Hacker News Reader

Comment by beambot | Hacker News Reader

"Energy bills would be essentially zero," he says. I'm not sure how this statement is accurate (or relevant). Flywheels do not produce energy, they store kinetic energy. Other comments...

(1) I seem to recall an automobile developed in the 60s or 70s by another National Lab that siphoned off stored power from a flywheel (driven by a smaller motor). IIRC, they had hellish issues with handling since the flywheel created gyro effects as the car turned. [Sorry, I couldn't find a link to the project. ]

(2) The failure modes for flywheels aren't pretty. This seems to be the big detractor from other technologies like the MIT microturbine [1]. You wouldn't want one of 'em to experience catastrophic failure when attached to the cellphone in your pants. To get sufficient shielding, you have to substantially up the weight of the power storage unit, thereby killing your power-to-weight ratios.

[1] http://web.mit.edu/newsoffice/2006/microengines.html

View on HN

The story leaves a lot to be desired in terms of defining technology and specifying methods.

Depending on how you buy/sell power, and if you have access to TOU metering, you could spin up flywheels with cheap power and sell it back when rates are higher. If you've got your own generation on site (solar PV, wind, tidal), you could bank excess energy with flywheels.

In terms of storage, flywheel round-trip efficiency is very, very good, above 90% for well-designed systems. The problem has always been storage density and cost. Beacon Power produces 25KWh flywheels priced at around $1000 /Wh.

Add into this the failure modes (Beacon's systems are installed underground in concrete culvert pipe with a several-ton lid cap), and remaining problems with precession (even the Earth's rotation around its axis peturbs flywheels), and they have some significant hurdles. For mobile uses (automobiles, etc.) these problems are compounded.

My money's on large, cheap batteries for grid-scale storage (molten salt or liquid metal), though flywheels for short-cycle load/generation matching may play a role.

Beacon Power produces 25KWh flywheels priced at around $1000 /Wh.

Is that really a $25mil flywheel? Or should that have been $1000/KWh?

For mobile uses (automobiles, etc.) these problems are compounded.

I'm probably demonstrating considerable mechanical naivety with this question, but could you balance multiple flywheels in different directions and orientations to mitigate these problems?

Looks like I had my numbers a bit off. It's $10,000/KWh, putting each flywheel at about a quarter million. The cost data aren't clearly stated and I didn't find my earlier posts on this until just now: https://plus.google.com/104092656004159577193/posts/21sJzeSw...

Flywheels are used in cars, and the current generation of Formula One hybrids includes flywheel regenerative braking designs. While you can offset gyroscopic effects with counter-rotating flywheels, you still have the issues of loading the bearings of the individual flywheels due to vehicle vibration, movement, rotation, etc. Some of this can be mitigated by gimbal mounts, but that introduces increased volume requirements to the design, and avoiding gimbal lock requires four (rather than three) axes of rotation, one of which must be driven. It's one thing to have the flywheel last for the duration of a race, another for it to stand years of use in a private vehicle.

>Flywheels do not produce energy, they store kinetic energy.

The article says, twice, that his design both stores and generates energy. Have they got something wrong? I'm aware that they're using the term 'flywheel' pretty broadly, and his newer device might not be a traditional flywheel. They use scare-quotes a couple of times when using the term flywheel, so they seem to be saying it's not a flywheel.

> His latest project -- perfecting a prototype of his lightweight "flywheel," a cylindrical, rotating battery -- could revolutionize the energy industry, storing and creating energy better than any conventional electrobattery on the planet.

> Post has been called the "father of the modern flywheel," a title he dismisses as "too generous." The rapidly spinning devices not only store energy with minimal losses, but they also generate it.

The relevance is also mentioned in the story - he wanted to "apply physics to help solve the world's energy problems" and banks of these in power stations might help. Isn't storing the excess power generated by solar farms during the day a problem? Couldn't something like this help with that?

You can't just generate energy from nothing. A flywheel stores energy, it doesn't create it. So yes, they got something wrong.

"We said, 'To heck with electromagnetic, we're going with electrostatic,'" Post says, laughing.

It's clearly not just a normal flywheel, perhaps it can convert some other energy form into electricity? (We call them "generators" don't we?)

Other form of energy? Like what?

Using electrostatics to pump up the flywheel is certainly interesting (I wish the article had even the slightest detail on it), but it still doesn't create energy. (Or collect energy from some untapped resource.)

I'd guess that phrase refers to how energy is put into and taken out of the flywheel: using electrostatic force rather than electromagnetic.

I'm intrigued by Dr. Post's statement regarding electrostatic rather than electromagnetic force.

When I think of the "motor-generator" component of a flywheel-based energy storage device, coils and magnetic poles and all that electromagnetic to/from mechanical energy conversion stuff comes to my mind.

Electrostatic to/from mechanical energy conversion is known [1] [2], but not to my knowledge known for sustained high power and energy density appropriate for practical flywheel-based energy storage.

Intriguing to consider that Dr. Post's research might be going in such a non-traditional direction as electrostatic motor-generators. Perhaps there are some incremental efficiency gains or design simplifications to be had in using the flat plates, good insulation and high voltages characteristic of electrostatics instead of the big coils, cores and high currents of electromagnetics.

That said, there's no free lunch: No matter which electrical-to-mechanical-to-electrical energy conversion mode used may be, the energy stored remains rotational kinetic energy, thus limited by the same critical physical limitations (failure modes, maximal energy density, maximal power density) of traditional (electromagnetic, to date) flywheels.

--------

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

[2] https://en.wikipedia.org/wiki/Electrostatic_generator

The Oil Drum has a pretty good reference on flywheel technology and reasonably recent developments as well: http://www.theoildrum.com/node/8428

The Wikipedia article at https://en.wikipedia.org/wiki/Flywheel_energy_storage claims that, in comparison to batteries, flywheels are not as adversely affected by temperature changes, can operate at a much wider temperature range, and are not subject to many of the common failures of chemical rechargeable batteries. They are also less potentially damaging to the environment, being largely made of inert or benign materials. A flywheel can potentially have an indefinite working lifespan. Flywheels built as part of James Watt steam engines have been continuously working for more than two hundred years. Another advantage of flywheels is that by a simple measurement of the rotation speed it is possible to know the exact amount of energy stored. Flywheels with Magnetic bearings and high vacuum can maintain 97% mechanical efficiency, and 85% round trip efficiency.

Why aren't these in wide deployment yet as local storage for renewable energy sources? The biggest problems seems to be overall mass, cost, and space efficiency. None are great concerns for long term energy storage solutions for off-grid buildings. In short, these things sound fantastic. Anyone know where to buy one large enough to store power for a small draw, solar sourced home?

This site explains a lot more about the concept, and engineering considerations behind two functional prototypes: http://home.earthlink.net/~fradella/homepage.htm

Flywheels in vehicles were the subject on an article in the Economist in late 2011: Road tests have shown that, thanks to modern materials and clever design, a flywheel as small as a hockey puck can reduce fuel consumption by more than one-fifth. In electric hybrids only 35% of the kinetic energy lost during braking is retrievable. With flywheels more than 70% is. http://www.economist.com/node/21540386

And someone just got funded on Kickstarter: http://www.kickstarter.com/projects/1340066560/velkess-energ... .. they claim Energy storage flywheels have been in use for over 100 years and are some of the highest performing energy storage technologies ever developed. Traditionally, they have suffered one fatal flaw... they are extremely expensive! Velkess has invented a way of making energy storage flywheels at such low cost it will be dramatically less expensive than even the most economical energy storage technologies available today. Phys.org discussion of this effort: http://phys.org/news/2013-04-velkess-flywheel-flexible-energ...

The biggest problem is they don't store enough energy. Or in other words, they don't store enough per how much they cost.

The energy capacity goes up linearly by mass, but goes up by the speed squared. So all the effort has gone toward increasing the speed.

But the trouble is that the force on the flywheel also goes up by the speed squared. The only materials strong enough are wound carbon fibers - and they are very very costly.

If you increase the mass instead you need very very large and strong bearings, and they tend to fail. For high speed, low weight, you barely need any bearing at all - magnetic bearings are preferred if you can manage it.

Thanks for your summary.

The kickstarter project was targeting 15kWh which should be more than enough for a low draw residential setting. For a one-off prototype, they asked for US$50,000 funding.

Let's guestimate that with mass-manufacturing the same product could be built for US$30k, and that's going to provide a probable rated lifespan of around 20 years with zero maintenance.

Contrast batteries, which might cost 30% of that (random figure taken from a solar site) but only last 3-6 years. Sure, they are portable, but they are horrible for the environment and a pain to replace.

I still see a huge market for these.

See also a 1979 presentation from an MIT guy: http://adsabs.harvard.edu/abs/1979STIN...8010639M Technical and economic performance analyses indicate that, contrary to general thought, a flywheel system will be competitive if not superior to more conventional systems utilizing either present day or advanced batteries. This derives from the ability of the flywheel to perform the functions of dc-to-ac inversion and optimal impedance matching between the PV arrays and the load in addition to providing energy storage.

> This derives from the ability of the flywheel to perform the functions of dc-to-ac inversion and optimal impedance matching between the PV arrays and the load in addition to providing energy storage.

We don't need this anymore. This stuff is easy with modern power electronics, so if this is what makes it competitive, then it's not.

I also suspect the reliability is not there yet for 20 year operation.

And it's not like people haven't tried - I've read about a large number of companies that have tried, and yet they all seem to vanish.

Not all of them disappeared:

http://afstrinity.com/

It looks like they pivoted to other storage systems and are now finding their way back to flywheels.

They are one of the early ones: http://www.nytimes.com/1994/06/22/business/flywheels-to-powe...

But it's been 20 years and no flywheels ever made it to market.

Actually, they did sell a whole pile of them for ride-through power in hospitals and other facilities where power availability is critical.

For land installation right? Not cars?

That's still cool though - are they still in use? Why don't they sell them now?

Definitely not for mobile applications. I don't know why they don't sell them now, probably they got overtaken by developments in UPS technology. Since they started working on that we've seen 2 battery technology revolutions and a whole new family of semiconductors. On the plus side there are now usable passive magnetic bearings based on the Halbach principle and high tensile carbon compounds to help maintain structural integrity at high velocity.

Flywheels have always intrigued me, the planetary rotation is testimony to how much energy can be stored in one :) A couple of billion years and it's still spinning.

A few years ago AFS had a car prototype with a cluster of bottle sized high speed flywheels in the trunk and under the hood, it was quite a neat package but very expensive and limited range.

The most practical use of flywheels for transportation was probably the gyro bus, http://en.wikipedia.org/wiki/Gyrobus another concept that just won't die until someone brings enough garlic, silver and wooden stakes.

Another link on cars (trams in Holland actually, with backing from Alstom): http://sciencewriter.org/flywheels-spinning-into-control/

> We don't need this anymore

My understanding is that AC/DC conversion is lossy and rule of thumb figures online quote 10-15%.

> I also suspect the reliability is not there yet for 20 year operation.

A lot of products already on the market have 20 years lifespans quoted, eg. Socomec FLYWHEEL, Vycon.

> I've read about a large number of companies that have tried, and yet they all seem to vanish.

Sources? I am genuinely curious.

> My understanding is that AC/DC conversion is lossy

We don't need to do this all that often, and the flywheel is lossy too. And you can get 96% efficient power supplies http://en.wikipedia.org/wiki/80_PLUS

Plus you can use http://en.wikipedia.org/wiki/Rotary_converter and http://en.wikipedia.org/wiki/Motor-generator which are basically the same thing, minus the energy storage part.

> A lot of products already on the market have 20 years lifespans quoted, eg. Socomec FLYWHEEL, Vycon.

This could be new - I haven't been following this recently.

> Sources?

When people first started focusing on automobile energy efficiency flywheels were the thing. They eventually gave up and switched to hybrids, and/or batteries.

Note the date: http://www.nytimes.com/1994/06/22/business/flywheels-to-powe...

I don't really have any links from stories from years ago, but this one is high profile and recent:

http://www.washingtontimes.com/news/2012/apr/9/feds-call-ban...

The stores would have a pattern: Some young new company proudly showing off its new technology to the fawning journalist - and then that would be the last you would ever hear of them. (Someone once told me to put their names in a google news alert then forget about it - most of the alerts would never trigger a second time.)

> We don't need to do [AC/DC conversion] all that often, and the flywheel is lossy too. And you can get 96% efficient...

Sure, even looking at that page 10-15% does seem a regular rule of thumb though. Many devices in the home (computers, displays, etc.) require DC. Photovoltaics produce DC, but not necessarily at a level suitable for variable, direct load by devices. So in a sustainable source powered situation, you have to pay for this loss somewhere: either in storage or inversion or both. It would seem that the 1979 author was refering to the flywheel's potential toremove the 'double tax' for storage and inversion. The listed 'motor generator' benefits (load and spike isolation, etc.) would also seem to apply.

> this one is high profile and recent

Beacon are often referenced. It seems they were revived and the major facility they built is still in operation. They cited a combination of DOE policy and 'the current political environment' as their cause of temporary bankruptcy.

Because you can't shake it.

Surely if it's buried, as seems to be the general consensus for cheapest mode of safe deployment, then it doesn't shake? Also, these have been deployed successfully in cars and on various other types of vehicles (cranes, etc.) so the physical stability issue is certainly not a big one. My high school physics is terrible but the keyword vacuum might have something to do with that. Data centers would be concerned about safety in earthquakes, too, but flywheels like this one seem widely deployed: http://www.socomec.com/range-dynamic-energy-storage_en.html?...

"Surely if it's buried, as seems to be the general consensus for cheapest mode of safe deployment, then it doesn't shake?"

The ground occasionally shakes, depending on where you are.

The Earth also rotates on its axis. And with the speeds and energies with which grid storage flywheels operate, that's a significant engineering concern.

The real killer though is cost / KWh of energy storage. For Beacon's designs it's about $10,000/KWh capacity (not $1000/Wh as I wrote above). Which is high. Liquid metal or molten salt batteries, or simple thermal storage (again, likely salt) could likely compete with this.

> The Earth also rotates on its axis. And with the speeds and energies with which grid storage flywheels operate, that's a significant engineering concern.

You can align the angle of the flywheel to match the angle of the earth at your latitude, which helps with that. However if you do that the flywheel is titled, so the bearings are uneven, and then you need some shielding above the flywheel - you can't just have the earth be the walls.

But it means that the bearings experience much less force, so it can be worth it.

> I seem to recall an automobile developed in the 60s or 70s by another National Lab that siphoned off stored power from a flywheel (driven by a smaller motor).

One possibly feasible application that comes to mind are automotive turbo chargers. Compared with current approaches, a flywheel does not actually seem that convoluted.

Reply on news.ycombinator.com