Hacker News Reader

Comment by plorg | Hacker News Reader

When I was in school for power systems planning the the accepted notion was that pumped hydro storage is cool and great, but like hydropower, it is extremely geographically dependent, and we have basically developed all of the places where it would work (at least in the US), or at least, by order of magnitude, there are not enough remaining untapped hydro-related resources to make a dent in projected storage demand. Is that a wrong understanding?

View on HN

Entirely wrong. All pumped storage requires is a big high spot and a big low spot*. Take a look at this, for example:

https://www.gordonbuttepumpedstorage.com/

There are a lot of good pumped storage sites out there.

*exaggeration, but close enough.

Unfortunately that rules out a lot of the US Midwest, where coal is a huge source of electricity generation.

That's the US wind belt - the Texas panhandle north to Canada, and east into Illinois.[1]

[1] https://www.kgou.org/energy/2013-09-03/oklahomas-wind-energy...

> All pumped storage requires is a big high spot and a big low spot

I think that's what the GP meant by "extremely geographically dependent." Finding a 1,000 ft height difference (using the Gordon Butte Project as an example) between reservoirs will be really challenging in huge swaths of the United States and elsewhere.

It also needs to be near a natural source of water, near population centres and is preferably not a pretty national park.

Dinorwig power station [0] is in the middle of a national park. A pretty one, even.

[0] https://en.wikipedia.org/wiki/Dinorwig_Power_Station

Yes, and add you can see it's already been developed - for over 60 years. That's not to say more can't be developed, but the rate of new development has been pretty low, and the prevailing belief was that this was the result of all of the best resources being used up.

...and even then they built it mostly inside the mountain, next to an old slate quarry. Which is going to help national park acceptability but not the unit cost of construction I suspect.

Doesn't need to be near population center. The majority of the energy I've used in my life comes from a dam over a thousand kilometres from where I live.

https://en.wikipedia.org/wiki/James_Bay_Project

There are many possibilities globally. 1000 feet is not needed.

https://re100.eng.anu.edu.au/global/

It is ridiculously uneconomical to create the volume for pumped storage. For context, with 100 meters of pressure head, you'd need 25,000 olympic swimming pools worth of water to store 12 hours of electricity from a 1 GW electric plant with 70% roundtrip efficiency. At a cost of $216/cubic yard of excavation, that's over $17 Billion. If you can make 3 cents per kwh profit buying at low times and selling at high, it would take 129 years to cover just the costs of digging the hole at the high spot. Then you need to dig another hole of the same size at the low spot so you can reuse the water.

There are really only 3 options - dam up a watershed so you can get a huge volume without much structure, build your resevoir on top of a mountain so you can get much larger pressure head and thus require less volume, or reuse a hole you were digging anyways such as an open pit mine. In all three cases you are heavily constrained by geography.

It is a good thing then that nature has already done the excavation for us by providing high and low spots.

Yep, now we just need to excavate the reservoirs on those high and low spots, a mere 100% of the work required.

This is nonsense. We have lakes already. We don’t dig two giant holes in rock. Look at real work pumped storage costs.

That's the "geographically dependent" part.

Except, not.

Lakes aren't geographically dependent?

Don't need any lakes.

Mind the thread context.

"This is nonsense. We have lakes already."

Maybe you could read up, instead of guessing.

Dikes are a technology older than writing.

I am curious where the 216/yd cost comes from. I had a rock driveway put in at 10/yd, and I assume the stone was excavated from somewhere.

Usually the rock is waste from excavation. When buying it you pay the cost of transport, group that did the excavation might have even paid for the rock to be removed.

Excavating a circle of area A costs O(A). Building a wall around a circle of area A costs O(A^0.5).

I think you'll find the attached link is project currently in development that does exactly what you're saying is economically impossible. It clearly is not impossible to build a relatively shallow dikes on top of a butte.

Yes, that is the whole concept.

The question is, is this something that has been re-evaluated? Are there more site locations available than previously thought? Have the economics changed significantly? Was this a known-wrong assumption 15 years ago? Has the public perception changed re: the environmental costs vs benefits of this land use?

I would much rather we build a bunch of pumped hydro than a bunch of chemical batteries, on the belief that it is a much more sustainable technology. But I have seen a lot of online discussion about pumped hydro that treat siting, operational, economic, and land use concerns as if they don't exist, when often they are the primary reasons these projects don't get built.

It is not "more sustainable". It is just, likely, cheaper.

Siting, operational, and land use concerns for pumped hydro are, generally, trivial. If one site is bad, another is good. There is absolutely no shortage of sites, so only the best need be even considered.

It's geographically dependent if built on rivers. Beyond rivers the possibilities expand enormously.

Reply on news.ycombinator.com