Why non-crosswind type is not popular?

I’m new in AWE thing. Why most of AWE stuff are based on crosswind type. How is the autorotation helicopter/multirotor strapped to tether (skyWindpower style) power generation compared with the crosswind type (Like most of player in the market)?


Welcome @mingaero ,

Indeed Sky WindPower is classed in rotational AWES with electricity generation on the flying device, classification on Figure 2 page 3 on the pdf (Electricity in the air: Insights from two decades of advanced control research and experimental flight testing of airborne wind energy systems) on the link on the message below:

For now it looks like crosswind devices produced more electricity during experimental tests:

Figure 15: average 92 kW, 12 m/s wind speed.

Hi @mingaero
Good question.
Any blade generating power from the wind, generates more power when it increases the apparent wind. (up to a point)
Travelling a path perpendicular to the wind increases the apparent wind over the blade.
Sky Windpower used a frame with 4x rotors on generators to send electrical power down a conducting tether. The tether connected to the frame with bridles.

Here is an introduction:

Airborne Wind Energy - Producing Energy with Kites

Sky Windpower style lifts the entire wind turbine into the air, which is heavy, and can’t sweep as much area, so can’t produce as much power, it’s not even close IIRC. I think that’s the main reason it is not pursued more, as the goal is to produce electricity. Also the electric cable to the ground becomes heavy. Lifting heavy things needs a strong wind so your capacity factor, what percentage of the time you’re generating electricity, becomes lower. There’s also a lot that can be said about outdated assumptions and realities of flying in the jet stream.

Edited to add (ETA).

[ETA: Moving] Crosswind also has disadvantages, flight control is difficult, a long fast moving tether introduces a lot of drag, and with ground generation maximum power is intermittent (about a quarter of the time IIRC) and the tether wears out.

If you [ETA: move] crosswind but put generators on the wing, you now do have continuous power but again your cable becomes heavy.

Everyone has biases about what they like best. I like multiple kites flying crosswind connected to a single tether best, but then flight control presumably becomes even more difficult.

Here are some videos of the Sky Windpower system in action:

I think most wind power have foils travelling more perpendiclular to the wind rather than more downwind.

The yoyo AWE rig is kind of an outlier as it does indeed travel downwind through reeling out the tether.

I am not sure if the recent definition saying some «AWE using rotors is not crosswind» is a useful way of putting things.

Anyway, on a pure theoretical level, a large sail travelling downwind only provides little force per area of sail. A sail/blade travelling crosswind will generate more force per wing area because lift force is proportional to relative airspeed squared. A high force on a small wing area is usually the more cost effective alternative.

To answer the question of popularity, I personally think at this point that any design should be analyzed on its own merits. There has been a huge focus on Flygen (Makani/Windlift) and Yoyo (Kitemill/Ampyx) in the outfits which are better funded. I think one should interpret this as most people find this the most probable avenue of success, though certainly «most people» may not be correct. Eg flies on a cowshit… but also they may be on to something…

Personally I think the rotor concept and TRPT (rotating tensile shaft kite networks) which are not crosswind (!) have been overlooked in the «industry» to an extent that doesnt make sense relative to the merits of these designs

1 Like

@tallakt , would you go so far as to suggest that Kitemill develop TRPT Daisy rotor kite networks with Windswept? If so, that would seem like an idea worth digging into, what do you think @Rodread ?

Nice @PierreB
Where do I sign?

By what reasoning would anyone say a normal rotor is not crosswind? They’ve been crosswind for 1000 years and counting… Even rotors at an angle make power in proportion to how crosswind they are. This is the silly thing about AWE - no matter how long the hype drags on, the practitioners themselves seem hopelessly unable to grasp the most basic aspects of wind energy. It’s always just more “blah blah blah”. Skygen? Groundgen? Flygen? Crosswind? Downwind? What do these basic terms even mean? For anyone who doesn’t know by now, maybe that is a sign. This is like a darkened room with everyone trying to find the door for over a decade, and still banging on the walls yelling for someone to open the door, wondering who turned out the lights.

A normal wind turbine. A normal wind turbine is stationary, not moving across the field. Sorry, I always forget to add the “moving” to the “moving crosswind.” I’ll edit my comment.

That’s my personal way of thinking about it. Perhaps also that of the topic starter and outside of literature. I think of a wind turbine as static when it is not moving across the sky or across a field and as moving crosswind when it is moving across the wind.

The reason I’ve been thinking about this is that I think the crosswind type is very difficult to scale. I think the crosswind type occupies more area than the hovering one because it needs to move around.

I’m a UAV programmer myself, so I think the fast dynamics of the crosswind system are a challenge for continuous production. (I won’t stand near it)

So it comes to the idea that I’m going to tether a helicopter or quadrotor and make it hovering autorotation windmill. It is a simple solution: hover, less dynamic, and self land. It might generate less than the fast crosswind type (or more?) but it can scale up because it doesn’t move as much. so we can place more of them in the same place. Also, the hovering one should have fewer regulation problems because it won’t be a completely kill zone for helicopters or airplanes that fly in the zone.

Once again i’m very new to this and very welcome to any idea

Fixing a defined place for your rotor to operate, confined within a space, is responsible and laudable.
The trade off is efficiency.
You want your blade materials to be sweeping as fast as they should (~2/3 of their unrestricted speed I think)
Here’s where a typical rotor set to a hub at the blade root is problematic. The speeds at the tip and root are very different.
A typical autogyro rotor is still a cantilever design and carries a lot of mass
I recommend getting rid of the hub and connecting blades of your autogyro as kites around a perimeter rim.

1 Like

In other words: very difficult to put close together.

You could have them fly like swarms of birds for example, something with which your experience might come in handy. You could program them to avoid obstacles, like the ground, or other kites moving in the same direction. You could try to make them fully self-flying. But yes, making them reliably fly and not hit the ground and then making them fly closer together and not hit each other are apparently difficult.

Both move around. A stationary kite is a line downwind, a kite flying crosswind is more of downwind cone. But the direction of the wind changes so the size of the cylinder you need to reserve is the same. The longer your line, the taller and wider the cylinder, or hemisphere. Or you could argue that you only want to reserve the cone that is occupied in the current wind direction.

But scaling and spacing is not important now. Can you sell me something that has a reasonable capacity factor (i.e. is not too heavy), and can, eventually, in principle, compete with solar on price? I’ll worry about where I’ll put it and if I want to buy more than one unit.

Here: Kitewinder update is one initiative that is somewhat similar to what you’d like to do. (This is more similar).

Some more info: Physics of tethered autogyros, autogyro rotors, and windmills at an angle of incidence to the wind

Yes. I think there is three main reasons: kite speed increasing the risk of collisions and their severity; low elevation angle (about 30 degrees) in order to mitigate cosine loss; space not used (swept) within large figure-eight or loop. The last one can be mitigated by using Low radius loop.

Conversely stationary kites can fly close to each other thanks to both no speed, high elevation angle (e.g. about 60 degrees for Kitewinder kiwee).

The video below shows how a multitude of stationary kites can fly with low spacing. People can walk between stationary kites, but not between crosswind kites. Crosswind (steerable) kites cannot fly in the same area: a specific area is used due to problems evoked above and that are well known by kite festival managers. I had to use it for my demo, even for a small crosswind flygen AWES.

I remember also that nobody could stay below the tether of a crosswind AWES in demo (some kW range) at AWEC 2013 Berlin.

So Power to space use ratio is a significant parameter. If crosswind kites are used, the whole swept area should be maximized or/and the site being away from any dwellings. It is not impossible but different architecture should be studied, comprising crosswind, crosswind-rotary (Daisy), stationary AWES (@Kitewinder Kiwee).

1 Like

OK here is the well-established requisite terminology for aviation, aerodynamics (fluid dynamics) and wind energy:.

  1. crosswind: your working surfaces travel across (perpendicular to) the wind direction. A regular wind turbine rotor is of course so configured by design. The turbine as a whole may be stationary, the blades move crosswind, usually at many times the ambient windspeed.

  2. lift: the force generated on the suction side of an airfoil. This force got its name from its original description of holding an airplane up, but may be in any direction such as down for a racing car.

  3. lift-based wind energy system in wind energy: any system where the working surfaces move faster than the wind speed in the direction of load driving. This traditionally includes regular propeller-style wind turbines.

  4. drag-based wind energy system in wind energy: any system where the working surfaces move slower than the wind speed, in the direction of load driving. This traditionally includes high solidity farm windmills for pumping water, and turbines with working surfaces that are “dragged” downwind such as Savonius turbines, cup anemometers, etc…

  5. cross-axis, vertical-axis etc., describing a wind turbine: A rotating wind turbine where the axis of rotation is oriented perpendicular to the wind direction.


This isn’t really right. For a start no stationary kite will be making power.
Yes a flown-cone moves around with the wind and covers a 360 deg field volume.

So being more efficient with the length of that cone or cylinder as it rises and grows is important.
The more power we can generate per length the less material we have to use, the less land we affect, - Lower LCOE, EOREI…
A more efficient cone or cylinder packing might be able to fly a long line beside other cones or cylinders.
Like WESD - Large-eddy simulation of airborne wind energy farms
Needs lots of control.

A rotary downwind line still cuts a cone, and can have multiple generating elements on it. So it may be generating using many blades per length … A multi blade yo-yo type can possibly also generate using multiple blades on a single line. (See @rachel Optimal control of stacked multi-kite systems for utility-scale airborne wind energy or Kiteswarms) Also a rotor can be used to drive a yo-yo design

Scaling and spacing - That’s an issue, - you’ve just recognised it. It is important now. It’s not a good idea to ignore it. It is important now.
You can’t just sell 1 solar panel stacked on top of another… recognising that before deciding to make solar panels would change how you started to design them

1 Like

It’s not important now. It might become important later, if later ever arrives. The important thing now is getting something to fly at all, and reliably. The rest is down the list. Generate revenue to buy yourself time to improve, or more likely completely change, your design.

That’s like being prepared to chase down a blind alley knowing it’s a dead end.
Or like firing endless junk into space hoping it won’t crash into your precious satellite.

We don’t have time for that luxury.
You can’t possibly justify the argument that efficiency isn’t important.

AWES has a mission.
Bite the bullet and get on with it.

Nah some people can’t be seen to change their mind @Windy_Skies
It might suggest they’d been un-scientific-ish.
There is a chance however that some people would be able to recognise the freak chance discovery of a design with remarkable properties which is able to operate efficiently from the ground up and with mechanical autonomy.

Not saying all yo-yo folks are NUTS. Would be worth still testing a few of them surely. The focus of AWES academic work has been MILES OFF COURSE for too long


Did I quote the wrong part? I’m not talking about efficiency, unless that’s about the size/efficiency of the reserved sky your design needs. I’m talking about scaling and spacing. Efficiency with materials would influence the cost of your single unit so that would be important, not now but sooner than scaling and especially spacing.

You could conceivably try to sell a system that uses a lot of space, but you’d quickly go out of business if your system broke every hour or week.

Efficiency with materials directly affects scaling ability and reliability.

As far as I can tell - all single line, single kite designs have gone wrong here.

Sharing lines reduces drag.
Limiting the scale of your blades matches them to realistic wind speeds and lowers control needs.
Nets of blades remove control mass.

Efficiency with airspace and ground space use is essential from day 1

The disclaimer I can offer you is
My view may be a product of my environment
It’s fairly windy at ground level too

1 Like

When I said TRPT had seen too little effort relative to yoyo and flygen, that is mostly due to the lack of effort and not necessarily that TRPT is a more promising direction.

Each method comes with its own challenges and drawbacks. I dont think anyone on this planet has a clear picture of what the outcome will be…

Also I must state again that these are my personal views, not necessarily those of Kitemill

1 Like