodynamic thrust with “B” number of blades and noting that each of “B” blade elements spend
Δθ/π percent of their time in the stream-tube,
stream tubethus:

𝐵∆𝜃
𝑇𝑎𝑣𝑔 = 2( 𝑇𝑖 ) (3.41)
𝜋

Considering swept area of the turbine for single stream-tube A  HR sin , one can easily get the non -
dimensional thrust coefficient as:

𝑇𝑎𝑣𝑔 𝐵𝑐 𝑤 22 cos⁡( 𝜃)
𝐶𝑇 = = ( )( ) (𝐶𝑡 − 𝐶𝑛) (3.42)
0.5𝜌𝑢2∞(𝐻𝑅∆𝜃sin( 𝜃) ) 2𝑅 𝑢∞ 𝜋 sin⁡(𝜃)

The average torque on rotor by “B” number of blades on the complete interval (0 < θ < 2π) and blade length
“H” is given by:
1
[ 𝜌𝑤2( 𝐻𝑐 ) 𝐶𝑡∗𝑅]
𝑄𝑎𝑣𝑔 = 𝐵 ∗ ∑𝑁𝜃 2
𝑖=1 (3.43)
𝑁𝜃

The torque coefficient can be defined as:

𝑄𝑎𝑣𝑔
𝐶𝑄 = (3.44)
0.5𝜌𝑢2∞( 𝐷𝐻) 𝑅

2
(𝑤 )
𝐵𝑐 𝑁𝜃 𝑢∞ 𝐶𝑡
𝐶𝑄 = ( ) ∑𝑖=1 (3.45)
𝐷 𝑁𝜃

𝐶𝑝 = 𝐶𝑄 𝜆 (3.46)

3.4.1.3 Double Actuator Disk Theory

The main disadvantage of the previous model is the inability to make a distinction between the upwind and
downwind part of the turbine. To make this possible, two actuator disks are placed behind each other, connected
at the center of the turbine Figure 3.8.

34
 Figure 3.8 Schematic of the two actuator disk behind each

other
models the

stream tube
In a similar way, as discussed in section 3.3.1, in this model, for the two actuator disks, velocities are
determined by two interference factors, a and a’. The induced velocity on the upstream will be the average of
the air velocity at far upstream and the air velocity at downstream equilibrium. The air velocity at the downstream
equilibrium can be expressed as:

𝑢 𝑒 = (2𝑢𝑖 − 𝑢 ∞ ) = (1 − 2𝑎)𝑢 ∞ (3.47)

3.4.1.4 Double Multiple Stream-tube Model (DMST)

The double multiple stream-tube model described by Loth and McCoy, 1983 and Paraschivoiu and Delclaux,
1983 combines the multiple stream tubes model with the double actuator disk theory. It allows to model velocity
variations in the direction perpendicular to the free stream flow and between the upwind and downwind part of
the turbine Figure 3.9. The previous models were not able to calculate the influence of the upwind part on the
downwind part. It is easily understood that the wind velocities at the upwind part are larger than these at the
downwind part, because the blades have already extracted energy from the wind.

Figure 3.9 Schematic of the double multiple stream-tube

model
The turbine’s interaction with the wind in the upwind and downwind passes of the blades is treated
separately. The assumption is made that the wake fromotherthe upwind pass is fully expanded and the ultimate wake
velocity has been reached before the interaction with
modelsthe theblades in the downwind pass. The downwind blades
therefore see a reduced ‘free-stream’ velocity. This approach more accurately represents the variation in
stream tube
flow through the wind turbine.

35
 The DMST model simultaneously solves two equations for the stream-wise force at the actuator disk;
one obtained by conservation of momentum and other based on the aerodynamic coefficients of the airfoil (lift
and drag) and the local wind velocity. These equations are solved twice; for the upwind and for the downwind
part of the rotor.
Upwind half  / 2   3 / 2: (w, α) = f (a)

Downwind half 3 / 2    / 2: (w’, α’) = f (a)

Once the new relative wind speed and angle of attack are found using the new induction factor; the thrust
coefficient, torque coefficient and power coefficient can be calculated.

Figure 3.10 Iterative

Procedure used to calculate
the flow velocity in DM ST
model

This method requiring some numerical calculations to solve the induction factor equation and to
numerically integrate the torque to obtain the power is very straightforward and easy to implement.

One of the most popular software work based on this algorithm is QBlade software. The software
QBlade is developed as an open source framework for the simulation and design of wind turbines. QBlade
utilizes the Blade Element Momentum (BEM) method for the simulation of horizontal axis- and a Double
Multiple Streamtube (DMS) algorithm for the simulation of vertical axis wind turbine performance. For the
design of custom airfoils and the computation of airfoil lift- and drag coefficient polars the viscous-inviscid
coupled panel method code XFOIL is integrated within the graphical user interface of QBlade.
Additionally, a module for the extrapolation of airfoil polars, beyond the stall point, for a 360° range of
angles of attack is integrated. The resulting functionality allows the use of QBlade as a comprehensive tool
for wind turbine design. QBlade is constantly being maintained, validated and advanced with new
functionality (Marten, 2013).

36
3.4.1.4.1 DMST model validation using QBlade software

The DMST model validated against experimental data measured in Low turbulence wind tunnel at the
“Politecnico di Milano” (Castelli M. , 2011) . The experimental setup consisted in a straight-bladed Darrieus
rotor, developed for experimental purpose and made of woven carbon fiber wrapped around a wood core,
with the following geometrical characteristics:

 Turbine Diameter = 1030 mm

 Turbine height = 1456.4 mm
 Number of blades = 3
 Airfoil type = NACA 0021
 Blade chord = 85.8 mm
 Solidity = 0.25

The turbine was modeled and simulated using QBlade software. The comparison showed good
agreement between the Qblade results and the measured data for low tip speed ratios see Figure 3.11, when
the tip speed ratio got higher the model over expect the coefficient of power due to the several disadvantage
of momentum based model discussed before.

QBlade wich its model DMST is very useful in parametric study because it takes short time to perform
each simulation. Other methods like CFD could take months to perform the same study. We used QBlade
in our parametric study beside CFD described in details in chapter four.

0.7

0.6

0.5

0.4
Cp

Qblade
0.3
Exp

0.2

0.1

0
0 1 2 3 4 5 6 7
TSR
Figure 3.11 QBlade DM ST model power coefficient comparison measured and
simulated data

37
3.4.3 Viscous models – Computational Fluid Dynamics (CFD)

Modern engineers apply both experimental and CFD analyses, and the two complement each other. For
example, engineers may obtain global properties, such as lift, drag, pressure drop, or power, experimentally,
but use CFD to obtain details about the flow field, such as shear stresses, velocity and pressure profiles, and
flow streamlines. In addition, experimental data are often used to validate CFD solutions by matching the
computationally and experimentally determined global quantities. CFD is then employed to shorten the
design cycle through carefully controlled parametric studies, thereby reducing the required amount of
experimental testing.

Solution procedure

1. A computational domain (geometry) is chosen, and a grid (also called a mesh) is generated; the
domain is divided into many small elements called cells.
2. Boundary conditions are specified on each edge of the computational domain (2-D flows) or on
each face of the domain, (3-D flows).
3. The type of fluid (water, air, gasoline, etc.) is specified, along with fluid properties (temperature,
density, viscosity, etc.).
4. Numerical parameters and solution algorithms are selected.
5. Starting values for all flow field variables are specified for each cell.
6. Beginning with the initial guesses are solved iteratively, usually at the center of each cell.
7. Once the solution has converged, flow field variables such as velocity and pressure are plotted and
analyzed graphically.
8. Global properties of the flow field, such as pressure drop, and integral properties, such as forces
(lift and drag) and moments acting on a body, are calculated from the converged solution.

As it is known, CFD resolves the fluid dynamic equations and it is certainly more realistic than the
1D models like momentum based models but there are many other problematic issues like stall and
turbulence modeling, unsteady rotational effects and long computation time. Some works were found in
scientific literature (Castelli M. , 2011) (Biadgo, 2013) (Sabaeifard, 2012) (Vassberg, 2005) (Gupta,
2010) (Ferreira, 2007) regarding the application of CFD modeling on VAWTs. The problem in general
was the power overestimation due to the arduous prediction of stall phenomena using fully turbulent
RANS models for low Reynolds numbers.

Lanzafame (Lanzafame, 2014) in his paper he presents the strategy of generating a 2D CFD model to
predict H-Darrieus rotors performance and solve such issues. First step was to generate a simplified CAD
and optimize the computational domain. Then a grid independent solution study was done, refining the
mesh till the results obtained with different grids were negligible. The most important part of the work was
to optimize the SST Transition turbulence model, modifying the local correlation parameters through a
series of 2D CFD tests on wind turbine airfoils. As the unsteady rotational effects are very important in
VAWTs, an unsteady Sliding Mesh Model with non-conformal mesh was used. Time step of the transient
model was optimized also. A comparison between fully turbulent SST k-ω turbulence model and SST
Transition model was done to demonstrate the superior capability of the modified Transition model to
predict VAWTs performance and underline the important role of laminar to turbulent transition modeling.
The use of the URANS (Unsteady Reynolds Averaged Navier Stokes) Transition model leads to a good

38
prediction of trend of mechanical power and power coefficient with an overestimation of 6-8 % due to the
well-known limits of the 2D models that do not take into account the 3D effects like tip vortex.

We followed Lanzfame (Lanzafame, 2014) strategy using Transition SST turbulence model and
validated with (Castelli M. , 2011) rotor described before, also we investigate using K-epsilon realizable
model recommended by Castelli in his paper.

3.4.3.1 CFD turbulence model validation

The process of generating the 2D CFD model was done inside the ANSYS Workbench multi-physics
platform where it is possible to develop a workflow, starting from CAD generation to post-processing of
the results. Particularly, the Finite Volume Fluent Solver was used to solve the Navier Stokes equations and
capture the unsteadiness like the continuous change in the aerodynamic angle of the blade with rotation.

The workflow of the model was as follow:

 Generation of a simplified 2D CAD and computational domain;

 High quality meshing of the domain to meet the specifics of the turbulence models and reach grid
independent solutions;
 Setting the Fluent Solver and calibrating the model;
 Optimization of the Transition Turbulence model;
 Post-processing results;

3.4.3.1.1 Experimental rotors features

(Castelli M. , 2011) Rotor used in this study as described before.
Number of blades: 3
Blade Airfoil [-] NACA 0020
Blade length (L) [m]; 0.000
Blade chord (c) [m]:0.0855
Radius (R) [m]:0.515
These rotor was chosen to generalize the CFD model. As there are no provided information on the
shaft, the CAD of the rotors were simplified eliminating the shaft and representing only the blades. The
experimental data were extrapolated in terms of Cp versus λ curves to correctly compare numerical results
with experimental one.

3.4.3.1.2 Computational domain generation and optimization

Generating the right computational domain for a Fluid Dynamic problem is an important task of the
modeling process. It is necessary to take into account different requirements. First of all, the domain should
not be too small to correctly reproduce the flow around the rotor and it should not be too large to not
uselessly increase cells number of the grid and hence computation time. Furthermore, the domain has to be
suitable for adequately reproduce the rotation. At last it has to be taken into account the requirements of the
meshing in terms of quality and first cell positioning near the blades. The advanced turbulence models used
in this work need a very fine mesh near the wall so that y+ < 1. Also important is to define the correct
boundary condition for the physical problem to reproduce. The computational domain was generated using

39
the CAD interface ANSYS "Design Modeler". Different domains were meshed and then tested with the
default parameters in Fluent. The best compromise was found in a box with a rotating ring that reproduces
the rotation see Figure 3.12. The domain has 3 separated sub-domains in order to use the Unsteady Sliding
Mesh Model. Only the ring is in motion while the box and the interior circle are stationary. The rotating
ring has to be as small as possible in the radial direction so that the mesh motion does not modify the real
flow-field of the wake. The rotor was placed five radius from the inlet and eighteen radius from the outlet
to correctly reproduce the wake effect. The use of the symmetry BCs allows to reduce the distance from
the lateral wall as the rotor was placed only four radius from it. It was verified that using too great a domain
would not lead to better results but only to an increase in cell numbers and hence in computational time.

Figure 3.12 Computational domain and Boundary Conditions

3.4.3.1.3 Transition Turbulence model optimization

The most problematic issue using CFD for modeling airfoil behavior at low Reynolds number is to
capture the stall phenomena. This is a well-known problem because of the inability to capture the boundary
layer separation caused by adverse pressure gradient. Above all at low Re, an important part of the boundary
layer is laminar so, the use of a classical fully turbulent model does not adequately predict the real boundary
layer behavior. Laminar boundary layer in fact is quite sensitive to adverse pressure gradient and this leads
to an earlier separation if compared to a turbulent boundary layer and, ultimately, in an unphysical
simulation of the incipient and deep stall. As the VAWTs work at low Re, stall phenomena are of crucial
importance for their modeling. For this reason, the use of a calibrated transition model should lead to a
more realistic prediction of the airfoils aerodynamic behavior and consequently a better prediction of the
VAWTs performance. The two tested used models were Transition SST (4 equations) turbulence model
and K-Ɛ realizable (2 equations) model with enhanced wall function.

The turbulence model determine the grid based on Y+ value required by each model. Y+ is
dimensionless and indicates the distance between first layer and wall and important in accuracy.

40
3.4.3.1.4 Grid Generation
A. Transition SST model
- In SST transition model, the Y+ should be around 1 which makes a very fine inflation layers
around the blade, first layer thickness was 0.001mm that’s why the high number of elements.
- Hybrid mesh used (unstructured & structured) Figure 3.13.
- The total number of elements was 1200,000 elements.

Figure 3.13 M esh around the rotor and blade for transition SST model simulation

B. K- Ɛ realizable model
- In K- Ɛ realizable model it has an equation which solves the flow around the wall “near it
“it’s enhanced wall function and it requires Y+ to be 30:300 which means no inflation which
helps in decreasing the number of elements.
- Unstructured mesh used only Figure 3.14.
- The total number of elements was 138,000 elements

41
 Figure 3.14 M esh around the rotor and blade for K-epsilon realizable model
simulation

3.4.3.1.5 Boundary Conditions

Boundary conditions (BCs) Figure 3.15 were defined as:

 A WALL type BC was used for blades.

 The line on the left of the box was defined as a VELOCITY INLET type BC. The velocity was
9m/s as wind tunnel experiment was. The turbulence viscosity was set to be 10 for inlet and outlet.
The turbulence intensity was set to be 3.46% for K- Ɛ model and 4.1% for transition SST model
also for inlet and outlet.
 The line on the right of the box was defined as a PRESSURE OUTLET type BC. The pressure was
set to be atmospheric pressure 0-gauge pressure.
 A SYMMETRY type BC was used for lateral lines of the box. The Symmetry BC is useful because
it allows the solver to consider the wall as part of a larger domain, like a ‘free shear slip wall’,
avoiding in this way the wall effects. Finally, it can substantially reduce the size of the box.
 An INTERFACE type BC was used for the contact region between the 3 sub-domains.

42
 Figure 3.15 Fluid domain Boundary conditions

3.4.3.1.6 Solver settings and calibration

To take into account the unsteady effects, particularly dynamic stall and interactions between blades
motion and wake, it was necessary to use the fluent solver in a transient version. The solver was set as
pressure based with absolute velocity formulation. The boundary conditions to activate the Sliding Mesh
Model (SMM) were set for the rotating ring cell zone, defining interfaces and rotational speed. The
rotational speed was gradually increased from the lower value to the upper value for each rotor to make the
simulation process less "aggressive" and correctly develop the flow field. The BCs in inlet and outlet were
defined fixing the wind speed in inlet at on-design value so that the calculated torque is function of rotational
speed only and thus it was simple to obtain a Cp versus λ comparison. Following a precedent work of the
authors and scientific, literature turbulent boundary conditions were optimized also, A Second Order
Upwind spatial discretization algorithm was used for all the equations (pressure, momentum, and
turbulence), and Pressure velocity coupling was PISO scheme, Hyberied initialization, second implicit in
transient. Because of the great variation of rotational speed and the complexity of the flow field, it was
impossible to establish a priori an optimal time step for the transient solver. Above all at high rotational
speed and high wind speed the time step should be very little to adequately simulate rotation and capture
the little time scale of turbulence. Moreover, the time step influences the numerical iterative process of the
solver that means that too great a time step leads to unphysical results, while too little a time step leads to
great increase in computation time. Basing on these considerations, time step was optimized performing
several simulations at different rotational speed. Establishing an angular step of 5° of position angle it was
calculated the relative time step as follows:

5
∆𝑡|5° = 360 ∗𝜔 (𝑟𝑝𝑠) (3.48)

43
3.4.3.1.7 Coefficient of Power Calculations

- Average coefficient of momentum (𝐶𝑚) was monitored, after ten revolutions of the turbine
(reach steady state) we get the average coefficient of momentum from revolution nine to
revolution ten,

Cp = Cm|avg * TSR (3.49)

- By changing the rotational speed of rotating ring we change the TSR and get the C m|avg then Cp
for each case, then draw Cp vs. TSR chart.

3.4.3.1.8 Results

Figure 3.16 relation between Cp & tip speed ratio

0.45

0.4

0.35

0.3

0.25
Cp

Exp
0.2 Transittion SST
K-e realizable
0.15

0.1

0.05

0
0 0.5 1 1.5 2 2.5 3 3.5 4
TSR

Figure 3.16 CFD numerical and experimental data

44
 Table 3.2 URANS Turbulence models comparison

SST transition turbulence model K-Ɛ realizable model

 Y+ should be around 1  It was good at low TSR

 Needs inflation  When TSR increased above the
 Large number of elements optimum it gave an over prediction
 It was good in determining the TSR we should work at because K- Ɛ realizable can’t work
 At TSR higher and lower than optimum TSR it gives on the flow behind the blade
inaccurate results because of wake
 At opt. TSR it gives over prediction for Cp because the  K-omega is good in studying the
case is in 2D and the tip losses are neglected and arms, effect of fluid impingement on the
hub and shaft which create a drag force are neglected blade walls.
too  Less time consuming for
 Very high computational time computation than other models.

As we see the K-Ɛ realizable model was the optimized model for our case study due to its fast and reliable
results despite the sort of error in high tip speed ratios. We used K-epsilon model in our parametric study.

3.4.3.2 Flow driven rotor simulation

In contrast to a simulation with the TSR given, the rotational speed of the blade () is not specified in
an unsteady flow driven rotor simulation. In a flow-driven rotor simulation, the turbine blade rotates around
its axis at a certain rotational speed by balancing the hydrodynamic moment and the moment of inertia
(Quang, 2014).

3.4.3.2.1 6-DOF solver Coupled with CFD

6-DOF refers to the motion of a rigid body in three dimensional spatial domain, namely the ability to
move forward/backward, up/down, left/right (translation in three perpendicular axes) combined with
rotation about three perpendicular axes (pitch, yaw, roll). As the movement along each of the three axes is
independent of each other and independent of the rotation about any of these axes, the motion indeed has
six degrees of freedom. The method is always used in robotics, game controller, etc.

The principle of using the 6-DOF model to simulate the motion of a VAWT is mainly based on the law
of rotation, which is shown in Eq. (3.50). The law indicates that a net external torque on a rigid body will
cause an angular acceleration. The rotation of the rigid body will accelerate until the net torque become to
zero or fluctuate about zero periodically.
𝑑𝜔
∑𝑀 = 𝐽 (3.50)
𝑑𝑡

In order to simulate the rotation of a rigid body in the fluid flow, 6-DOF method should be coupled
with CFD. The principal procedure is as follows (Wang, 2011):

1. The flow field at the time step “i” is simulated based on RANS equations. Note that the realizable
k-ε turbulence model is applied;

45
 2. The torque acting on the rotor from the fluid can be obtained. Based on Eq. (3.50), the angular
acceleration of the rigid body is calculated;
3. The rotation speed at the time step “i+1” is available according to Eq. (3.51);
𝑑𝜔
𝜔𝑖+1 = 𝜔𝑖 + . ∆𝑡 (3.51)
𝑑𝑡

Where Δt is the increment between two time steps.

4. Based on the new rotation speed of the rigid body, the new flow field needs to be simulated;
5. Repeat the above procedures 2-4 until the rotation speeds of the rigid body at neighboring time
steps fluctuates about a constant periodically.

3.4.3.2.2 6-DOF method validation

For the validation of this method we used the same strategy as before in turbulence models validation
using mesh motion method except now we use 6-DOF method instead of mesh motion method. The steps
we followed were:

1. We used the computational domain and mesh settings for K-epsilon realizable model.
2. We used K-epsilon realizable turbulence model with enhanced wall function.
3. The boundary conditions were the same for the 6-DOF simulation.
4. The same solver settings.

3.4.3.2.2.1 6-DOF parameters

The three-blade rotor is set to the rigid body for 6-DOF analysis. The out far field and the inner
stationary zones were set to be deforming zones.

The mass moment of inertia of the turbine was required to do this simulation. The mass moment of
inertia was not specified by (Castelli M. , 2011) in his paper, so we used different values for mass moment
of inertia in range (0:1.3) with increment step of 0.05 kg.m2 .

3.4.3.2.2.2 Time step calculations

The variable time step was determining in this way:

1. Initially the time step was set to be 0.001 second

2. After 20000 iterations we calculated the rotational speed visually from solution animation
3. The new time step was calculated using Eq. (3.48)
4. Repeat this procedure until the rotational speed changes are negligible.

3.4.3.2.2.3 Coefficient of power calculations

The coefficient of power was calculated as follows:

1. Calculate the rotational speed in (RPS) visually using solution animation. When the turbine
end entire revolution in several time steps (n) use the equation
w = 1/(n*Dt) (RPS) (3.52)

46
 2. Calculate the tip speed ratio using Eq. (3.2).
3. We monitored the coefficient of moment the entire simulation, then we take the average Cm
from the steady state part.
4. Coefficient of power now can calculate from Eq. (3.49).

3.4.3.2.2.4 Validation results and discussion

0.45

0.4

0.35

0.3

0.25
Cp

Exp
0.2
6-DOF Method
0.15 Mesh Motion Method
0.1

0.05

0
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5
TSR

Figure 3.17 CFD 6DOF method numerical results and experimental data

The numerical results of both methods 6-DOF and Mesh motion compared to the experimental results.
Showed good match for low tip speed ratios and over predicting the performance in high tip speed ratios
due to turbulence model sort of errors and 2D simulation error.

47
3.5 VAWT Design analysis methodology

The design strategy is consisting of four main steps as follows:

1. Determining the forces acting on the blades and rotors

2. Wind turbine construction
3. Material selection for different components of the wind turbine
4. Stress analysis
5. Vibration analysis
6. Cost analysis

We applied this strategy on our design process which showed good results and reliable design.

3.5.1 Forces acting on VAWT blades and rotor

There are three types of forces acting on the turbine blade (i) the aerodynamic forces (ii) the centrifugal
force (iii) weight force. The aerodynamic force acting on each rotor blade can be decomposed in a
component tangent to the chord, the tangential force FT, and a component perpendicular to the chord, the
radial force FN , see figure 3.18.

Figure 3.18 Aerodynamics forces acting

on rotor blade

The centrifugal force is defined as

Fc = m*Rrotor*ω2 (53)

Where m is the mass, Rrotor is the rotor radius and ω is its angular velocity. This force acts in the opposed
direction of the radial force, always pointing outwards with respect to the rotational axis.

In next chapter we will discuss the design case using CFD to determine the aerodynamic forces and
centrifugal forces. Using the CFD Mesh motion method to determine the aerodynamic forces and using
CFD 6-DOF method to determine angular velocity which cause the centrifugal force.

48
3.5.2 Wind turbine construction

VAWT is consisting of several main components:

1. Blades
2. Arms
3. Hub
4. Shaft
5. Bearings & bearings housing
6. Tower

Each component was drawing and modeling using AutoCAD 2018 & SolidWorks 2016 software.

3.5.2.1 Blade construction

The blade was constructed of main spar carrying ribs, blade shells and top/bottom plates. The ribs
function was to give the blade its airfoil shape other ribs were to connect the blade to the arms with bolts.
The shell was to give the blade its final shape and the top\bottom plates to reduce tip losses see figure 3.19.

Figure 3.19 Blade

main components.

The location of ribs which attached to the arms was optimized to give the minimum bending moment
on the blade.

49
3.5.2.2 Turbine Arms

The main purpose of the arm was to connect the blade to the hub. The arms was made of metal duct.
The duct then cut to three pieces and welded together.

The idea of inclined arms was to reduce the length of the hub and shaft which reduce the cost. When
the shaft is short the vibration and the loads on bearings will reduce.

Figure 3.20 VAWT arms

3.5.2.3 Turbine hub

The hub function was to connect the power from blades to the driven shaft. Hub connect the power to the
shaft by set screws. It was consisting of two flanges welded to center cylinder figure 3.21.

Figure 3.21 VAWT hub

3.5.2.4 Driven shaft

The man function of shaft is to connect the power from the blades to the generator. The shaft was made of
metal machined to take the final shape. The shoulder in shaft was to deliver the weight force to the upper
bearing by this we can eliminate the danger of buckling in shaft.

3.5.2.5 Bearings and bearing housing

We used two deep groove ball bearings. The housing was machined to meet the specific bearing design
and dimensions recommended by the bearing manufacturer.

Figure 3.22 VAWT Shaft

with bearings and housing

50
3.5.2.6 Tower

The tower was consisting of two metal bases connected to three metal tubes by set screws. The function of
tower is to support the turbine and eliminate vibrations as possible.

Figure 3.23 VAWT

construction on
SolidWorks 2016

51
3.5.3 Material selection

The material selection was based on many factors:

1. Material availability in the local market

2. The material cost
3. Strength and other mechanical properties of the material
4. Machining ability
5. Effect on mass moment of inertia

We will discuss the materials and its cost in details in next chapter.

3.5.4 Stress analysis

After we calculated the different loads acting on blades and rotor and select the proper materials for
each component. We had to test these materials against the loads they carrying and see if the stress was
higher or lower than the strength of the material.

Due to the complexity of the turbine shape, we used Finite element analysis (FEA) to analysis the
stresses on the turbine. We used ANSYS Mechanical software to perform this analysis.

The steps we followed to perform this analysis was:

1. Insert the geometry to Design modeler software

2. Applying material
3. Grid generation
4. Modeling the different loads on the turbine
5. Specify the supports
6. Post processing

3.5.4.1 Geometry

We divided the turbines into two groups separated from the bearings location see figure 3.24, then insert
them in design modeler:

A. Rotating group (blades, arms, hub and shaft)

B. Stationary group (bearing housing and tower )

Figure 3.24 VAWT

FEA geometry groups

52
3.5.4.2 Material insertion

For each part we insert the selected material by its name and its mechanical properties.

3.5.4.3 Grid generation

We used automatic meshing for both groups where group (A) total numbers of elements was 9000000
element while for group (B) it was 2000000 element.

3.5.4.4 Loads modeling

For group (A)

1. Aerodynamic forces: after it calculated from design case using CFD it was applied on the blade
surface by its magnitude and direction.
2. Centrifugal force: after the angular velocity calculated from design case using CFD. We applied
the rotational speed to all the group (A) parts
3. Weight force: it modeled by applying standard earth gravity for all the components.

For group (B)

1. Reactions on bearings: After we performed the CFD design case we calculated the bearings
reaction by hand calculations discussed later in next chapter. The reactions were modeling by
applying forces on the housing inner faces.
2. Weight force: it modeled by applying standard earth gravity as group (A).

3.5.4.5 Specify the supports

For group (A) the supports were the location of bearings located on shaft and it were a fixed supports.
For group (B) the supports was the end of the tower near the ground.

3.5.4.6 Post processing

We monitored and determined the total deformation to see the amount of displacement each part will make
when we applied the load on the turbine. We also monitored and determined the equivalent and the
maximum principle stress. This two values helped us to determine if the material selection was good or
bad. The results will discussed in next chapter.

53
3.5.5 Vibration analysis

Vibration has to be considered as it is important to avoid dynamic instability in the form of resonance,
where there is coincidence between the natural frequency of the blade and the rotational frequency.
Resonance can cause fatigue damage and rapid failure (Burton, 2001).

Any object vibrates in six modes of vibration, each mode has its natural frequency. We used ANSYS
modal software to perform this analysis to get the first mode natural frequency for Blades, arms and shaft
separately. By determining the First mode natural frequency for the most important rotating parts (blades,
arms and shaft) and compared to the operating rotational speed we calculated from CFD design case. The
operation frequency must be less than the natural frequency to avoid resonance.

Modal analysis using ANSYS modal is a FEA approach using the same steps of FEA using ANSYS
mechanical except in modal analysis there is no external forces applied because we aim to calculate the
natural frequency.

The results will be discussed in detail in next chapter.

3.5.6 Cost analysis

The total cost of raw materials and the manufacturing process is important to be reasonable.
Considering we are living in developing country. A full cost analysis will discussed in next chapter.

54
 Chapter

Results & Discussion

4
4.1 Introduction

In this chapter we will discuss the results we found following the different methods discussed in chapter
three for designing our turbine.

This chapter includes:

A. Parametric study
B. CFD design case
C. Material selection
D. FEA stress & vibration results
E. Cost analysis

4.2 VAWT Parametric study

The parametric studies were performed to optimize the turbine design to ensure that we get the
maximum power coefficient, but optimizing with varying all the parameters is a time consuming process
so we will fix some of the parameters and then see the effect of other parameters on the turbine performance.

4.2.1 Turbine swept area

We will fix the swept area to 3 m2 and that will reduce our initial cost of construction. The initial
coefficient of power was set to 0.2 with air density at sea level of 1.225 kg/m3 . We chose the design air free
stream cut in velocity to be 5 m/s it was chosen based on the site where we will install our turbine, In
Alexandria, Egypt the average wind speed is 5 m/s (Windfinder, 2018)

55
 From power correlation from chapter 1 we can get a power of 46 Watt. The output mechanical power
is considering low but by trying to increase the coefficient of power and in case of windy days the power
can tremendously increase because it is proportional to the cubic value of wind speed.

4.2.2 The aspect ratio effect on performance using DMST QBlade

The ratio between blade height and rotor radius (aspect ratio) is play an important role in design process
of VAWT, Basically high aspect ratio means that we get high rotational speed than rotor torque for the
same power and vice versa. The factors which indicate the best configuration of aspect ratio is the power
coefficient and rotor torque.

We need to fix the other parameters to see the effect of aspect ratio only. We chose the turbine solidity
as 0.4 which makes the turbine capable of self-starting and have a good power coefficient (Brusca, 2014).
The airfoil selected to be NACA 0020 which have the ability of self-start. The number of blades was three
blades to reduce to our initial cost of construction. The cases we used in our simulation is listed below in
table 4.1.

Table 4.1 aspect ratio simulation input values

Turbine Height Rotor Diameter Rotor Radius (m) Aspect ratio (H/R) Chord length (m)
(m) (m)
3 1 0.5 6 0.067
2.5 1.2 0.6 4.167 0.08
2 1.5 0.75 2.67 0.1
1.5 2 1 1.5 0.133
1 3 1.5 0.67 0.2
0.5 6 3 0.0833 0.4

We chose the case number three for our design. Case three has lower coefficient of performance than
case 1 & 2, but it has a higher rotor torque than them Figure 4.1. The rotor torque is important factor to
drive the alternator shaft.

Figure 4.1 Aspect ratio effect on turbine performance

56
4.2.3 Solidity effect on performance using DMST QBlade

A solidity is an important parameter in design process, it indicates the turbine ability to self-start.
Solidity higher than 0.4 is good for self-starting, but in limits proportional to power coefficient.

Parametric study was performed to indicate the best solidity for our design with constant aspect ratio
2.67, number of blades 3 and airfoil type NACA 0020 and changing the chord length as shown in table 4.2
to maximize the power coefficient.
Table 4.2 Solidity and chord length

Solidity ( σ ) Chord ( m ) Solidity ( σ ) Chord ( m )

0.8 0.2 0.4 0.1
0.7 0.175 0.3 0.075
0.6 0.15 0.2 0.05
0.5 0.125 0.1 0.025

As shown in Fig. 4.2 as the solidity increases the torque increases but the TSR at maximum torque reduces
that will reduce the power coefficient, also the torque peak get narrow which will reduce the turbine capability
of producing power over a wide range of tip speed ratios.

For our design we chose solidity of 0.4 which has the maximum power coefficient.

Figure 4.2 Solidity effect on turbine performance

57
4.2.4 Airfoil selection using DMST QBlade

Airfoil selection analysis will specify suitable airfoil for VAWT according to optimum values of angle
of attack AoA, lift and drag coefficient (CL/CD ), Torque and coefficient of power CP .

We chose NACA [0012-0015-0018-0020-0021-4415] airfoils to investigate between them. NACA

airfoils are airfoil shapes for aircraft wings developed by the National Advisory Committee for Aeronautics
(NACA).and analysis will be calculated by XFOIL algorithm and DMST simulation.

Initially we study effect of each parameter on all types of airfoils and determine the better values of
parameter for which type of airfoils and compare between better values and types of airfoil to select suitable
airfoil.

4.2.4.1 Optimum (AoA) for every airfoil

Angle of attack must achieve high lift coefficient CL and less drag coefficient CD and have high Cm
momentum coefficient. Then we choose AoA according this concepts. The analysis calculated CL/ CD value
and Cm when AoA change is shown in Figure 4.3.

Figure 4.3 airfoil type effect on lift, drag and moment coefficients.

58
 From figure we can see that the symmetrical airfoils have a higher coefficient of moment than the
cambered airfoil NACA 4415 but on the other hand the cambered airfoil has much higher lift to drag ratio,
to indicate which one is better for our design we need to optimize using the coefficient of power at each
case.

The optimum AoA for every airfoil from graph is 10 degrees, that angle of attack will increase the lift
force on the blade, which will increase the power coefficient.

4.2.4.2 Airfoil type effect on turbine performance

An optimization study was performed to simulate the airfoil type effect on the turbine performance, our
objective is to select the best airfoil type to maximize the power coefficient Figure 4.4.

Figure 4.4 Airfoil type effect on performance

Maximum Cp is 0.59 attributed to NACA 0015 airfoil and slightly change in others between (0.57) to
(0.58) at optimum TSR of (3.2) and find minimum Cp is 0.52 attributed to NACA 4415.Max torque is 6.6
N.m attributed to NACA 0015 and slightly change in others between (6.4) to (6.5).

59
 Simulation analysis refers to select NACA 0015 which has max Cp & torque, but we select NACA 0020
for its ability of self-start and the higher thickness can reduce the stresses on the blade. NACA 0020 has Cp
as 0.58 & torque as 6.45 N.m Slightly change compared with NACA 0015 airfoil.

NACA 4415 has a high lift force but due to low coefficient of momentum, it has lower torque and
power than other symmetrical airfoils so it was proven that in designing of VAWT cambered airfoils are
not suitable, also this supported by Premkumar (Premkumar, 2015).

4.2.5 Optimum tip speed ratio using CFD Mesh motion method

As discussed before in chapter 3, TSR is the ratio between blade speed and the wind free stream
speed, when it is very low you will not be able to extract all the possible power from the wind, and when
it is very high the blades will block the wind from entering the turbine and that will make you lose power
too. Every rotor have an optimum tip speed ratio that maximize the coefficient of power.

Performance study performed on our turbine using CFD mesh motion method. The computational
domain was drawn using CAD by the same ratios as described in model validation. We used the same
mesh settings in our turbine that the total number of elements was 220000 element. Using the same solver
settings and boundary conditions and turbulence model as in case of model validation.

We determine the optimum tip speed ratio Based on the error percentage in validation of the CFD
model the optimum tip speed ratio found to be 2.5 see figure 4.5.

For the optimum TSR the angular velocity we should reach to have the maximum output power are:

 Wind speed = 5 m/s rotational speed = 159.15 RPM

 Wind speed = 7 m/s rotational speed = 222.82 RPM
 Wind speed = 9 m/s rotational speed = 286.48 RPM

0.45

0.4

0.35

0.3

0.25
Cp

0.2 CFD mesh motion method

0.15 Without error

0.1

0.05

0
0 0.5 1 1.5 2 2.5 3 3.5 4
TSR

Figure 4.5 our turbine performance curve

60
4.2.6 Pitch angle Effect on performance using CFD 6-DOF method

The effect of pitch angle investigated using CFD 6-DOF method where we fixed the mass moment of
inertia on 1.39 kg.m2 . The value of mass moment of inertia came from hand calculations based on material
selection – discussed later -. We investigate the turbine ability of self-starting in the simulation.

We used different values of pitch angles described in next table with simulation results:
Table 4.3 pitch angle effect on performance

Pitch Wind Rotational Starting

Torque Power
angle speed speed TSR Cm|avg Cp|avg time from
[N.m] [Watt]
[degree] [m/s] [RPM] rest [sec]

5 52.63 0.8267 0.04928 0.0407 0.6 0.8489 9.577

0 7 125 1.4024 0.08082 0.1133 0.32 2.7288 71.40

9 250 2.1816 0.09987 0.2179 0.24 5.5741 291.885

5 52.63 0.8267 0.05396 0.0446 0.48 0.9295 10.495

4 7 133.33 1.4956 0.08875 0.1327 0.32 2.9965 83.633

9 285.71 2.49 0.09417 0.23 0.24 5.2560 308.093

5 52.63 0.8267 0.05678 0.0469 0.49 0.9781 11.036

6 7 133.33 1.4956 0.09156 0.1369 0.32 3.0914 86.283

9 285.71 2.4933 0.09949 0.2480 0.24 5.5529 332.2

5 52.63 0.8267 0.05857 0.0484 0.49 1 11.389

8 7 125 1.4024 0.09190 0.1289 0.32 3.1029 81.24

9 285.71 2.49 0.08867 0.22 0.24 4.9490 294.69

We found that pitch angle of six degree was the optimum pitch angle to having the maximum output
power see figure 4.6.

From the values of time that the turbine takes to move we can deduce that the turbine having the ability
of self-starting which achieve one of the project objectives and solve the problem of H-Darrieus VAWT in
starting from rest see figure 4.7.

The output power at wind speed of 5 m/s is 11 watts which is very low to power a generator. For place
like Alexandria, Egypt having average wind speed of 5m/s this turbine is not applicable to install. From this
we can conclude that the turbine mass moment of inertia was high to meet the aerodynamic moment
generated by air on blades. We cannot reduce the mass moment of inertia than 1.39 kg.m2 because of the
limitation of the materials in our local market in Egypt.

61
 0.3

0.25

0.2

0 AOA
Cp

0.15
4 AOA
6 AOA

0.1 8 AOA

0.05

0
0 1 2 3 4 5 6 7 8 9 10
wind speed [m/s]

Figure 4.6 different blade pitch angle performance curves

0.7

0.6

0.5
TIME [ sec ]

0.4 0 AOA
4 AOA
0.3
6 AOA
0.2 8 AOA

0.1

0
5 7 9
wind speed

Figure 4.7 VAWT starting time from rest

62
4.2.7 Turbine characteristics data sheet

From this chapter we could decide our turbine main design using parametric studies to ensure we get
the maximum power coefficient and suitable rotor torque from DMST model using QBlade software and
from CFD using ANSYS fluent coupled with mesh motion solver and 6-DOF solver.

Table 4.4 Turbine main design data

Turbine swept area 3 m2

Turbine diameter 1.5 m
Blade length 2 m
Number of blades 3 blades
Solidity 0.4
Chord length 0.1 m
optimum Tip speed ratio 2.5
Airfoil NACA 0020
Fixed pitch angle 6 degree

63
4.3 Design case using CFD mesh motion method

We design our turbine to withstand forces at wind speed of 9 m/s. From CFD 6-DOF simulation with
pitch angle 6 degrees the rotational speed was 285.71 RPM. We used rotational speed of 350 RPM to be in
safe side in this case and in FEA stress analysis case.

Using ANSYS fluent mesh motion method we give the rotating ring a velocity of 350 RPM and in
boundary condition we put the inlet velocity to be 9 m/s.

After the 9th revolution is completed we calculated tangential force (tangent to the blade rotation circle
in the direction of rotation) and the normal force (perpendicular on tangential force towards center of
rotation) acting on single blade see figure 4.8, also calculated the total torque acting on the turbine three
blades see figure 4.9, for each time step until we reach the 10th revolution. All the results were stored in
Excel data sheet.

150

100
Force [N]

50
Tangential Force
0 Normal Force
0 50 100 150 200 250 300 350 400
-50

-100
Position angle [degree]

Figure 4.8 Blade loading aerodynamic forces

50

40

30
Rotor torque [ N.m ]

20

Rotor torque
10

0
0 50 100 150 200 250 300 350 400
-10

-20
Position angle [ degree ]

Figure 4.9 Turbine rotor loading aerodynamic torque

64
 The maximum normal force on blade conceders in the same direction of centrifugal force, by that the
blade will suffer the maximum bending moment coming from summation of normal force and centrifugal
force.

FN max on blade1 = -68 N θ = 260 degree FT = -3.7728912 N

FN blade 2 = 7.5085378 N θ = 20 degree FT = -2.26296 N

FN blade 3 = 110.210718 N θ = 140 degree FT = 23.711964 N

The Max average torque T = 43.80421579 N.m

4.4 Material selection

As discussed in chapter three the material selection depends on factors. We tried to satisfy those factors
in our selection. The materials we selected were easy to found in Alexandria local market with reasonable
price.

To reduce the mass moment of inertia we try to use light materials for the parts far from the center of
rotation – having long radius of inertia -. We found that the wood is suitable material for the blades. Wood
have lower density than any other material in the market, but having some drawbacks like its strength
consider very low so we choose wood for constructing our blades.

The arms, hub, shaft and tower all these parts are made of metal. Using light metal for the arms to
reduce inertia and strong heavy metal for hub and shaft because they have low inertia – short radius of
inertia -. The mass moment of inertia was hand calculation.

The shaft design and material selection was based on hand calculations using ASME equation, also the
reactions on bearings and bearing selection calculated using hand calculations. All the hand calculations
are in the appendix B.

The material selection for each part and its mechanical properties listed below in the table

Table 4.5 M artials & properties

Mass moment
Part name Material Mechanical properties of inertia
[kg.m2 ]
ρ = 0.655 g/cc
Blade spar Teak wood E = 12.28 GPa
Su = 54.8 MPa
ρ = 0.13 g/cc
Blade wooden ribs Balsa wood E = 3 GPa 0.85958
Su = 7-14 MPa
ρ = 2.7 g/cc
Blade metal ribs 5052-H12 Aluminum E = 68 GPa
Sy =180 MPa

65
 Su = 230 MPa
ρ = 0.13 g/cc
Blade shells Balsa wood E = 3 GPa
Su = 7-14 MPa
ρ = 7.9 g/cc
Bolts btw. Arms and E = 190 GPa
M5, Property class 10.9 0.05020
metal ribs Sy = 830 MPa
Su = 940 MPa
ρ = 2.7 g/cc
E = 68 GPa
arms 6063-T6 Aluminum 0.42806
Sy = 210 MPa
Su = 240 MPa
ρ = 7.9 g/cc
Bolts btw. Arm and E = 190 GPa
M10, property class 4.8 0.00486
hub flanges Sy = 310 MPa
Su = 340 MPa
ρ = 7.9 g/cc
hot rolled AISI 1025 E = 190 GPa
Hub 0.04430
carbon steel Sy = 250 MPa
Su = 450 MPa
Total torque capacity of 114
Set screw btw. hub 2 X M12 set screw -
N.m
and shaft
ρ = 7.8 g/cc
Hot Rolled 1040 Carbon E = 190 GPa
Shaft 0.00021
Steel Sy = 320 MPa
Su = 570 MPa
2 Chromium steel bearings Static load = 7.8 KN
Bearings SKF 6205-2RSH/C3 Dynamic load = 14.8 KN -
bearings Limiting speed = 8500 RPM
ρ = 2.7 g/cc
E = 68 GPa
Bearing housing Aluminum -
Sy = 210 MPa
Su = 240 MPa
ρ = 7.9 g/cc
Bolts btw. bearing E = 190 GPa
M10, property class 4.8 -
housing and tower Sy = 310 MPa
Su = 340 MPa
ρ = 7.9 g/cc
from Hot Rolled 1025 E = 190 GPa
Tower bases -
Carbon Steel Sy = 250 MPa
Su = 450 MPa
ρ = 7.8 g/cc
E = 190 GPa
Tower tubes Steel ASTM A53 Grade A -
Sy = 205 MPa
Su = 330 MPa

The total moment of inertia is 1.39 Kg.m2 after adding the inertia of the generator.

66
4.5 FEA stress and vibration analysis results

After the material selection and the loads calculated from CFD design case. We perform the stress
analysis as described in chapter three.

By comparing the results of maximum principle stress from numerical simulation and yield, ultimate
strength of the materials the comparison showed that the material selection was successful see figure 4.10.

Figure 4.10 FEA stress analysis using ANSYS mechanical

67
 The vibration analysis performed using ANSYS modal software as discussed in chapter three. The
results were good depending on the design rotational speed.

The design rotational speed was 350 RPM or 5.83 RPS ≡ 5.83 Hz. That is our lower limit, if the natural
frequency of the parts was lower than that limit that’s mean we will have resonance.

The numerical results from our vibration analysis is sorted the next table

Table 4.6 Vibration analysis results

Part 1st mode Natural frequency [Hz]

Blade 45.452
Arm 25.91
shaft 320.64

The results showed that at operating design rotational speed the resonance phenomenon will not
happened. In the next figure 4.11 the total deformation of the three parts in case of resonance. These figures
showed that in case of resonance the turbine will have severe damage.

Figure 4.11 Total deformation in case of resonance

68
4.6 Cost analysis

As discussed in chapter three the cost management is important for any project. The output power
should cover total cost of the project. That cost of raw materials and its manufacturing process listed in the
table below.

Table 4.7 project cost analysis

Manufacturing Manufacturing
Part name Raw material cost Total cost
process cost
Spar 100 LE Saw cutting 60 LE 160 LE
Wooden ribs 114 LE Laser cutting - 114 LE
Metal ribs - CNC milling 285 LE 285 LE
Blade shell 336 LE Cutting - 336 LE
Saw cutting
Arms - welding 200 LE 200 LE
drilling
Bearings 120 LE - - 120 LE
Casting
Saw cutting
Hub, Shaft
Drilling
Bearing housing - 1840 LE 1840 LE
Welding
tower
Quenching
Turning
Bolts 42 LE - - 42 LE
Paint 35 LE - - 35 LE
Paint thinner 15 LE - - 15 LE
Sand papers 10 LE - - 10 LE
Sheet sticker 90 LE - - 90 LE
Adhesives 200 LE - - 200 LE
Concrete base 225 LE - - 225 LE
Cutter knife 20 LE - - 20 LE

The total initial cost of the project was 3700 LE. The cost conceder reasonable for the project size. The
running cos of the project is low because the material we select. The only thing need maintenance is the
bearing after its operation life ended in five years.

69
 Chapter

Conclusion & Recommendations

5
5.1 Conclusions

The aim of this work was to design and manufacturing H-Darriues VAWT to operate in low wind speed
conditions and to be self-started turbine. We used different methods to have the optimum design for our
turbine. From our research we conclude these points:

1. Performance analysis and parametric studies using momentum based models and QBlade
software proves it can be Useful tool in the design process as it gives quicker and qualitative
results than other methods.
2. Viscus models – CFD in designing process of VAWT need more effort than other models,
though it can give you more accurate prediction of performance.
3. CFD 6 DOF method proves its importance in prediction the performance of the turbine and,
it’s a good indicator about the power generated from our turbine.
4. The mass moment of inertia is the most dominant design parameter when you design wind
turbine to work in low wind speed conditions.
5. Wood is a suitable material for blade construction in small-scale wind turbines.
6. Self-starting ability depending on the airfoil type, turbine solidity, fixed pitch angle and the
mass moment of inertia.
7. The low wind speeds generate low aerodynamic torque on our turbine blades.
8. We cannot reduce the mass moment of inertia because of the material limitations.
9. FEA is useful tool in stress analysis for complex shapes like VAWTs.
10. Vibration analysis and resonance phenomenon is important to includes in design process
11. The total cost of the project was reasonable for the VAWT size and for the economic
conditions in our contrary.

70
5.2 Recommendations

The work on this project is not done yet. A series of testing and experimental work should be performed
on the wind turbine

The work on converting the output mechanical power to electric power useful to use in our daily life
need further research and efforts.

71
 Appendix

Hand calculations
A
1. Shaft design hand calculations

Before designing of the shaft we need to understand the forces and where is there action done
on the shaft.

Y
X

72
  Centrifugal force [Black arrow] when translated to the center the three forces eliminate
each other. So the effect of centrifugal force on shaft cannot be considered.

 Aerodynamic normal force [blue arrow], the three forces are not equal so we should
consider the max normal force and work on it in our design.

 Aerodynamic tangential force [red arrow], the three forces are not equal so it will be
included in design and its average torque T.

 The weight force [orange arrow] will be included also as an axial force and moment M.

 The tangential force on gear [green arrow] will be considered in design.

 The radial force on gear [purple arrow] will be considered.
We will try now to study the position that have maximum load as moment and forces and torque
on the shaft.

Calculations
1. Aerodynamic normal force

After force analysis of three forces at center C and D the value of total force = 63.425 N in x

Moment @ C = 63.425 * 420 = 26638.5 N.mm

Moment @ D = -63.425 * 420 = -26638.5 N.mm

2. Aerodynamic tangential force

After force analysis of three forces at center C and D the value of total force = 11.246 N in x
direction

Moment @ C = 11.246 * 420 = 4723.32 N.mm

Moment @ D = -11.246 * 420 = -4723.32 N.mm
That case will not give us the max Torque on shaft but it will give us the torque when the normal
force in the same direction of centrifugal force so we well design the shaft on the max torque
capacity max torque T = 43804.21579 N.mm

3. Weight forces
After force analysis the total axial force will be @ D = 125.55 N in direction Y

Moment @ C = 6227 N.mm

Moment @ D = 6227 N.mm

73
4. Tangential force & radial force on gears

The design of belt is not suitable for our project:

1. Flat belt requires very large belt width 120 mm
2. V belt requires high central distance
The solution here is using sprocket and chains or using pair of spur gears with velocity ratio (4) to
ensure that at low wind speed the generator can charge 12V battery easily
I mean if the wind speed was 5m/s and your rpm around 100 rpm you will need at least 300 rpm
so the generator can produce 12 Volt.
 The pinion will have 17 teeth and that is the minimum no. of teeth to prevent interfacing
 We will select module 2mm to reduce the center distance
 The velocity ratio will be 4
 We will choose pressure angle of 20 because 14.5 have a weaker tooth

So the diameter of pinion = m*Np = 2*17 = 34 mm

Diameter of gear = VR*D p = 4*34 = 136 mm

Tangential force on gear = T/ (D gear/2) = 644.1796 N
Radial force = Ft * tan (phi) = 644.1796*tan (20) = 234.4622159 N the plane of radial will have
less max moment so we will include tangential only in our plane.
We will assume that the radial force in the direction x, that will maximize the load but this case
could be happen if the direction of blade and generator and gear matched together.
******************************************************************************

The total moment @ C = 37588.82 N.mm total moment @ D = -25134.82 N.mm C.W +

Force in X direction @ C = 74.671 N

Force in X direction @ D = 74.671 N Force in Y direction @ D = 125.55 N

Force in X direction @ E = 644.1796 N https://beamguru.com/online/beam-calculator/

74
HA = -125.55 N

RA = -748.00 N

RB = -45.52 N

75
Shaft design ASME equation

2
16 𝛼⁡𝐹𝑎⁡𝑑0 (1 + 𝐶 2 )
𝑑03 = √(𝐾𝑚 𝑀 + ) + (𝐾𝑡⁡ 𝑇)2
𝜋⁡𝜏𝑎𝑙𝑙 (1 − 𝑐 4 ) 8

1. Let d0 = 20 mm
2. C =0 solid shaft
3. Km = 1.5 rotating shaft gradually applied
4. M = 53269.73 N.mm
5. 𝛼 = 1.046
6. Fa = 125.55 N
7. Kt = 1 rotating shaft gradually applied
8. T = 43804.21579 N.mm

𝜏𝑎𝑙𝑙 = 58.1946 = 0.75*0.3 Sy or 0.75*0.18 Su

Sy = 258.64266 MPa
Su = 431.07 MPa

Select material (steel) have sy > 250 MPa and Su > 411 MPa

 The shaft will be made from Hot Rolled 1040 Carbon Steel
1. Density = 7.8 g/cm3
2. Modulus of elasticity = 190000 MPa
3. Poisson’s ratio = 0.29
4. Tensile yield strength = 320 MPa
5. Tensile ultimate strength = 570 MPa

76
2. Bearing selection

 Fr = 1.122 KN
 Fa = 0.188325 KN
 D sh = 25 mm
 N = 352 RPM
Co = F.S * Po = F.S * (Xo Fr + Yo Fa)

F.S = 1.5 for excellent condition of operating smoothness

Fa/Fr = 0.1678 < 0.8 so Xo = 1 & Yo = 0

Co| calculated = 1.683 KN

Go to catalogue select bearing 6205-2RSH deep grove ball bearing

Co| catalogue = 7.8 KN C| catalogue = 14.8 KN Fo = 14

---------------------------------------------------------------------------------------------------------------------

C = Ks * P * (L10 )1/∈

Ks = 1 uniform load
∈= 3 Ball bearing

L10 = time in hr * 3600 * N (rps) * 10^-6

Time in hr = 50000 hr for 24 hours continuous operation

L10 = 50000 * 3600 * 5.86667 * 10^-6 = 1056 million revolution with reliability of 90%

P = X * V * Fr + Y * Fa
V=1 inner ring rotating

Fo* Fa/Co| catalogue = 14*0.11769 / 7.8 = 0.338 ------------------------ so e = 0.2187

Fa/Fr = 0.1678 < e X= 1 & Y=0

P =1.122 KN
C| calculated = 11.42564 KN < C| catalogue = 14.8 KN

So select 2 bearings for booth A&B of type SKF 6205-2RSH/C3

77
 BIBLIOGRAPHY

Abramovic. (1987). Vertical Axis wind turbines: asurvey and bibliography. Wind engineering, 334-343.

Alaimo. (2015). 3D CFD analysis of a vertical axis wind turbine. energies, 3013-3033.

Allet. (1997). On the aerodynamic modelling of VAWT. Wind Engineering, 351-365.

Battisti. (2012). Gli impianti motori eolici. Battisti Editore.

Battisti, L. (2016). Analysis of different blade architechtures on small VAWT performance. Journal of physics
conference series.

Biadgo. (2013). Numerical and analytical investigation of vertical axis wind turbine. FME Transaction, 49-58.

Brown. (2010). Design and Analysis of A vertical axis thermoplastic composite wind turbine blade. Plastics, Rubber
and Composites, 111-121.

Brusca. (2014). Design of a vertical axis wind turbine: how the aspect ratio affect the turbine's performance.
international journal Energy environment engineering , 333-340.

Burton. (2001). Wind energy handbook. Sussex: John Wiley & Sons.

Carrigan. (2012). Aerodynamic shape optimization of a vertical axis wind turbine using differential evolution. ISRN
renewable energy, 16.

Castelli. (2013). Numerical Evaluation of the Contribution of internal and Aerodynamic Forces on VAWT blade
loading. Aerospace and mechanical engineering.

Castelli, M. (2011). The Darrieus wind turbine: Proposal for a new performance prediction model. Energy, 4919-
4934.

Castillo. (2011). Small-scale vertical axis wind turbine design. Tamper university of applied scince .

Chang. (2005, october). Advanced topics in environmental engineering - wind power. Retrieved from University of
New Brunswick: http://www.ece.unb.ca/powereng/courses/EE6693/index.html

Claessens. (2006). The design and testing of airfoils for applications in small vertical axis wind turbines. Delft:
Delft university of technology faculty of aerospace engineering.

Darrieus. (1931). Turbine having its rotating shaft transverse to the flow of the current. Us patent.

78
Duncan. (1962). An elementary treatise on the mechanics of fluids. Edward Arnold Ltd.

Edwards. (2008). Wind tunnel and numerical study of a small vertical axis wind turbine. 46th AIAA Aerospace
Scinces Meeting and exhibit. Nevada.

El.Ali, M. (2007). Comparison between solar and wind energy in lebanon. 9th international conference on electrical
power quality and utilisation. Barcelona.

Ferreira. ( 2007). Simulating dynamic stall in a 2D VAWT:modeling strategy, verification and validation with
particle image velocimetry data. Journal of physics, Vol. 75.

Golding. (1976). The generation of electricity by wind power. New York: John Wiley.

Gorlov. (1998). Development of the helical reaction hydraulic turbine. US department of energy.

Gupta. (2010). Computational fluid dynamics analysis of a twisted three bladed H-Darrieus rotor. Journal of
Renewable and sustainable energy.

Hill. (2009). Darrieus turbines: the physics of self starting. jpurnal ofpower and energy, 21-29.

Islam. (2008). Aerodynamic models for Darrieus type straight bladed vertical axis wind turbines. Renewable and
Sustainable Energy Reviews, 1087-1109.

Islam, M. (2008). Design analysis of fixed-pitch straight bladed vertical axis wind turbines with an alternative
material. Windsor: Mechanical, Automotive and materials engineering department, University of windsor,
Canada.

King. (2004). Farmers of forty centuries: Organic Farming in China, Korea and Japan. Courier Publications.

Klimas. (1981). Effects of blade preset Pitch/offeset on curved-blade Darrieus vertical axis wind turbine
performance . Sandia National Laboratories.

Lanzafame. (2014). 2D CFD modeling of H-Darrieus wind turbines using a transittion turbulance model. Energy
procedia, 131-140.

Lazauskas. (1992). Three pitch control systems for vertical axis wind turbines compared. Wind Engineering, 269-
282.

Marten. (2013). QBlade: An open source tool for design and simulation o f horizontal and vertical axis wind
turbines. International journal of emerging technology and advanced engineering , 264-269.

(2005). MECH 4010 Design Project Vertical Axis Wind Turbine. Dalhousie: Department of Mechanical Engineering
Dalhousie university.

Muller. (2009). Vertical axis resistance type wind turbines for use in buldings. Renewable energy, 34.

Mulugeta. (2015). Numerical and analytical investigation of vertical axis wind turbine. Belgrade: Faculty of
mechanical engineering .

Paraschivoiu. (2002). Wind turbine design. 148.

Premkumar. (2015). Numerical studies on the effect of cambered airfoil blades on self starting of vertical axis wind
turbine part1: NACA 0012 and NACA 4415. Applied mechanics and material, 250-254.

79
Quang. (2014). Flow-driven rotor simulation of vertical axis tidal turbines: A comparison of helical and straight
blades. International journal of naval architecture and ocean engineering , 257-268.

Sabaeifard. (2012). Determinatin of vertical axis wind turbines optimal configuration through CFD simulations.
International conferance of future environment and energy OPCBEE vol.28.

Sathyajith. (2006). Wind energy fundamentals, resource analysis and eco nomics. springer.

Schaffarczyk. (2014). Introduction to wind turbine aerodynamics. Springer, 1-12.

Shepherd. (1994). Historical development of windmill . wind turbine technology: Fundamental concepts of wind
turbine Engineering ASME, 1-46.

Strickland. (1975). The darrieus turbine: a performance prediction model using Multiplestream tubes. Albuquerque:
Sandia National Laboratories.

Strickland. (1986). A review of aerodynamics analysis methods for vertical axis wind turbines. 5th ASME wind
energy symposium, (pp. 7-17). New Orleans, USA.

Templin. (1974). Aerodynamic performance theory for the NRC vertical axis wind turbine. Ottawa: National
research council.

Tong, W. (2010). Wind Power Generation and Wind turbine design. WIT Press.

Torresi. (2008). Accurate numerical simulation of high solidity wells turbine. Renewable energy, 735-747.

Touryan. (1987). Electric power from vertical axis wind turbines. Propulsion and Power, 481-493.

Tullis. (2008). medium-solidity vertical axis wind turbines.

Vaishnav. (2007). An investigation on the aerodynamic performance of A vertical axis wind turbine. Durg: Bhilai
institute of technology.

Vallverdu. (2014). Study on vertical axis wind turbine using streamtube and dynamic stall models. Chicago: Illinois
institute of technology.

Vassberg. (2005). Revisiting the vertical axis wind turbine design using advanced computational fluid dynamics.
43rd AIAA Aerospace scinces meeting and exhibit, (pp. 12783-12805). Nevada .

Wang. (2011). 6-DOF numerical simulation of the vertical axis water turbine. Joint fluids Engineering conference.
Shizuoka.

Wilson. (1994). Aerodynamic behavior of wind turbines. wind turbine technology: Fundamental concepts of wind
turbine engineering, 215-282.

Wind Atlas. (2005). Retrieved from New and renewable energy authority:
http://www.nrea.gov.eg/Ar/Technology/Windatlas

Windfinder. (2018, July 6). Wind statics-Alexandria Airport/Nouzha. Retrieved from Windfinder:
https://www.windfinder.com/windstatistics/alexandria?fspot=alexandria_beach

80
 Yang. (2013). Effect of moment of inertia to H type vertical axis wind turbine aerodynamic performance. 6th
international conference on pumps and fans with compressors and wind turbines. Lanzhou: School of
energy and power engineering, University of technology, Lan zhou, China.

Zhang. (2014). effect of blade pitch angle on aerodynamic performance of straight -bladed vertical axis wind turbine.
Journal of centeral south university press and springer, 1417-1427.

81

View publication stats