Hacker News Reader

> Storage is a solved problem.

Lol no, it really isn’t.

> Pumped hydro has been in use for decades around the world,

Pumped hydro works if you have mountains, valleys you can flood, and dams. It works great in the Alps; good luck doing that in the UK or in the Netherlands.

> battery systems are becoming commonplace as lithium ion battery prices have plunged

They are nowhere near the right scale. Bear in mind, in Europe they’d need to store energy over several weeks or months. The continent is prone to winter anti cyclonic conditions that go on for weeks, with no wind and no sun light. It’s quite different from, say, most of Australia.

> alternative battery technologies have matured

I’d like to know which one, because as far as I know none of the alternative are anywhere near maturity. Progress is being made, but you’re overselling it.

> Several Iron-flow chemistries are now commercially deployed.

These are no silver bullets and have several drawbacks, including the “flow” bit, which requires pumps and valves. Efficiency is also not that great. There are sulphides, lithium-air, and Na-ion designs that are more promising. But again, none of that is mature, or even really produced at scale.

> I don't understand why every time wind/solar are mentioned on HN, someone feels the need to say "tHe WiND doEsn'T alWAys BloW" while utilities are actively deploying storage, and not just for "when the sun isn't shining", but to help smooth peak usage to avoid transmission upgrades, provide resiliency from infrastructure failure during weather events, and increase flexibility for service/maintenance that doesn't require outages.

It’s the counterpoint of all these “lol renewables all the way” posts that crop up any time we discuss energy policy.

Most people would say that there are very good use cases for renewables. But they won’t be enough. If you read analyses at the grid level (some of that is in the IPCC reports), constant base production is very much required, otherwise you need to vastly overbuild and your grid has one huge point of failure (yes, freak meteorological events do happen).

Just a minor nit -- and some links for those who might not know where to find them. I of course agree with you.

> Pumped hydro works if you have mountains, valleys you can flood, and dams. It works great in the Alps; good luck doing that in the UK or in the Netherlands.

It works in Wales and Scotland -- but not in England where it is needed the most.

https://en.wikipedia.org/wiki/Cruachan_Power_Station https://en.wikipedia.org/wiki/Dinorwig_Power_Station

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

https://en.wikipedia.org/wiki/List_of_pumped-storage_hydroel...

Transmission lines work fine in England. Wales and Scotland are hard by.

(... as people actually say, in England. Or did.)

> Pumped hydro works if you have mountains, valleys you can flood, and dams.

This falsehood has been corrected numerous times on HN. It need not be repeated.

Pumped hydro is in no wise dependent on any of those. All it needs is a hill. Even England has hills. Even Australia has hills. The Netherlands are in Europe, which has hills.

Another falsehood is that "you need to vastly overbuild". This too has been frequently corrected. We use, today, NG for backup. We will have no need to demolish any of that infrastructure, even after storage is built. Overbuilding renewables to charge storage makes compelling economic sense, just because it is so cheap, but the NG backup remains. Eventually, synthetic fuels will be cheaper, for such occasions, imported from solar farms in the tropics.

So, renewables have no such "single point of failure". This falsehood need not be repeated.

Finally, pumped hydro and batteries are just two of numerous practical storage technologies being deployed. Their round-trip efficiency is of minimal importance when they are charged from free surplus. Storage with NG backup provides fully ample 100% service with exactly zero "constant base production" needed or wanted.

My I-think-conservative WAG for the cost and land-area-use differences between England and Switzerland, for this kind of storage, would be about 50x and 5-10x, respectively, for similar amounts of stored energy. Both elevation and grade of the surrounding land matter a lot for how much energy you can reclaim per stored liter, and having two or three sides of your reservoir provided by nature in the form of steep-walled valleys has to provide serious cost savings.

Like of course any water you push up a hill is gonna gain potential energy, but I think a reasonable reading of "works" in that post was "works efficiently enough that it's actually worth doing", not "is technically possible".

Your "wild-ass guesses" for costs are worth as much as you say. Grade, less: it affects mainly how long the penstock must be, but pipe is cheap.

An earthen berm is a very cheap construction project. Classically those were achieved with hand labor worldwide, for millennia, eventually augmented with wheelbarrows starting in China, and oxcarts. They cost less today.

Nature providing one, two, three, or four walls (steepness optional) reduces construction cost, but the turbines and pumps are a substantial part of the total.

Opex is of course extremely low. Bonus points for solar farms floating on the reservoirs. Round-trip efficiency of pumped hydro is near the top of all alternatives, although efficiency is relatively unimportant when the power being stored was free.

Not seeing how the physics can possibly work out for the real costs and land-area-use being too much lower than my guesses to achieve the same total stored energy, but sure, maybe there's some engineering magic I'm not aware of. Could be. In particular, maybe I'm overestimating how much more energy one can efficiently extract from, say, a liter of water dropping 5,000 feet over three miles, versus a liter of water dropping 600 feet over three miles.

Anywhere pumped hydro would cost more than another of many storage alternatives, the other alternative will reliably be chosen. England is, e.g., preparing hydrogen storage in depleted NG cavities. Pumped hydro will be built in the places where it is cheapest, favoring high hills and variable topography.

Costs for many storage alternatives are falling fast, and it is unclear which choices will win. Adding pumps to existing dams has lately been cheapest, by far. Future people will choose according to local conditions without asking us.

Right, I get that there are alternatives that would be used instead, what I was surprised at was characterizing "hydro only works in mountainous areas" as flatly false, if we take "works" to mean "is worth doing", which is a common usage and I think a fair reading of it, and since it seemed to me like there would be large differences in the economics of it, depending on local geography.

Low hills, where the hilltops are not otherwise in use, may suffice. Some places have very deep gullies or underground cavities that would serve in place of or augment hills.

The point is that whether a storage medium is practical in an area depends more on the landscape of choices than on the physical topography. Even Nebraska might have deep cavities that would serve in place of hills. But the extrema are extreme.

Pumped hydro also works if you have water at all, which my country doesn’t. So, some kind of new molten-salt reactor which doesn’t need water for cooling and can be built in the desert could work very well for us.

> Storage is a solved problem

Why do people just assert things rather than spending five minutes actually looking up some numbers doing arithmetic?

Storage systems and theirs costs, discharge rates, cycle endurance, etc: https://www.energy.gov/sites/default/files/2019/07/f65/Stora...

What the wind power production rate over time looks like for a country: https://www.gridwatch.templar.co.uk/

Then you can look up country power consumption and wind cost per KWh and half life.

Then it's just some arithmetic.

You're not helpless consumers of data, you CAN look this up and figure it out.

To save you some trouble, the bottom line is that for countries like the UK pure renewables would triple electricity costs. Is that an acceptable price? Maybe.

Nobody expects, demands, or plans "UK pure renewables". So all of the above, conditioned on it, is specious.

Renewables are the cheapest power generation the world has ever seen, by a large and growing margin. Storage in many practical forms is being deployed.

Everything that has always worked still works, and will be abandoned only when it is not needed anymore.

The issue is the amount of storage required. It depends a lot on the country, but here in Germany I think everybody agrees that we need to store enough energy to power the whole country for at least 2 weeks before we can get rid of fossil fuels. That's a lot of batteries. And currently there's not enough storage for a single hour...

In these dimensions, resource availability becomes a limiting factor. Pumped hydro requires a lot of space and geography that many countries can't provide. It may be an option for Norway or Austria, but not for Germany. And even if Germany would buy the planned world production of lithium batteries for the next 10 years, we still wouldn't have enough. The math just doesn't work out for both. Hydrogen looks more realistic, but is not a mature technology at the required scale.

> we need to store enough energy to power the whole country for at least 2 weeks before we can get rid of fossil fuels.

Maybe this is self-obvious, but "get rid of" seems to be carrying a lot of weight in that statement. I personally am not offended or concerned by keeping natural gas generators around for decades, even, as insurance for catastrophic disasters, if in practice they're unused 99% of the time because storage infrastructure has reached 24 hour capacity. We reach diminishing returns on increased storage.

There is, in fact, no need to "get rid of fossil fuels", and no plan or desire to do so. So all the remaining claims are irrelevant. NG backup for renewables will remain. More storage mean burning less frequently, which is a pure good.

Pumped hydro storage does not, in fact, require "lot of space". Hydro power generation needs a watershed. Pumped hydro can reuse existing hydro infrastructure, but does not depend on it, and does not suffer its failings. Pumped hydro needs only a hill.

There will be no need to repeat the cited falsehoods.

Finally, a multitude of practical storage methods are being deployed. Batteries are the most expensive, so are only ever used for very short-term storage. NG backup for renewables works and will continue working; storage forestalls need to use it, decreasing opex.

So simply plan really, overbuilt massively on renewables. Then build a while bunch of storage. And then also have inactive low utilization natural gas plants and infrastructure.

Germany will spend like 40 years building this massively complex production system with massive overcapacity of solar and wind, quite a bit of storage, natural gas backup and lots and lots of distribution networks to move around power everywhere requiring a massively complex to operate country wide grid.

Man o man, France in 70/80s solve the same problem with 60 technology and it took them less then 20 years. The mostly build the nuclear plants on sits of existing oil plants and didn't even need to make their grid complex. And they didn't even want to produce green electricity, they did it by accident.

Really hard to see what the better plan is. Germany could have literally had 100% green energy by now if the 'Energiewende' had gone with nuclear, but instead they will spend the next 20 years shutting down their remaining nuclear, then shutting down all their remaining coal, and also building more gas plants.

Brilliant plan really.

If you take a moment to read what you replied to, you will see that your each assertion is directly disposed of there: there is no need for "massive overbuild" of renewables, or for much storage. The backup NG infrastructure is already in place and in use, and will continue working.

Germany's only problems just now are that Russia turned out to be an unreliable supplier, and their renewables build-out is incomplete. Both are being dealt with.

> Pumped hydro has been in use for decades around the world

They were used last time in Europe when there were an energy shortage. Almost all was decommissioned quite quickly afterward when they became economical unsustainable.

They are generally expensive. They are generally located in places where there already exist a lot of hydro power. They generally need a lot of land and connecting power infrastructure. They generally tend to compete with hydro power. They generally need a lot of subsidies.

With energy prices skyrocketing in eu there is some lukewarm interest to restart those old pumped hydro (those that still exist and with existing power infrastructure in place). They are however not a drop in replacement for existing nuclear power plants nor existing natural gas power plants. There is however a energy shortage and so any source might be economical viable if used correctly.

> They are generally expensive.

For the foreseeable future, electricity storage will stay expensive when you compare it with fast to cycle natural gas generators. This will only change with a carbon tax or after we consume most of the natural gas available on Earth (looks like it would take 2 or 3 decades on the shortest predictions).

That applies to any kind of electricity storage.

Compared to the alternatives, pumped hydro is currently a cheap highly efficient storage. It gets the best qualities of synthetic fuels (low cost) and batteries (efficiency). But it's not available everywhere. Things are changing fast, and I do expect pumped hydro to not be competitive when we fully deploy storage, but it's currently the best option you will see.

1) There are countries that simply cannot rely on hydropower.

2) I am Italian, so I will talk about Italy. The Italian plan to get to 100% renewable requires 70x the current world production of lithium.

3) Rare earth mining for batteries, panels and wind turbines has a non-negligible impact.

4) You are implicitly stating that the cost of wind power is not that, but 3/4x higher, since you also need storage, and therefore it is much more expensive than nuclear the "real" cost.

5) You are talking about distributing storage over the electrical infrastructure. So you're talking about completely redoing existing grids. Other additional costs.

In short, the reality of the facts is very different from saying "renewables are cheap."

Where is Japan going to build the reservoirs to support pumped storage? They would literally need to blow up some mountains to do it.

Or build it at the top of said mountains?

Well, usually you have many dams in sequence on a river and you can often turn them to pumped hydro by pumping the water from the lower dam back up when you need to store some power.

And Japan has a lot of such cascades - even just in Kurobe there are at least 4 major dams one after another.

But all that capacity is built out. Japan has about the same as the USA in pumped storage capacity...which is a bit amazing considering Japan is just the size of California.

I do think hydrogen is the answer, and I think it will be economical. Not just for energy, but for ammonia for fertilizer.

Hydrogen is part of the answer, for steel and perhaps concrete manufacturing. And also to fuel container ships, in the form of ammonia. But as a storage medium it is not very efficient. At least not now.

Round-trip efficiency is not very important for a backup storage medium when the marginal cost to produce is zero.

Other considerations drive choices, then, e.g. capex, transportability, usefulness for other processes.

And once you realize that nearly all renewable energy production has a marginal cost of zero, then it quickly becomes the overwhelmingly best idea for energy storage.

The infrastructure needed for hydrogen production and storage is far, far, from zero.

Yet, the marginal cost of producing hydrogen using that infrastructure is zero: the only inputs are water and energy, and the marginal cost of that energy is zero.

Will be fantastic when we have all those hydrogen production plants who only run with 30% utilization because they only make sense when you massively overproduce renewables.

To "massively overproduce" renewables is cheap. Build more, get more, sell the overproduction.

> To "massively overproduce" renewables is cheap.

No it actually isn't really.

> Build more, get more, sell the overproduction.

You will have to pay to get ride of it in many places. Chemical plants want reliable electricity to have reliable production.

Hint: there is no cost to disposing of excess solar and wind energy.

When you produce chemicals yourself, you may operate on a schedule of your own choosing.

> Storage is a solved problem.

And yet, the presumably intelligent people in charge of Japan's energy policy still think nuclear is the way to go. Do you know something they don't?