Wind Turbine Final Report
Wind Turbine Final Report
Wind Turbine Final Report
Vertical Axis Wind Turbine to
Power a Well Pump
Wind Driven Well Pump Group 2
Adam Rusenko
David Hontz
Andrew Zhang
Josh Norton
ENGR 493
Professor Mike Erdman
December 15, 2014
2
Abstract
The group was tasked with finding a socially and environmentally sustainable way to
power a mechanical well pump in a developing community. After research into a variety of
vertical axis wind turbines, a combination of the Savonius and Darrieus designs was chosen. By
utilizing the Penn State wind tunnel, the design can be tested for power output at specific wind
speeds. A scale prototype was constructed so the design can fit in the wind tunnel. Although
more work is needed before a full scale design is implemented, the test results will provide
valuable details into improvements and future steps in the process.
3
Table of Contents
i. Abstract…………………………………………………………………….2
I. Introduction………………………………………………………………...4
II. Research……………………………………………………………………5
III. Our Design…………………………………………………………………6
IV. Prototype…………………………………………………………………...7
V. Budget……………………………………………………………………...8
VI. Results……………………………………………………………………...9
VII. Conclusions……………………………………………………………….10
VIII. Future Recommendations…………………………………………………11
IX. Works Cited………………………………………………………………12
4
Introduction
A major issue facing many communities in the developing world is access to water.
Water is a vital resource to not only health and survival, but also to agriculture and industry.
Lack of such a necessary resource can hinder the development in these areas. Without a
centralized infrastructure, these communities are forced to use other techniques like rainwater
collection to attain water. Still others need to travel a long distance to a well. By making the
process to acquire water easier and more efficient, we can provide the first steps to solving this
problem.
The initial guidance for this project came from a joint
IsraeliPalestinian organization, COMETME, that implements similar
infrastructure solutions to underprivileged communities in West Bank.
Examples of some of these projects include solar panels and wind
turbines used to pump water from a well to a holding tank. The challenge
that was posed by COMET and Penn State project advisor Andras
Gordon was to create a vertical axis wind turbine that would be able to
power a mechanical well pump. Additionally, the design should be easy to construct and
maintain by these communities where it is
applied. The goal for the project was to create
a solution that is both environmentally and
socially sustainable. Although the resources
are available for a highly technical project, it
would not be a maintainable solution in
developing areas. By focusing on the people
rather than the details of the system, the
possible influence of the project could be seen
on a global level, rather than a single
application.
5
Research
As four undergraduate engineering students without any prior experience in the field of
aerodynamics, we needed to spend a significant period of time researching these topics. We
started our preliminary research on general principles of converting wind into energy and how an
efficient wind turbine functions. Professor Andras Gordon gave us guidance as to which types of
vertical axis wind turbine would generate enough power to run the water pump. He provided us
with many resources about different types of vertical axis designs. For each type of design we
conducted extensive research to decide which designs
would be ideal for the conditions of the target area.
Factors that we considered when evaluating each design
were discussed and established with the guidance of
Professor Gordon. In order to have guidelines to
evaluate these designs by, we contacted COMET asking
for any of their data and they sent their comprehensive
report for their turbine for the past year. The report
included key information about wind speeds and power
produced versus the power needed to efficiently pump
the water. A low cutin wind speed (the speed at which
the turbine starts to rotate) would be necessary for the
turbine because the wind speeds were low and steady.
Their pumping efficiency was reported to be around 1.5
cubic meters per day to a holding tank that is ten meters
tall. We decided that our goal would be to generate
enough power to pump 3 cubic meters of water per day
in the same conditions of a 10 meter head and 3 m/s average wind speed. With these criteria, we
were able to accurately evaluate the designs.
Once we had a firm grasp on the general concepts of the project, we moved into research
about specific aerodynamic principles. Specifically, we worked with the Bernoulli principle of
lift and the properties of airfoils. We worked to design an airfoil that would maximize the power
output of our turbine. Once we had a general shape in mind, we met with Dr. Susan Stewart to
get an expert’s opinion on our airfoil design. Dr. Stewart received her PhD in Mechanical
Engineering from Georgia Tech in 2003 and has since been conducting research in the field of
aerodynamics at the Pennsylvania State University. Dr. Stewart gave us valuable guidance in our
design process of the airfoils and after meeting with her, we collaborated with Andras and
decided on a final design of our airfoils.
6
Our Design
After careful individual evaluation of each vertical axis design, we collaborated within
our group and with Professor Gordon on a final design. Our design can be broken down into two
distinct parts: the Savonius and Darrieus designs. The inner portion of our turbine will house the
Savonius aspect to the design. Flush to the central rod about which the turbine spins, two
halfcylindrical shells propel the turbine in a
circular rotation. On the outsides of the design
are the two airfoils that make up the Darrieus
aspect to the design. Each of the design aspects
add specific advantages to our hybrid design.
The two shells close to the central axis of
rotation will help the turbine to have a low
cutin wind speed. This is because the Savonius
turbine is designed specifically for areas in
which the wind speed is not very high. The
Savonius shells also ensure that the turbine will rotate regardless of the wind’s direction due to
the circular nature of the design. As the turbine spins with the initial push from the Savonius
design, the airfoils on the outside of the perpendicular axis will generate a higher tipspeed ratio,
propelling the turbine’s rotation faster than the Savonius. A greater tipspeed ratio means that
the blade will move faster in relation to the wind
speed, generating more power along with it. The
airfoils are modeled after the GOE 435 design.
With slight modifications to the design, we
made sure that the connection of the airfoils to
the perpendicular axis would be strong to
withstand a strong wind. We installed the
airfoils in a way that allows us to change the
pitch of the airfoils in order to experiment and
find the most efficient angle of attack and
direction of the lift force. This Darrieus design
will help generate the higher power output that
we require in our goals for the project.
In order to achieve the goal of 3 cubic meters of water pumped per day, the turbine will
have to provide at least 7 Watts of power to the pump. By calculating the available power of the
7
wind, assuming 8% efficiency of the turbine, the necessary dimensions of the full size design
will be two meters tall by three meters in diameter of rotation. The calculation is shown below.
W ind P ower = .5 × Air Density × Swept Area × W ind Speed3
T urbine P ower = 7W = .5 × Density × Area × (3.34 m/s)3
Area ≃6 meters2
Prototype
Building our prototype based on our design was a tricky task to say the least. One of the
factors that made it so difficult was the budget. Our job was to build the working prototype in
the most inexpensive way possible because the targeted area is 3rd world areas. This lead to us
having multiple challenges with finding materials to build with. Our biggest concern was the
airfoils so that was where we started. At first we had the hopes to 3d print them, but this had 2
problems. First, this feature is inaccessible for 3rd world countries so it wouldn’t help our end
result, and second it is extremely expensive to 3d print the size airfoil we needed because it was
too big for the school 3d printer and we would have to have it shipped in from a separate
company. This led us to how we designed it in our final prototype. We carefully cut small
sections of the airfoil and connected them all with a rod. We then wrapped them in a metal sheet
to allow for it to be completely covered, and the inside being mainly hollow allowed for them to
be lightweight.
From here we moved on to the savonius part. We needed 2 half cylinders 12 inches in
height and 8 inches in diameter. Finding this specific dimensions were not easy so we went with
a different approach. We found a long empty cylindrical shaped piece of metal that was 8 inches
in diameter. We then cut it into 2 pieces of 12 inches in length to give us our base of the
savonius. From here we cut 4 blocks of wood that would act as the top and bottom of each
savonius piece and nailed them together.
Finally, we took our connection pieces, that would connect the airfoils to the center rod,
and cut a hole in the middle that would allow for the connecting pieces to slide onto the center
rod. From here we had all the pieces built and ready to be assembled. To start the assembly we
first put our ball bearing on one end of the rod and secured it using the screws that came in the
bearing. From here we measured the distance from the bearing to where the start of the PVC
fitting, that holds our savonius and airfoils together, and we slid the fitting over the center rod
and bolted it in place. From here we slide on the connecting pieces and then attached one end of
our airfoils to the connecting piece using adjustable bolts. These adjustable bolts allow for us to
easily tighten and loosen the airfoil in order to change the pitch without tearing through the
wood. Next we attached the savonius parts by overlapping them around the center rod and to
hold it all together we put bolts through the bearing to the other side of the wood piece holding
8
the savonius part together. This allowed for all the pieces to be attached and move as one while
holding it steady on the center rod.
We then repeated this process backwards for the second PVC fitting. The final ball
bearing will be attached once we are ready to test in the wind tunnel and the extra length on the
center rod is to allow for part of the rod to exit the wind tunnel. This part the exits the wind
tunnel will have a small hole drilled into the end where we will insert cords of a DC motor. This
will allow for us measure the volts generated from our turbine rotating. The ball bearings will be
drilled directly into the walls of the wind tunnel to hold the design in place.
Budget
The funding for this project was provided by Lockheed Martin. The goal for the
prototype was to keep the price as low as possible using cheap, local materials. However, two of
the pieces could not be made using the given resources. The two ball bearings and the two PVC
fittings were ordered online from the manufacturer. These pieces accounted for around half of
our budget. The other half of the budget consisted of the materials needed to build the prototype.
However, many of the materials bought were only available in far greater quantities than we
needed. Another prototype could likely have been constructed with the materials already bought,
so the cost of construction for one prototype was less than the existing budget. Nonetheless, the
budget for the project, including all of the purchases, is listed below.
9
Item Quantity Price
Grainger Flange Mount Ball Bearing 2 $54.36
9V Magnetic Cylindrical Mini DC Motor 1 $5.62
Spears 856 Series Flange PVC Pipe Fittings 2 $14.84
10 ft x 10 ft Aluminum Flashing 1 $8.78
Gorilla Glue Epoxy 1 $5.48
8Pack #6 1.5 Inch Nuts and Bolts 1 $1.24
8” x 24” Round Metal Pipe 1 $5.20
Balsa wood sheets 2 $9.98
Wood circle 1 $2.49
Xacto knife 1 $4.49
Wooden rod 2 $5.56
Wooden block 1 $4.47
Wooden plank 1 $7.40
Carpet tacks 1 $1.30
Wood plank 1 $2.17
Screws 3 $3.54
1.5 inch nails 1 $1.30
1.25 inch nails 1 $1.30
Wooden moulding 5 $4.65
Wood screws 3 $3.54
Oak Dowel Rod 2 $7.56
Oak Dowel Rod 1 $1.14
10
Project Total: $160.07
Results
After testing the design in the wind tunnel, the results turned out to be largely
inconclusive. This obscurity was caused by improper test conditions. The plan to test the power
output of the turbine was a small, DC motor with a turn shaft connected directly to the central
axis of the turbine. However, the DC motor, and electrical power, is only useful at large
revolutions per minute values. Due to the size and orientation of the vertical axis turbine, it is
very difficult to create the necessary RPM to output a noticeable voltage from the motor. The
horizontal axis turbines are much more efficient for electrical power generation. For the
application of a mechanical well pump, the best design is a vertical axis orientation because it
generates the necessary torque. There was no way to test the torque of the machine, which is the
most important value for the application. Regardless, some simple measurements were taken,
and are shown in the table below.
Volts Velocity revolutions sec rpm
(ft/s)
As seen in the table above, a small voltage was created by the motor as the wind speed
increased. Another problem that arose from the motor was the load. Without a high enough
RPM, the load attached to the motor, a resistor in this case, has little to no effect on the turbine.
To obtain valuable results for COMETME, the design will need to be tested with a mechanical
load to test for torque.
Additionally, the startup wind speed of the turbine was larger than expected. The turbine
did not begin to turn until the wind in the tunnel reached nearly 17 mph (8 m/s). With an average
wind speed in the area of application for COMETME at just over 3 m/s, the full scale design
should start at a very small wind speed. Even with the scale speeds in mind, the prototype should
ideally start at a lower wind speed.
Lastly, after seeing the turbine in the tunnel and a more detailed analysis, tunnel operator
Rick Auhl believed that our airfoils, that were meant to increase the tip speed ratio using the lift
force, were not functioning as expected. The lift force of the airfoils did not provide any further
rotational speed for the design. Instead, the only benefit the airfoils provided was a “deflector,”
pushing the wind toward the Savonius part of the design. Since the turbine rotates on its axis, the
orientation of the airfoils is constantly changing. Ideally, the orientation and pitch of the airfoils
would be constant in relation to the direction of the wind. With the airfoils constantly moving, it
would only obtain the desired lift force an one instant over a whole rotation. While the
expectation of the airfoil influence on the rotation was small to begin with, the lack of any sort of
influence was unexpected and disappointing.
Conclusions
In conclusion, we have been tasked with designing a wind powered water pump that is
intended for use in rural areas. We chose the design after extensive research into different
vertical axis wind turbine designs and combined two separate existing designs so as to make our
12
design more efficient in both low and high winds. The design had to be simple and easy to
construct as rural areas do not have many tools or good equipment. It must also be able to be
constructed using cheap, local materials as the areas it is intended for will often not have stores
where they can buy materials but will only have spare items or repurposed materials. We feel
that our design is easy to construct with materials that can be found lying around anywhere like
wood and sheet metal or plastic sheeting. We faced many challenges in choosing our design as
we had to make sure that it met all of our criteria. We looked at many interesting designs but
many of them were too complex and difficult to build. Our design is socially and
environmentally sustainable because it can be built with cheap, local materials, uses only
renewable energy with no emissions and can be easily built to help communities get water easily
and effectively. We also tested our prototype in the wind tunnel and measured the voltage that
the prototype produced as well as the RPMs at different speeds. Our start up speed was about the
same for all of our tests and our results showed that our prototype had fairly linear RPMs with
respect to wind speed which is good because it will be consistent and you will know what you
are getting from it. Our design worked fairly well and we were very pleased with how it worked.
However, there is room for improvement in our design as we did not have the time to fine tune
our design to make it the best that it could be. We also did not have the time to do a detailed
analysis of our design so we do not know if our design is as efficient as it can be.
Future Recommendations
Moving forward with this project, there are certain improvements that we would hope to
implement. With respect to the design, certain changes to the Savonius design would increase the
efficiency of the rotation. As our design stands now, the two shells are perfectly circular. This
creates a force of the wind on the trailing shell when the wind is directed perpendicular to the
shell. A solution to this would be making the shape of each shell more aerodynamic than a
13
perfect semicircle. Another factor that this could help with would be to diffuse the wind off of
the ends of the shell toward the airfoils, generating a stronger lift force and tipspeed ratio.
Ropatec, an Italian wind turbine company, has perfected the shell diffuser in their designs of the
hybrid SavoniusDarrieus turbine, the TVision.
The means of changing the pitch of the airfoils is fairly crude and imprecise on our
prototype. A device that could automatically change the pitch would greatly increase the
efficiency of the turbine. The device would have to have a means of measuring the wind’s speed
and direction and then automatically change the angle of attack of the airfoil. Airplanes have
achieved this implementation with ringlike structures around their propellers. The dual airfoil
design also changes the purpose of the airfoils to more of a wind deflector rather than generating
lift. More lift could be generated if the angle of attack of the airfoils were more precise and
aerodynamic. Much of the future work on the efficiency of the design should be focused on
making the airfoils more aerodynamic.
Teams in the future could benefit from some insight into the testing process. As
discussed in our results, the DC motor was not successful in calculating any of the generated
power because it measures revolutions per second. After discussions with Rick and Professor
Gordon, the solution decided upon is to use a torque measuring device. Because the force is
directly applied to the mechanical pump, torque is more important that a high power output.
Since an automated torque measuring device is expensive, it is not practical without department
approval to purchase one specifically for this purpose. Rick proposed the idea of a manual
device that would measure the torque: something that would apply a resisting force to the axis of
rotation and measure the time it takes for the rotation to stop. Makeshift ideas that were
brainstormed included a brake pedal with a force sensor or making a de Prony brake as a
dynamometer. With these devices, an accurate measurement of how efficient the turbine is can
be used to evaluate what aerodynamic properties need to be changed.
In the future, we would like to see this design implemented successfully by companies
like COMETME and implemented globally with many different applications. In order to
successfully achieve this goal, we would plan to work with a lamination school in Maine to
develop a way to easily construct certain parts of the turbine. The plan as of now is to package
the each turbine into a kit that would make the assembly simple enough for a person without an
engineering background to assemble with ease. The main parts of the turbine would come
already assembled with the help of the lamination teams in Maine. Preliminary requirements for
the lamination would include a very detailed 3dimensional CAD drawing of the airfoils and the
Savonius shells. If these steps are taken, then we can certainly make a positive impact towards
helping developing worlds thrive.
14
Works Cited
Dotan, Noan. Pictures Published by Comet Middle East. September 2013.
GOE 435 AIRFOIL (goe435il)
"GOE 435 AIRFOIL (goe435il)." . Airfoil Tools, 2014. Web.
Dr Magdi Ragheb
Ragheb, Magdi. "Vertical Axis Wind Turbines." . University of Illinois
UrbanaChampaign, 19 July 2014. Web.
ROPATEC
"Ropatec Takes Advantage from the Wind." . ROPATEC S.r.l., 2013. Web.
Penn State Personal Web Server
"Susan Stewart Personal Page." . The Pennsylvania State
University, n.d. Web.