[throwaway5752]

There's a video somewhere of a failing flywheel, it's a fairly energetic explosion. I follow the energy space, also, and ... while I didn't expect Li batteries to have grid scale roles, I didn't think flywheels were going to succeed there. They have properties more like ultracapacitors (discharge rate suitable for continuous backup, clean up messy grid sine waves). I thought pumped air/pumped water was more likely because it scaled up so well and was old tech. Fuel cells, too, seems to fall more into the backup/spot generation/remote generation+NG access niches.

Looking on the bright side, anything that drives Li efficiency/density research is probably a very good thing.

[dwyerm]

> it's a fairly energetic explosion

So is a lithium battery explosion. So is a gasoline explosion. So is a flour mill explosion. So is a wind turbine explosion.

Energy is energy, and losing control of energy is never great, no matter where it comes from. So while I catch your meaning, it might be better to phrase it with respect to how controllable that energy is.

From what I know of flywheel storage, the problem mostly comes down to keeping the wheel from coming apart, and containing it when it does. The nice thing about using flywheels for grid storage is that you can bury them and make them large. The earth contains your explosion risk and the lack of jostling means that your bearings don't need to take as much stress and limits that failure risk.

[throwaway5752]

I would disagree with you slightly. Energy is not totally fungible. I can punch you in the face or shoot you with a gamma ray (edit: though of course, I would never do either), and it would have very different effects, potentially. Same for a fastball vs a small caliber weapon. Heat, light, and kinetic energy are important distinctions - particularly when you have a bunch of 30k flywheels in close proximity and one of them catastrophically fails. Now, I know that's the first thing someone would think about deploying a bunch of them and they would take precautions. Li batteries are less kinetic and more thermal and that's a bit easier to manage/less likely to cascade.

I thought the trend in flywheels was magnetic suspension and removing mechanical linkages? Admittedly I haven't kept up.

[microcolonel]

> that's a bit easier to manage/less likely to cascade.

What gives you that impression? Seems to me that it explodes if you contain it, and if you don't contain it, it can spout jets of thermal energy at virtually any angle. With flywheels you need to arrest it in bulk heavy objects that don't tend to sustain fire. That seems a lot simpler to me.

The bigger problems with flywheels are cost of manufacture and (depending on the technology used) efficiency for overnight storage.

> I thought the trend in flywheels was magnetic suspension and removing mechanical linkages? Admittedly I haven't kept up.

IIRC flywheels with limited motion gimbals (to reduce the tolerances on the wheel) are becoming more popular, still magnetic bearings.

[throwaway5752]

What gives you that impression?

Insulation is cheap, effective, and very compact. And it's easy to transfer heat quickly, also (either via injecting lots of cold extinguisher or flush lots of hot oxidizing gas)?

Cool, thanks for the update on flywheels. I ... am not trying to create a false dilemna, here. Fuel cells for stranded methane deposits are great. Flywheels have outstanding responsiveness and energy density. Li / compressed air / pumped water et al scale well. They all fit into a more resilient grid storage strategy that permits a transition to periodic sources of input from non-renewable base load.

[wbl]

Except we've had a carbon free baseload tech for decades now.

[throwaway5752]

It's kind of bizarre how nuclear had achieved a sort of counterculture renaissance. It has many, many drawbacks too numerous to go into here that are technical and sociological in nature.

And anyway, you tried to be too clever. I said "non-renewable" and not "carbon free" just to avoid this conversation. Unless you can start synthesizing utility grade quantities of well-behaved fissile material at a net energy surplus then it's not renewable even if we have decades/centuries of supply.

[samatman]

Batteries don't like insulation, in fact preferring active cooling.

Which is itself a good thermal-runaway damper, to speak to the second sentence of your second paragraph.

[throwaway5752]

The other kind of firewall: https://en.wikipedia.org/wiki/Firewall_(construction)

[reaperducer]

>So is a flour mill explosion

Having witnesses a corn silo explosion, I was unprepared for the ferocity of that ignition.

>So is a wind turbine explosion

Off to YouTube...

[dwyerm]

The Hornslet collapse[1] is the one I was thinking about in particular. It is especially daunting now that I've toured a site for myself. Those things are huge!

[1] https://en.wikipedia.org/wiki/Hornslet_wind-turbine_collapse

[vvanders]

Ever seen a tire come off of a semi?

Assuming the flywheel keeps its integrity it's much harder to predict the "blast radius" of where that thing is going to go.

[blevin]

Are buried flywheels used in practice? Is the flywheel axis aligned with the Earth rotation axis? Do they pull a vacuum around it?

[dredmorbius]

Yes, in cases, to the first two. Containment is frequently buried concrete vaults.

Not AFAIK for the last, though precessional torque bearing load is a nontrivial consideration.

[blevin]

Thanks for this response. My untrained guess is that precessional torque would cause more loss than air friction, assuming the wheel is cleanly symmetric (ex. no bolt heads sticking out). It's fascinating how many paths there are to push something up various kinds of potential energy gradient in a way that it mostly stays there by itself but is still available to us to access in a controllable way.

Wikipedia claims flywheel loss rate circa 2013 of 5% per day. https://en.wikipedia.org/wiki/Flywheel_storage_power_system

That compares to recent estimates of Tesla li-ion loss rates at under 5% per month -- 0.16% per day.

Amber Kinetics is one company building fixed flywheel storage products. http://amberkinetics.com/

They have one 32 kWh, 5-ton, 98% steel flywheel installation on Oahu; pictures here: http://amberkinetics.com/hawaiian-electric-and-amber-kinetic...

[dredmorbius]

Frictionaal losses are part of the consideration, but just plain bearing wear is a bigger one AFAIU.

Angular momentum in 100 kWh - multi MWh rotational systems is large.

[ballenf]

Also flywheels don't give off noxious fumes during failure.

[naravara]

For now. If you want to use one for long term energy storage, though, you're going to need to invent some kind of room temperature superconductor and frictionless surfaces. Who knows whether those would be noxious or not.

[smaddox]

Why? Magnetic bearings and copper work just fine...

[jhayward]

Are you sure? Metal fumes and vaporized epoxies are usually pretty toxic.

[tossandturn]

...or during production (relatively speaking).

[Florin_Andrei]

> So is a lithium battery explosion. So is a gasoline explosion. So is a flour mill explosion. So is a wind turbine explosion.

No, they're not all the same.

Compressed air, flywheel, flour mill - very dramatic events.

Lithium battery - much more mild, typically.

Gasoline - it depends.

[3pt14159]

The core insight for me was when Musk talked about how little the cost of Li mattered in the unit economics of Li batteries. It was something like 3% of the cost. The rest of it was mostly manufacturing costs. With so much room for improvements at scale and so many potential applications, it seemed like a sure winner.

[bdamm]

That's a very useful insight. My "aha" moment came during a lecture given by DARPA at PARC about 5 years ago, outlining how wasteful current battery technology is in terms of density entirely because of safety requirements. They had at the time over a dozen different research projects on ways to improve density by changing how safety is achieved. It is only a matter of time until some of them get to market, and in fact I believe one of them has. They were all 2x-10x improvements.

[hinkley]

There’s a kind of Moore’s law in batteries, but the multiplier looks more like inflation than 2x.

Any tech that falls outside of the growth curve seems to run into issues with production or cost that delay it until it fits under the curve. Something cheaper and easier gets picked first.

The first modern EVs had lead acid batteries. More sophisticated than your starter battery, sure, but lead acid all the same. Which is why Tesla was a big deal. We talked about LiPo for something like fifteen years before it showed up in consumer electronics, and then they started catching on fire.

All of this stuff is painfully slow. The big story in EVs is how crazy efficient the motors can get. A company I used to follow (whose name is escaping me now) had a motor that was 95% efficient in its sweet spot. They had scaled up the design to 100 HP.

[snovv_crash]

Yeah electric motor efficiency is awesome. Back in 2005 I had a film-canister sized motor capable of 300W at 90% efficiency, and magnets and motor design have just improved since then.

[jammi]

Not only does materials overall matter little, but Lithium is also super cheap compared to the other materials in the batteries. There's a lot more of Nickel in a Li-Ion battery and it's much more expensive per volume and weight as well.

The batteries would be called Nickel-Carbon or something like that rather than Lithium-anything, if they were named by the amounts or costs of materials in them.

[mrfusion]

Why can’t we bury flywheels to protect from any explosions?

[maliker]

I've seen utility installations done like this. More expensive but safer.

[gitgud]

Well yeah you can bury anything to keep it safe, but at what cost?

[maliker]

The utility flywheel installations I'm aware of are underground. More cost, but safer.