New fusion reactor design

[dragon]

Not to be left supposedly power as cheap as coal.

Fusion is back in the news, and not just because Lockheed is claiming to have solved the world’s every problem with magical million-dollar truck-sized generators. Historically, fusion research was all about possible strategies and mechanisms that might one day allow for controlled fusion; today, researchers are now at the point that they know specifically which breakthroughs must occur to make the technology viable — and, as always occurs in such situations these days, the upgrades have been coming in quickly. Several large, well-funded teams are competing to be the first to generate one net joule of energy from fusion, but an academic team from the University of Washington may be doing that work that ends up winning. They’ve dreamed up a new fusion power design called a Dynomak, and it could make fusion power stations cost competitive with coal.

What does “cost competitive” mean? Well, relative to prior fusion projections, a Dynomak facility could be built for about a tenth the cost of competing fusion reactor designs and produce up to five times as much power. This lets it catch up to the price-per-watt of coal, though only at the gigawatt scale; a 1GW Dynomak reactor might cost $2.7 billion, to a modern average of $2.8 billion for comparable coal plants. It’s all theoretical of course — this team has presented a major improvement to reactor design, but it will be up to larger, better funded research teams to actually make use of it. What’s the big innovation, then?
The team's test rig, called HIT-SI3, in full resolution. Has only three "helicity injectors." The final version, HIT-SIX should have, uh, six.

The team’s test rig, called HIT-SI3, has only three “helicity injectors.” The final version, HIT-SIX should have, uh, six. Click to zoom in.

There are two (major) schools in fusion tech right now: magnetic and laser confinement. In both cases you have to shrink down a sample of your fusion fuel — usually a mix of hydrogen and helium — so it get super-pressurized and thus super-heated, beginning the fusion process, and in both cases we need incredibly fine control of our inward pressing force. Whether it’s a complex magnetic field or the combined force of trillions of watts of laser light, getting strong-enough and fine-enough control of confinement is undoubtedly the biggest hurdle facing consumer fusion tech. The University of Washington’s work could make that process far more economical. By wading into the mathematics of magnetic confinement, they may have negated the need for magnets altogether.

Read: 500MW from half a gram of hydrogen: The hunt for fusion power heats up
Here's a basic cross-section of a spheromak rector. Note: still very complex.

Here’s a basic cross-section of a spheromak rector. Note: still very complex.

Probably the most widely publicized design for a magnetic fusion generator is based on a tokamak, a huge, donut-shaped magnet. The precisely shaped magnetic field it creates has been proven to be capable of containing a fusion reaction (just not while using less energy than the fusion itself creates). The costs are also prohibitive, as with the test model for ITER’s 30,000 pound super-conducting Slinky which recently arrived. Just as we’re finding with MRI machines, super-cooled magnets are a limiting factor for fusion power, and so we then developed another design called a spheromak — a magnetic fusion machine that creates its confining field by running current directly into the sphere of plasma at the power station’s heart.

This is obviously a huge improvement to the tokamak, but as you might imagine, just pumping electricity into a sample undergoing fusion isn’t very precise. Spheromak designs had the theoretical advantage in cost and efficiency, but struggled to show that they could actually work. Then, two years ago, this University of Washington team published an idea called imposed dynamo current drive (hence “Dynomak”), a proposed model for predicting the magnetic field based on the injection of outside magnetic fields. This was hailed at the time as having huge implications for fusion power, and now we’re seeing some hard numbers to that effect. [Research paper: doi:10.1016/j.fusengdes.2014.03.072]
JET's tokamak

Here’s the tokamak at the JET fusion lab in the UK – a smaller version of the tokamak that will eventually be installed at ITER

Essentially, the Dynomak is a precisely controlled spheromak reactor that uses imposed dynamo current drive to control the magnetic field it creates. This means we might be able to keep heated hydrogen isotopes in one spot without having to leverage several national economies to do it, or further deplete helium reserves and kill the birthdays of tomorrow.

That’s great, but this team’s test rig is only about 10% as big as it ought to be and uses only three of the final six “helicity injectors” that make the Dynomak possible. Until a full-scale test run is conducted, this will be just another fusion fairytale. Still, given the incredible cost benefits we’d enjoy if it is correct, this seems like an idea very worth exploring.

link

[Borgholio]

Huh, so two new fusion designs coming out this close to each other? Color me excited. A bunch of new ideas all at once can't be a bad thing for an idea that has been pretty much stagnant for the last 50 years.

[Jub]

It's too bad that it costs so much and takes so long to test new ideas in fusion reactor design, but the fact that the ball seems to be rolling a bit faster than normal with this and Lockheed's new design is great. I just hope that we get a breakthrough one of these days and the dream becomes reality.

[Guardsman Bass]

We'll have to see on this one. A big killer with fusion projects is scaling - something seems to work well at Y-size, but when you have to scale it up to Y-500 (the hypothetical level when it should go farther than break-even and get sustained ignition-and-burn) you run into a ton of problems.

[Starglider]

Huh, so two new fusion designs coming out this close to each other? Color me excited. A bunch of new ideas all at once can't be a bad thing for an idea that has been pretty much stagnant for the last 50 years.

Once again, you have absolutely no idea what you're talking about. Innovation in fusion reactor design has been continuous since the 1960s. Media coverage has been highly variable. You would know this if you read any fusion newsletters or even a single general audience book on the subject. Or even took 60 seconds to remember polywell and all those cold fusion press releases before blurting out innanities.

I'd note that the Wendelstein 7-X stellerator finally completed assembly and is due for first plasma next year. It's a very German (complicated, overengineered) way to attack the problem, but quite a promising one (stellerators are inherently stable, unlike all tokomak and derrived designs).

[Borgholio]

Once again, you have absolutely no idea what you're talking about. Innovation in fusion reactor design has been continuous since the 1960s. Media coverage has been highly variable. You would know this if you read any fusion newsletters or even a single general audience book on the subject. Or even took 60 seconds to remember polywell and all those cold fusion press releases before blurting out innanities.

The general public doesn't give a rat's ass about any of that. They care about what they can actually see. When you ask a layperson about fusion power, what are they going to say? That it's been "right around the corner" for the last two generations. And they would be right! All the incremental improvements and advancements don't mean shit to the people who they're going to have to sell the bill to. To most people, fusion is just a pipe dream. It HAS been stagnant as far as the public is concerned. All this new media attention is good because it's making progress public.

[Purple]

Out of curiosity. Do you guys think that fusion power will be a major source of energy in the next say 20-30 years?

[Darth Tanner]

Do you guys think that fusion power will be a major source of energy in the next say 20-30 years?

Even if they get the prototype working tomorrow and it turns out to cost £1/MWh with no side effects, no waste issue and near perfect engineering reliability we would still not see them as a major energy source inside 30 years. The construction time and investment decision to commit on new technology would simply be too great.

The fact we are still decades away from a working prototype, have no idea on its end run costs let alone resolved issues like reliability and sustainability means we will likely not be getting anything from fusion for 40-50 years on the industrial scale. Same as we were 40-50 years ago as the old jokes goes. And that ignores that no one has yet achieved net energy gain which is quite important for energy generation.

[Guardsman Bass]

The big killer in that case is any infrastructure built to replace coal and gas. It's not going to do your fusion reactor much good if you finally get a commercial design working in 2060, and there's already been a massive build-out of solar/wind/etc plus storage to replace coal and gas anyways - you won't even be able to claim environmental benefits. It would have to be a lot cheaper than any other power source out there aside from hydro.

[Purple]

So in other words fusion power is a dead end barring a miracle?

[Jub]

So in other words fusion power is a dead end barring a miracle?

Not at all Purple, even if we've already replaced Coal/Oil/Gas plants with cost effective alternatives, fusion is great for certain types of spaceships/space probes. There is also the fact that we don't know what the final cost per watt will end up as, fusion could be huge if alternative energy storage and generation turn out to be expensive solutions that dwindling petrochemical supplies force us into.

Also, I swear that you read the threads here and don't really comprehend them. Your question here sounds a lot like your questions in the global warming thread and show that you don't seem to research much before dropping in to ask random questions.

[Purple]

Also, I swear that you read the threads here and don't really comprehend them. Your question here sounds a lot like your questions in the global warming thread and show that you don't seem to research much before dropping in to ask random questions.

You know what. I could be answering the other part of your post and explaining what I meant but I feel addressing this is far more important. People here seem to have this rather strange idea that I just have to dispel. You feel that asking questions is something that is done AFTER doing research. And this is not only not true but it's frankly paradoxical.
Research and asking questions are in an exclusive or relationship. You either do one or the other. The entire point of asking questions is to avoid having to do any research. It's a way to check if someone has already done said research and can digest it in a way that's effort free for the person asking. The reasons for choosing to do this are myriad and include but are not limited to the fact that the person asking might not have the time, resources, scientific background, education, or quite frankly does not care enough to graduate physics just so that he can get an answer to a question that's relatively meaningless and will be forgotten by the end of the week.

Here is a graph to show you what I mean.


And you only do the research step if no one can help you AND you really, really, really care about a topic.

[Jub]

You know what. I could be answering the other part of your post and explaining what I meant but I feel addressing this is far more important. People here seem to have this rather strange idea that I just have to dispel. You feel that asking questions is something that is done AFTER doing research. And this is not only not true but it's frankly paradoxical.
Research and asking questions are in an exclusive or relationship. You either do one or the other. The entire point of asking questions is to avoid having to do any research. It's a way to check if someone has already done said research and can digest it in a way that's effort free for the person asking. The reasons for choosing to do this are myriad and include but are not limited to the fact that the person asking might not have the time, resources, scientific background, education, or quite frankly does not care enough to graduate physics just so that he can get an answer to a question that's relatively meaningless and will be forgotten by the end of the week.

Here is a graph to show you what I mean.


And you only do the research step if no one can help you AND you really, really, really care about a topic.

Your method of ask first, fail to understand the answer, complain when people tell you to do some research is frankly lazy. Plus, unless you have at least some background in a subject, the answer you get might not have any meaning to you, this happened repeatedly in the global warming thread where you admitted you had no knowledge and did no research, even as you continued to put the onus on us to educate you. We're not your teacher and this isn't a classroom, asking a question without knowing the first thing about the subject is both lazy and rude, doubly so when you then bitch and moan about the answers you get.

[Darth Tanner]

So in other words fusion power is a dead end barring a miracle?

Definitely not... even if we have achieved a build out of solar/wind/tidal and the hugely expensive storage to keep the lights on when its not sunny/windy these are very likely to be hugely expensive and inefficient in terms of demand response and land use - if fusion can come along at a cheaper rate and undercut them all we would be free of the constraints put on us by a renewable future... although if its too late we may well be committed to renewables to the point the grid can no longer handle a large single point generator which would add further issues to a fusion build out... can you imagine a flat base load coming on-line in an age when wind can swing from providing 3% of national demand to 400%.. the price instability in the market would hinder anyone coming on without a subsidy similar to Hinkley.

Far more likely though is that fusion would be competing with coal, gas and nuclear in 50 years time just as it would be today but at reduced levels of coal/gas depending on how many people take climate change seriously enough and reduced nuclear depending on how many Greens are in political power. Of course we might still be 40-50 years away from working fusion in 40-50 years... you can't put a time line on a technology until you actually get it working.

[Ziggy Stardust]

Your method of ask first, fail to understand the answer, complain when people tell you to do some research is frankly lazy. Plus, unless you have at least some background in a subject, the answer you get might not have any meaning to you, this happened repeatedly in the global warming thread where you admitted you had no knowledge and did no research, even as you continued to put the onus on us to educate you. We're not your teacher and this isn't a classroom, asking a question without knowing the first thing about the subject is both lazy and rude,

I disagree with this part, generally speaking. I think my response to this can be best summed by the Dude: "You're not wrong, you're just an asshole." Purple is far from the only person on these forums that asks honest questions about topics they don't understand. I am one, as well. I have frequently asked Broomstick, Kuroneko, Starglider, and others for details on subjects I barely understand and haven't extensively researched on my own. While it's true this isn't a classroom, it is a community that is open to discussion. While it's better to do some cursory research to make sure you aren't asking patently ridiculous questions, I don't think it's inherently wrong to ask people on these forums to break things down for you. Hell, there are times where I try to do the research, but without a proper understanding of the correct technical terminology used within a given field it is impossible for me to find useful introductory level information unless someone with familiarity answers a couple of simple questions for me first. This is basically what happens anytime there is a thread about theoretical physics around here. Now, that said...

doubly so when you then bitch and moan about the answers you get.

This is the part I DO agree with, and I think the crux of the reason people got mad at Purple in the global warming discussion.

Purple: there's nothing wrong with asking questions. I disagree with Jub about it being rude and improper. I think you are perfectly right to be asking questions. The problem is the perception people had in the global warming discussion that you seemed to either be fundamentally misinterpreting the answers or arguing that those answers were somehow insufficient, which led to the inevitable dogpile. That is, people weren't interpreting your questions as simply innocent attempts to expand your knowledge, but rather as some passive aggressive attempt to refute global warming or something. I admit, that's the impression I first got when reading your posts in that thread, though I now think that's the wrong way to interpret your posts.

In short: nobody (well, maybe Jub) begrudges you just for the act of asking the questions, but I think that the global warming thread involved some sort of miscommunication or disconnect tied to the WAY you were asking the questions and responding to the answers.