Hacker News new | past | comments | ask | show | jobs | submit login
Salters Duck (wikipedia.org)
84 points by atilev on Jan 30, 2015 | hide | past | favorite | 39 comments

Total available wave energy relative to present human energy consumption is minuscule.

Realize: wave energy is solar power twice removed. First solar heating converted to wind energy, then wind to waves. If you can capture the initial solar flux, you're better off. Wind is your second-best option, and where that's locally high, the EROEI (energy returned on energy invested) is quite high.

Marine installations are notoriously high maintenance. Water, salt, corrosion, marine life, snagging, navigations hazards, and other factors, all play a role. Few man-made (or natural) materials survive well, and those which do tend not to play a big role in power generation.

Even tidal energy, with both more available power and greater concentration than wave energy, has at best limited potential. As an example that may be relateable to HN, the energy avaiable from daily fluctuations in the San Francisco Bay, at the Golden Gate. In 2006, EPRI (the Electric Power Research Institute) established that the Golden Gate has a tidal energy potential of 35.5 megawatts of total extractable average energy power (852 MWh/day, 310 GWh/year), of which 15 to 17 GWh might be practically exploitable at costs comparable with current wind and natural gas projects.

The total tidal energy represented at the gate is 237 MW (average). Capturing the entire amount would require blocking the Bay to any and all shipping traffic, or installing locks or other means to allow shipping in and out of the Bay. The total daily energy (5,688 MWh) is still well below the 18,000 MWh consumed by San Francisco daily), though it is a healthy fraction at 32%.

Note that that's the power draw _of San Francisco alone_. Not the greater Bay Area.

Yes, all that water flowing in and out of the bay daily and the "immeasurable energy" it represents? About 1/3 of the City of San Francisco's electricity usage.

Human beings use a LOT of energy.

http://www.sfenvironment.org/article/hydro/tidal-energy http://oceanenergy.epri.com/attachments/streamenergy/reports...

What waves and tidal energy represent, though, is something which might be a lot more capital efficient.

Yes you're much closer to the power source using photovoltaics but they're only 20% or so efficient right now commercially. Further, the sun only shines on where I live between 8 and 16 hours a day, depending on the season. The useful average power I can expect to extract on average is about 3-4 hours per day.

Waves, on the other hand, are an excellent store of energy and they persist for a long time. Easily overnight and actually over many days. So if you can reliably produce power from waves for the same capital cost per watt installed, you might get 3x-5x the energy production from the wave plant versus the solar plant. And that's before you take into account the storage necessary for solar!

In other words you might be able to spend 10x on a wave power plant as a solar plant and break even. If it only costs 2-3x as much per watt, well, you're saving money. And the batteries (the ocean) are free.

Energy storage is a challenge. The most effective (and most used) form of bulk storage remains pumped hydro -- it's something like 95% of all energy storage. Round-trip efficiency is high, technology is (relatively) simple, highly proven, capital costs are (relatively) low, capacities are high, response time is rapid.

The disadvantage is that there just aren't all that many places which are well-suited to pumped-hydro placements. Areas with lots of water frequently have little geographic variation, and vice versa. And lakes with daily level swings of several to tens of meters are a bit dodgy for various reasons.

The rest of the storage landscape reads as a list of odds and sods and bad options: batteries (small-scale, expensive, high maintenance, but fast-reacting and proven, some interesting large-scale potential), flywheels (small scale, really expensive, lots of engineering challenges, but very fast reacting), capacitors (minuscule scale relative to grid, though useful for power conditioning, horrendously expensive, blazingly fast response), compressed air (Boyles law), thermal (sizing and other issues, though potentially national-scale capable).

Then there's power-to-fuels. It's not very efficient -- you lose about 50% of what you put in going to fuel, Carnot's law robs you of all but 35-40% on the way back, leaving you with 17.5% - 22.5% round-trip. If you factor in solar -- 20% efficient panels at 20% capacity factor, 90% inverter efficiency, 55% spacing factor, you're down to per-unit-area storage efficiencies which are lucky to rival biomass -- growing plants and burning them. Main difference being that you don't need to water solar panels -- in theory. In practice, washing the dust off of them helps, though they tend _not_ to get attacked by insects or birds.

I've been meaning to tabulate various energy potentials, though Tom "Do the Math" Murphy's got a pretty good list of options. I really don't see wave as worth the effort.

Which technology do you think is the most worthwhile one? Nuclear power?

"Worthwhile" needs unpacking.

For all practical purposes, we'll be solar powered in the future, though my question is at what level of technology.

Woodfuel's proven robust for millennia, though there's the forests before, deserts after problem.

In terms of abundance and accessibility, I'd generally say direct solar (PV or thermal), either wind or hydro, geothermal, and biomass, in about that order.

It's far less a question of what's worthwhile, than what's a sustainable scale.

It helps to understand what life is, and how it has transformed Earth over the past 3.5 billion years. Humans are part of that process, and, from the PoV of Earth have largely been engaged in one of the fastest redistributions of mineral resources ever seen, post primary-bombardment.


As for nuclear generally: conventional plant designs suffer from limited fuel (80 years at present rates of use, 6 if scaled to all present human energy demands), breeders entail considerable risks, proposed thorium MSR designs haven't even landed on, let alone left, the drafting sheet, and there are complex long-term systemic risks for which I find sgcollins's "Nuke Skeptic" and "For Rohan" videos quite good. If you want more depth, I recommend Charles Perrow, The Next Catastrophe and Normal Accidents.

http://fixyt.com/watch?v=KOBEloymLQA http://fixyt.com/watch?v=S5hvY3atwUg

http://www.powells.com/biblio/65-9780691150161-1 http://www.powells.com/biblio/62-9780691004129-1

Solar: purchase panel and leave it in an incline for 15 years. If you have net metering you don't need to even think about storage.

Wave: develop massive steel structures in the ocean and transmit power to land with low power density.

It's telling that, given the possibility to pick anywhere to develop wave power (even at the most convenient locations) the tech has barely any projects.

At some point someone has to consider storage (or alternative provisioning, or dispatchable load) for solar-backed grid. Even if it's not the homeowner with panels on their own roof.

This is where I'm hoping the combo of electric cars and solar will pay off. The car can provide a local buffer when the family is at home during the night, but isn't needed when the sun is shining.

I think the missing piece of this puzzle can hopefully be filled by something like Ambri's (previously LMBC) liquid metal battery tech [1]. A shipping container or two per suburban block, or urban building should be able to provide enough buffer, and then the load is sort-of evenly distributed, and hopefully makes the power management of the grid a little bit simpler as a side effect.

[1]: http://www.ambri.com/technology/

Theoretically with solar panels you could have a round-the-clock, round-the-globe balanced network requiring no storage. It would be a network with the same capacity of panels on each meridian. Of course this is not very practical.

Not only not very practical but it would require entirely new transmission technologies: the maximum practical length of a 60Hz transmission line is only a few thousand miles. To go from the bright side to the dark side of the earth is worst-case 12,000 miles so new tech would be needed, or you'd just have to suffer 30% (or more) transmission line loss.

The tech has had many projects, but none of them made it in the longer term.

I like this concept in particular, because it has no mechanical parts in contact with the water: https://www.youtube.com/watch?v=gcStpg3i5V8

Edit: It seems however that tidal power is a more interesting research area [1]. I still find it's a neat idea, it also makes awesome noises: https://www.youtube.com/watch?v=IK65S0sPtsg (loud)

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

A friend of mine worked with Prof. Salter. He's an incredibly brilliant, smart and kind scientist. Unfortunately, much of his research was 'blocked' due to someone miscalculating the cost by twice the amount, starving the research field of funding until recently. There are quite a few related spin-outs (e.g. Pelamis Wave, website currently down). Edinburgh also just got a huge wave tank for more experiments (http://www.bbc.co.uk/news/science-environment-27702506).

That's the problem with cheap oil, like we have now with fracking, it kills all market incentive to develop clean energy. If we're ever going to get serious about climate change we need a dynamic carbon tax that keeps the price of fossil fuels stable and high.

If we had such a carbon tax right now, that kept gas price at the previous price of $4/gallon, it would be a huge windfall that could be used to spend on clean energy incentives as well as low-income energy subsidies.

well no, because most renewable energy sources aim to produce electricity. Very little electricity is generated using oil, so the oil price should have little impact on most renewable energy production. The culprits are cheap gas and coal - especially coal, which really should be heavily taxed because of the huge health and environmental impacts from mining and burning coal.

A few corrections.

Within the US, little electrical generation comes from oil (about TK%), but outside the US that level's rather higher -- something I only learned recently myself when trying to rebut the statement that oil was significant in electricity generation elsewhere.

Hrm. Still not all that big. 256 Mtoe per the IEA in 2012, of 6,497 MTOE, about 4% - 6% depending on how you read net global imports / exports. (How does the world import/export to itself?)


Though in many parts of the world, electricity comes from gasoline or diesel generators on small scale rather than central generation plants.

There's also the fact that renewable energy can produce liquid or gas fuels via the Sabatier process (resulting in natural gas) or Fischer-Tropsch fuel synthesis (liquids). These aren't used in bulk now, but could be, and represent a possible carbon-neutral gas and liquid hydrocarbon fuel source in future.

>Very little electricity is generated using oil

~80% of Hawaii's electricity is generated using petroleum.

Lower electricity costs vs. petroleum costs also make electric cars a much better deal.

It's all linked because fracking produces both gas and oil. But I agree coal is the biggest culprit. That's why the tax should be on carbon, not a specific fuel. But the tax should be designed to be dynamic somehow to act as a stabilizer to keep prices from going to high or to low too fast.

"DUCK EFFICIENCY on COMPLIANT AXIS" is a great title for a graph. It would make for a pretty amusing piece of art.


Wasn't it the title of an ELP album from the 1970s?

Wave energy is one of those ideas that 100's of clever people have had independently over the years and not a single one has been able to make it work reliably and cost effectively.

We've seen a wide variety of prototypes, some even made it to net power out but absolutely nothing came close to winning the prize. (Though that didn't stop the government from pumping large amounts of subsidy into some of these.)

When you compare 'wind' to 'waves' (or tides, for that matter) the environment is even more hostile, the power transportation problem more acute and the engineering challenges incredibly more formidable and the maintenance headaches harder still.

Having mostly stationary waterborne structures is hard enough and commands skills that are rare and very expensive (think oilrigs). Now change that to dynamic structures which drive generators, add in storms, salt water and interference with shipping and you're looking at a very expensive and complex mixture. It will take extremely deep pockets to pull this off successfully, with the inbuilt assumption of a number of false starts and resulting re-engineering.

Elon Musk will likely walk on Mars (or die on impact) before we will see widespread adoption of tidal energy for electricity generation on Earth. Which is a pity, because it seems such a good idea.

This one is in use in Scotland: http://en.m.wikipedia.org/wiki/Islay_LIMPET

500 KW (now derated to 250KW), it's neat but that's piddly little bits of power for an installation that size. A single present day wind turbine will produce up to 3 MW continuously rather than pulsed.

Pity I didn't know about it when I toured Scotland a couple of years ago looking at their renewable energy installations.

This one was by far the most impressive:


The problem with wave energy is not to make a device that makes energy from waves, but to make one that can withstand the storms that invariably are present wherever there is wave power resources, and at the same time remain efficient. Most wave power prototypes have ended up as wreckage during the pilot period.


The Pelamis, based on technology evolved from the Salters Duck at the Edinburgh labs, was installed in pilot installations for a few years through the company has sadly since gone bankrupt

I couldn't find any pictures or videos of how this works.

Interesting overview of Edinburgh duck R&D. https://www.youtube.com/watch?v=_bdeNuRF-yE

A lot of wave and tidal motors were built along the California coast in the late 1800s and early 1900s. It's an interesting history. Cheap fossil power made them uneconomical, though.



Still never understood why they don't concentrate on using areas where currents are well known, for Europe that would be the Straight of Gibraltar. There is energy in the outflow from the Med into the Atlantic at the bottom layers, surely capturing energy like that is not as difficult as surviving the forces waves put out.

I've read another article about a similar device in Brasil:


Tidal generators are in production. There's a pilot program installing them in NY harbor. They're fully submerged and I don't think they're ducks.

this is not about tidal power

Does it generate DC or AC? I think there are losses if you want to transport and connect the power generated up to the grid

It generates pulses of AC power, but that can be overcome by feeding those pulses into a flywheel or superconductor based short-term storage system.

What's more interesting is that Salter was Studying AI at the university of Edinburgh

It sounds like a brilliant idea, how come these things aren't all over the North sea?

Because waves don't transfer that much energy. There's a reason why ships had sails, even though winds were highly variable and waves somewhat less so. Also, maintenance on anything you put in the ocean tends to be high.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact