[rwallace]

I think I am missing something here, because I don't quite get how the wind does continual work on the device to generate power. I mean, at the start the device goes downwind unreeling a tether, so it is pulling on the tether and doing work, so far so good. Then it reaches the maximum extent and... starts more or less staying where it is, staying aloft? Okay. But it's not systematically moving, so no longer doing work. What am I missing?

[myself248]

I couldn't find a good explanation either, but my guess was different from some others I've seen here:

I think it doesn't generate continuously. I think the plane spends some energy to get up into the fast-moving air, while the tether pays out some of its length easily to let it climb. I think it only does this "once", or as infrequently as possible.

Once the plane is in the fast-moving air, the generators kick up their field current to extract energy from the tether being unreeled further, as the plane turns broadside to the wind to haul the tether out despite this drag. It's generating power as the tether pays out.

Then the tether runs out.

So the plane dives/glides back towards the base, while the generators run in reverse, reeling the slack back in. But it doesn't return so far as to get out of the good wind. So as soon as most of the tether is reeled back in, the plane turns to haul on it, the generators go back to generating, lather, rinse, repeat.

I think it only lands when the wind dies, or if it gets too strong to remain safely in the air.

[brandmeyer]

This is one scheme that runs the generator on the ground. Makani and Ampyx both run generators in the air and transmit power down to the ground on the tether.

Here's an (unkind) summary where the players are: https://cleantechnica.com/2014/03/03/airborne-wind-energy-pl...

[brandmeyer]

Imagine for a moment that you built a wind turbine generator with just one blade. Instead of two other blades, you just had a counterweight. Then replace the tower and hub with a tether, and drop the counterweight.

That's effectively what these technologies are doing.

In point of fact, all of the blades on a commercial wind turbine are shaped like wings, such that the reaction force of the air flowing over the wing acts to drive the wing in the direction of motion and adds a drag force away from the tether.

In the classical tower arrangement, you extract energy by opposing the rotation of the blade about the tower. In the energy kite arrangement, you extract energy by running the "propellors" in reverse as generators to oppose the forward thrust.

[jacquesm]

Single blade turbines approach the maximum possible under Betz' law regarding power extraction from a moving air mass but suffer from resonance and material fatigue to the point that it offsets all advantages. Twin blade rotors suffer from thump and material fatigue as well, three seems to be the optimum and is what almost all commercially available turbines use.

Kites are a complication that nobody needs.

[brandmeyer]

The single-blade model reduction is just a rhetorical device used to guide the reader on the principle of energy generation, not on viability.

As you point out, single-blade turbines have been done but they have their own suite of mechanical problems. A more practical way to save capital cost is to under-size the electrical system relative to the blade area. You may still only get 30% capacity factor of the blades and tower, but you could get 40% or better on the electrical system for a small win.

[rwallace]

Thanks! That actually reminds me of something else I was wondering: why only three blades, leaving so much empty space for wind to pass through unimpeded? I understand adding a fourth or fifth would give you slightly less energy per blade, but wouldn't it give you more energy per windmill, for greater overall cost efficiency? Or is it the case that for some reason I'm not aware of, most of the cost is in the blades themselves, such that energy per blade is the most important factor?

[brandmeyer]

The error in logic is the part where you see so much empty space for wind to pass through. The horizontal velocity of the wind is quite a bit lower than the circumferential velocity of the turbine blade. So only a relatively short distance passes before another blade comes along.

I had a great animation in mind that shows it clearly, but I can't seem to find it right now. To get a static idea, See the corkscrew graphics in figures 4.18 and 4.19 of https://scholarworks.umass.edu/cgi/viewcontent.cgi?article=1...

Basically, there is a relationship between number of blades, blade tip speed, and flow speed that is optimized. For the same flow speed, power, and area, a two-blade turbine has to spin faster to reach optimum and makes more drag. A four-blade turbine has to spin slower to reach optimum, but it is only barely closer to the Betz limit than the corresponding optimized three-blade turbine. The difference is small enough that its difficult to measure or simulate - some other inter-blade losses rise in importance as well.

edit to add: Here's a deep dive into tip speed ratio as it applies to wind turbine design: https://cdn.intechopen.com/pdfs/16242/InTechWind_turbines_th...

[jacquesm]

It would give you more torque, so you will see this in windmills that require a lot of starting power such as water pumpers. For other purposes efficiency of extraction is more important than starting torque. Wind does not 'pass through unimpeded', the speed of the mill is carefully calibrated to slow down the maximum amount of air without causing it to pool behind the machine. That is precisely why there is a maximum amount of energy that can be extracted to begin with.

[brandmeyer]

> It would give you more torque, so you will see this in windmills that require a lot of starting power such as water pumpers.

This is the classical rationale, but having worked on tons of fluid systems, I don't buy it. Its true that you will get a lower tip speed ratio, and therefore a lower RPM for the same power output.

But in practice, that just changes the gearing required for the fluid pump.

IMO, the real reason that fluid applications used four blades and more was that humanity didn't know any better. The theory behind wind turbine optimization wasn't fully developed until the early 1980's.

[mbreese]

I don't know anymore than what was on their site, but my reading of it was that it works by moving the airframe (plane? glider?) around in the wind-stream. The tether would then be continuously moving back and forth (or in and out). I think this the meaning behind the figure-eight pattern they show.

If it's anything like some ocean wave generators, it might be able to generate power when moving in either direction, but I'm not sure about that.

[NohatCoder]

The kite goes sideways relative to the wind, spooling out the tether, generating power, and speeding up. Just like a high-speed sailing boat it can reach several times the speed of the wind this way. Then when time comes the kite turns into the wind, letting the tether spool back in at no real energy cost (but not generating power during this period). The kite then turns again to repeat the process.

Note that while there are several different techniques, they all rely on the kite moving fast, this allows it to sweep a large area relative to its size, thus working around the fundamental limit that you can't extract more energy than some percentage of the directed kinetic energy contained in the air molecules moving through the swept area.