Turbine Orientation

But as the connection would be made by the angle pulleys rather by the hubs, some significant adjustments would be required. I think a preliminary work would be building the different variants (pulleys, bevel gears, universal joints, angle pulleys like for Kiwee, and so on) of multi-rotor facing the wind or the apparent wind.

An example of connection by the hubs is given below. Unfortunately the apparent wind would be different according to the side of the turbine.

propellers belt and rope-drive

The frame would be a bar like this, the turbine being correctly oriented:

And as the rope-drive transmission avoids the weight of the generator in flight while the system scales up, a single larger rotor would be simpler.

I like your idea of using bars to connect the oriented turbines. The problem is to connect the cable drive loops without interfering with the propellers. My idea, in the attached sketch involves a lot of pulleys. Any other ideas? The bars must have bends at the end to satisfy the orientation. Perhaps these bends can be adjustable to accommodate various tether angles.
DOUBLE PULLEYS.pdf (59.5 KB)

Hi @gordon_sp,

Your stepped axis with pulleys looks to be a workable solution. The question is balancing between the expected benefit by the orientation of the turbines, and the weight and complexity of the device.

According to the report below, SuperTurbine ™ looks to be efficient (see § 5):

I remake my sketch (see below) because the rope-drive is useless in the SuperTurbine ™ part. The rope-drive connects the lower rotor to the generator at the ground in Kitewinder patent way.

I think the Superturbine tests were done with 25 deg angle. This results in a 25.6% loss in power. For high altitude we have additional wind speed gains. The additional weight can be compensated by a larger lifter kite.

You are right about the rope-drive section. The question remains, how do you orient the turbines to be horizontal if they are connected to the bars? I can think of universal joints and sails (inefficient) or pulleys. Any other ideas?

Variants with universal joints, double universal joints, bevel gears has been discussed.

I think we have to keep it simple. All these variants, comprising pulley system, lead to additional weight and frictions. Certainly there are some (limited for a low angle) losses for SuperTurbine ™ but also some lift, and it is simpler and lighter.

And also an option for multi-rotor on a tether (instead of a bar) can be studied in yo-yo mode.

I think the focus on % loss is a little misguided. For any AWE, it is quite likely that zero elevation angle is plain impossible.

Conparing with a HAWT (regular windmill) yes you have losses, but you have benefits as well.

Cosine loss only means that you need more wing area to achieve the same power. This is bad, but other factors may be limiting the scale also, like tether strength, power rating of the ground eqipment etc.

And, for a superturbine type og windmill, if you point the rotors horizontally, you need to add an extra force to counter tether downwards pull.

Also remember higher elevation would mean more altitude and higher average windspeeds.

Im not sure cosine losses is the best place to start a design…

1 Like

Unfortunately the cosine cubed losses are more than the gain from increased wind speed. For example, increasing the tether angle will cause a cosine cubed loss of 46% whereas a 10% increase of wind speed will increase the power by 33%.

It is also possible to increase the altitude without modifying the angle of elevation by lengthening the tether.

But other issues would occur such like higher space use, and tether weight.

My example suggests increasing the tether angle from 30 deg to 45 deg. It would be even worse for greater tether angles.

Yo-yo mode requires crosswind action. Why not replace the turbines with arrays of kites which are lighter and cheaper. Kites will naturally turn to face the wind during crosswind action providing more power. This is similar to your Orthokitebunch system with a different power takeoff.

The parachute on chinese umbrella description.pdf (1.3 MB) does (did) not use crosswind action during yo-yo phases (The most basic airborne wind energy system).

Multi-rotor could make the same, enjoying a higher lifetime as the rotors are rigid, and also perhaps easier to implement. Indeed thin rotors are around the tether without transfer issue, only thrust. So the rotors are not connected and are adapted to different wind conditions according to altitudes.

It is true. The same is not easy for rotors excepted for facing rotors like for Makani or FlygenKite.

Like something you’d see sketched in the margin of a junior high school kid’s notebook

Indeed, because as you just wrote:

Yes Pierre, it definitely has the advantage of simplicity, and I’m probably one of the kids who sketched it in the margin of my notebook way back in the '60’s or '70’s. Also underwater versions. By the time I got to college all my margins were full of wind-energy-themed art, mostly just very vague themes.

However my sketch has something still wrong because the Kiwee bevel pulley system has fixed parts on both sides of the rotor, but the axis of SuperTurbine ™ rotates with the propellers. So putting said fixed parts can be a not easy task.

Effects of Yaw Error on Wind Turbine Running Characteristics Based on the Equivalent Wind Speed Model

1 Like

Following up on Pierre’s idea I thought of a modular system where cable drive systems can be linked together while all turbines are oriented to face the wind. Each module will contain a turbine, two pulleys and four idler pulleys. These idler pulleys should not cause much efficiency loss because the contact angle is only about 45 deg. The frame can be aerodynamically designed for minimal air resistance and elements can be canted to provide some lift to the module.
Alternate modules can be made to rotate in opposite directions to neutralize transverse torque. This can be achieved by crossing the cables above and below each module. To prevent the cables from rubbing together we can offset the idler gears and also cant them slightly to align them properly.

1 Like

This module arrangement looks to be correct. After the weight can perhaps be a difficulty, as well as launching and landing for each module.

In the other hand the pulleys driving the respective rope-drives have a common axis with their respective propellers, unlike the right-angle return pulley used by @Kitewinder. So the motion of the rope-drives is not aligned with wind force among other problems. As a result, the maintenance of the cable on its pulley seems random, as well as the power transmission ratio.


The orientation of the frame is fixed and not random. Visualize a frame where the tether is attached to symmetrical opposite sides. The position of the frame is determined by the tether angle and the location of the attachments. In our case we locate the idler pulleys in the location where the frame is horizontal and the turbine faces the wind.