Floating wind turbines could rise to great heights | Hacker News Reader

Floating wind turbines could rise to great heights

economist.com

135 pointslxm99 comments

Reminds me of the incredible engineering done by Makani, as told in this 2 hours long autodocumentary [1]. They shut down in 2020 because they didn't see a path towards commercializing their airborne wind turbine. They made a non-assertion pledge on their patents, and open sourced much of their code (simulations, firmware) [2].

[1] https://www.youtube.com/watch?v=qd_hEja6bzE

[2] https://github.com/google/makani

You just beat me to this! I deleted my comment. I can't agree more. If you are at all curious about this topic, the documentary is well worth the viewing!

A whole bunch of their equipment started showing up on auction sites recently - including a few fire engines. I was extremely tempted to pick up a $3,000 fire engine.

Those sell for a quarter million used, I think that's worth it barring major mechanical issues.

Yes, that's what I thought too. The only wrinkle was the logistical problem of where to park it.

Spoilers follow!

I usually don't cry at movies. I cried when Mufasa died in the Lion King, and I nearly cried when the last prototype of Makani crashed into the sea in the linked documentary.

So much good engineering, and effort, and hopes and dreams and sweat went into the whole project. I really felt for the team there.

What a great project and so incredibly nice of them to open source at least something.

There was another airborne wind turbine project that was based on a blimp filled with hydrogen/helium, the name escapes me right now.

Some motor biker drove around the big empty spaces in Alameda, put it on youtube, and this is where their HQ was:

https://youtu.be/mfDNWkwWk9Y?t=3066

https://twitter.com/ramez/status/1418261837255118849

"1. No need for ocean bed fixation hardware. 2. Ultimately lower tons of steel & cement per MW (not true now, but coming). 3. Much lower EPC costs. Build in port, tow to destination. (No jack-up ships.) 3. Lower O&M (tow back to port) 4. Better wind resource, meaning higher AEP per year (amortizing capex over more MWH). 5. Better wind resource also means higher capacity factor -> increased value, slower value deflation.

Those wind turbines are upwards of 10 MW each. That is massive. People dont get the scale of it.

If the dynamics, control amd reliability are proven, then the only hurdle would be drawing high power carrying cables through corrosive sea waters. And, build cables that withstand constant movement and bending.

Edit - Wind turbine capacity.

I believe most of the huge ones you see on the interstate are ~2MW, so imagine ~5x the scale

woooo

> imagine ~5x the scale

For wind turbines power is proportional to the swept area, and to the square of the diameter. Doubling the diameter quadruples the available power.

So for the same wind speed the difference between a 3MW (commonly installed today) land-based turbine and a 15MW turbine (largest offshore class today) would be a little over twice the diameter. Of course, the offshore monsters are also taller to allow greater swept area, and to take advantage of better resource at height.

> Those wind turbines are upwards of 10 MW each

Assuming a 35% capacity factor (taken from land-based wind turbines, offshore should be higher), that's 84MWh/day.

The average California single family home uses 20 kWh per day, which means that that capacity factor this turbine could power about 4200 homes per day.

For comparison, a 10 megawatt solar project, assuming a standard 20% solar panel efficiency and one kilowatt per square meter insolation, would require a five square kilometer solar array. The capacity factor of PV solar is inherently limited by sunlight hours, but it is around 28% in California.

If offshore wind is available 24 hours at such a huge scale, it could change the renewable energy mix considerably.

In that world the biggest advantage of solar would be lower maintenance costs and greater visibility at smaller scale (for example, coupled with batteries to form micro grids).

I think that you made a mistake with the solar farm area comparison. 10 megawatts per square kilometer (annualized, already accounts for night time and clouds) is a decent rule of thumb for solar farms. So 84 MWh/day (3.5 MW) would take 0.35 km^2 of solar farm. If solar farms were completely tiled over with 20% efficient panels achieving 28% capacity factor, 3.5 MW would occupy only 0.0625 km^2.

But offshore wind still has advantages in that it doesn't occupy on-shore real estate and can supply power at night without batteries.

> 10 megawatts per square kilometer (annualized, already accounts for night time and clouds) is a decent rule of thumb for solar farms.

Oops, you're absolutely right, I didn't square the power, and it messed up the answer by a factor of 1000!

In India, a 10 MW installed capacity solar plant occupies 52 acres, about .25 sq.km

You 5 sq.km figure, even for Cali is suspiciously high.

Maybe you missed a decimal?

Edit

https://en.m.wikipedia.org/wiki/Topaz_Solar_Farm

Topaz farm is 550 MW on 19 square Km.

Unit problem: such a turbine could power about 4200 homes; no "per day" involved.

I should have clarified: the energy produced by one such turbine in a day could provide for the 1 day electrical energy needs of 4200 CA homes.

Yes, but if there's a unit in both sides of the division, you can remove it.

I understand that, but I purposely left it in to express it in terms of energy, not power.

5 sq km is 5 million square meters, so 5 sq km would give 1GW according to your assumptions :)

Offshore capacity factor can be very high:

https://www.equinor.com/en/news/20210323-hywind-scotland-uk-...

"With an average capacity factor of 57.1% in the twelve month period to March 2020, the floating offshore wind farm set a new record in the UK."

In France 2019 nuclear capacity factor was 68.7% (379.5 TWh produced out of 63.1 GW installed capacity).

Edit: nice sortable table here https://energynumbers.info/uk-offshore-wind-capacity-factors

Thanks for the corrections on solar output! I was off by a factor of 1000 because I didn't square from square meters to square km.

> then the only hurdle would be

(In New England) the lobsterman, the fishermen, and their political allies, who see offshore wind as a threat and an opportunity, respectively.

I wonder if they're wrong to see it that way. Given these would "defend the area" from birds, that might actually increase fish stocks without really getting in the way of fishing in nearby waters. Also just marking some areas as off limits to fishing tends to improve fish stocks over time and benefit everyone.

I'd be very curious to see if fish are capable of learning that they can use the turbines to shelter from both humans and birds. They're not that smart, but life tends to find a way.

Fish are already being attracted to floating stuff, so who knows? Maybe these will also attract them.

Source: https://en.wikipedia.org/wiki/Fish_aggregating_device

Maine recently passed a law preventing offshore wind in our state waters :( . Our lobstering industry is very politically strong because they are very well united and put other political issues aside to improve the lobster industry.

However, state waters only legally go out a few miles, and our (democratic) governor, while signing the law, specifically called out wanting offshore wind power in the federal waters off the coast, but we also have a shitload of mountaintops that could have wind power, and a few installations already, as well as our biggest state university having a hugely and important wind energy lab and facility

my brother is working on an off shore wind farm, of the cost of Normandy (should 10 km from the shore, near Courseulles, I think) and currently the biggest opposition is the local fishermen too

That, and actual proper ocean storms.

California's renewables mix is currently solar-heavy. Water depth increases steeply off its coast so traditional fixed bottom offshore turbines don't have many viable locations. Floating wind would be a great complement to solar there if the technology matures.

Much of the West Coast is in the same boat.

It's interesting tech, but a big part of offshore wind power is bringing the power back to shore. Onshore wind and solar are going to have the advantage there for the foreseeable future.

I live on the Central Coast and it never struck me as very windy around here, at least compared to other places I've been. I looked it up and as I thought, the winds are further North:

2020 Offshore Wind Resource Assessment for the California Pacific Outer Continental Shelf https://www.nrel.gov/docs/fy21osti/77642.pdf

It gets plenty windy just a few miles out into the Pacific Ocean. Pay attention to the marine forecast for a few months and notice how often they include small craft advisories.

Most of the wind is 4-12 miles offshore, as the wind rolls down the coast it's blocked by the mountains that run along the coast. At 25 miless offshore you get into mesoscale wind patterns which are extremely stable and aren't impacted nearly as much by day/night cycle, and can be forecasted with great reliability days and sometimes weeks in advance, particularly in the summer once the north pacific high settles in between CA and HI.

See Duck curve https://en.wikipedia.org/wiki/Duck_curve

Essentially, solar is great from 8am to 4pm, and California has replaced about half of their load during those hours with solar. However, they have negligible renewable capacity during other times of the day, including peak demand from 6-10pm.

There's a more up-to-date version of California's Duck Curve here: https://www.energy.gov/eere/articles/confronting-duck-curve-...

Solar is so cheap that it's now worth it to point solar arrays towards the 5pm sun angle, for utility-scale deployments that get time-of-day pricing.

>"Principle’s uses a buoyant steel triangle that has water-filled cans at two of the vertices. These ballast tanks balance the weight of a turbine at the third vertex, with water pumped around inside the triangle to trim its stability."

Do they mean trimmed in real time, like active control?

It's been a while since I watched this NREL talk, but my fuzzy recollection is that there is some amount of active control:

https://youtu.be/SavaSnOOrvw

Yes, since the turbine creates different force vectors for different wind directions/speeds and the power generated is greatest when the mast is fully vertical, they actively re-trim when the wind direction and speed changes.

I wonder if balloon wind turbines might not be a better approach in those circumstances. You have to accept smaller blades but I'd think the other advantages would be worth it.

Wind turbines output grows with the size of the blade (it's squared). http://large.stanford.edu/courses/2011/ph240/parise1/ edit: When you take into account that the blade size increase means a higher tower (and thus higher wind speeds), the increase is more staggering, because wind speed is cubed in the final output:

> Combining the effects of height and blade length, doubling the height and doubling the blade length will result in an increase of power over 22 fold.

This is why companies try so hard to build bigger blades, higher.

Current blades are just massive https://cdn.vox-cdn.com/uploads/chorus_asset/file/10382023/W...

The wingspan of a 747 is 68 meters. The newest GE turbine has a diameter of 220 meters, each blade is 107 meters. One blade is 50% longer than the whole wingspan of 747. You can fit 3x747 planes inside the area swept by a current turbine.

This is why Google's Makani project never took off...I really hoped for that project to succeed, it was very cool, but there is no way to compete with the bladed giants.

edited to address cjbest and baq valid comments

I think you meant it grows as the square of the blade length (and the cube of the wind speed)

From your link: > Thus, the power passing through the blades of the turbine, and thus in turn the power of the electricity the turbine generates, is proportional to the square of the length of the blades: a doubling of blade length leads to a quadrupling of wind power passing over the blades.

Doesn’t change your conclusion of course. Thanks for the interesting link!

you're right, and I roughly knew that, I just didn't knew how to express it well enough in a short manner. When you increase the blade you also have access to higher winds, so the total output is much much bigger. I was focusing on the quote below and I didn't knew how to express it better.

> Combining the effects of height and blade length, doubling the height and doubling the blade length will result in an increase of power over 22 fold.

> grows exponentially with the size of the blade (it's cubed).

sorry, couldn't resist nitpicking. you're contradicting yourself here :)

If only it were exponential... build one REALLY big blade and power the whole earth.

The newest GE turbine has a radius of 220 meters, each blade is 107 meters.

My last maths class was decades ago, but ... 'diameter'?

ha, I think I need more coffee this morning. Or maybe i was just too excited to type that out, finally something I know a little bit about on HN

I did a research project on this.

Aerial wind turbines have two key advantages over terrestrial ones: access to steady high-speed winds at altitude, and limited ground footprint. But floating or flying imposes limits on their size and efficiency.

So they only fit niche uses in nomadic settings, like a temporary research camp. But then it's probably cheaper and more reliable to just use a fuel-powered generator.

Also they're a bit of a hazard to aircraft (and each other if not steerable), are more at risk to storms, and the floating ones require expensive or explosive lifting gas.

Too bad, because I think they're beautiful!

Yes, floating ones won't rise higher than to the surface.

Off topic: I see a lot of interesting Economist articles posted on HN. Is it worth subscribing to?

Current offers show $94.50 for first year and then $189 from second year for digital access.

Their audio versions are very well done. The reading is clear and non monotonous. You can play back faster obviously. They are a very good magazine. They have their stance, but it is consistent. For knowledge of what's going on in the whole world, it's very good. I recommend it. Those prices sound high though.

If you're interested in international news, politics, and economics with extensive analysis, then there's no better source.

If you're interested mainly in domestic news, mainly tech/science/culture, or "just the facts" without analysis, then probably not.

Heavily disagree Le Monde Diplomatique or Voltaire Network are just infinitely better for acquiring a better macro perspective of the world, if anything Economist is just renowned for bounding the scope of debate

Wow, "infinitely" better? That's quite a strong opinion.

But I'm curious how you think they give a "better macro perspective"?

Is it in the range of stories covered, and so what kind of stuff do they cover that the Economist doesn't?

Or is it the "debate", and I'm curious what positions you don't find in the Economist? Especially since their articles are very often in dialectical format, presenting competing interpretations of events as seen from the different sides involved in an issue.

I mean it's true they're not going to apply a European democratic socialism lens to American political issues... but then they're looking at European issues through a European lens.

It is one of the few subscriptions to media / news sites that I really value. Apart from their Editorials section, all articles have a balanced point of view, presenting all sides of an argument. They are all well written, and manage to explain complex topics (science, economics) in a fairly accessible way. And is also a good way to keep up to date on what happens around the world as each world region is covered and articles are often a well argumented summary of any relevant topic in that area in the past week(s). Their Christmas double issue is joy to read. But you can buy the print version at your newsagent for a few weeks as a trial.

When it comes to it, I would generally discourage from doing it, Le Monde Diplomatique is just heads and shoulders above the Economist on both breath of reporting as well as scope and coverage

They also have Spanish and English editions besides French

https://mondediplo.com/

Perhaps submit some articles you think HN would appreciate. So far not many have been submitted, and only one in 2013 got some conversation going: https://news.ycombinator.com/from?site=mondediplo.com

I considered an _Economist_ subscription twenty years ago, but I was put off by their endorsement of George W Bush in 2000. I now notice that they have not endorsed a single Republican for US president since then, so maybe I should give them a chance now.

$94.50 for first year and then $189 recurring? I would suggest staying away from this ridiculously unashamed rip-off at all costs.

Turn of javascript when loading the economist website and you can read all the article with no subscription. You miss out on the audio version though, which is excellent.

This is actually an easy problem. (Anyway, easy given the size of the things.) Make a ship that can nest the top of the tower close to horizontally, axis up. Winch the top of the tower down into its nest. Fix the turbine using cranes attached to the ship. When finished, unwind the winch, and let the tower right itself.

There will be complications. You might need to release (one or more of) the tower's mooring lines after it has tilted some ways, and re-attach them on its way back up. You will need a cable attached to the top of the tower to winch it down with, either left there from construction or attached at need. The ship probably needs a boom sticking out aft to the base of the tower, with a grabber that can swivel up as the tower comes down.

The nest and boom need to be customized for the tower design operated on, but the ship can get those mounted before it goes out. The nacelle probably has to be re-designed from scratch to be worked on from that angle.

For some definition of easy :-).

Consider the following, in order to work, the turbine (which has a fairly high mass) has to avoid being blown out of position by the very winds it is trying to harness. Because it is quite high, the distance between the turbine and the base form a moment arm of great length. So the entire system understands the torque moment on the turbine and has in place the ability to avoid tilting.

Compare that with the displacement mass of pretty much any ship you might send out there to work on it. I'm guessing that you're more than likely to winch the maintenance ship out of the water than your are going to get that turbine to tilt over. (disclaimer here, a spent a lot of my youth sailing and got a pretty good appreciation both for how "hard" ocean winds could try to tip you and how much a boat with a serious keel could resist that attempt to turn them over.)

You might also need to attach a float at the bottom of the tower, and inflate it while winching down. As I said, there will be complications. But that's what engineers are for.

Anyway, ships are often astonishingly massive.

Not disagreeing, I just tend to flinch at the word "easy" for complex engineering tasks :-).

Certainly a ballast transfer system would be useful, and there are examples of that with deep sea drilling rigs that transition from horizontal to vertical.

"Easy" is provocative. But the Economist publishing an article on an engineering problem by someone who evidently didn't bother to talk to an engineer is more provocative.

To an engineer, there little space between "easy" and "too expensive for you".

Why not assemble onshore and tow it out fully assembled rather than messing around with tilting the turbine?

Wind turbines need maintenance. Periodically the whole generator section needs to be lubricated and aligned and there are other wear and tear that needs to be mitigated as it comes up. In California, when you drive through some of the wind farms here you will see them out doing maintenance, not a job I would want given the height at which these folks are working.

But surely that would be best done at height rather than tilting the tower. Climb up from a boat or fly in on a helicopter.

Some companies are doing this, you need specially built ships. They actually assemble turbines at the dock-side, with the floating part in the water (shorter crane needed).

Videos on YouTube if you look.

Yeah but... if the barge base the tower is attached to isn't of the order as large on each side as the tower is tall, the first gale will capsize it.

It could also use an adjustable centerboard like some sailboats.

Centerboards are really only effective when the boat is moving. For static stability you need a keel with enough ballast to provide a sufficient righting moment.

Don't forget that although the base floats it is also tethered to the seafloor with chains and anchors! The mooring provides most of the stability, and if it were not tethered then it would not stay in place at all.

Can someone explain if they are not anchored to something, are they self positioning? i.e. have some propulsion to put themselves back where they were placed initially? Otherwise, these will be floating nightmares for shipping and sailing.

They appear to still be anchored, but just with a rope rather than solid steel.

I would imagine they are anchored to the seabed the same way floating oil rigs are.

The article explains all of this, and a lot more. It's absolutely worth the read if you are interested in this stuff.

This diagram shows some of the proposed anchoring systems.

https://www.economist.com/img/b/1000/679/90/sites/default/fi...

You need the power cable anyway. So it must be anchored.

BIT OFF TOPIC sorry, but - What about the crazy rain fall that's happening right now?

Could some sort of devices be placed into subway tunnels and storm drains that pushes a turbine to generate power?

Sorry if that sounds dumb, I'm just curious.

Not really. Micro-hydro is a thing (YouTube is your friend here) but there is surprisingly little energy to be harvested from small rivers with only moderate drops. Great for a someone living off the grid who wants to generate power overnight, but not worth it otherwise.

I guess turbines will slow down the water, leading to more floods.

Wouldn't it increase basin lag and so reduce flooding?

Some pipelines have passive devices to slow the content of pipes e.g. if your city water supply is coming from a higher region. A recent trend is placing turbines that generate electricity rather than heat from this process. A double win if the heat was unwanted and lead to energy being used for cooling.

Would this reduce the environmental impact? Particularly noise pollution during the pile driving for traditional turbines, that affects whales and such.

This is about wind turbines floating in the ocean[0]

There are also ideas for wind turbines floating in the air[1]

[0] no paywall : https://www.bloomberg.com/news/articles/2021-07-15/shell-joi...

[1] : https://en.wikipedia.org/wiki/Airborne_wind_turbine

Sans paywall:

https://archive.is/T5tic

> taller they get, the harder they will be to repair

If they’re floating couldn’t we have a mechanism to sink them to a reasonable height during repairs?

Or it wouldn’t seem to be that hard to have ships with large ladders or a crane for repairs.

(I couldn’t read the article because it’s completely paywalled.)

It’s technically possible, but this things are barely cost effective already. A system for sinking them and raise them for repairs would be too expensive.

As for ships, they already use custom barges, but if they’re deployed somewhere it’s because there’s heavy winds, so most repairs need to be workable even in (somewhat) adverse conditions. This also makes matters worse, because you don’t want your ship to be tied to a massive windmill if the wind picks up.

> but this things are barely cost effective already.

This really needs some citations. Wind turbines at sea typically have well in excess of 15% return on investment over their expected lifespan and are being financed without any government subsidies these days. 15+% is not "barely cost effective" territory for this kind of capital intensive infrastructure, that is "how can I reinvest this torrent of income quick enough".

“Wind turbines” and “floating wind turbines” are not the same. Floating wind turbines are not as cost effective as their stationary counterparts.

Interesting idea. I think the problem is with getting salty water everywhere. Salt is the main killer out on the ocean. Also, it you lower it enough you will end up with the blades in the water, I really don't think you want that during wavey weather.

Anyway, this extra height would solve a non-problem. It makes little difference if a human has to climb 100 or 300 meters. It's still work at heights - same kind of protocols and gear. If it's something that the humans can't fix (replace the gearbox or something massive) - then you need a crane anyway. Maybe the only issue would be with the waves moving the turbines enough to make the humans seasick at that height. Not sure, I can't see the whole article.

It's basically a fiberglass tube. Corrosion from a quick dunk is a trivially solve-able issue since everything that could corrode will already need to be protected enough fora decades long service life in salt spray. I can't see sinking it for routine maintenance being cost effective. Even at a hundreds of dollars per man hour it's still probably not gonna be cheaper than making the maintenance people climb a ladder.

That said, being able to lay over the main tube at like 60deg would make it way easier to construct them since you could use a much smaller and cheaper crane to assemble and it would minimize the side to side movement caused by rolling. You basically just need a cleverly engineered chamber to pull that off since the ship can have all the pumping hardware.

> lay over the main tube

In my opinion, the main tubes are nowhere near strong enough to support the weight that way! They are weaker than they appear. The companies can built them so high because they gave up the idea that the tower can withstand any wind thrown at it and still oppose the wind enough to produce energy. Instead, smart computer programs turn the propeller angles independently, so that when the blade reaches the top (where it exerts maximum torque trying to topple the tower), it is angled such that it won't catch too much wind. Then when the blade is near the bottom, it's turned again in order to capture most wind.

The article talks about how oil service crane ships are designed for heavy, not high. There are prototype crane ships for the task, but they cost $250 million minimum which hampers growth of offshore wind installation. The issue is the blade length and positional stability, so sinking isn’t much of a solution. My takeaway is that this is promising but a few engineering feats from being economically sensible.

Some offshore turbine companies have developed a lift that attaches external to the base and carries people up. There’s a lot of interesting technology in the space - the maintenance cost is the most difficult (at least, in my experience) so there’s lots of research and innovation happening to lower that cost. When I stopped working with them, the company I worked for was using drones to take high res images of turbine blades to better plan for what to work on, rather than just climbing up and looking.

What toll are all these wind farms going to take on global weather patterns if they continue to multiply?

Basically rounded down to none. While we continually underestimate our unintended impacts, I don't believe that's the case here. Even if we sourced 100% of the world's energy needs from wind it would be slowing a miniscule fraction of the total atmospheric movements. Think of the three dimensional space occupied by turbines versus the vast volume of air with no turbines.

I'm thinking they may have a double impact on combating global warming. Not only do they displace fossil fuel emissions, but they pull excess energy out of the atmosphere.

However, the scale they operate at is a rounding error on a global scale.

Reply on news.ycombinator.com