▲ 107 points
back
30 comments
When I read this article, I was blown away by this passage:
> Typically, outsiders cannot comprehend how the massive expenditures never manage to yield energy. Typically, insiders cannot comprehend how little is being invested in a project that presents such immense technical obstacles and also such potential. A graph commonly passed around among the insiders—an enduring scrap of twentieth-century budgetary ephemera—depicts the 1976 federal plan to build a working thermonuclear reactor. The graph tracks various scenarios for attaining fusion energy. The “maximum” effort, the most expensive up front, with initial spending as high as nine billion dollars a year, was projected to yield a reactor by 1990. The “moderate” effort, with spending never exceeding four billion dollars in a year, would take fifteen more years. The fusion community might be easy to criticize for its many unmet milestones, but for decades the United States has never come close to even the moderate effort. In 1977, when the American fusion budget was at its peak, government investment in the research, adjusted for inflation, was seven hundred million dollars; by 1991, this had fallen by more than half. It is now half a billion, not appreciably more than the Korean budget.
The chart was produced by Energy Research and Development Administration (ERDA), which was later subsumed into the Department of Energy and presumably represented the expert wisdom at the time. (Please correct me if you have info that most experts were in fact more optimistic.) It can be found, in 1976 dollars, as figure 1 here:
http://www.21stcenturysciencetech.com/Articles_2010/Winter_2...
I am having a hard time getting over the importance of this. For all the discussion of global warming, energy prices, population crunches, etc., etc., fusion is always ignored as a solution for the simple reason that people think it's unfeasibly costly. And they think that only because they think it's had a history of being grossly more expensive than promised. But if we've really known how much it would cost to get commercial fusion power since 1976, then this reasoning is completely faulty. So then if we just haven't spent the modest amount of money on it...isn't this insane? Someone please tell me what I'm missing.
From the article, ITER is currently projected at $15B which is not that much for a construction project of this scale. The crossrail project in London is costing ~$25B. Heathrow terminal 5 cost ~$6.5B. Also, it's not unusual for the costs of a major infrastructure project to be double or triple the estimate. I think they are doing just fine.
Not everyone thinks the problem is the R&D cost. I mean if that was the only problem, then yes, I would agree, give them a blank check.
It's my opinion fusion just isn't a good way to generate power on the scale at which 21st-century civilization operates (i.e. a fusion reactor wants to be very much larger than a gigawatt) so we would be better off to forget about it and spend the money on either fission (which can actually do everything fusion promises except have emotional appeal - and I suspect the real source of the emotional appeal of fusion power is that it exists only in people's imagination) or if people won't politically accept fission, just keep working on solar, wind and biofuels.
Are you arguing that we shouldn't pursue fission because we won't need as much power in one place as is provided by efficient large-scale fusion designs? Because that seems very weird to me. Power is great. Our civilization thrives on power. The costs of making things energy efficient are large.
Now, the current thinking is that available power sources either cause harm (fossil fuels) or are expensive and intermittent (renewables). And that thinking is prudent. But if we can solve all this with fusion...that would be fantastic! Think of all the things you could do if you had cheap power, and all the regulations you could dispose of. Is your argument really that we won't find a way to make use of cheap, clean power?
IF petawatt-scale power is possible (and that's a big if), one of the cool things you could do is bottle it and send it where it's needed.
Literally.
How?
Turn it into oil. Crack seawater to get hydrogen and separate out CO2 that's dissolved as gas and carbonate. Fisher-Tropsch glue the two together.
With a petawatt of power, even if the process were only 50% efficient (which is roughly what US Naval Research Lab, Brookhaven National Lab, and MIT research suggest, you'd be synthesizing about 2,580 billion barrels of oil annually. The US currently consumes about 6 billion barrels, and total global extraction in 2013 was 317 billion barrels. That's 368 bbl/year per person on the planet (US consumption is 19 bbl/person-year).
So, yes, that's overkill, but you could put the additional energy to use, oh, say, sequestering carbon from the atmosphere (which, say, the surplus oil you're synthesizing pretty much is).
And if you burn the oil somewhere else later, it's actually pretty much OK. The stuff'll be clean (it's pure hydrocarbon, no contaminants), it's carbon-neutral (though if you're sucking seawater out of a single spot on the globe that might be an issue (we know that messing with ocean pH is a Less Than Desirable thing.
But it turns out we've got a pretty considerable global infrastructure for doing just that with oil.
Mind: I don't think fusion power will happen any time soon, possibly not ever. But if we actually did have that much power to play with, we could probably sort it reasonably well.
Okay, so let's build a petawatt-scale fusion power plant. I can actually think of practical ways to build one if it can be that large.
But that's not going to happen. A petawatt-scale fusion power plant is not going to be economically or politically feasible this century, no matter what progress we make on plasma physics and no matter how cheap it would be per watt.
Every power source has a preferred scale. A fuel cell can be the size of a mitochondrion, but a combustion engine, in order to be efficient, really wants to be at least big enough for a lawnmower, and an efficient fission reactor wants to be at least a good few megawatts.
My conjecture is that a fusion reactor wants to be at least a petawatt, and even if we solve the plasma physics problems, a gigawatt fusion reactor will cost more per watt than the equivalent in solar plus appropriate attached storage - which makes it pointless to work on fusion at this time.
There is no such thing as an 'efficient' combustion engine. There are only combustion engines that are more efficient than other combustion engines but all of them are inefficient compared to other types of engines.
Fair enough, I guess I just question your two premises. Can you link me to anything that would support the Petawatt number? If China can make use of a 20 GW plant (the Three Gorges Dam) and the Congo can make use of a 40 GW plant (the proposed Grand Inga Dam) at locations determined by nature, don't you think that many countries on earth could make use of 100-1,000 GW plants with significantly more freedom for location?
Thanks for your thoughts.
I don't know whether the limit for economical fusion power is 1GW or 1TW or 1PW either. But if it turns out to be the higher side of that, then I'd be leaning against it. Sure, we could use the power, but right now, our power needs are low enough that we might build only one or two of them to run whole countries or even continents off of. I don't really like the idea of having One Huge Power Plant for most of a continent Seems too vulnerable to all sorts of failure modes to me. Imagine if some important part deep in the machine breaks because of some mistake somewhere, or even sabotage. The whole continent's power could be down, and it might take weeks or months or longer to make a new widget and get it installed, given that there's so little demand for them with only a couple of plants.
I'm afraid I can't; it's my personal guess, and with an admittedly large margin of error. You make a fair point that if it turned out to be more like a terawatt, that might not be out of the question.
Well it might eventually be shown that it is just to unstable to be usable.
It is a big gamble to put money in to unproven stuff, if money was not a parameter global warming could be fixed whenever we wanted.
Tying two threads together
"So far, the vast machine exists only as 1.8 terabytes of digital information..."
A 512GB SD card was on the front page here today, 4 of those would hold all the plans for powering the planet for the next 30 million years. That's basically the size of half a playing card.
After reading the article, I wonder if the real issue to building something like this is as much a managerial science problem as it is a hard science problem. Complex endeavors like this should be hotbeds of social science research.
Very rarely do I read every word in an article posted here, but this was one of those times and I'm glad I did. The article itself was very well written. You could feel the tension engineers must feel and the mind-numbing scale of the project. I think it captured the atmosphere at ITER very well; a mix of impending doom and unescapable existential claustrophobia counterpoised by a sense of mission.
Neat in-vessel assembly video: https://www.youtube.com/watch?v=pt70mO2nQac
wow, that's some hugely complex build mechanism. A lot of the cost of building it is going to go into the design, and manufacture of scaffolds, and custom struts etc, just for the conveyor belt.
Would it have been cheaper to not have the assembly done in such a fashion?
From what I remember, they need absolutely crazy tight tolerances in every single join and gap. Something like <0.1mm for every component involved in the reaction chamber. And it's huge too. Otherwise the whole thing will start falling apart as soon as they turn it on. I imagine that's what's requiring the unusual build mechanisms.
That really illustrates the scale of the operation.
There has been a previous discussion, if interested, at https://news.ycombinator.com/item?id=7291008
Excellent article. I worked in fusion years ago and I even learned a few things about the politics of the ITER team that I did not know.
The comparison with Apollo or the Manhattan Project always comes up when discussing ITER. I truly hope that when it all comes together, that it is more of a Manhattan than an Apollo. The difference being, that once we built one atomic bomb, we kept building more and more and now it seems we may never be rid of them. Whereas with Apollo, we harnessed the political will to land on the Moon (multiple times even!) but once the political will dropped away, so did the path to the Moon, because Saturn Vs don't come cheap. If ITER ignites, but does not result in a path to cheaper, commercial use of fusion energy, then the next few generations may be in the same position with regards to fusion energy as my generation has been to space exploration.
This is why I tend to fall in Bob Hirsch's camp with regards to the faster, cheaper, smaller development of multiple fusion reactor concepts. Imagine if instead of developing the Saturn V for Apollo, that NASA had gone for many concepts from big rockets (e.g. Saturn V) to air launches (e.g. Pegasus) to suborbital balloons (e.g. JP aerospace) to space guns (e.g. HARP). Sure, the Russians would probably have beaten us to the Moon, but if a cheaper alternative to space launch had been found back when it was a national priority, we might have a Moonbase today.
The thing about gathering political will is that you have to have a clear target. The dangerous thing in the case of ITER, is that once that target (ignition) is achieved, the political will to do more beyond that goal may not be there. And ITER ignition alone does not answer our energy issues.
Is that Robert Hirsch of SAIC? References?
Yes, that is correct. Long before he was known for his views on Peak Oil (which I'm not really in agreement with), he was known for his work in fusion. I find it interesting that his wikipedia page shows little detail about his early career.
Here are some references: http://www.aero.umd.edu/html/sedwick/presentations/S9P1_Robe...
http://www.issues.org/13.4/hirsch.htm
http://fire.pppl.gov/fusion_critic_response_stacey.pdf (section V)
Found a current picture of the reactor:
https://www.iter.org/doc/all/content/com/img_galleries/final...
The project's main website looks pretty cool too:
Awesome article. But I gotta wonder - given the sheer size and cost of the ITER reactor, it's hard to see how we'll ever be able to build a commercial version of the reactor that's competitive with anything conventional. I don't suppose anyone here has any insight on that?
Besides the whole prototype status, ITER itself also has many problems which conflate the cost (and time) required to build. Every country involved wants to make a "significant" contribution, so you get many parts built in different countries, which may need to tool themselves to build these parts, and that need to later fit together. I even think one part had to be rejected because it didn't fit the rest.
The scale is also crazy, inside the tokamak the magnets can push against each other with forces twice that which the space shuttle's launch creates http://www.iter.org/newsline/269/1593
Once ITER is built though, and if it works, then we'll have the knowledge, and tooling required to make more of the parts to build a second (and that knowledge and tooling is spread amongst many countries). The cost will decrease a lot for the first few, and much more after.
The article also made me wonder how much of the cost is just due to the way the project is managed and split for multiple countries?
One example from the article: "A single manufacturer should build ITER’s vacuum chamber, a high-precision device that must operate with perfect symmetry. Instead, it will be constructed in nine segments, two in Korea and the rest in Europe. The design calls for certain features to be welded, but the Europeans decided to use bolts, which are cheaper."
The tokamak is the fusion community's choice of device for getting energy out of a fusion reactor, but smaller configurations may also be promising.
There's talk of exploring compact fusion configurations as a faster path to commercial viability; many private companies in North America are working on compact fusion and Lockheed's Skunkworks has expressed interest. Smaller size of these devices means research and operating costs are lower. The end product might be something more easily integrated into power grids (~100MW versus ~1GW of DEMO) and would hopefully have a competitive cost of electricity.
Path to Market for Compact Fusion Cores: http://fire.pppl.gov/FESAC_WP_Path2Mark_SWood.pdf
http://www.icosa.co/2013/02/lockheed-skunkworks-develops-com...
There's a plan to get from ITER to a commercial version. Next in line after ITER is "DEMO", which will have higher output and actually produce power (ITER will just vent off the heat). PROTO would then build further on the first two, further refining all the improvements made along the way, to provide a template for building commercially viable fusion power stations.
I'm going to assume the cost is the novel research behind this. Trying to create fusion in the first place has been researched for decades. If this works, and they can do it at scale, a new reactor wouldn't have the same cost as all the previous research attached to it.
>Thirty-five countries, representing more than half the world’s population, are invested in the project, which is so complex to finance that it requires its own currency: the ITER Unit of Account.
Oh lord. This is a perfect example of a bike-shed problem. They couldn't agree to use one country's currency or another, so now they're doing it in a totally custom currency.
> > so complex to finance that it requires its own currency
> This is a perfect example of a bike-shed problem.
I disagree, I think your post is a perfect example of the "anything I don't understand outside my field must be easy" problem.