Wind Turbine Technology

I invented a wind turbine technology (US Patents: 8,049,351 and 8,178,987) that can harness instantaneously maximum energy from very low wind speeds and with a very fast response to wind speed variations.  Recipient of the prestigious 2009 Popular Mechanics Break through Awards and 2010 Edison Gold Award for Best New Product. 

Solar energy creates ground temperature variations which in turn heats up the air and subsequently induces air motion and speed variations.  At any given site, the local wind speed can vary considerably.  When statistical processes are applied to wind variation a Rayleigh distribution is produced as shown in the graph below.  The graph represents the percent of hours the average wind, measured at 33 feet above ground level, is at a given speed.  What is shown is Class 4 and Class 6 categories out of seven categorized wind classes.   For each wind class a Rayleigh distribution annual average wind speed is also given. 

These seven wind classes help define general wind speed trends at a given geographical site.  Below is a wind map of United States.  The map indicates that wind turbines should be sited East of the Rockies and in the Great Lakes.

In developing a wind turbine, a clear understanding of what average wind speed means is needed, as it critically impacts turbine basic design.  

Nassim Nicholas Taleb gives a clear definition of what an average means ( Nassim Nicholas Taleb, Antifragile, Things That Gain from Disorder, Random House): 

 “You have just been informed that your grandmother will spend the next two hours at the very desirable average temperature of seventy degrees Fahrenheit…Excellent, you think, since seventy degrees is the optimal temperature for grandmothers.  Since you went to business school, you are a “big picture” type of person and are satisfied with the summary information.

But there is a second piece of data.  Your grandmother, it turns out, will spend the first hour at zero degrees Fahrenheit…and the second hour at one hundred and forty degrees…for an average of the very desirable Mediterranean-style seventy degrees…So it looks as though you will most certainly end up with no grandmother…

Clearly, temperature changes become more and more harmful as they deviate from seventy degrees.  As you can see, the second piece of information, the variability, turned out to be more important than the first.  The notion of average is of no significance when one is fragile to variations- the dispersion in possible thermal outcomes here matters much more.”

While the first graph describes the seven wind classes each with its annual Rayleigh average wind speed, the grpah below shows the significant variation of wind speed measured at a fixed anemometer location.   Since wind power varies to the third power with air speed, the chart below reflects a significant variation of available wind power in a given short parcel of time.  

My technical design objective was to build an innovative turbine technology that can harness the maximum energy from such instantaneous wind dynamics.  The starting design goals were:

- Low mechanical resistance

- Effective aerodynamic lift forces at low wind speeds

- Optimal total mass for fast response to such instantaneous wind speed variations

To reduce mechanical resistances, sealed permanent neodymium iron boron magnets were located at the perimeter of a lowmass, high-strength wheel.  The magnets pass with high speed at the wheel perimeterthrough the patented stator windings.  This novel design eliminates the need for gears that help increase the magnet/stator relative speed to produce the design voltage according to Faraday’s Law.  This design enhances simplicity and minimizes preventative maintenance and future costs.  

The following Navier-Stokes governing equations were solved using ANSYS to model and optimize the blade designs and flow fields. and they were critical to guiding the empirical development and testing of the turbine:

Where ρ is air density, u is air velocity vector, p is dynamic pressure,  r geometry length, FL lift force,  A is blade plan area, FD is drag force, CL is lift coefficient, and CD is the drag coefficient.

The schematic shown below represents an optimized six foot diameter wind turbine technology design that has patented high solidity blade matrix and perimeter electrical energy generating system.  It starts generating power at approximately 0.5 miles per hour and responds to wind speed variation within one second.  It has side deflectors that passively orient it to be always perpendicular to the incoming wind direction.

Simms Electronics were contracted to design a battery charging system whose design goals I set at an input voltage ratio of 10:1, high storage efficiency, and very fast response time.  The SmartBox, as it was called, has proprietary pulse width modulation (PWM) software that was custom designed to closely match the turbine fast response time to wind dynamics.  The PWM extracted power using dynamic maximum power point algorithms.  

This turbine technology was developed and optimized using a large high power blower fan that could deliver steady state average air speeds of up to 20 mph.  This was a very practical engineering solution to establish the maximum steady state power curve that is needed for the final optimization of the battery charging system and its certification under UL1741.  

The lab wind tunnel was used to calibrate the wind turbine under steady state wind speeds up to 20 mph.  A second order polynomial graph was fit to the data and then theoretically extrapolated the turbine power output to 40 mph.  This steady state power curve graph is shown below.  Also indicated are the cut in wind speed at approximately 0.5 mph, the plate power of 1500 Watts at 32 mph and the cut off power at 40 mph of approximately 2400 Watts.

This steady state turbine power curve represents the maximum power at these wind speeds and it represents the steady state energy content of the wind at this speed plus the inertial energy content at the moment of PWM extraction.  This turbine absorbs energy from the wind and acts as an energy inductor (the energy within the flywheel is directly analogous to the magnetic field within an inductor) and consequently could exhibit higher values at the low winds speeds than that predicted by the steady state assumptions in the thermodynamics model of the Betz limit.


Ford F150 truck was subsequently equipped with a steel cage where the wind turbine was suspended.  An anemometer was installed in front of the turbine in line with the center of its radius.  The SmartBox battery charging system together with a set of deep cycle batteries were installed in the truck bed. The truck then moved at different speeds and the winds speeds and the turbine power output were recorded on the SmartBox data logger.  The graph below shows  the truck turbine power output in comparison to the steady state power curve.  The variations observed on the truck turbine power output reflect the actual non steady state variation in truck speed on the road.  The SmartBox DC charger has a built cut off maximum power when the wind speed is in the range of 38 to 40 mph.  This test was created because large scale wind tunnel tests are prohibitively costly.  Also field tests take a long time and reflect only the test site wind profile.  The truck tests, which produce on-demand high wind speeds, affirm the steady state calibration power curve.


The turbine and the SmrtBox DC charging system were tested by Intertek, a Nationally Recognized Testing Laboratory (NRTL), for compliance to various UL Standards.  The SmartBox DC charging system was certified to UL1741 compliance.  In the testing process the wind turbine was connected mechanically to a high power dynamometer.  The dynamometer was then set at various RPMs and the turbine power output as measured by the SmartBox was recorded.  A graph of the turbine output at different turbine DC output voltage was then generated and certified.  The graph below shows the turbine power output graphs for the initial steady state wind calibration, the steady state power curve extrapolated, the truck test power output, and the Intertek dynamometer certification against the turbine DC voltage output.  There is significant correlation between these methods as seen in Figure 7 that affirm the turbine design  and calibration methodology.


Computation Fluid Dynamics using the Navier-Stokes equations listed above were performed using ANSYS CFX.  CFD movies were generated for my turbine and a similarly-sized three-bladed turbines under the same free stream air flow 12.5 mph average wind speed of Class 4 wind regime. It is very clear how fully my turbine design engages with the free air stream in comparison to the standard three-bladed wind turbine.  Please go to ANSYS CFD movies comparing this turbine to a very common three blade turbine under identical air flow conditions.

As described earlier, natural wind varies dynamically in time and is seldom steady state.  Furthermore, the average wind speed is seldom meaningful due to instantaneous wind speed changes.  When this turbine technology is installed in the free air stream, it will produce power at 0.5 mph and higher.  More importantly, it will produce a varying amount of power at the same wind speed depending on how long the wind stays at this speed.  For example, if wind gusts to 10 mph for a fraction of a second, then the turbine power output will only be a fraction of the steady state power output at 10 mph.  The longer the wind gust stays at 10 mph, the closer the wind turbine power output approaches the maximum steady state calibration curve.  

The following is a graph showing the turbine power output at an installation site.  The blue dots are one second turbine power output data points in Watts.  The Red dots are the maximum steady state power curve. The wind at this site did sustain wind speeds up to 18 mph long enough so that the turbine output did reach the calibration curve.  In some wind gusts, the power output exceeds the stead state calibration, emphasizing the role of inertia in gusts and sudden wind forces.  The SmartBox has an energy accumulator so that the total energy can be computed by integrating all power points in real time over an extended period. 

The graph below shows the power output of the wind turbine verses wind speed at a different site installation.  The blue dots are the instantaneous turbine power output and the red dots are the instantaneous wind speed values.  The SmartBox integrates the power over a long period of time to give the accumulated energy extracted from the wind.  

Wind turbines need to be installed where global wind maps indicate a good annual average wind speeds, Class 4 and larger.  The smaller the turbine diameter, the more critical this becomes.  But even when they are to be installed in such high wind speed zones a detailed site assessment must be undertaken to ensure good wind energy harvesting.  Free stream air flow modeling using ANSYS was used to determine the effect of a tree on this six foot turbine.

The following ANSYS CFD graph shows where best to place the small turbine in the vicinity of a 40 foot tall tree.  As shown, it is best to install the small turbine some 10 feet above the tree for best performance results.  The tree wake can have a wind power attenuation up to some 150 feet as shown.  It is imperative that even average site wind speeds be measured at the actual site and not at typical nearby airports.  Wind is a very local phenomenon.

In theoretical aerodynamics, the Prandtl one seventh power law correlates the effect of ground and solid surface on free stream wind speeds due to the presence of a non-slip boundary layer.  The free air stream slows down as it nears the ground surface.  The slowing down of the air stream near the ground surface depends on the surface roughness, irregularities, shrubs or trees, and man-made obstacles as part of the building design and construction.  Typical city ground roughness can reduce the air speed by 55% or greater.  Conversely, wind speeds do not attenuate as much, over water due to surface smoothness.  The Prandtl power law is modified empirically to become:

U(h) =  U(10) . (h/h10)^α

Where U(h) is the site wind velocity at height h, U(10) is the site velocity at 10 meters height, and α is the Hellmann exponent.  For example, the Hellman exponent can vary between 0.06 for natural air over water to 0.6 for stable air above human inhabited area.

The diagrams below show the free air stream ANSYS analysis for the best installation of a small wind turbine on a typical home.  It is clear that even the optimal installation of the small wind turbine on the roof of a home can be compromised by the oncoming wind direction.  To minimize the wind speed attenuation due to roof geometry and other interfering structures it is best to install the turbine at least 20 feet above the roof and at the location shown.  Installing the turbine on a pole away from the house or trees would give best results in a good Class 4 wind zone.


The ANSYS solution below shows the best installation of a small wind turbine atop a 10 story building. The turbine should be installed on a pole at least 20 feet high to maximize its potential power output in good Class 4 wind zone.



It is very important obtain wind speed data for an extended period of time at the very site and elevation intended for a turbine installation.  While this might be costly and takes a long time, it is this data that should be used to assess the business potential for installing a wind turbine.  This is because site wind variability, as clearly discussed above, has little to do with any published average wind speed.  Furthermore, in site assessment it is very important to install a wind speed anemometer within three feet radius of the intended installation site and elevation.  The graphical solution shows that the pressure field of the free air stream approaching this small turbine begins to be distorted due to the turbine presence in the vicinity of three feet radius or under. The ANSYS analysis shown below depicts this pressure field variation for 10, 20, 30, and 40 miles per hour.  

Wind turbines, small or large, need to be installed in at least a Class 4 wind zone, where electricity is expensive, and with good state or government subsidy to offer a modicum of a return on investment (ROI).  There are some very good wind zones in the United States, namely East of the Rockies and the shores of the Great Lakes. However, the very low cost of energy in the United States compounded by very low feed in tariff rates, if indeed they exist, together with low  subsidies renders small turbines an unattractive renewable technology.  Most small wind turbine, like Skystream or Bergy, have an ROI of some forty years or greater.  Large wind turbines have had only a modest success in states where state subsidies have been significant.  The challenge in North America is that the dependence of wind energy on unpredictable State or government policy can lead to a feast or famine for any fledgling wind technology company.  It is far better to leverage advanced wind technology for the World markets where the wind speeds are high, the energy costs are high, and the potential subsidies are large due to the local critical needs of energy and the lack of local fossil or nuclear resources.


The Navier-Stokes equations were solved using ANSYS to generate a computational fluid dynamics (CFD) movie (please press the play icon to start the CFD movie below) that shows a very high degree of itnerface between the wind at an average Class 4 speed of 12.5 mph and this novel turbine.  The CFD movie clearly shows the very high degreee of interaction and therefore energy extraction potential of this novel wind turbine design:

The Navier-Stokes equations were then solved using ANSYS to generate a comparative CFD movie (please press the play icon to start the CFD movie below) that shows a low to moderate interface betwenn the wind at an average Class 4 speed of 12.5 mph and a standard three-bladed wind turbine: