Salient White Elephant

February 24, 2009

Scalable Tower for Very Large Wind Turbine

The diagram above has been simplified for clarity. It appears to depict a downwind turbine whose tower is supported by only two sets of guy wires – one set upwind and the other set downwind. In reality, the tower is supported by anywhere from three to five sets of guy wires that are radially dispersed in the expected fashion (i.e., if there are 3 sets of guy wires, then an aerial view of the machine would have one set at 12 o’clock, one at 4 o’clock, and one at 8 o’clock). Some guy wires are attached to the tower at points just beneath the nacelle. The blades do not hit the guy wires because their lengths are continuously adjusted by the wind turbine controller. As the turbine yaws, the controller simply lengthens the downwind guy wires to make room for the blades to pass.

It may be desirable to keep guy wires taut when moving them out of the rotor path. This way, they can still enhance the stiffness of the tower, even if only by a small amount.

Disadvantages of the Traditional Tower

  • The size and cost of the tower increase rapidly as the size of the turbine is increased. For this reason, increasing the size of the turbine provides diminishing economic returns.
  • The size and cost of the tower are determined by the overturning moment generated in wind storms, hurricanes, and so forth, (or possibly by the overturning moment generated during the highest winds at which the turbine produces electricity). Therefore, reducing the weight of the nacelle does not result in a lighter (lower cost) tower.

Advantages of the Scalable Tower

It may turn out that the size and cost of the scalable tower do not escalate as rapidly with increasing turbine size as do the size and cost of the traditional HAWT tower. If so, the tower may ultimately lead to larger turbines, and this could lead to a reduction in the capital cost per installed kilowatt of a wind farm via the well-known mechanism of economies of scale. That is, a 10 megawatt wind plant constructed with two 5 megawatt turbines has only 2 transformers, 2 concrete pads, and so on. But the same wind plant constructed with four 2.5 megawatt turbines has 4 transformers, 4 concrete pads, takes longer to build, takes longer for technicians to change the oil in all of the turbines, and so on. Theoretically, the larger the turbine the greater the cost reduction. However, if some costs (such as the cost of the tower) increase at a greater than linear rate, then these exponentially increasing costs can compromise the benefits of scale. Truthfully, I do not know whether current turbine designs are burdened with these exponentially increasing costs, but I suspect that they are, and that is the reason for proposing the alternative designs presented in this document.

Keep in mind that in addition to the direct benefits provided by a tower that is more easily and more economically scaled, there may also be secondary benefits. For example, as I understand the current state of affairs, there isn’t much benefit to reducing the size, cost, and weight of nacelle components, because these reductions do not lead to a corresponding reduction in the size, cost, and weight of the tower. But if the scalable tower allows for the design of a very large wind turbine that does not have a massive tower, then perhaps reducing the cost and weight of nacelle components would allow the tower to be lighter still. One of the reasons for considering the ideas presented here is that they might lead to these kinds of “cascading benefits”.

Leaning Turbine of Pizza

Another variation of the ideas presented here has the tower attached to the ground through a universal joint. All of the guy wires are slightly longer than they would have to be to hold the tower in a perfectly vertical position. When the wind has significant velocity, it pushes the tower so that it leans in the downwind direction. This may seem like a crazy idea, but note that it has two big advantages over the vertical machine whose controller alters the length of the guy wires. First, because only natural forces are employed, it is impossible for a controller failure to have any impact on the state of the tower and its guy wires. Second, tilting the tower simultaneously moves the rotor blades away from the guy wires, and allows the guy wires to be pulled in towards the tower. Because both of these effects occur simultaneously, it may be possible to design a machine with a tower that is tilted by only a relatively small angle.

The nacelle of the leaning turbine yaws about the top of its tower just like a traditional horizontal axis machine. The blades are tilted with respect to the tower so that the rotor disk is approximately vertical in spite of the fact that the tower is tilted. The yawing motion occurs about an axis that runs right down the center of the tilted tower. If you imagine the nacelle rotating through 360 degrees, the tip of the lower blade in the above diagram would maintain a constant distance from the tower. For this reason, as long as both the tower and the rotor are in the downwind position, the rotor disk will always be vertical (at a right angle to the direction of the approaching wind), regardless of which way the wind is blowing. One advantage of this embodiment is that the blades will seek the downwind position not only because of the aerodynamic drag force, but also because of the force of gravity. To see this, imagine rotating the nacelle in the above diagram by 180 degrees, and you will see that the blades are higher in that position than they are in the downwind position depicted in the diagram. I cite this advantage because experience has shown that downwind turbines sometimes need the help of a small yaw drive in order to maintain the correct yaw angle.

Yawing the Entire Turbine

Yet another variation would yaw the entire wind turbine. In this embodiment, the nacelle does not move with respect to the top of the tower. Instead, the tower is “permanently tilted”, and is able to rotate about a vertical axis drawn through the center of the concrete pad upon which it rests. The guy wires that attach to the tower at a given elevation would attach instead to a kind of ring that encircles the tower, and that is able to rotate about the tower. Like the previous idea, this one sounds pretty crazy, but it has similar benefits. In this embodiment, it may turn out that the machine does not need a yaw drive. Instead, perhaps the force of the wind itself would be enough to keep the machine yawed to the downwind direction. However, if a yaw drive were required, note that it would be on the ground rather than in the nacelle.

If it is desired to use natural forces to yaw the machine, then perhaps a tail fin could be added to increase the yaw moment. The following diagram shows one option for using a tail fin to yaw the machine.

The tail fin is mounted on an extension of the rotor shaft. The tail fin is able to rotate about this extended shaft, but it has a weight near the lower part of the fin that keeps it in the vertical position shown in the diagram. In other words, the extended shaft rotates at the same rpm as the rotor, but the tail fin remains in the vertical position because it is able to rotate with respect to the extended shaft, and because the lower half of the fin is heavier than the higher half. The fin is also symmetrical about the extended shaft so that when the machine is not properly aligned with the wind, the aerodynamic force that is applied to the top half of the fin is equal to the aerodynamic force that is applied to the bottom half. Because the aerodynamic force is symmetrical about the extended shaft, it will not cause the fin to rotate away from the vertical position shown in the diagram. Of course, wind gusts will cause the fin to bounce around a bit, and the wind rotation imparted by the rotor will also have an effect, but small deviations from the vertical position should not significantly impact the yaw moment. If necessary, a compact weight could hang from the bottom of the fin on a thin cable.

Perhaps you’re wondering how a machine without a yaw drive would untwist its power cables. One option would have the controller send an email to the service department whenever the cables are twisted. A technician then drives to the turbine and attaches a long pole to the bottom of the tower. The pole extends horizontally from the bottom of the tower to a distance just inside the radius of the guy wire anchor points. The end of the pole that is away from the tower has a cable attached to it. The technician attaches this cable to the back of her service truck, then drives around in a circle until the power cables have been untwisted.

An alternative to yawing the entire turbine would be to yaw the nacelle and the guy wires as a unit. In this case, the tower would be vertical and would remain stationary when the machine is yawing.

Drawbridge Turbine

In this embodiment, the guy wires are replaced by long tubes (or perhaps tube-like lattice structures). Like the guy wires, the tubes are only required to support a tensile load. The reason for using tubes is so that these supporting structures can support their own weight. This allows any downwind tubes that are in the path of the rotor blades to separate and move out of the rotor’s path. This action is reminiscent of the action of a drawbridge. (Tubes or lattices that can also support a compressive load may be desirable if the designer wants greater control over tower resonance.) The disadvantage of this embodiment is that the controller would have to manage the separation and reconnection of these supporting structures as the machine is yawed.

The following diagrams illustrate variations that require the anchor point of the supporting structure to move.

Some Kind of Wind-Up Guy Wire System

Some Kind of Wind-Up System For Guy Wires

Tricycle Turbine

The diagram appears to show a tower that rests on a single supporting lattice structure. Actually, there are two supporting lattices resting on wheels, and they form an upside down “V” shape whose vertex is just beneath the nacelle. The third leg of the tricycle is anchored to the ground, and is able to rotate about the yaw axis. The advantages of this configuration include:

  • None of the supporting structures need to be moved to make way for the blades.
  • The nacelle does not rotate with respect to the tower.
  • The yaw system (which drives the tires) is on the ground, so the nacelle is lighter, and the tower supports less weight.
  • The tower can have as many supporting guy wires or lattice structures as necessary to withstand high winds and other extreme loads. This allows a very large turbine to be built without requiring a proportionally massive tower.

Generator and Gearbox on Ground

Some of the turbines described here have nacelles that are fixed to the tower. If the nacelle does not rotate with respect to the tower, it may be possible to devise a way to transmit power to a gearbox and generator that are located at ground level. There are a number of different ways of doing this. Of course, it is right to be concerned about any design option that introduces losses into the system, and I doubt that it would be possible to transmit power to the ground with near 100% efficiency. But I am discussing the option here because I want to remind you of the possibility of discovering “cascading benefits”. The general idea is to discover design options in which an improvement in one area ripples throughout the system, and leads to possibilities for improvements in other areas that would not have been possible with conventional designs. For example, suppose one of the scalable towers allows for tower requirements that are no longer a function of the drag force of storm winds. Suppose further that the generator and gearbox are able to be located at ground level. This may lead to a nacelle that is so light in weight that much longer blades become feasible, and now the power that is lost on its way to the ground may not be such a big drawback. You can see how it might be easy to dismiss the option of putting the gearbox and generator on the ground because the efficiency of transmitting power to the ground is unacceptably low. And yet, if this opinion were reached before considering the benefits of a lighter nacelle, a good design option may be missed.

Collapsible Tricycle Turbine

I wonder if it would be possible for the tricycle turbine to fold up in a way that would bring the nacelle near the ground. Options for doing this include:

  • One of the rolling legs is able to fold in half or collapse like a telescope. Both rolling legs move to accommodate this action.
  • The leg that is anchored at the yaw axis is able to collapse like a telescope. The other two legs roll to accommodate this action.
  • The leg that is anchored at the yaw axis folds approximately in half, causing its mid section to rise. After folding, the anchored leg looks like an upside down “V”. The other legs roll to accommodate this action.
  • The two rolling legs are able to move away from each other until the nacelle is near the ground.

If the tricycle turbine can be collapsed in some way, then maintenance and repair of nacelle components can be carried out near the ground, and expensive heavy duty cranes would not be required during construction. Collapsing would also render the machine virtually indestructible in storm winds, and would lead to tower specifications that are no longer a function of the extreme loads inflicted by high wind conditions.

Here is one final idea on bringing the Tricycle Turbine nacelle to the ground.

Scalable Tower for Upwind and 3 Bladed Turbines

Many of the ideas presented here may be equally useful for upwind turbines and for 3 bladed turbines.


Leave a Comment »

No comments yet.

RSS feed for comments on this post. TrackBack URI

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

Blog at

%d bloggers like this: