The Evolution of Small Wind Turbines
By Ryosuke Ito
President, Zephyr Corporation
“Project-Z” began in the summer of 2002 in Japan with the goal of developing small wind
turbines that could really generate practical and efficient electric power with true rigidity. This
was the project which gave birth to the “Airdolphin Mark-Zero”, which is gradually gaining
In this report, I would like to discuss why only a few consumers currently use small wind
turbines, the reasons why large wind turbines cannot always be an ideal model of small
ones, and how small wind turbines need to outperform large wind turbines in order to be
practical. In addition, I would like to explain why the performance of the Airdolphin is
outstanding, with a special focus on its rotor and tail which are at the heart of the Airdolphin’s
range of unique technologies.
1. History of Small Wind Turbines
Small wind turbines were first industrialised in the U.S. since 1973, which was the year of
the first so-called “oil shock”. More than 30 years then passed with no real further market
progress as an industry.
Zephyr was established in 1997, at the same time that the Kyoto Conference on
greenhouse gas emissions was held at the Third Conference of Parties to the United
Nations Framework Convention on Climate Change (COP3). Because of the Kyoto
Protocol, many ordinary people became aware of the need for reducing carbon dioxide
emissions. Some in Japan began adopting new lifestyles, including using small wind
turbines at home. Companies, schools and local governments all became interested in
installing small wind turbines, demonstrating the wide appeal of the concept.
The Small Wind Turbine Committee (SWTC; Mike Bergey, Chairman), a subordinate
organization of the American Wind Energy Association (AWEA), released the “U.S. Small
Wind Turbine Industry Roadmap” in 2002. This report describes the future of small wind
turbines until 2020.
The report offered the optimistic view that, although sales of small wind turbines within
the U.S. were only 13,400 units in 2001, by 2020, they would contribute 3% of U.S.
electrical consumption, or 50 million kW (i.e. 50 million 1kW wind turbine machines). The
report also indicated the issues that needed to be solved in order to achieve that target.
The main issues were:
1. Economic efficiency
2. Technical innovation
3. Improved turbine reliability
4. Installation convenience
5. Easing of various regulations
6. Government funding
The report also mentioned that 50 million kW of energy output could generate annual
economic growth of $1 billion and 10,000 jobs. The small wind turbine industry expected its
products to become a major new category of home energy appliance. The report went on to
suggest that there was great market potential not only in the U.S. but also abroad, since 2
billion of the world’s 6.5 billion people were living without electricity.
It is evident that the demand for small wind turbines is growing globally, but the market
has yet to show its full potential. This is because truly practical small wind turbines still do
not exist. Project-Z was launched to develop a “real” small wind turbine – a practical answer
to the market’s needs.
*Takashimaya Department Store in Yokohama, Kanagawa Prefecture
2. Problems with Large Wind Turbines
Before discussing small wind turbines, let us first examine the performance of large wind
turbines. Is it a practical answer to develop high-performing small wind turbines by simply
copying large ones? The answer is no by any means.
*Tarifa, Spain (The Straits of Gibraltar)
The Airdolphin continues to generate power, while large wind turbines halt during strong winds.
(1) Yaw Errors
The active yaw controls on large wind turbines are usually electrically powered. This is
done by basing each 5-10 minute period’s rotor angle on the average wind direction of the
previous period. This is unavoidable, since huge rotors cannot change direction in seconds,
and it also means that the rotors cannot track wind direction in real time. Unfortunately, a
30-degree discrepancy in wind direction reduces power generation by 50%. We know from
experience that sudden wind direction shifts can happen under any air flow conditions.
Thus, we know that if we could keep track of wind direction in real time, then generated
output would increase significantly. This is possible indeed with the passive yaw control
used for small wind turbines. (A horizontal axis design is another possible method, but we
did not use this technology due to inferior rotor convergence efficiency.)
*A mountain lodge on Mt. Yatsugatake, Nagano, Japan
(2) Poor Dynamic Wind Capturability
Although we have talked about wind direction and wind speed, there has been little
discussion about short cycles in wind pressure, such as turbulence or gusts. To cope with
these more efficiently, the new concept of “Dynamic Wind Capturability” was born during
the R&D process of this project. This is the ability to follow dynamic wind speed fluctuations
– for example, from 10m/s to 30m/s in less than a second, then suddenly back to 10m/s –
and convert them into energy. To accomplish this, we have drastically reduced the mass of
the Airdolphin’s rotors, both static (mass) and dynamic (inertia moment), while keeping the
rotors highly rigid. While small wind turbines can generate power during dynamic wind
fluctuations, large wind turbines cannot fully respond and let a great deal of wind pass by.
(3) Low Capacity Factor
Capacity factor (CF) is the ratio of actual generated output (Wh) to pre-measured output
(Wh) under rated wind speed (12.5m/s) over a certain period of time. The Capacity factor in
Japan is considered between 10% and 30%. Machine failures, periodical inspections, and
shutdowns due to strong winds may be causes of these low rates.
There are also reports in Japan of insufficient power at certain wind speeds. One reason for
this may be that average data includes the wind speed of strong gales that halt the wind
turbines. Another reason may be the low Dynamic Wind Capturability in the
turbulence-prone Japanese climate.
(4) Excess Mass
Large wind turbines have an enormous mass per rated output of around 70-200
grams/watt. For superior Life Cycle Assessment (LCA), this needs to be far lighter. Mass
per rated output for small wind turbines is usually 100 grams/watt, though there are some
heavier turbines with 1kg/watt. However, because of its outstanding lightness, the
Airdolphin boasts 17.5 grams/watt.
*Ashikaga Institute of Technology, Tochigi Prefecture
(5) Installation Limitations
Places with a stable and constant wind supply are considered the best for wind-power
generation. Other factors that can limit the installation options for wind turbines are: too
many gusts, abnormal weather such as typhoons, strong winds over 20m/s, impacts by
birds, noise, harmony with the landscape, equipment transport and capacity of the electric
power system. As opposed to large wind turbines, efficient small wind turbines resistant to
turbulence and storms can eliminate these limitations. This opens up a variety of new
possibilities, such as miniature wind farms.
If small wind turbines become less costly and perform better than large ones, we can
expect a considerable expansion in natural energy usage. The Airdolphin was developed
by learning from tough issues of large wind turbines. Our results far exceeded our
3. Development of the Airdolphin – small wind turbines outperforming large ones
The three-and-a-half year Project-Z turned out to be very complex, combining academic
research, fine materials and traditional Japanese craftsmanship - the cornerstones of
Japanese technology. The following is a documentary-style report of the process.
Super-light weight, high rigidity, quick response, high output, cost effectiveness, safety,
longevity, zero maintenance, good design… it was no easy task to achieve Zephyr’s desires
for its new small wind turbine. Project members from a variety of fields - industry,
government and academia – each brought their expertise to develop “the world’s best small
wind turbine”. As a result, we overcame considerable challenges and created a new
The Newly Developed Basic Technology Used for the Airdolphin
Special new blades designed with the Multi Staggering system (patent pending)
Super-lightweight, high-rigidity, long-life blades using the CFRP (Carbon Fiber
Reinforced Plastic by Toray Industries/Torayca) (patent pending)
Silent Disrupter noise reduction blades (patent pending)
Swing Rudder tail for quick response to wind direction changes (patent pending)
Power Assist rotor starter (patent pending)
Stall control electronics for fault-tolerant operation without using any pitch control
mechanisms (patent pending)
Operation control by the Zephyr Power Management System software (patent
“No screw” hub and body with a Block Puzzle (traditional Japanese handicraft)
structure (patent pending)
Heavy duty motor generator (newly developed, maximum output of 4.5kW)
Web-enabled communication system by the Zephyr Ecommunication System
Two Major Technologies for the Airdolphin’s Superior Performance
(1) High-performance rotor
A great deal of time has been spent on developing the blades. The National Institute of
Advanced Industrial Science and Technology (AIST) and the University of Tokyo were in
charge of scientific analysis and engineering. Toray industries and others developed the
blade materials and manufacturing methods. 3D-CAD was used to make artisan moulds.
Through repeated prototypes and tests, the blades and screw-less hub began to take shape.
They were strong enough to endure several tons of centrifugal force.
*Block Puzzle structure
*Blades and hub cover
Through extensive tests, we confirmed that no fluttering occurs during wind speeds from
2m/s breezes to 50m/s windstorms. The blade rigidity showed good response to pulsing
wind speed changes, and the wind turbine’s high “Dynamic Wind Capturability” was later
proved by power generation data. Braking tests have shown that even at a wind speed of
20m/s, the Airdolphin can halt within 0.5 seconds. This is also effective for braking to the
stall mode in storm conditions, demonstrating the superiority of lightweight blades.
Sudden wind pressure increases cause the power to rise at the same time, so we needed
to substantially reinforce the control circuit. Through repeated track tests on the AIST’s
circuit, we gained valuable data which could not be achieved merely by wind tunnel tests.
In the final stages, we found out that the Airdolphin withstood continuous operation under
wind speeds of up to 50m/s. The blades are tremendously silent, thanks to a design based
on owl wings. We also achieved extremely quiet rotors.
*SD Blade and Owl Wing
(2) Swing Rudder – a unique discovery
The Swing Rudder, which was inspired by the tail fin of a fish swimming upstream,
completely outperformed our expectations. Previously, yaw errors were thought to be
caused by the rotor face being pushed out of position by turbulent flows. The wing rudder
was thus designed to prevent this. The swing rudder, with its tail down and continuing during
turbulent flows, looks rather comical, but the improvement in yaw error was simply superb.
Surprisingly, our tests also demonstrated another advantage. As wind speed increased, the
nacelle made small irregular swings while the rudder stayed still. However, the nacelle
wasn’t overshooting. Its swinging was caused by swiftly keeping track of the wind pressure’s
main course as captured by the high rigidity rotor. In addition, the rotor followed exactly the
highest power point, and unlike the fixed rudder, did not interfere with movement. When we
saw the power meter shoot up from 500W to 2kW, it was like nothing we had ever seen
before. That was when we coined our original and new technical concept of “Dynamic Wind
Airdolphin Mark Zero Output Features
As the name “Airdolphin” implies, this wind turbine has an endearing way of moving.
However, more importantly, we are extremely pleased with how strongly it meets our original
objective of increased output.
Airdolphin verification test
We then moved the completed Airdolphin to the verification stage. We collected test data
on a total of 30 Airdolphins in the following places:
Tarifa, Spain; at large wind turbine wind farm (where 30m/s average wind speed is quite
Soria, Spain; at CIEMAT, a Spanish national test site
Shetland Islands, Scotland (famous for their constant gales)
Ocher Plateau, China (famous for its thick clouds of yellow dust)
Cape Erimo, Hokkaido and Cape Tappi, Aomori Prefecture
both of the above are said to have the strongest wind speeds in Japan
Ashikaga Institute of Technology, Ashikaga, Tochigi Pref.
Odaiba, Tokyo (official test site operated by the University of Tokyo)
Takashimaya Department Stores, Takasaki, Gunma Pref. and Yokohama, Kanagawa Pref.
Mt. Akadake, Yatsugatake mountain range, Nagano Pref.
Mt. Nonobori, Mie Pref.
Daisen large wind farm, Tottori Pref.
Yoshino River, Tokushima Prefecture
Kyushu University, Fukuoka
Miyagi Island, Okinawa Prefecture
Head office, Zephyr Corporation, Hatsudai, Tokyo
Live images and output data of the experiment are currently delivered in real time via the
Zephyr Internet Homepage. < http://www.zephyreco.co.jp >
*Hatsudai, Tokyo. 20kWh output in 10 days. Average wind speed: 3.9m/s
The team members of the Project Z
Private Sector Firms
Toray Industries – Blade modelling
NEOMAX Co., Ltd (former Sumitomo Special Metals Co., Ltd.) – Power generator
Unitec Corp. – Power generator/motor
Yokogawa Electric Corp. – Communication systems and maintenance
NTN Corp. – Rotation parts including bearings
Moriyama Giken – Metallic body
Oume Denshi – Electronics printed circuit board
First Energy Service Co. Ltd. – Market development
Iwasaki Electric Co. Ltd. – Application for streetlights
Nitto Kako Co., Ltd. – Vibration absorption parts
Ministry of Economy, Trade and Industry, and the (National) New Energy and
Industrial Technology Development Organisation(NEDO)
Dr. Hikaru Matsumiya (Kyushu University), Dr.Tetsuya Kogaki, Rotor Engineering
- Basic blade design and implementation tests
The University of Tokyo Graduate School
-Dr. Chuichi Arakawa, Professor, Engineering & Dr. Makoto Iida, Engineering
- Blade analysis and basic design
C 2006 Zephyr Corporation