Makani had huge amounts of capital from Google to work with. As you point out there are other much smaller players in the field who keep the generator on the ground and generate power from the upward pull of the tether. I worked at one of these startups. But after Google canceled Makani it became very difficult to find funding because angel investors figured that if Google got out of the game it must not be viable. And so we folded.
Making flying things lightweight is a means to an end, not the goal or a hard requirement. Makani experimented with a wide variety of solutions before settling on their final model. The initial work on soft kites had serious durability issues, IIRC. What they ended up with looks a lot like a very lightweight airplane, for good reason.
[Former Makani software engineer, here. I wasn't involved in the early stuff, or the design trades, but I heard a lot about it.]
I'm realizing my phrasing was too critical. I think it's a very good thing that Makani explored that "design path" and shared the knowledge acquired with that level of details.
I would also say that rigid wing has more potential for AWE, but I am pessimist toward 'fly-gen'/'energy-kite' solutions, because when dealing with a gas as bearing medium reducing weight seems a reasonable and trending rule of thumb.
It's difficult/too-vague to compare designs with substantially different ramifications, some of which are still hypothetical, but it's maybe worth if it leads to a better understanding or general awareness.
I am curious, do you know why Makani went 100% 'fly-gen'?
From by biased point of view I would say that it was easier to bootstrap (more control agency, easier take off and landing) but harder to scale.
I tried to skim through the 1k+ pages 3 part report available from Makani (https://x.company/projects/makani/), but it doesn't seem to speak much about the considerations which when into that early design choice. Do you know more?
Is it possible to summarize them or it's an organic set of reasons which cannot be untangled nor simplified like often happen when dealing with complex systems?
Because Makani was by far the best funded startup in this space and they got very far. Very easy to understand why I'd say.
As to fly-gen, it has several benefits, like continuous power delivery and not least an easy way to launch and land - you already have the propellers.
The main downsides are complexity, weight, and the weight and girth of the tether needed to transfer the power. The drag from the tether is significant. Plus the sound of the blades.
What's the reason for the emphasis on weight? From reading the technical reports (from Makani), there were larger issues with the design unrelated to the weight.
As an example, Makani's Y-bridle design to attach the tether to the kite introduce stability issues during hover, and limited their control when the kite was aloft making circles, which in turn limited their power generation.
> What's the reason for the emphasis on weight? .... As an example, Makani's Y-bridle design to attach the tether to the kite introduce stability issues
It's quite possible that if their kiteplane was much lighter that these tether issues would have been much less of a problem. Remember that the Makani tether also had to include some copper cable for power transmission to the ground. That makes the tether itself heavy as well.
It seems to me Makani's problem is they wanted larger scale power generation, and of course can't rely on constant wind, so needed a way to settle and recover. Of course that's not a problem for smaller-scale builds like the OP.
Very good documentary on the project. It's amazing what you can do with basically an unlimited budget. I worked at a much, much smaller kite power startup that folded soon after Google stopped funding Makani (investors lost interest because they figured if Google wasn't funding in this space it must not be viable). Anyway, looking at that documentary it seemed clear that Google must've poured hundreds of millions of dollars into that company over it's lifetime. So many engineers and so many resources. I was kind of envious, but then again the ready funding might have been part of Makani's downfall, you're forced to make different design decisions when money isn't so plentiful.
Well that's just utterly overcomplicated and overengineered. No way they could produce that at scale for dirt cheap for underdeveloped remote areas, which is what this sort of thing would be the best application for.
Compared to this project that's essentially a few rings and a bike transmission.
This is cool, but I’m not seeing how this could be useful in unattended application. It will need relaunching on a regular basis, and will wear a lot. Even string wears Internally just flying in the wind.
My guess is that except in some specific applications like winter use in polar regions, solar power will end up producing more energy for the same investment over time.
Cheap solar is hard to beat, and the new printable panels promise to make it even harder.
Solar is just a lot of energy compared it to kinetic sources. It’s counterintuitive.
A typical single solar cell from a panel (5 watts) does the work of lifting a 5kg weight 300m high every hour. It’s kinda mind blowing.
Automated relaunching could handle the unattended part. As for applications perhaps it could work as a compliment to solar panels for deployable night time power generation? Given general trends of economies of scale, huge tower wind turbines perform better on a permanent basis but kites would be far more dispatchable and transportable in a casual way. Might be roughly similar in niche to a gasoline or diesel generator aside from the obvious limitations exchange (not burning fuel for power vs being usable indoors or underground with proper ventilation, smaller footprint).
Raising a mass of m = 5 kg by a height of h = 300 m under gravitational acceleration g = 9.8 m/s^2 represents an increase in gravitational potential energy of E = mgh = 14700 joules.
Power is the movement of energy over time, measurable in joules per second, or watts. Moving 14700 joules in 1 hour = 3600 seconds represents 14700 J / 3600 s = 4 watts of power.
If your body was perfectly efficient at converting calories into gravitational potential energy, the calories in a single can of coke would be enough to carry you 1km into space
Is there a rule of thumb for the true efficiency of that calculation? I.e if I drink one can of coke, how far can I climb a ladder on the energy I metabolized from it?
> My guess is that except in some specific applications like winter use in polar regions, solar power will end up producing more energy for the same investment over time.
Or energy generation at night and in cloudy areas.
A portable kite turbine, tested on the field and that is a finished product, not a prototype. 24V DC, 100 watts guaranteed with 25 km/h winds, a very conservative estimate
Nice! Basic description: Parasail kite with a small bladed turbine hanging off it. 100 Watts, 4.5 kg, no tools to assemble. €900 is not cheap, but the price should come down eventually.
I seat the office next to the team of people who designed it and who are currently producing it. Cost is mostly because of the current low production volume : they can't buy full containers of screws, bearings, motors, they can;t order thousands of injection molded parts.
They assemble it themselves. They make the control board mostly themselves, apart from the PCB which is made in China. They also produce the plastic (molded resin I think) and metal parts (we have a big CNC mill) themselves, so they can handle low production volumes without a big upfront investment.
They recently tested it on the Brittany coast under heavy winds, it did fine. It was tested a lot on Aquitaine coast.
That all sounds so familiar. I was involved with design and low-volume production of a complex optical assembly long before AliExpress. We tried using surplus lenses and prisms, which were practically free compared to new or custom stuff, but the inability to standardize was ultimately more costly in terms of re-engineering, searching for parts, etc. Balancing all that out was an ongoing challenge. I remember finding a factory in India that could make us prisms at a very reasonable price and it was a huge relief. They came in individual handmade pasteboard (cheap frangible cardboard with surface layers of very thin paper) boxes with hand-lettered labels glued on. Old school.
I think these will be very appealing for campers, off-grid, emergency and other backup situations when they can hit a somewhat lower price. I understand solar panel costs are dropping, but it isn't always sunny everywhere!
Agreed. Large-scale might be a stretch, but the low cost and weight, open source design, and the ease of transport would make this ideal for temporary installations, eg what you mention, emergencies or even in the developing world.
This is interesting for replacing small wind turbines, which often break before they recovered their embodied energy[1], but I have a hard time seeing this technology replace large scale commercial wind turbines. Kite based wind power has been talked about for years, but somehow it fails to materialize. I assume cost and safety concerns play a role.
Launch costs. Kites work well enough when the wind is blowing but a lull, or even a quick change in direction, can see them crash to the ground. Then someone has to launch them again. I cannot see them ever being practical outside of those very rare places where the wind literally never stops.
Makani's kite is steered. This one just needs to stay aloft. Launching it is just a matter of a wind-up spool on a windvane-like small rotating platform (no more than a few m high)
> Makani's kite is steered. This one just needs to stay aloft
The implication is that you are suggesting that autonomous steering is neat and the other person is suggesting that not needing steering is neater.
Also, Makani was steered into the ground literally and figuratively. It's been cancelled. It will never do what it said it could do because its creators won't keep doing it. It's over.
I'm simply agreeing with the guy that said that launching is expensive. It's indeed the hardest part for any kite energy system, I know this from experience, I spent several years in this industry.
If you could live with having to launch and land the system manually, that would change the equation completely.
The autonomous control of the kite is one of the big issues. It was basically not possible when the idea was first presented in the 80s - there weren't computers small enough with enough computing power needed for the task. Also there were materials problems. Now we've made a lot of progress on both fronts, but still, the control software is a huge issue. We were planning on using reinforcement learning to learn a controller for our kite (when I was in the biz), but getting enough data to learn on was an issue - the kite design was an iterative process - the kite design would change and that meant we needed to recollect data to enable us to model the new kite changes, wash, rinse, repeat.
Fly them over failed nuke plant reservations. Those have no-fly zones marked on all the maps, so you can go up where the wind is really blowing without creating a navigation hazard.
That is pretty great! But GP is right in the sense that to really extract a lot of energy you want to go higher into the air too, so why not make use of those no-fly zones for scaling this up?
Energy available from wind goes up as the third power of wind speed. Wind speed rises with distance from the ground, up to a limit. So available power goes up as the 4th power of distance from the ground.
There are very sound reasons for wanting to get your energe extraction farther from the ground.
Sure, but the higher up you can go the more wind there generally is. It's one of the big potential advantages of kite windpower over fixed blade systems.
Regarding their use of YouTube, the best way of working around that problem (at least without using a privacy-preserving provider) is to embed from youtube-nocookie.com instead of youtube.com.
My brother has one of those Cannondales with a single sided front fork. He’s also got a dynamo hub in his front wheel.
With some sturdy ground stakes to pin his upside-down bike the ground, this could make a neat source of power when your off-bike, in the field. What’s the smallest possible clone of this that you could bike-camp with, I wonder?
notice this is a vastly different project though: it's not a high-flying kite with complex steering logic; it's a small wind wheel that transmits torque to the ground. If this things crashes to the ground there is basically no expensive damage.
I was referring to the cost of the equipment getting destroyed by crashing: makani's (and others) kites are pretty expensive and crashing one is a Big Deal, it does not seem the same in this design.
Potential for damage and energy are probably correlated, in the sense that this design seems to attempt to harvest low-altitude winds (30m), while makani and others wanted to get the high-altitude stronger and more constant ones.
This design does not have an heavy high flying turbine, it has a small lightweight one, and they plan to have multiple tethering ropes so I think the risk you face is in fact in the same ballpark as a hobbyist kite.
IANA wind-electricity-physics person tho, so take everything I say with a large grain of salt.
The soft parts on hobby kites get quite hard when they are strained by the wind, though. As I said - it should depend on the energy they can harvest - the same energy can also force them into the ground, or onto your head.
Flying conventional wind turbines aren't that good, they suffer from most of the downsides of both normal wind turbines and kite turbines.
The idea with kite energy (in serious projects) is to eliminate the need for a tower and the inner parts of the blades.
The kites can be seen as the tip of the blades, which is where almost all the energy is generated.
Kite energy can dispense with the tower because they are tethered with a largely horizontal cable, so there is no lever involved and the foundation can be much smaller. You could sort of do that with the flying turbines, but that wouldn't really work in a city environment.
I was actually thinking about an alternative energy source from the sea... it would float on the surface and be like a set of interlinked chains in a fabric with dynamos in them that the simple harmonic motion of the ripples of water constantly could be used to create electricity.
Let's suppose out at sea each peak to trough is 60cm and the movement up and down causes the internal dynamo to be driven by 30cm for each one - you probably need two lines into each one, positive and negative serially and you could get a pretty good average amount of electricity probably.
No idea if it would work or the output would be too uneven but there is a lot of sea and it seems a bit easier to build and repair than something that sits statically in the water.
A lot of work has gone into wave power, but broadly it's not got anywhere - there's a lot of complexity in building machines which are robust enough to deal with all sea states and the hostile marine environment, and the amount of energy which can be extracted hasn't made any of the trials so far particularly cost effective.
Looks very interesting.
Didn't find a word about birds' safety on your landing page.
This is even less visible to birds than traditional turbines, if I am not mistaken.
2.4 billion birds are killed by cats every year. 600 million by flying into glass panes. 200k by flying into wind turbines. (these are US statistics, so not global numbers)
The bird death argument is completely meaningless in several aspects. I'd rather make cats illegal than put any kind of breaks on wind power expansion.
Your statistics are of course true, but cats and wind turbines kill different species of birds. It does make sense not to put wind turbines where endangered birds are nesting.
Ok, then either post better ones or say why you don't like them, or what is the reason for distrust. Just saying "I don't trust these numbers" adds absolutely nothing to the discussion. Imagine if I said "I trust these numbers" - is that helpful?
In any case, those are probably just napkin calculations. And I bet I can find a wild variety of estimates for birds killed by wind turbines, too - depending on whether the person doing the estimating wants turbines to succeed or not.
It seems likely they don't kill the same birds as wind turbines, either. And not in the same locations.
All in all the comparison might just make not much sense to begin with.
How many birds are being killed and eaten by humans, for example? Probably several billions more, but they are not the same birds that are being killed by turbines.
Actually, yes, that would indeed be a good use of the concept - anchor the generator to a submerged bouey (so as to not disturb shipping) and it will always be pointing in the right direction and generating. Nice!
Morphologically interesting here is these are HAWT at the top and VAWT at the bottom.
A kite isn’t the only way to do this, could alt be; some kind of semirigid textile mushroom; suspended between buildings or across a chasm; integrated into a built structure; an inflatable; an aerostat; and so forth.
For small scale energy production small VAWTs [0] are relatively cheap, they offer more W/m2, and they are much more easily serviceable and installable.
For large scale production it's orders of magnitude less efficient, plus having multiple kites on the same plane brings in the wake effect [1] in full force. That's why regular turbines are generally laid in lines rather than wide surfaces, or with significant separation between them.
Okay, but why is everything wind-related so impossibly big? What's the physics behind needing blades so big that they can't fit into a cargo container? Instead of several very large turbines, why not 1000 or 10,000 of them?
Regarding the size x quantity, I believe the reason lies in where the best wind is, which is usually high above the ground. The roughness of the terrain and obstacles generate shear and turbulence, which translates into more stress for the components. The higher wind has a more uniform distribution across the rotor and is higher in magnitude than in lower heights. So for small wind turbines to have access to the best wind, you would have to build expensive structures to reach there, making it infeasible. Hence kite approaches like the one posted and Makani (with different principles).
And not only is there more output, it's more valuable per unit.
Higher winds are much more stable. Both in having less completely still days, and having the median be much closer to the maximum. A wind turbine that is built big enough starts having a large part of it outbut be effectively baseload instead of intermittent, and as more renewables are built out, baseload capacity is increasingly more valuable than intermittent.
Yes, it's not exponential at all. Larger blades also need to be thicker, and the tower stronger etc. But still, it's economies of scale like everywhere else.
I wonder if you were to design a conventional turbine that was made out of a telescoping rod. The blades were CF with fabric like an old wind mill, and then gearbox/motor/charging circuit... say it was only 10 ft tall, compare it to this (power).
Flying things need to be lightweight.
Electrical generators are heavy.
Makani: Ok I will build a flying generator ... :thinking:
Related Keywords:
- Airborne wind energy (AWE) (the name of the field)
- Fly-gen (mechanical to electrical conversion happens in the sky)
- Ground-gen (mechanical to electrical conversion happens on the ground)
Other resources worth of attention:
- On the edge, possibly near to commercialization: https://www.ampyxpower.com, ...
- Commercial solutions: https://thekitepower.com/, ...
- Project similar to the one linked (rigid wing): https://www.someawe.org, it also have a nice conceptual map style overview of the field https://www.someawe.org/awe-map-the-someaweorg-airborne-wind...
- Forum with knowledgeable people on AWE: https://forum.awesystems.info/