Fall 2011‎ > ‎ME43 - wind‎ > ‎




The WhaleWurbine is a three-bladed horizontal axis wind turbine (HAWT) designed to charge a cell phone battery. The power rating of a typical cell phone battery is 3W; so we designed the WhaleWurbine to generate 3W under steady 10 m/s wind speeds. We designed the WhaleWurbine for the required Mechanical Engineering Senior Design Project for The School of Engineering at Tufts University. The goals of our project were (1) to  understand the hardware and software requirements of a small scale wind turbine, (2) analyze specifications for a suitable generator, and (3) build a prototype for testing and presentation. We conducted multiple iterations of experimentation, assembly, and testing to achieve our final product.

The wind tunnel testing results show that the power output of the final WhaleWurbine design exceeded our original power output goal of 3W by over 100%. The WhaleWurbine generates 7.31W in conditions of 10 m/s wind speed and maximized resistance load of 3 ohms.

Features and Design

In this section we discuss the salient features of the WhaleWurbine. Pictures are included in the images section.

1) Built-in Rotor Encoder:

The WhaleWurbine incorporates a fully-integrated photo interrupter which measures instantaneous blade speed. The encoder is connected to an Arduino micro-controller and it is physically attached to the rear of the generator. Such an arrangement provides a seamless and fully enclosed setup. The Arduino uses the output from the photo interrupter to:

  • Calculate and display speed in rad/s on an LCD screen.
  • Control the speed the blades and generator.

Image showing the attachment of photo-interrupter and encoder disk to the rear of the generator.

2) Control System

The purpose of the control system is to reduce the effective load resistance across the generator so that the generator can reach it's desired speed of 4500 RPM. If the the load resistance across the generator is low, the generator will spin faster compared to when the load resistance is high. Reduction of load resistance is achieved by rapidly opening and closing the charging circuit; a technique also known as PWM (Pulse Width Modulation).

The Arduino micro-controller sends a PWM signal to a solid state relay, IRF510, to reduce the effective load resistance. The PWM signal has a magnitude or duty cycle that is proportional to blade speed: low PWM duty cycle (0) implies low effective load resistance and low resistive torque on generator whilst high PWM duty cycle (255) implies high effective load resistance and highest resistive torque on generator.  The Arduino reduces effective load resistance only if blade speed is less than 4500 RPM; if blade speed is above 4500 RPM, the Arduino sends a PWM signal of magnitude 255 to maximize effective load resistance. The Arduino software the we used can be downloaded here.

Electrical circuit for WhaleWurbine

3) User Interface


The hardware interface consists of an Arduino, push buttons, and an LCD screen which are enclosed in a white acrylic box. The Arduino generates the information that is displayed on the LCD screen and takes user input through push buttons. In addition, the Arduino sends data  about speed of blades, voltage across the generator, and current generated by the generator to a computer over USB.
User-interface box


The LabView user interface is easy to use and clean. The purpose is to graph certain electrical parameters generated by the WhaleWurbine. These include, but can be modified for others, Voltage, Power, Current, and RPM. There are buttons that display each graph individually, so as to mimic the feel of the interface, as it would be physically implemented into our turbine system.  The graph display could also be changed to a number display to consume less space for easier integration in future developments.

WhaleWurbine software user interface

The graphs are updated live, pulling a set of data every few milliseconds from the Arduino.

4) Blades

The blades feature the NACA 4412 airfoil shape. We rotated the angle of attack along the length of the blade to optimize the tip speed ratio at a value of 6-- this means the tips of the blades turn at a speed that is six times the speed of the wind. We performed multiple iterations that focused on redesigning the blades of the turbine because this largely affects the performance and power output of the turbine in the wind tunnel. Iterations were first centered on blade design and blade width, and then geared towards structural stability at the hub-to-blade interface to prevent failure at high rotational speeds. The blades were first created in Solidworks by importing the coordinates of the NACA 4412 airfoil and then rotating these curves at the desired points along the blade.  They were 3D printed out of ABS plastic.  The three blades and the hub were each printed as separate parts and then assembled after bring printed.

                                                                             WhaleWurbune blade and hub

5) Enclosure

The enclosure is an aerodynamic structure that surrounds the generator, shaft, coupling, tachometer, and a portion of the controls system. The enclosure provides protection for the generator, tachometer, and controls system. The enclosure consists of two parts: the bottom part attaches to a cylindrical fixture to hold steady in the wind tunnel and for presentation; the top part fastens to the bottom part with two screws, one located on each side of the enclosure. The two parts join together in a sleek, compact design. We designed the enclosure to fit tightly around the assembly to reduce vibration that could lower the efficiency of the turbine. Holes along the top, front, and back sides of the enclosure allow heat to escape from the generator to avoid overheating.

Exploded view of the enclosure

Two types of models were created for the system, one mathematical and one software based.

Our mathematical model for estimating torque was derived from the equations involved with the Betz limit, power, and lift and drag force equations. The equation used to estimate torque in was the following:

From previous work done with our chosen airfoil the optimized angle of attack of 4 degrees was used in the blade design. To find the proper coefficient of lift a linear fit was applied to the chart below and used:

Our apparent velocity was calculated using our assumed efficiency for the Betz limit and our setting angles.  Using MATLAB the torque was integrated over each section (corresponding r and l values) and multiplied by three because we had three blades. These calculations showed in increase in torque along the blade, with the most coming from the last section before the tip. Our mathematical model predicted a torque of .64 N-m.

After the mathematical model was created, the software model WTPerf was used as a data check for eventual comparison with experimental results. By coding the WTPerf input files for blade design (blade twist, blade radius, hub radius, etc), basic operating conditions (air density, etc), and desired outputs (power, torque, etc), we obtained an expected power output of 8W based on software modeling.

Feed our fish please!

The gadget spec URL could not be found

Subpages (2): Data Comparison Images
Michael Bai,
Dec 12, 2011, 5:51 PM
Michael Bai,
Dec 12, 2011, 5:51 PM
Michael Bai,
Dec 12, 2011, 5:50 PM
Michael Bai,
Dec 12, 2011, 5:50 PM
Michael Bai,
Dec 12, 2011, 5:51 PM
Jonah Kadoko,
Dec 19, 2011, 9:57 PM
Jonah Kadoko,
Dec 19, 2011, 10:08 PM