Hacker News Reader

I've had this argument so many times in that last month... renewables outside of hydro are just not viable as baseline power until batteries improve cost and density by several orders of magnitudes. Nuclear fission is really the only choice we have right now. We can't avoid climate change by relying on technologies like fusion power & next gen batteries that haven't even been invented yet, let alone mass manufactured.

Jean-Marc Jancovici is a French engineer/entrepreneur who study energy and climate change for about 30 years. There is a single talk I know he gave in China that is in English: https://www.youtube.com/watch?v=gRC5-hAbUGA Unfortunately most other talks are in French(auto subtitles maybe). His style is a little arrogant and may hit a nerve. But past this, all he says is backed by research and physics.

He says renewable can't provide the level of energy we have without decreasing drastically our consumption. And nuclear fission is the only path we have to reduce our CO2 emissions.

I would love renewable or fusion to cover all our needs, and keep my comfortable life as it is today. But I didn't find any inspiring talks or readings so far to balance his arguments.

> I would love renewable or fusion to cover all our needs, and keep my comfortable life as it is today

I've been reading Jason Hickel's book "Less is More" [1], which talks a bit about energy use, in general, and how it must be reduced, regardless of what technologies we wind up with. And one of his points is examining how much energy is just flat-out wasted. ~50% of food is just wasted, for example—whether because it's ugly, or left in a refrigerator too long, and rots. Think about how most American cities are designed, and how it's just normal for most Americans to own and drive personal vehicles for hours each day. Those resources and the energy needed to extract those resources is largely just unnecessarily wasted.

Consider countries like Costa Rica, Finland, Taiwan, etc., where quality of life is better than in the US. Life expectancy is higher. Infant mortality is lower. Education is better. And they need a tiny fraction of the energy the U.S. does. Clearly there's room for a lot of improvement with how we use energy.

[1] https://www.jasonhickel.org/less-is-more

This is similar to the point Vaclav Smil makes when he looks at our dismal prospects for decarbonization, if humanity is going to survive, then we are simply going to have to consume less. I think this concept of reduced consumption is in conflict with a system of high inequality today that is often justified by the prospect of (slowing) growth.

https://www.theguardian.com/books/2019/sep/21/vaclav-smil-in...

For whatever reason people have a hard time hearing this. I've similarly had the same argument over and over again.

Also for whatever reason people love to talk about how mining trucks will become EVs, and they always link me to the same Komatsu truck which they claim is an EV, but actually has a diesel engine and uses an electric power transmission.

In terms of scaling up, renewables + batteries seems to have a clearer logistical and economic path. Batteries can be mass produced and we need to increase production capacity anyway for electric cars. And there is already a massive industry around battery technology from consumer electronics. Solar panels in particular are ridiculously modular and offshore wind can rely on existing oil industry experience.

I just don't see a similar pathway for nuclear? Particularly in less stable parts of the world without an existing industry.

I disagree. The UKs National Grid see a role for nuclear in its Future Energy Scenarios (https://www.nationalgrideso.com/future-energy/future-energy-...), laying out 4 different energy scenarios in 2050, but wind will dominate electricity generation.

Also, the current generation of EVs will provide a substantial amount of flexibility resulting in £billions of savings for the UK grid and consumers (see https://www.kaluza.com/blog-busting-the-myth-ev-smart-chargi...). The addition of Vehicle-to-Grid chargers will further allow increased renewables.

With the cost of these technologies decreasing, the importance of nuclear power will reduce over time.

Maybe you should write an article about that, and put it up on HN, given that it's such a common point of contention/misunderstanding.

There are people more knowledgeable then me who have made these points better than I could, I'd recommend Vaclav Smil as a great source on this. The problem is that his material makes for grim reading, and I think many people researching this subject are drawn to techno optimism even when the numbers don't add up because the subject as a whole is so grim.

Do you have specific recommendations among his writings?

http://vaclavsmil.com/2017/05/08/energy-and-civilization-a-h...

I also recommend looking through his list of books on the sidebar.

The whole "baseline" thing is partly a myth, and there is an article about that

https://energypost.eu/dispelling-nuclear-baseload-myth-nothi...

This is a good one http://www.withouthotair.com. It looks at whole energy plans. Not just wether technology X is cool.

Isn't that the site that argues the UK cannot get by with renewables, but assumes large amounts of biomass are needed to be grown locally?

It's from 2007 and will not get an update since the author has died. I do like it for taking a quantitative approach even though the assumptions used are increasingly outdated.

Here's the summary page where the author asked if the UK could supply its energy needs from renewables alone:

http://www.withouthotair.com/c18/page_103.shtml

The answer was "not quite."

However, he assumed that utility scale solar farms would be using 10% efficient panels. 20% are now widely available. So let's make these two quick edits:

- Assume that solar farms provide double the energy from doubled panel efficiency.

- Drop biomass, because it's unneeded with better solar farms.

With these changes the renewable energy supply of 206 kWh/person/day exceeds modeled demand of 195 kWh/person/day. Making some other simple updates like assuming that car transportation is mostly electrified would further increase the margin of safety.

> are just not viable as baseline power until batteries improve cost and density by several orders of magnitudes.

This is an outrageous falsehood. Play with the numbers at this model site to see:

http://model.energy/

Also, why is energy density relevant for stationary storage? The cost of land to site batteries is trivial.

You're assuming "batteries" which is just wrong. There are a lot of economical grid-scale power storage systems. Batteries are low-latency storage used to stabilize load - they are expensive because they are so quick to change the power they provide. Frequency, voltage, current, whatever you need a battery can provide in milliseconds. But you don't want to run it for days.

If you want to store power overnight you want kinetic storage, or possibly power to gas. There are lots of technologies available. Some of them are cheaper than nuclear, but none of them are economical yet because there is no market for day-to-week-scale storage until solar/wind start to approach 50% of the power mix.

> You're assuming "batteries" which is just wrong. [...] you want kinetic storage, or possibly power to gas.

...neither of which are cheaper than batteries or we'd be doing that

Yes, we are doing pumped hydro, but only in a few convenient locations. I recently watched a video about Ireland's new pumped storage facility[1], it mentions some pretty tight constraints for what we currently consider economical (e.g. hill slope). Bill Gates' new book[2] doesn't go into as much detail but also doesn't sound like we got this figured out any better than batteries, basically saying we need breakthroughs or this is going to simply be a prohibitively expensive undertaking for non-rich countries. Specifically about hydrogen (the "to gas" option that has the most buzz at least) [2] says that there are high losses in conversion (without saying how much) as well as hydrogen leaking from metal tanks and electrolyzers being very expensive. There's hope for breakthroughs but.. they do have to happen in order for it to be better (cheaper) than today's batteries.

At a bare minimum, it's not obvious that what GP wrote is "just wrong". This isn't something we have a definite answer to.

[1] https://www.youtube.com/watch?v=JSgd-QhLHRI

[2] "How to Avoid a Climate Disaster: The Solutions We Have and the Breakthroughs We Need" ISBN 978-0-593-21577-7

> At a bare minimum, it's not obvious that what GP wrote is "just wrong". This isn't something we have a definite answer to.

I agree that we don't have definite answers, but that's why what the GP wrote is just wrong. They're reducing a very complicated set of technologies to the economics of lithium ion batteries. While yes, pumped hydro is something of a cop-out, it's clear there are a variety of techs that are cheaper than lithium ion.

The economics of this are also a little unclear to me. In the presence of sustained negative electricity prices, it seems obvious in principle that lots of storage technologies would suddenly become economical. But it's hard to imagine how that would work given the way we currently finance these things. (And by extension, it's not clear to me that this is actually a technical problem so much as a political one.)

Building pumped hydro is not going to suddenly be more feasible or cheaper as demand for it increases. What else can store energy for more than a week at a cost that approaches reasonable?

What's an example of one that's cheaper than nuclear?

The nuclear waste argument is a purely American one. European countries (France specifically) have a multi-reactor chain where each successive step uses the "waste" from the last step. While this doesn't eliminate the need for long-term storage completely, it does reduce the potency of the leftovers since they're not being discarded after using ~3% of their total stored energy (a la the US).

Additionally, fission isn't the only way to get energy out of uranium. Nuclear batteries, such as the ones powering Curiosity and Perseverance, can't melt down like a traditional reactor and will continue working until all the energy stored inside it is gone. This is the type of nuclear power I'd like to see gain prevalence for many reasons, foremost of which is the ability for each household to have its own source of electricity independent of a power grid.

Solar in space is all well and good, but without a way to beam that power back to Earth, a space-based solar collector is as useful as if nothing were there at all.

> foremost of which is the ability for each household to have its own source of electricity independent of a power grid.

While I’m all in favour of each home (or apartment block) being energy independent, I very much do not want widespread access to even RTG material.

Consider: average power use is about 1 kW, 1000 J/s, and that the “dead in 1-2 days with treatment” radiation absorbed dose is merely 30 J/kg, so any domestic power supply is capable of killing an adult in 2-3 seconds of full exposure… or, more likely, a tiny flake imbibed or inhaled due to either malice or incompetence would have the same effect during one night of sleep.

> Solar in space is all well and good, but without a way to beam that power back to Earth, a space-based solar collector is as useful as if nothing were there at all.

I’m definitely against beamed power, at least until we get a single planetary government.

Anything which beams with a higher power flux than sunlight is a potential weapon. Anything which isn’t, requires so much more land than PV that it’s always better to use PV even when you have to put the PV in your nearest desert and beam it to you from there using whatever future tech you’re thinking of to get from orbit.

(I assume one world government would mean that government could authorise itself to build a weapon pointed at itself without having to worry it might be secretly planning to shoot itself; I don’t see how either the USA or Russia could ever trust the other if either built one).

So what you're saying isn't entirely true. I believe you're referring to spent fuel processing, which is just one part of it.

I specifically mentioned UF6 (uranium hexaflouride) as a hazardous byproduct of centrifuge enrichment. Now France (and some other countries, including the US up until 2018 at least) do process UF6, but only to put it into a more stable form (eg UF4). That more stable form still requires long term storage.

These arguments are all moot. Even assuming that humans can't be trusted with each of these points, then still you are exchanging hypothetical lives and hypothetical damage to the environment for real lives and real damage to the environment.

Who cares about storing fuel and toxic byproducts perpetually? We are destroying our environment right now. It's ridiculous we're even still having these conversations.

It doesn't matter if fission has a future, we could start to build a hundred power plants today, and the first ones would be finished in 5 years, and it would fucking save our planet.

Nuclear dangers and issues aren’t theoretical. The Chernobyl Absolute Exclusion Zone is quite literally 1,000 square miles.

And humanity is greatly affected by the loss of use of a patch of land? I'm saying it's inconsequential.

How many people are dying from coal plants? How many are dying from the exhausts of ICE cars. How many more from the rising tides, the swelling storms and the blistering droughts?

The dangers of flying aren't theoretical either. Malaysia Airlines Flight 370 is still missing 7 years later.

Yes and the fatal accident rate for commercial airlines, based on some quick googling, is 0.27 incidents per million flights.

We’ve had at least 3 nuclear disasters (Chernobyl, Fukushima and There Mile Island) and Google suggests there are 440 reactors in the world (current; so add a few more for decommissioned).

Of those, only Chernobyl can be reasonably compared to the loss of life of an airline crash.

Three Mile Island wasn't even a disaster, except for the wallets of the investors.

Working at Fukushima for 40 years was safer than your average construction job, even if you worked during the disaster.

And if you have to unfairly compare nuclear safety to airliners, also consider that airliner disasters have become 40 times less frequent than they were when Chernobyl happened.

So what's the point in absolute statements like "the dangers aren't theoretical". So what if they aren't?

Everything is about tradeoffs, and the collective decision to keep coal, with all its attendant problems, is a worse tradeoff.

It seems like it would be easier and substantially cheaper to take existing fracking/drilling tech and go for deep-well geothermal, rather than trying to launch a Dyson swarm.

Why not both ? :) Also you don't necessarily need to dismantle all planets during Dyson swarm construction- you can keep a few as a historical curiosity. Sure, it's horribly inefficiency use of mass but still. :)

> I'm not sure which issue is more polarizing on HN: nuclear power or Bitcoin.

...and every time there's the same litany of poorly researched talking points repeated ad infinitum.

Almost nobody ever points out the systemic risks of highly centralized power plants in case of war or civil war or terrorism.

When we get that solar panels in orbit, how do we transfer the energy back to Earth?

Here's an article from the Department of Energy that outlines a few options: https://www.energy.gov/articles/space-based-solar-power

Microwaves and lasers are the two options they describe.

That's a lot of cooked goose.

In all seriousness, that is a pretty terrifying industrial project. I'm not sure how you'd make the failsafes strong enough to match the risk of cooking thousands of people alive.

The power density wouldn't be near that bad. From geostationary, the footprint would be several square miles, and birds could fly through the beam without harm. Even getting that much focus from phased-array transmitters would require a reference signal from the ground, so if the beam went astray it would spread out much more.

Despite the low power density the ground station would still be cheap, since it'd mostly be wire antenna.

(Source: the book The Case for Space Solar Power)

Moreover, how to keep that from being intentionally weaponized, assuming that it is strong enough to cook people alive.

I imagine a large solar power-sat in orbit could be easily crippled by missiles launched from outside its aim-zone, or even well-aimed orbital debris being slung around.

I'd worry more about such an installation being hacked/sabotaged by someone who doesn't care if the billions-of-dollars investment gets turned into confetti within a few hours.

This is why software is kept out of control systems whenever possible.

“They hacked the web interface that tells me how much power the diagnostic thinks is being generated so now it doesn’t agree with the receiver” isn’t as scary as “they redirected the death ray to a nearby city”.

> The long wavelength of the microwave requires a long antenna, and allows power to be beamed through the Earth’s atmosphere, rain or shine, at safe, low intensity levels hardly stronger than the midday sun. Birds and planes wouldn’t notice much of anything flying across their paths.

It doesn't look like they mention a specific wavelength in the article, but the reference to "long wavelength" and "long antenna" seems to imply frequencies much lower than what I'd consider microwave.

Generally, the lower the frequency is, the less harm it causes to living things. Though I either way I wouldn't want to live right next to the ground station. I suppose it would be similar to living right next to a radio transmission tower or a radar station.

I generally expect sub 100 MHz to be called radio rather than microwave, but I shouldn’t make assumptions.

So you need a quarter wave to be longer than a building to make it not a concern, practically it needs to be much longer if you’re sending serious power and want to keep the density down.

Let’s aim for a 1 km long antenna in space with voltage high enough that ohmic losses are not a concern, or a very large superconducting antenna and all of the apparatus that goes with it. Millions of volts likes to arc when you have debris just floating around everywhere in space. Superconducting is much more attractive regardless of the extra apparatus.

I don’t see many points of concern, though you could still damage infrastructure if you aren’t constantly aligned. Blowing up terrestrial grids isn’t as bad as cooking people alive.

It would still be the most dangerous thing we've ever put in space by a pretty wide margin. Keep space filled exclusively with small tin cans is not an accident and not entirely due to technical or economic limitations. It is a very sensitive political area. Putting anything that could be dangerous in orbit makes it dangerous to everyone. So your adversary across the ocean blows up a reactor? Cool. As long as they don't blow me up.

> It would still be the most dangerous thing we've ever put in space by a pretty wide margin.

I don't know about that. We set off a 1.4 megaton nuclear warhead in space back in 1962 just to see what would happen. (What happened is that we created a temporary radiation belt, knocked out a bunch of satellites, damaged some of the electrical and communication infrastructure in Hawaii, and caused some impressive light shows.)

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

That's why the microwaves approach might be preferable to the laser approach: "Safe, low intensity levels hardly stronger than the midday sun. Birds and planes wouldn’t notice much of anything flying across their paths."

Or you can microwave cities.

Pump few gigawatts of microwaves over a country, and revert it to stone age in minutes.

As other commenters note, there are methods of getting power down from orbit and I understand them to be viable, at least in theory.

But my point was primarily as a means of producing energy for industries and people that are stationed off Earth. For a space habitat, solar is just ridiculously good compared to the alternatives.

Lasers or microwaves are the current candidates.

I guess we do not. The only things one might want to export from space to Earth are things that absolutely cannot be made on Earth. Everything else would almost certainly be produced cheaper on Earth.

Creative ideas and all, but we're already struggling to build, transport and store enough solar energy down here on earth. I'm guessing that in the timeframe this needs to deliver the cc mitigating benefits, this might be less feasible...

LFTR addresses almost all of those issues that you cite. Those are "old nuclear". Proliferation (doesn't make weapons isotopes as part of its fuel cycle), 99% fuel use and breeding (even can use old "spent fuel"), low/zero meltdown risk with a simple plug that melts and liquid fuel, scalable, startable/stoppable quickly.

But I agree that fission (and fusion currently) is current noncompetitive and not worth investment (aside from research) until the economies of scale that solar/wind/battery are currently undergoing stabilize or plateau.

>It's because humans ultimately can't be trusted with

You can already see what effect nuclear has when the safety is not done well (Chernobyl and Fukusima). But even those cases kill less people than coal done well.

Ok a lot of nonsense here but just to address one egregious point:

There is no waste problem, stop spreading misinformation [1]

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

The finns building a long-term storage site for waste produced domestically does not equate to their being no global issue with long-term nuclear waste sequestration. There absolutely is.

The only site in the entire world currently intaking waste is WIPP, and it's been dogged by problems: https://en.wikipedia.org/wiki/Waste_Isolation_Pilot_Plant

The only two other sites that have ever actively taken in waste, Gorleben and Morsleben in Germany, are now closed and both have severe site stability issues requiring ongoing remediation into the billions of euros to ensure the sites don't fail.

This project has definitively proved that it's a political problem and not a technical one. The solution is to build a deep unmarked hole and bury it, duh.

I think it's not that controversial of a statement to say that the main problem with nuclear energy in general is political. As evidenced by the fact that this thread exists and full of people who are trying really hard to find problems with our only extant technology for carbon-free large-scale energy production. And that we keep shutting down perfectly good nuclear power plants. There's just no basis on fact here.

Meanwhile, the water is lapping at our feet.

The waste argument should be made in comparison to the alternative baseload option: carbon.

Carbon waste is worse than nuclear waste, yet is given a free pass because it floats and we can’t touch it or see it. But it is very real.

Even the visible carbon ash is stored, toxic and radioactive too. You don't want that contaminating water supplies either.

The argument against fusion is that containment vessels aren't there yet. So this becomes a debate on the speed of containments vessel research.