Fusion Foolery
dothemath.ucsd.edu
dothemath.ucsd.edu
"In any case, the public reaction to the fusion story tells me a lot about our collective psychology. To me, it speaks to a sense of desperation. I think people sense that the “bad news” side of the ledger is overcrowded of late, and it’s starting to dawn on people that the future could possibly be worse than the present. This causes a cognitive dissonance in that our cultural narrative is one of progress, growth, and innovation. How can these competing visions be squared? News of fusion has the effect of temporarily permitting people to shed the anxiety and embrace the dream all the more strongly."
Case in point: my local news just published an article saying that global warming is disrupting spider romance. I, for one, am pretty excited about that!
Looking for something to extinct mosquitoes though.
And of course, a room temperature superconductor can have better or worse qualities than existing conductors in other ways. If it were much heavier, or required more expensive raw materials, or had lower critical current, it could in favt be worse than the existing solutions (the extra construction costs could offset the extra energy output).
Even today steam turbine plants produce higher cost electricity that natural gas combined cycle plants.
Perhaps because it's boring. Solar is basically the same tech as it was in the 70s. The only major difference is that it is literally more than a million times cheaper.
It's cheaper to install and operate solar than it is just to operate a coal plant.
Sure there are challenges with wind and solar, but that's all they are: challenges, not showstoppers. When something has such a compelling cost advantage, there is lots of margin to throw at the challenges.
It'll be difficult to power a grid with 100% wind and solar so that's all you hear. But on the flip side, it'd be quite straightforward to power the grid using 90% renewables using the existing plants for the last hard 10%. It'd be both cheaper & cleaner! Why the heck aren't we celebrating that? Add short term storage and it'll be 99%.
And this isn't theoretical or anything. We are doing it. We're currently installing solar & wind at about 5% a year, and that 5% is increasing by about 50% annually.
There is way more good news than bad news about solar. But good news is boring and bad news gets clicks, so you only see the bad news.
When you say this you need to also talk about how much it costs to build enough energy storage for solar to become some given percentage of grid power. There is probably some % where it stops being cheaper than coal
Solar, hydro and wind when you have them, nuclear when you don't, gas when you don't have nuclear and coal when you don't have gas. It is really quite simple. And the best bit is that you can have your solar and wind installations up and running before the ink is even dry on the kind of permit and investment required for a nuclear plant and it is much cheaper to boot, as well as decentralized so far more resilient and friendlier to the grid (assuming the grid is well designed in the first place, which isn't always true).
And 'in a country' is the wrong level to be thinking at. You should look at much larger areas than that, and across both longitude and latitude.
> It'll be difficult to power a grid with 100% wind and solar so that's all you hear. But on the flip side, it'd be quite straightforward to power the grid using 90% renewables using the existing plants for the last hard 10%. It'd be both cheaper & cleaner! Why the heck aren't we celebrating that?
It's technically cool, yes, but we need electricity now, and tomorrow, not next century.
I'm excited for fusion as a human accomplishment, as high tech research, maybe as a power source for future spaceships.
But I think it's likely the time may never come when it's used to generate power for the electric grid, and expecting that to come soon is just naive. We don't even have a single fusion plant that produces power yet, let alone the hundreds of them that would be needed for the tech to have a real-world impact.
Fusion energy companies like CFS (tokamak) or Helion (colliding FRCs with direct energy capture) might start producing electricity to the grid in less than 10 years.
We won't even focus on storing much of it but just build more and more.
The story on that is surprisingly positive, but it's quite easy to FUD by parties with a vested interest (carbon lobby, nuclear lobby) because some people will hear:
* "well, the sun doesn't shine at night. the wind doesn't always blow! check mate!"
E.g. https://news.ycombinator.com/item?id=37161473 (posted after I wrote this)
or, if it's aimed at a slightly higher intellectual level, things like:
* "pumped storage is a nice idea but realistically there just isn't enough space for all the pumped storage we need".
E.g. https://news.ycombinator.com/item?id=37161790 (posted after I wrote this)
* "we need 1-2 weeks worth of storage to deal with the inherent instability of wind/solar"
And they'll believe it because it all sounds plausible enough, even though it's wrong.
And then they'll repeat it all over hacker news lol...
Pumped storage and batteries for short term energy storage - hours to weeks. Australia is already building one pumped storage battery which should provide them with roughly 350GWh - roughly half of the short term storage they'd need if they had a 100% solar and wind based grid. With one plant. You can't do this in, e.g. Florida but the geography to build this is more than plentiful enough in most of the world (this topic is pretty well settled: https://www.nature.com/articles/s41467-020-14555-y but for some reason people keep disputing it).
Hydrogen for seasonal storage - for weeks when the wind doesn't blow and the sun doesn't shine. Roughly 2-3% of power will need to be stored this way on a 100% solar/wind grid. It's not efficient and expensive to generate hydrogen from electricity and then turn it back into electricity but still cheaper than generating and using nuclear power at the point of generation. It is cheap to store enormous amounts of power for long periods this way though.
But it seems hardly replicable say in central Europe.
(I am speaking about central Europe or more specifically Poland)
In Europe there's some geography known as "the alps" which has been used for years this way by the countries surrounding it.
Poland's environmental record is atrocious on almost every level. I guess the coal lobby must run the government or something because nobody else in Europe that reliant on fuel that dirty. They're a model of how not to do anything right.
Well, that is hardly solution for Poland.
https://en.wikipedia.org/wiki/Ludington_Pumped_Storage_Power...
Though
> The plant takes advantage of the natural steep sand dune landform of eastern Lake Michigan
- Ukraine conflict's impact on gas prices
- Cost of renewables from over 10 years ago when Germany jumpstarted the solar industry using subsidies.
- An ill-advised by highly popular switching off of nuclear plants. The Germans preferred higher energy prices now for less hassle and worries due to having Nuclear plants.
None of these points have any impact on the argument that solar is the clear winner of becoming the dominant energy source until 2050.
Which part is wrong? Except that we need more than 1-2 weeks worth of storage to rely on solar/wind as anything more than occasional addition.
https://reneweconomy.com.au/much-storage-needed-solar-wind-p...
"Graham says that the CSIRO modelling showed that at very high levels of wind and solar, a maximum of half a day’s average demand was needed for storage. In some areas of the grid, only around three hours might be needed."
Or:
https://pv-magazine-usa.com/2018/03/01/12-hours-energy-stora...
The amount of short term storage to get a grid between 80-95% running on solar/wind is measured in hours.
Electrolyzing and storing hydrogen in an underground cavern can buffer the rest and be stored easily for years if necessary.
Which one of reports at linked http://www.energynetworks.com.au/projects/electricity-networ... is one that you refer to here?
For the second one - is "more responsive loads" code for "load shedding" which is code for brownouts and blackouts?
So what if we roll another Eyjafjallajökull in one part of the world, simultaneously with a war breaking out in another, and some large-scale maintenance ongoing in another part still, while everyone's grid is designed to keep "oh half a day's tops" and not a watt more, because ain't anyone gonna pay for it? With globally reduced capacity for, say, a week, that "half a day's average demand" worth of storage may start running dangerously low, and what then? Will we somehow scramble to fire up long-closed fossil fuel plants in that time?
The goal is to reduce Co2 emissions and replace fossil usage.
Well, empirically, we only observe that countries heavily relying on renewable (Denmark and Germany are the poster children) (i) have a much worse average CO2/kWh ratio, (ii) still need to import energy when, well, the wind does not blow and the Sun does not shine.
But feel free to elaborate on where this is wrong.
117 looks ok to me. They're probably one of the most improved countries in the world in the last 10 years, although overall France's is likely still lower thanks to all their 1970s nuclear power plants - decarbonizing decades before anybody gave a damn about global warming.
For reference Poland is at like 650. They use an ungodly amount of coal. Environmentally Poland are an absolute a disaster compared to every country in Europe (even poorer ones), but, they didn't shut down a couple of aging nuclear power plants so they're on the good side of the American nuclear power lobby and get relatively little stick.
> they're on the good side of the American nuclear power lobby and get relatively little stick.
I'm French, I don't care about American nuke power lobby – and frankly, given the pitiful state of electricity CO2/kWh in the US and how much coal is burned, it seems that this lobby must not wield that much influence.
Exactly why they haven't enraged the nuclear lobby. The nuclear lobby couldn't give a damn about the environment and neither does their supporters.
Surely california has tons of taxpayer money in excess what is preventing them to do all that?
Look at these graphs: https://gridwatch.templar.co.uk/
You need to either overbuild 10x or you need lots of storage.
Even a 3x overbuild means you need weeks of storage if you want to avoid once a year brownouts.
And electricity storage still costs 100-400$/kWh per 10 years. An extra £60 billion per year while ALSO building 3x more solar and wind than we will use during peak times is a lot of money for the government to spend.
People used to say that you can't make baseline power with solar and wind. Slowly - and often grudgingly - they've come around and now realize that it isn't about 100% availability, that will always be a mix. The question is simply how much of that mix can be offset by renewables and the answer is 'much more than you originally thought'.
And as the price of solar and wind drops further and further that fraction only increases.
Every KWh that is produced by solar and wind does not need to be generated by fossil.
No one is ignoring it. We're rapidly deploying both technologies all over the place. It's exciting that real money is flowing into the industry. Fusion has speculator money, and Fusion gets extra attention because it's sold as basically, the cure to all energy problems, forever and ever.
Renewable optimists consistently under-estimate the storage problem by orders of magnitude. We don't have enough batteries on Earth to hold one single day's worth of electricity. And for us to actually rely on solar, we would need to store multiple months of electricity to survive the winter.
Solar is cheap because you're comparing a generation source that works on occasion with a generation source that works 24/7.
I'm not anti-solar at all. But I am realistic about what we can and can't do with it. The thing that will make a solar grid practical _today_ is focusing on variable loads. When the sun is shining, we should be cranking the AC of every home and office, smelting aluminum, filling dams, etc.
Wind can be very low for days and low for weeks.
Hydro is ecologically devastating and there aren't enough sites for it.
We absolutely are not headed towards extinction as a species. We are like cockroaches and the human species will survive. Using "oh no, the human race is dying out altogether and we must preserve it all costs" is a ridiculous argument.
> [..] it makes sense to cause local ecosystem damage in the form of flooding.
It isn't just the (risk of) flooding. It's hubris to believe we can actually foresee all the second-order ramifications from such large-scale terraforming projects. (Hydroelectric) dams have devastated the local wildlife, transformed the surrounding permaculture for the worse, and set off horrible chain reactions with devastating consequences for the local flora and fauna which have global ramifications down the line, as we are discovering today. Nothing happens in a bubble.
This same argument was once made of fossil fuel emissions (when we thought that smog in the biggest industrial cities was our biggest problem and outsourcing them to developing nations was a great win-win /s). That obviously wasn't true. Why would you wish to repeat the same mistake?
I recommend reading actual IPCC projections and analysis rather than whatever media hype you've been fed.
Maybe not the national grid, but there are plenty of places in the US where the sun shines a lot more than 40% of the day during the winter. Start there. If Southern CA, AZ, NM, TX, FL, etc.. which all get a ton of sun no matter the season can reduce their reliance on the federal grid, that's a hell of a start, and we should be celebrating that as a great first step and building off of that.
I thought the whole point of alternative energy sources is acknowledging that certain locales have potential for different energy sources (solar, hydroelectric, wind, etc..) and we should as a society should take full advantage of that? We know oil isn't going away any time soon, but why aren't we saying "oh City XYZ or State ABC is abundant in (alternative energy source), let's make use of that as best as we can"?
It looks like Los Angeles only averages about 37% for the year as a whole, and Albuquerque averages about 38%, so I'm skeptical that there's anywhere that gets "a lot more than 40%" in winter. You have to take clouds into account as well, remember. Cloudy days exist even in desert areas.
https://en.wikipedia.org/wiki/List_of_cities_by_sunshine_dur...
Nobody is pretending that solar can provide electricity via night time generation. That's what storage systems and alternative sources are for.
The daylight story in ABQ is this:
> "It is sunny 78.9% of daylight hours. The remaining 21.1% of daylight hours are likely cloudy or with shade, haze or low sun intensity. "
If you can generate solar power 80% of daylight hours, that's a pretty good place to start.
Again, it's not an end-all-be-all solution. But you're talking about some very highly populated areas that could be using solar energy a hell of a lot more than they are now. Phoenix, Tucson, San Antonio, Dallas, LA, San Diego, Houston, Miami, Orlando, Las Vegas....that's a lot of places that could be utilizing solar power in a way that does move the needle. That's some 30-50 million people (if not slightly more) that could reduce their use of the federal grid in a measurable way.
I am optimistic that Santa Fe will soon see a new utility scale PV generation install that ought to generate power (with storage!) corresponding to roughly 1/3 of the city's residental use. The standard NIMBY-but-I'm-all-for-solar-have-some-myself crowd is out in force, however.
The OP appeared to be suggesting exactly that.
It's like the (many) solar advocates who pretend that a solar nameplate capacity of x watts is equivalent to a conventional nameplate capacity of x watts, when in fact you're going to need 2x watts at a minimum, with an expensive storage system on top of that.
Your last paragraph is something different -- that's not an implication that solar can generate at night. Note that levelized cost of energy does take into account that consideration (that why it's "levelized"). And renewables are trouncing new construction nuclear on that metric.
Trying to sell it as "no biggie" will not work well.
Where?
Maybe we should try a more global approach to energy.
Almost the entire world? The sun is up for less than 9.5 hours in winter unless you are very close to the equator.[1][2][3][4]
And that's assuming 0% cloud coverage. In reality, places like Germany only get an hour or two of real sunshine a day in the winter[5].
[1] https://www.timeanddate.com/sun/usa/new-york
[2] https://www.timeanddate.com/sun/usa/san-francisco
[3] https://www.timeanddate.com/sun/spain/madrid
[4] https://www.timeanddate.com/sun/china/beijing
[5] https://www.statista.com/statistics/982758/average-sunshine-...
Global transmission of electricity isn't a "problem," it's non-existant.
Russia must be confronted with economic, political, and yes, military incentives to behave very differently in the future, or at least to take more responsibility for the reckless actions of their leaders. In the meantime, working towards independence from fossil fuels has become even more important, even for those who aren't otherwise known for advocating good climate stewardship.
See [1] for a very short English description and [2] for the German original.
[1] https://deepresource.wordpress.com/2017/09/16/blueprint-100-... [2] https://www.ise.fraunhofer.de/de/veroeffentlichungen/studien...
And if you combine the graphs for solar and wind from this: https://aleasoft.com/european-solar-and-wind-energy-producti...
You can nicely see that they are complementary in terms of energy production.
That's the conceptually simple solution, but ofc there's more practical ones. Mix in wind, some nuclear, different storage solutions (Ev2grid, pumped hydro, rocks-in-a-box, etc..), demand-side solutions (VPPs), efficiency improvements (insulation, heatpumps), a few of the hundreds of advanced projects being researched (solar in space, enhanced geothermal, iron air batteries, SMRs, etc..). Combine the 1000s of solutions being proposed, researched and scaled out and you have a very solid plan of attack.
The latter is crucial. The cost of providing "synthetic baseload" in Germany doubles if you don't include it.
See https://model.energy/ for optimization of an energy system using these to provide synthetic baseload, using various tweakable cost assumptions and historical weather data. The cost is not outrageous, likely well below new construction nuclear.
In a place like India, you don't even need wind and hydrogen. PV and batteries will do just fine.
Solar and wind cannot be used in this role and at this scale due to their massive variability
This of course has a lot of political implications, but if we wanted to do it we could.
IMO, the right move is for the government to subsidize power generation to the point of absurd excess, like we do for agriculture. It's a matter of national security, and it ends up acting as a subsidy for industry of all kind.
Wind power requires an enormous amount of epoxy and the windmills wear out and then you have to throw an enormous amount of epoxy away and it is not green.
Storing energy using batteries from the above sources requires an enormous amount of nickel and cobalt mind out of the Earth by third world countries and is not green.
Fusion power requires mining lithium and is not green.
Nuclear power requires changing people's minds and when it does fail it causes two-headed fish. While perhaps green, it is not politically feasible at this time.
Here's an idea. Take public transit instead of using that Tesla that required a bunch of imported rare earth minerals from sweatshops in China and microchips from Taiwan -- which may or may not be available to us in the future in this geopolitical climate -- and lithium and cobalt from death trap mines down in Peru and Africa.
Burn natural gas in your stove instead of electricity which is generally coal-fired, or if it isn't coal fired, it comes from so solar and wind and is therefore not green.
My point is that the original paragraph stands. It's going to get worse before it gets better and no one wants to believe this.
Where can I buy this machine?
Yeah, having mine next to your house sucks, we need to get better at managing that kind of thing. But it's a local suck. It's in a totally different category than having the climate shift into a mode that no longer supports our food crops, for instance.
Wow, you've concatenated a string of lies into a single sentence. What a density of mendacity!
Solar requires none of those things, and is quite "green". Solar doesn't even need concrete for footings -- steel earth anchors, or even laying modules flat on the ground, are both practical and cheaper.
Sadly incorrect. You can see the variation of wind and solar we get here in the UK at: https://gridwatch.templar.co.uk/
As you can see, fossil fuels are about a 1/3rd of power and we expand renewables by maybe another 30% before we start getting excess power. After that point, you start to have excess power while no longer reducing CO2 by much as you still need the Nat gas for the cloudy and still days.
In practice you might be able to get to 75% non-fossil fuels before you're firmly into the realm of diminishing returns without a big storage breakthrough or a doubling/tripling of electricity costs (its triple to go pure wind and current tech energy storage because you need up to weeks of storage to not have blackouts, but with a nuclear power expansion youd only need a doubling).
But even if you can't easily get to 90% why not celebrate the fact that 75% is easy. Sure the last 25% is hard, but 75% good news and 25% bad news is not what I see in the news articles.
Breeders are from an obsolete vision of the future where we'd have an order of magnitude more nuclear power plants than we do now. In that future, uranium would start to be in short supply. But that didn't happen, uranium remains cheap, and breeders have no business case.
New offshore wind projects in the UK contract to produce power at about 40% the cost of new nuclear, about the same as gas before the Ukraine war interruptions. And that’s before the 45% construction cost overrun on Hinkley Point C that EDF is responsible for.
Onshore wind would be half the price of offshore but we’ve stopped building it because Tories.
Over provisioning a mix of renewables is likely the most cost effective route to greening the electricity supply.
Modelling for Denmark shows over 90% of its electricity consumption could be provided through modestly over provisioning renewables 1.5-2x along with 4-12h of storage and 99% with 3-4x over provisioning renewables and 48h of storage.
Most of the public can't grasp the mathematics of compounding change.
In the 1940s - cities smelled like sh!t, were so polluted you almost never wanted to go outside, healthcare was still largely snake oil, the average person was still working 10 hours a day 6 days week, human rights outside of straight white christian men were questionable, AND we were in our second world war in 20 years - at the advent of the nuclear bomb.
Look at where we are now.
If you don't see progress, and if you can't see how people back then thought the world was surely coming to an end, and you think it's worse now, I'm honestly amazed.
At no point in time, was progress a up and to the right with no impediments. Every advance we've had, people have worked HARD for. So thank your lucky stars so many people get up every day and keep working for it.
Because it isn't easy. But it's what we do. Always have. Always will.
Eventually, we will reach the physical limits of how good technology can get. But we are laughably far away from that point in so many major aspects of life, that we've got a long way to go before the 10 & 20 year future might not be better.
> But it's what we do. Always have. Always will.
You can’t extrapolate from the past because never before in history we had the capacity to have impacts at that scale (e.g. atomic bomb). We are in uncharted territories.
I'm always amused by the doomsaying Malthusians. They're always so confidently wrong.
If you cut down a few trees to build a house then 50 years later the forest will reclaim the land. If you instead cut down the entire forest then that can't happen.
>We have been using nonrenewable resources for hundreds (tens of thousands?) of years, and yet we keep finding more.
Multiple civilisations in the past have collapsed at least in part due to resource shortages. Again though, because none of these civilisations were _global_ the collapse was local and could be recovered from using outside resources.
Seriously, we can't even consistently land on the moon but everyone is convinced we're going to have space mines operating by the end of the century. Keep dreaming.
The rate of technological progress has if anything _slowed down_ in the last 50 years. Compare the difference if you look at the 40s vs the 60s, then the 2000s vs the 2020s.
But i do not see how thats relevant
More and more technological solution I see nowadays are the solution to problems technology created in first place.
https://www.economist.com/leaders/2013/06/01/towards-the-end...
In the middle of an environmental crisis? Are we not going to factor that outcome into our measure of progress?
You have to prove the analogy, not just state it.
"Plumbing Facilities
In 1990, only 1 percent of our homes lacked complete plumbing facilities. But, things were much different in 1940, when nearly half lacked complete plumbing. Then, about ten States had rates approaching or exceeding 70 percent. In succeeding decades, the proportion of homes lacking complete plumbing dropped dramatically, falling to about one-third in 1950 and one- sixth in 1960. It is interesting to note the States with the lowest percent- age of such homes in 1940 were higher than Alaska, which topped the 1990 list."
Some can't be reasoned with.
Also population growth has declined more rapidly than predicted and in places where it wasn't expected. Which brings it's own issues most of which have to do with finance and sustainability of rentier capitalism, but not being able to feed and house people.
Partly due to policy makers fuckups around the 2008 financial crisis birth rates in US dropped. Which means today's 14 years olds are going to enter the best job market in three generations.
Physicists are no exception.
~there's a lot of a bad news around, so people are excited about potential good news~
as if before climate change there'd never been media hype over scientific breakthroughs
this whole article makes exactly one interesting point: the media hype around fusion fails to properly illustrate the actual energy being put in to the experiment. it explains this well once, then wraps it up in snark and verbosity and repeats and re-explains it 5 or 6 times
the useful information in this article could be imparted in one, maybe two paragraphs
In the case of these laser fusion stories, the effect described is at least real, but the press is happy to spin it in wildly optimistic ways.
The LM99 room temperature semiconductor circus was a very different phenomenon, mostly driven by internet people rather than journalists. And very short-lived. If mainstream journalists were as skeptical of fusion as a net power source as they were of LM99, people would be better informed.
"fast-charging, high-range" EVs are already here, so I'm not sure what you're on about. "non-degrading," no, but nobody is actually claiming that. machine learning also exists and does useful work. We can argue that it's overhyped. That's a different conversation than the one about whether fusion power can ever work.
"In a week there will be a lot of thinkpieces on how the internet got this so wrong. A lot of people just want to believe, full of hope, because their life situations (poverty, housing crisis, hot weather) are so dire ... I should say that even if it doesn't affect them personally, it creates their worldview when it is 90% of their news. Homeless crisis, war in Ukraine. Why wouldn't you want and allow some positive, apolitical, hopeful story about noble science to consume your attention instead?"
https://news.ycombinator.com/item?id=36895407 https://news.ycombinator.com/item?id=36895665
It was downvoted and flagged. Mainstream press like NYTimes got it right by ignoring this until replication. HNers fall for hype as much as the rest of the world, it's just different hype.
But the NIF was never “about” societal energy. Its primary purpose is nuclear weapons research. This pesky thing called the nuclear test ban treaty means we can’t just go around detonating nuclear bombs whenever we feel like it. Surely we did not run out of South Pacific island paradises to blow to smithereens. The NIF allows study of matter at extremely high energy density.
NIF was built by the Lawrence Livermore National Laboratory, a weapons research and development lab established during the Manhattan Project. Talk of laser fusion as a viable path to commercial fusion reactors is propaganda intended to further the budgetary aims of the nuclear weapons industry. The realistic path to fusion power lies through magnetic confinement reactors (eg. ITER, Wendelstein-7X, etc.)
https://www.osti.gov/biblio/50733
""" The National Ignition Facility (NIF) will enable us to produce energy densities (energies per particle) that overlap with the energy densities produced in nuclear weapons, yet the total energy available on NIF will be a minuscule fraction of the total energy from a weapon. This combination of low total energy with weapons-regime energy density will allow us to pursue, besides ignition experiments, many nonignition experiments. These will allow us to improve our understanding of materials and processes in extreme conditions by isolating various fundamental physics processes and phenomena for separate investigation. Such studies will include opacity to radiation, equations of state, and hydrodynamic instability. In addition to these, we will study processes in which two or more such phenomena come into play, such as in radiation transport and in ignition. Weapons physics research on NIF offers a considerable benefit to stockpile stewardship, not only in enabling us to keep abreast of issues associated with an aging stockpile, but also in offering a major resource for training the next generation of scientists who will monitor the stockpile. """
Anyway, in recent years at JET they have been doing a new round of deuterium-tritium experiments. The interior of the reactor has been refurbished with tungsten and beryllium inner surfaces, like ITER will have, and they have been testing longer reaction pulses. Sounds promising https://physicsworld.com/a/fusion-energy-record-smashed-by-j...
I think they have a good shot at commercial fusion by 2030s. The REBCO magnets reduce the size and complexity of the machine massively. Oddly, Brexit is a benefit to them; JET employees can no longer go work for ITER if they want a pay and lifestyle upgrade, but they can go a few hundred meters down the road.
They have delayed the announcement of the delay, but it's expected to be another 5+ years.
It's the most complicated machine ever built, with each part built by a different firm in a different country for political reasons. The first vacuum vessel sector installed was corroded. Korea used steel that didn't meet the specifications. 8 more sectors from four different nations to go, each one could bring its own 5 year delay. Or maybe they don't find any issues on the inspections but only find a leak after the whole machine has been assembled.
How do you know it's not the other way around:
That they're using the gigantic piggy bank of the military to fund actual fusion research?
I agree that ICF doesn't seem like a winning strategy. But surely it's not complete waste, either, right?
But you can achieve plasmas of equivalent density and temperature to those you get in an H-bomb explosion, which is invaluable to weapons researchers if they're not allowed to actually detonate any bombs.
Why is it so valuable to create those temperatures for researchers?
Isn't this wrong? LLNL was an 1952 off-shoot of Lawrence Berkeley National Lab, which in turn was founded in 1931. LLNL was not established during or by the Manhattan project, which ended in 1946.
> [Lawrence Livermore National Laboratory] was originally established as the University of California Radiation Laboratory, Livermore Branch in 1952 in response to the detonation of the Soviet Union's first atomic bomb during the Cold War. It later became autonomous in 1971 and was designated a national laboratory in 1981.
https://en.wikipedia.org/wiki/Lawrence_Livermore_National_La...
> [Lawrence Berkeley National Laboratory] was founded on August 26, 1931, by Ernest Lawrence, as the Radiation Laboratory of the University of California, Berkeley, associated with the Physics Department.
https://en.wikipedia.org/wiki/Lawrence_Berkeley_National_Lab...
> Although the Manhattan Project ceased to exist on 31 December 1946, the Manhattan District was not abolished until 15 August 1947
https://en.wikipedia.org/wiki/Manhattan_Project#After_the_wa...
Isn't the research basically done, as in "we can build big enough bombs to annihilate whatever we want"?
Existing nukes degrade due to various factors such as the natural radioactive decay of the uranium and plutonium they’re made of. Supercomputers are used to simulate this and make predictions about reliability, remaining lifetime, etc…
When it comes time to refurbish old bombs, ideally they should have the cores reprocessed into new, modern designs with better safety, reliability, and less fallout when they go boom (better efficiency).
All of this requires measurements, simulations, tests, etc…
In the past it was done with experiments on complete bombs, now it’s done with experiments on just small subsets that don’t go boom in the desert.
It's the gain factor of the fuel itself, not the entire system that achieved positive value. The point is that until 2022, noone was able to achieve any gain at all. So this was a breakthrough (alas many more needed to make it commercially usable) and just because some stupid people misinterpreted it, it doesn't mean it's not important.
And, like you said, nobody who was interested that I talked to failed to understand the pretty simple setup: overall cost is the asterisk, but gain factor is great! The best I can assume is that this guy deals a lot with students and maybe students didn't grok the whole situation (as all idealistic and naive - wonderful traits in students - are likely to do).
In any case, it just seems really pessimistic to say "really don't expect anything to come of this laser process", because of the obvious practical reality: other fusion researchers aren't all using lasers to create fusion, yet all of them can use the results from the laser fusion to make efficiencies in their own designs.
You can think of a this as test flight with a plane model vs. building the actual plane. We needed to test the aerodynamics and see if this flies at all, it worked, and now it's the time to build the actual plane. Someone has to do that test at some point. Would make no sense to build a plane without knowing if it's going to work.
The 100% yield scenario would yield 75MJ of energy.
Modern lasers that are 20% efficient would require 10MJ instead of 400MJ for the reaction.
In theory we only need a 13% yield with modern lasers to reach breakeven. 9% with 30%, 7% with 40%, etc
Note that this is just for this particular pellet they tested - larger pellets likely have better yields due to scaling laws, but would require a more powerful laser array.
I think the article is rather pessimistic, understandably so, but doesn’t really paint an accurate picture of the progress made. If anything, we are closer than we think.
Nowhere in this article does it mention the gains from using more efficient lasers, instead treating the 400MJ input as a constant. Bad reporting.
I knew I recognized the name so I checked and yup it was him. I've seen his stuff before. Summary: "Everything is futile so give up now." He would have been arguing for the impossibility of space flight in the 40s, or small computers in the 60s, etc. His approach is to "do the math" with the most pessimistic assumptions and then conclude it'll never work.
Thing is: if you take that position you will be right more than half the time... probably more than 2/3 of the time. Being a permanent curmudgeon about anything new is a great zero-effort way to seem prescient.
Fusion is obviously monumentally hard, but there is a steady march of gains toward higher and higher energy levels at lower cost. There is no known fundamental physical reason why fusion can't be done in a reactor, and given that it's a path to effectively infinite clean energy it'd be stupid to not keep working on it.
That's pretty bad. Fusion might be feasible, but this approach isn't. (And I would say the same about ITER, even though that's WAY more feasible.)
We shouldn't pay too much attention to approaches that, when 100% successful, are still failures.
That being said, sometimes technology can do surprising things. If some other line of research somewhere yields, say, lasers that are multiple orders of magnitude more efficient then suddenly things might change and ICF would become a viable path. The massive inefficiency of lasers is the largest single problem.
The main reason magnetic containment is so much better than ICF is that huge electromagnets are more efficient than lasers plus inertia at confining the plasma.
That said I am more optimistic we wont need to go back to hunter gatherer en mass.
From his about page. Not exactly what I'd call doctrinaire.
He does have qualms about whether "effectively infinite clean energy" would do us any good right now. Look at what we've done with a temporary surfeit of very cheap (but not clean) energy. We are kicking out the legs of the stool we're sitting on. Human beings perch precariously at the apex of a massive biological edifice, the foundation of which is fundamental to human life, but we have the unfortunate habit of thinking all we need is technology and ingenuity. Far more than new technology, we need the wisdom to apply our tools for the long term flourishing of earth-borne life, of which humanity is a part.
Calling him that name is IMO an attempt of a "character assassination" and not a valid critique. His blog is "Do the Math" and his arguments are based on the math. If you can contribute a single example where his math is wrong, I'd be more inclined to believe you. Otherwise, I'll do my best to ignore your future comments.
> Summary: "Everything is futile so give up now."
I'm quite sure you can't cite an actual text where he wrote anything like that.
Whatever your feelings about his pessimistic tone, regularly debunking the latest popular media baloney is a reasonable pastime, which society could certainly use more of.
> Fusion is obviously monumentally hard ... steady march of gains toward higher ... There is no known fundamental physical reason why ...
True. And limestone can be mined by hand, on top of Mount Everest. The cost per kilogram would be enormously higher than any normal commercial limestone quarry, but if we just invested enough...
Meanwhile, "aim your solar cells roughly toward the sun" fusion energy is available at scale, now, and is orders of magnitude cheaper than there's any reason to believe possible for a commercial fusion reactor. And human society does not have infinite resources, to invest in sounds-cool stuff with massively negative ROI's.
"We don't have an energy problem. We have an energy storage problem."
The blog post author seems to be concerned about the time and attention spent on something that is not likely (i.e. as his post states is off by several orders of magnitude at least) to realistically help with alleviating the urgent problems of planetary climate change and resource exhaustion. I agree with that concern.
Faced with finite resources and a time limit, prioritization is essential to ensure the best chance of success. Technologies like fusion detract from finding and implementing more realistic approaches that could help with long-term civilizational sustainability.
I also really dislike back-to-hunter-gatherer primitivism. It's an ideology of extremely privileged people who have never actually lived "close to nature" a.k.a. poor. It's a reactionary fantasy of a lost golden age that never existed and belongs in the same category as American right-wingers glorifying the 1950s or neo-medievalists glorifying the 1500s.
It's a brutally tough problem and it's likely that we already have dialed in a certain amount of climate change we will have to deal with, but it's not futile unless of course we just give up.
I haven't seen the author you criticize ever promoted either that or that he suggests that humanity as the whole should revert to "hunter-gatherer primitivism."
I see now you are summarizing his writing just as "just lots of pessimism". But the numbers don't lie. Unless you directly depend some specific miracle (and you should state which that is supposed to be) the numbers, if extrapolated assuming continuous growth, result in resource exhaustion. Continuous growth is provably impossible without some miracle involved.
And I guess the above observation you translate to "the whole thing is just impossible"? Would you define your "whole thing" as the "infinite growth and scientists will give us a miracle allowing that"? If not, what are you talking about then?
In the words of Tom Murphy, from the article:
"The physical reality is that we are living in an ecologically, evolutionarily untested paradigm that is very recent (on relevant timescales) and powered by patently unsustainable practices and resource use. The cost is rapid ecological degradation and global disruption to the biosphere. It seems quite clear that the track we are on does not lead to the stars, but to ignominious self-termination of this whacky mode called modernity."
Energy and Human Ambitions on a Finite Planet
2021 Murphy, Thomas W, Jr
For sure a lot of people don't know obvious truths about fusion, but a lot of people don't know obvious truths about a lot of things. That doesn't cause all CPUs to ignite and planes to fall out of the sky.
Most noise you hear in the news is about something generating more heat than power is put into the plasma. But that's a very misleading thing for commercial power generation because not all the power spent on heating the plasma actually goes into the plasma, not all the power that comes out of the plasma can be turned into power, and there's other things that also require power for the whole thing to work.
Unless you consider ITER commercial, I don't believe you. Most news about commercial fusion is ‘X had a big investment round’ or ‘Y made a really hot plasma’, and if we were in the world where a startup's tokamak hit ignition (we're not), we'd be in a world actually qualitatively pretty excitingly close to net energy, even if it didn't break even end to end.
We have enough money that we can do both.
It should only be a question of whether this NIF research is useful.
Surely, they are actually producing some value. There research is beneficial to fusion in general.
Is it useful enough? I don't know.
- decrease spending elsewhere
- tax people more
- borrow on credit
- print money
Once you take into account that you waste a lot of energy heating up the plasma, and that you capture less than 100% for energy production, and that there are all sorts of auxiliary costs like magnets, the picture is a whole lot less rosy.
I support research into fusion energy, but IMO it's very likely it'll never be used for commercial energy production. It might eventually make it into spacecraft and submarines, but I think before it becomes practical to build a powerplant, renewables will eat its lunch.
The startup costs of heating the plasma are only significant in the high density/high confinement time/low temperature regime of pulsed ICF devices, which use no magnets.
The notion that laypeople have an accurate idea of rosiness based off of Qplasma progress is not one I can treat seriously. The jump from Qplasma 0.1 to Qplasma 1 is similar to the jump from Qplasma 1 to Qplasma infinity. We burn the plasmas to find the minimally viable machine and we do the science and engineering to continually push it down (obviously the world can not run on NIFs and ITERs).
It consists of
* Room temperature semiconductors
* Useful fusion power
* Quantum computers something anything useful outside of a simulation
I'm asking this in part because I was thinking way too much about applications of superconductors during peak LK-99 hype and now think room-temp superconductors would be the greatest possible discovery (we wouldn't even need fusion about because solar cells in deserts and a global superconductors grid). I wonder if I got to that conclusion because I obsessed over superconductors for weeks and if I'm missing other equally amazing, possible, future technologies.
So there's a lot of money and glory at stake if you can demonstrate you're making headway toward any of these goals.
There's a parallel to alchemy in all this. Before we knew it wasn't impossible to create gold through chemical processes, it seemed like a very appealing quest indeed. You have figures like Newton spending an inordinate amount of time and effort trying to figure it out.
Artificial intelligence. By a gigantic margin. Fusion would be very expensive and likely not competitive with fission, due to vastly more complex and expensive reactors. So effectively useless. Quantum computers: Nobody knows what they would be practically useful for, not even experts like Scott Aaronson. (No, cracking RSA is not useful.) Room temperature superconductors: The most realistic example I have heard about them is ... smaller MRI machines. Great.
Which is very similar to inertial confinement fusion. This is a nice side effect, but don't confuse it with the existence purpose of the lab.
The serious sources have always portrayed NIF's work as technical achievements. But they are read mostly by scientist and engineer types.
Mass media which hypes things is read, well by the masses, who dont have the patience or inclination to delve into technical details.
This dichotomy will always exist. I remember once reading a Chekov story where two intellectuals discuss how the townspeople are more interested in silly affairs and scandals rather than recognizing intellectual achievements.
The author claims that cryogenic targets will always be too expensive. Why should they be? Mass production has brought down the cost of precision devices like CD drives and hard drives. Why should it be so difficult to do this for fancy ice? They claim cryogenic targets won't stand up to the heat in a power plant like environment. They don't need to for very long. If the pellets are shot into the chamber, the time they spend exposed to residual heat from the walls can be very short.
Their entire discussion of the economics of ICF power is superficial. There is a range of conditions in which ICF power may be profitable.[0] Repetition rates of kilohertz as claimed are unnecessary.
[0]https://royalsocietypublishing.org/doi/10.1098/rsta.2020.005...
"So this news is both good and bad. Hats off for cracking into single-digit yield! But that leaves less room to improve. Even at 100% efficiency, we’d get just 25 times more energy out, or 75 MJ. That’s still not enough to pay for the price of admission (400 MJ, just for the laser part)."
Even if you solve magnetic confinement of a superheated turbulent fluid in a fusion reactor (and that's a big "if"), you still lose energy and destroy your container through the loss of neutrons.
I'm skeptical of any energy "breakthrough" now, be that with fusion, batteries and superconductors. With LK-99 I refused to care until it was reproduced (particularly given the factor that at least one of the paper's authors had previously had to restract papers). So many "breakthroughs" are just about building reputation for the individuals and seeking grants and funding for their research. That's all.
Solar, in particular, is our future.
And while we're worrying about far-future tech like fusion, we're ignoring the very real problems of today. Like it or not, we have and will continue to have a dependence on fossil fuels for some time to come. So much so that the US hasn't built a significant refinery in 30-40 years. I get the naive opposition to this but a new refinery produces WAY less pollution than the old refineries we have.
This is set to change with a new refinery in Oklahmoa that will be 100% powered by renewable energy and produce 95% less greenhouse gas (per unit of fuel) than existing refineries [2].
[1]: https://lifeng.lamost.org/courses/astrotoday/CHAISSON/AT316/...
[2]: https://journalrecord.com/2023/05/25/planned-cushing-refiner...
Scientific "stunts" which author correctly points out pretend to be graceful but come at an extreme cost and failures. But anyone has done any science knows this far too well that this is how you push boundaries of science and make progress.
For me, that colors everything that was said before it, and causes me to reinterpret the objections on cost/efficiency as being rooted in "we're not there yet, and because we're at the end of scientific progress, we'll therefore never get there".
And politely pointing out that common human bias might be a better approach than pinning "blame" (for people kinda being suckers for the idea of fusion power reactors) on ideology / mythology. The latter often get more emotional and adversarial.
[1] https://en.wikipedia.org/wiki/Project_Orion_(nuclear_propuls...
https://news.ycombinator.com/item?id=37092212
And someone linked a podcast of a interview with the director (in german, http://alternativlos.org/36). And after listening, I do became convinced that with the Stellerator design, a working Fusion plant is possible.
Maybe still not in 20 years, because it is hellish complicated, but some day.
(till then I would bet on harvesting more our existing very big fusionreactor called the sun)
In either case, controlling fusion is awesome technology and research, with lots of potential applications and deserves further funding. But yeah, please more of civil projects like Wendelstein and less disguised weapon research.
https://brilliantlightpower.com/suncell/
Note that the wikipedia article about this company is policed by skeptics and has been in dispute for more than a decade.
Also, this one should be shouted louder for those in the back.
> "Many in our culture truly believe in “the amazing future,” uncritically extrapolating our fossil-fueled joy ride into ever-more impressive innovations and technologies."
The computer was in the same category 50 years ago, as was the ball point pen 100. Just making aluminum cheap, was a miracle.
And before you think aluminum isn't a biggie, it revolutionized so many industries. Including airflight, missles, and more at the time.
Things impossible, became possible. And now these things are trivial.
The only question is, when it becomes cheap and easy to use fusion power, will it be the optimal power source at that time?
Scientific advancement comes from exploring every path at once, quickly pruning failed paths and doubling down on successes, with enough stochastic jumps to keep us out of ruts for too long.
Fusion power is 100%, hands down possible. There is no question in this.
It's all about cost, and my point was, when, not if fusion power is a net positive, will other things be cheaper?
I find it incredible how people on HN, a more technical site, seem to consider 10 years a long time, or even 100. Or that money is "wasted" because the research and work will "vanish", which is beyond absurd. Every penny spent on fusion, and cold fusion comes back to us in many wsys, there is no waste.
And your presumptive statement presumes that nothing changes. What is difficult today, is simple tomorrow.
Just like aluminum, which used to be more expensive than gold.