Notes - Wind Turbine ISR-G
by Leon Santen
These are notes for my wind turbine independent study with Jeff Dusek.
- 1 11/12/2020
- 2 11/05/2020
- 3 Tower design
- 4 10/29/2020
- 5 10/22/2020
- 6 10/15/2020
- 7 10/01/2020
- 8 9/24/2020
- 9 9/17/2020 - Wind Profiles
- 10 9/12/2020 - I began Coursera Course
- 11 9/10/2020 - First Semester Call with Jeff
- 12 References
A tower design with four pressure-treated 4x4s or 6x6s that are connected via diagonal crossbeams on the outside and plateaus seem reasonable.
On this website you can find some inspiring images.
Small Scale Wind Turbines Optimized for Low Wind Speeds
Letcher, T. (2010). Small scale wind turbines optimized for Low Wind Speeds.
10 Wind Turbines That Push the Limits of Design
Design and Manufacture of a Cross-Flow Helical Tidal Turbine
Anderson, J., Hughes, B., Johnson, C., Stelzenmuller, N., Sutanto, L., & Taylor, B. (2011). Capstone Project Report: Design and Manufacture of a Cross-flow Helical Tidal Turbine. University of Washington, Washington.
- think about Gorlov blade design - Onix 1 3D printer - prints with carbon embedded - build volume 320 x 132 mm 154mm tall
- scale down based on reynolds number - relevant dimension: qart of the foil, dynamic viscosity paper
- look into motor shaft sealing water - propellor shaft seals - drive shaft seal
- final deliverable
- tower design
- email STI files to Tim or Daniela
Newly structured wiki page for wind turbine system
All research insights, prototyping steps, and other system information is documented in Wind turbine system.
Twisted Savonius V1
We attempted to prototype a twisted Savonius turbine to allow a more continuous application of force when the wind turbine spins. Ideally, both sails would be slightly offset to each other to allow airflow from one airfoil to the other. Such an offset can be seen in the following picture that inspired our design. As the Savonius turbine is based on drag, it has a high starting torque, which will allow for lower self-starting speeds.
With our next prototype, we seek to combine a twisted Savonius turbine with a Gorlov turbine (Rajdeep Nath Dr. John Rajan; How to Design and Fabricate a Vertical Axis Wind Turbine: Design, Analysis, and Fabrication using Gorlov and Savonius Blades (p. 75)). We can 3D print the blades but might want to scale down the prototype to decrease printing time.
How to charge our batteries
We seek to charge our battery system with the energy obtained from the wind turbine. However, our current system diagram connects the AC rectifier to the inverter, which outputs 110 V alternating current (US outlets). As the energy output from the wind turbine will vary quite a lot, we need to charge the batteries with direct current (DC). For that purpose, we might have to acquire a wind turbine battery charge controller that can handle high voltages (0-600 V from the AC rectifier) to charge our 24 V battery system. Tae Chang offers a Battery Charge Controller. We probably need a 600 VDC 24v wind turbine MPPT charge controller 2500W. Most charge controllers seem to be in the range of 200-300 VDC (example), which might be enough considering wind speeds in our area.
We considering building a wooden structure on top of the house to elevate the wind turbine. A tripod design that attaches to the two bump-outs and the gable would make the tower accessible. We are currently considering a tower height of 15 feet up to 22 feet.
Guy wires are an option to secure the turbine as it will be quite large to account for low wind speeds. However, a pole that is merely held up by guy wires is not accessible enough.
Materials - stress and strain
Forces on VAWT
Meeting at Farm - Agenda
- where does everyone's interest lie?
- go over the system diagram and questions
- what are our resources? Whom do you know? - Dylan and maybe folks from Appalachian State University
- look at to-dos
- find a meeting time - Tuesdays 5PM
- assign rough groups and projects (tower, shaft, blades - Jasmine, Seba; electrical system - Riley, Odalys; fluids analysis - Seba, Nicola)
Aurora 3kW inverter - data sheet
Aurora wind interface box/rectifier - manual
TCMG-3kW Taechang N.E.T. Generator - online specs
3kW Discharger Breaker -
3kW Dynamic Braking Resistor - instruction manual
Wind direction meter
Moving forward, we will do these things by next Thursday:
- choose design - CAD turbine - tower design thoughts and calculations
- alternating current research - how do you work with it? - location, what tower is going to look like, how strong should it be - tower mounting - transmission - 3D print several pieces, wrap-around fiberglass or carbon fiber - find a weekly meeting time - connect wind direction meter to pole - which type of wind turbine should we prototype? - how can we get a system up the fastest? Sheet metal, etc... - read through all manuals and understand the purpose of all components - how can we charge the 24V battery system in the house? - CAD our house and run fluids analysis - assign projects to people - timeline - analysis of the tower torque
System diagram - wind turbine
Anemometer mount on roof
The anemometer is mounted at an altitude, 8.85 m (29 ft), where the wind turbine could possibly go. However, it is slightly offset from the gable. Ideally, the wind turbine would sit on the gabble as the wind speeds up when it passes over that point. Furthermore, I added a real-time clock to the system and connected the Arduino to a power supply that is independent from the inverter, which is turned off every night.
Approximate size of vertical axis turbine
Mechanical power = 0.5 x p x A x V^3 x Cp
Cp for Savonius turbine = 0.05-0.1
Cp for Darrieus turbine = 0.3-0.38
Cp for Gorlov turbine = 0.3-0.35
During extremely windy days, we will wind speeds up to 15 m/s and a power-coefficients of around 0.08 with a refined Savonius turbine prototype. Thus, a Savonius turbine would have to cover an area of 17.83 m^2 (2m x 8.9m) to output 3000 W of mechanical power. In comparison, a Darrieus or Gorlov turbine with a power coefficient of 0.3 would only require an area of 4.7 m^2 (2m x 2.377m) and will lighter due to its mechanical features.
As the wind speed during normal days won't be higher than 6 m/s, we probably want to optimize for medium wind speeds.
area of Savonius turbine at 6 m/s: 278 m^2 area of Gorlov or Darrieus turbine at 6 m/s: 74 m^2
- how can we charge the 24V battery system in the house?
- mount wind direction sensor and determine average wind direction
- create CAD model of the house to run fluids analysis. How will the "bump-outs" affect wind from different directions? Will they channel the wind?
- kick-off with wind energy group on Saturday
- research sheet metal Savonius wind turbine options
- write up report with learning insights from Coursera course
- literature search
Questions for Elizabeth - Apalachian State Univerity
- do you have any turbine blades?
- could you help us with composites?
Wind speed measuring
We are using a Vortex wind sensor. One revolution per second equals 2.5 mph. Since our anemometer has a relay (a mechanical switch), it creates a switch bounce. Therefore, we need a debounce circuit.
What I did
- data logging to SD card with Vortex anemometer and Arduino
- site observation and assessment
- how does Aurora rectifier work?
- create datasheet list
- finish structural mechanics section
- write up report with learning insights from Coursera course
- apply learning insights to current project - site assessment
- improve portable stand for data logger anemometer
- look at the generator optimal RPM range (do I need a transmission?)
- find manuals and data sheets
- system diagram for electrical and mechanical components and identify pieces that we currently don't have
- does brake monitor RPM?
- look into bicycle components for gear shifting
Forces on the blade
Moving forward, I will:
- I will look into saving data on SD-card with an Arduino. I will also try to set up Arduino MATLAB.
- order tech components (sd card reader, sd card, voltmeter...)
- Mount new anemometers, and figure out how to read their outputs.
- CAD solar panel mount. (I just couldn't find the time to do that last week.)
- clear out the area for solar panel mounts and start locating post holes
- Reassess locations for turbines. I have been trying to find spots, but I will approach this question more professionally.
- find fluid dynamics book?
- take more quizzes
Wind turbine terminology
- rotor - generates aerodynamic torque
- nacelle - converts torque into electrical power
- tower - hold nacelle and rotor blades
- foundation - hold the whole turbine in place
Main degrees of conventional Darrieus turbine
- azimuth - rotation of rotor to generator
- yaw - rotation of nacelle about vertical tower
- pitch - angle/ pitch of blades about their lengthwise axis
- tubular towers
- lattice towers
- tripod towers
Foundations are oftentimes made out of concrete.
Wind energy extraction
The mechanical power of a turbine can be calculated with the equation to the right. The Betz limit on the power of a fast-spinning wind turbine (59%) describes the maximum energy that a turbine can extract from the wind.
Rotating lift-based machines
Horizontal axis wind turbine, Darrieus turbine, vertical axis wind turbine with blades,
Rotating drag-based machines
Vertical axis turbines - similar to cup anemometer; Savonius (efficiency 10%-15%) could be nice for aesthetically pleasing turbine
Flying lift-based machines
pulls up sail that pulls on a cord that spins spindel
Machines using flow-induced vibrations
Our revenue is the production of net electricity in Wh (Annual Energy Production): Capacity of Farm x 8760h x capacity factor 0.25
1 - Simple payback time (SPT)
- estimate of annual production in Wh
- annual revenue - production x energy sale price
- annual operating costs
2 - Net present value (NPV)
(revenue - costs) - original investment
if less than 0, probably not profitable
excel has NPV function
3 - Levelised cost of Energy (LCoE)
goal is to find cost per MWh that can be used for comparison:
(Capital Investment + Operational Costs + Decommissioning Cost)/ MWh
Wind drag on solar panel plate
Individual solar panels have an aspect ratio of 1:2 (62.2 x 31.8 x 1.4 inches). The current plan for the five panels is to mount them side by side resulting in an aspect ratio of 1:2.56 (62.2 x 159 x 1.4 inches).
F = 1/2 * p * v^2 * A * 2pi * sin(alpha); (p - density of air, v - wind speed, 2pi sin(a) - drag coefficient); density of 10 celsius air - 1.246 kg/m^3;
solar panel dimension - 62.2 x 31.8 x 1.4 inches - 1.58m x 0.8m x 0.0355; area = 1.264 m^2
Lift on all five panels: F = 0.5 * 1.246 kg/m^3 * (17.8 m/s)^2 * 1.264 m^2 * 5 * 2pi * sin(a) <= 7835 N
- Maximum Wind Speeds in the Southern States
- Paper on wind lift on photovoltaic panels - Researchgate
- The Flat Plate Airfoil
Remote Wind Speed Sensing
- two challanges: low noise to signal ratio, need to measure volume not single point
- LIDAR & SODAR Doppler shift measuring; LIDAR measurements are more precise
Statistical Analysis of Wind Speeds and Turbulence
- Navier-Stokes equation (evolution equation) to calculate wind speeds in space
- statistical analysis, spectra, turbulence intensity is obtained by mean wind speed and standard deviation of wind speeds
- every wavelength (after Fourier transform) can be thought of as a length scale
- wind vector (u, v, w)
- integral length scale; from time series, we compute the auto-correlation function
- turbulence spectra
- averaging periods of 10 min are commonly used; 30 min period for turbulence studies; the larger integral time scale, the larger should be averaging period; sampling frequencies should be much smaller than integral time scale
- you have to detrend te time series to get rid of high-frequency fluctuation
Transition to turbulence begins when some flow instability (such as the instability analyzed in sections (Bkc) and (Bkd)) leads to some fairly large scale disturbance(s) or “eddies” in the flow field. As these disturbances gather energy from the mean flow, they begin to spawn smaller disturbances or eddies which, in turn spawn even smaller eddies. This process ends because, eventually, the eddies reach a size for which viscous effects become important and the very small eddies are damped out by viscosity. Eventually, the spectrum of spatial or temporal eddy sizes reaches a “fully developed” state in which energy is fed from the mean flow into large eddies and then continually cascades down to smaller and then smaller eddies eventually reaching a size at which viscosity becomes important and damps out those small eddies. In this fully-developed state the disturbance energy for any one size of eddy becomes relatively constant though it can, of course, continue to change with the flow conditions.
- write up questions about auto-correlation function
- include lift coefficient in flat plate lift equation
- reading on the efficiency of vertical turbines vs conventional turbines
- interview farmer Ann again
- ask Ayden to mount anemometer
- CAD solar panel construction, include Jasmine
9/17/2020 - Wind Profiles
What I did:
- I measured wind speeds throughout the week. At a height of 4 m, the wind speeds vary around 0-4 mph. We had gust of up to 10-15 mph. However, wind speeds seem to be quite low. From talking to a nearby farmer I learned that wind speeds increase during the months November - January.
- I completed the first two weeks on Coursera, but I didn't take the quizzes due to limited time
- We have been working on a mount for the five solar panels. We will most likely build a solar-sail that rotates around its middle axis. Its angle of rotation can be manually adjusted for winter and summer times. We will have to dig at least four post holes to keep the construction sturdy during heavy storms.
- the surface roughness is important for wind energy estimations
- surface roughness is hard to determine
- orientation of the turbine
- mount anemometer higher up in the trees
- look into turbine tree mounting techniques
- finish week three and take week 2 quizzes
- work when it doesn't rain
- what's the surface roughness here at the farm?
- find possible turbine spots
- research boundary layers and atmospheric stability correction
- look into calculating the lift of solar panels (flat plate approximation - coefficient for lift and drag)
9/12/2020 - I began Coursera Course
- we have a hight surface roughness probably around z_0=0.4 (wind profiles)
- Eddies are swirls of a fluid and it's reverse current in a turbulent flow regime (Wikipedia)
Wind Resource Assessment
- due to roughness of the forest, wind speed increases
- The Wind Atlas Analysis And Application Program (WAsP) provides an upward and downard analysis of the terrain
- on top of hills, we have over speeding due to continuity
- What is Weibull distribution?
9/10/2020 - First Semester Call with Jeff
site assessment - take photography and topography
data collection with raspi anemometer
hard part: is there enough detail in the Coursera course?
deliverables: structural design of design the tower, beam bending, blade design --> lifting line theory
model for the amount of power --> modeling in Simulink --> amount of wind, size of turbine...
Grade? --> a few deliverables: weekly journal entry/blog, Coursera quizzes (weekly entry), site assessment, modeling,
Get started: Coursera, as you work trough (document built-in quizzes) Do Simulink onramp
To-do for next week:
- Write up ISR-G application
- First week or two of Coursera course, wind resources, tests, and measurement part
- Personal notes in a google doc, final Report - technical report in latex, final deliverable that is more public-facing
- Araya, D. B. (2016). Aerodynamics of vertical-axis wind turbines in full-scale and laboratory-scale experiments (Doctoral dissertation, California Institute of Technology).