Germany is turning on its monster stellarator
businessinsider.my
businessinsider.my
Im Zeitraffer: Zusammenbau von Wendelstein 7-X: https://www.youtube.com/watch?v=MJpSrqitSMQ (best @ 0.56x)
Previously: https://news.ycombinator.com/item?id=10437075
More: https://www.youtube.com/channel/UCAx243-m51RI-k_4lsAqidA/fee...
Kudos to these guys for sticking at it long enough and persevering to achieve their vision.
- *massive* success of coordination and cooperation
- illustration of potential humanity has
- that careful and methodical approach can make even such monsters of recursive complexity work
[Your level of surprise may vary]
Refactor it, preferably yesterday - but make a "shadow deploy". For each set of input / output parameters run both functions and compare them. If they differ, use the old one and go back to figuring out where it went wrong. After some time you can be quite sure that the new clean function works as it should - then you just switch.
http://www.amazon.co.uk/Working-Effectively-Legacy-Michael-F...
More seriously though, the kind of code you described will be what I imagine the first subject of study of the software archeology field in the future.
I've been working at the system to free it from that subsystem with an eventual replacement. There are all kinds of single-letter variables and Russian words, refactoring wouldn't be economical or sane.
can anyone speak to sort of considerations that lead to this value? 2.5T seems insignificant considering the events it will be confining
how sensitive is the field to magnetic interactions outside of the device?
(o) https://youtu.be/lyqt6u5_sHA?t=301 (i) https://en.wikipedia.org/wiki/Tesla_(unit)#Examples
This technique is mostly used in solid state and, surprise surprise, fusion experiments. Oh, and you can make an EM bomb if you make a big enough one, and wipe out a city's electronic infrastructure in a microsecond. https://en.wikipedia.org/wiki/Explosively_pumped_flux_compre...
And considering these devices are heavily shielded, I would doubt there would be much interaction outside.
EDIT: On reading further, it turns out I may have misinterpreted the article.
Today the numerical effort required for realizing it might seem not very impressive given the technological capabilities that we have, but you have to take into account that the first version (Wendelstein 7-AS) was built in the 1980s already.
That said I really hope that the experiment confirms the viability of this design and its (potential) superiority compared to the classical Tokamak fusion chamber.
True but that is largely true of the first of anything complex and expensive since it's Development Costs + Construction costs / 1 until they build another.
The issue is that you can not turn up the current more than a maximum level, and you have to keep the direction the same (so you can’t switch between turning off and on).
Stellerators instead solve this with complex geometry and the plasma creating its own continously changing magnetic field.
Do you have a citation for this claim?
You might be able to find some sources when you search for the reasons of Tokamaks pulsed usage.
If I misunderstand it disregard my comment entirely, although I'd like to know why.
- The simplest closed magnetic configuration is a torus. You can't smoothly "comb a hairy ball", but you can comb a hairy torus.
- For a toroidal plasma to be stable, it needs magnetic field components both the "long way" around (toroidal), and the "short way" around (poloidal).
- The tokamak generates the poloidal field from a current within the plasma, whereas the stellerator imposes that component with external magnets.
So you could make a "natural" argument either way, and I was asking the grandparent in what sense the stellerator is more "natural".
Edit: to add some additional context, the main problem with fusion reactors is confinement time and plasma stability. The general idea of a fusion power plant using magnetic confinement is that a gas is turned into plasma, confined magnetically, and heated until it can undergo significant fusion reactions. There are two core problems to creating fusion plasmas that could potentially produce power. One is pumping the plasma up to conditions that enable self-sustaining fusion by maintaining a high enough temperature from balancing the heat losses in the plasma against the heat gains from fusion reactions. The other is producing more energy from fusion reactions than was put into heating and containing the plasma. The longer the fusion plasma is contained the longer it has to continue fusion reactions and producing energy.
Tokamaks are easy to build but have a fatal flaw in that they have high plasma current, which makes them difficult to control with very unstable plasmas. Stellarators are vastly more difficult to build but they don't rely on currents in the plasma itself, which offers the potential for vastly longer confinement times.
Wendelstein 7-X is small early experimental reactor for testing stellator geometry for reactor. Research questions are similar to where JET (the largest working tokamak today) was 30 years ago. Stellator is more complex design but promises inherently stable containment.
ITER is solving different questions. ITER solves engineering problems for full-scale electricity-producing reactor that would be practical to build and operate. JET studied containment in Tokamaks. ITER is the last step before first industrial fusion reactor design DEMO (Demonstration Power Plant).
There is also lots of technical overlap of course. Vacuum vessels, cooling, magnets, breeder blankets, etc. ITER is developing these for industrial-scale reactor, even for stellator.
Um, what? I'm about to pass a nuclear fission plant on my way to work, and I surely hope there are no plasmas in that.
AKA, they mean fusion reactor.
They're not disputing the nuclear term, but fission vs. fusion.
Also, hello from Indian Point, Buchanan, NY! I can see a containment dome as I write this.
Next up on my Snake Plisskin-esque commute is literally going through Sing Sing prison: https://www.google.com/maps/@41.1506265,-73.8700067,613m/dat... (those train tracks are the Metro North Hudson line, among other things)
What kind of association does that make in people's minds, they already think nuclear = death...
Assuming it works as intended, the next step would be a research reactor that can, that would be a pilot project to generate net energy which could be connected to the grid (probably more for symbolic reasons at that point…) and breed tritium for further reactors.
Then we can start to plan building actual power plants. So, few decades, as always.
The current schedule is to have an operational power plant in 2050.
30-50 years is the respectable man's "I have no idea" estimate for anything.
Personally, I'd bet on other technologies being more cost effective for power generation in the next 20 years.
Also, my guess is neither of these "traditional" approaches will work to create actual fusion power. My guess is someone else's out of left field approach that will totally shock the scientific community will be the winner.
http://lawrencevilleplasmaphysics.com/ (Focus fusion)
http://www.lockheedmartin.com/us/products/compact-fusion.htm...
Helion is similar to Tri-Alpha but with a pulsed design instead of steady state. They have funding from YCombinator. Instead of boron they're shooting for a hybrid D-D/D-He3 fuel cycle (the D-D reaction produces He3). They say only 6% of the energy released would be as neutron radiation.
Sandia's MagLIF project is interesting. According to their computer simulations they can get 100x to 1000x energy gain by preheating a fuel capsule and crushing it in an upgraded Z-Machine. Things are looking good so far.
UW has the Dynomak project, which is fairly similar to tokamak but a lot smaller and cheaper. They need $10 million to test the concept.
http://www.technologyreview.com/article/541996/a-workable-fu...
Right now, almost all research is directed towards the easiest possible fuel that requires least powerful containment, that is, Tritium (H3) and Deuterium (H2). Deuterium can be separated from water, and Tritium can be bred from Lithium using waste neutrons.
T+D is inexhaustible, but also quite expensive as both separation and neutron breeding are relatively expensive. However, as a fusion reactor consumes tiny amounts of fuel per unit of energy, a T+D reactor could potentially still run profitably.
If we get better at containing plasma, other fuel choices become available. Most important being Boron + Hydrogen, which would be inexhaustible, cheap, and produces minimal neutrons during fusion, making it easier on the equipment.
However, free fuel doesn't mean free energy. The cost of all nuclear power, including current fission power, is utterly dominated by the capital cost of equipment.
As in, you want to partially disassemble and tweak the reactor regularly. Neutron activated metals will be radioactive enough after a while, no need to make things worse by adding tritium to the mix.
"The purpose of Wendelstein 7-X is to evaluate the main components of a future fusion reactor built using stellarator technology"
So I'm guessing it can't produce power as it is.
"Approval to go ahead is expected from Germany's nuclear regulators by the end of this month"
No, seriously. All these statements about German character or German mentality should stop. If the last few months haven't taught us anything else, at least it should have taught us that. First, Germans and the German government were praised as reliable in connection with the Grexit (as opposed to the Greeks), then a few months later, a German government owned car producers is caught in what might be the biggest corporate fraud ever.
I wonder if mentality and national traits even exist.
CEO resigned, 10 board members suspended, all non-management employees (engineers, programmers, etc) have full immunity, even if they worked on the defeat device.
That’s the most honorable reaction one could expect, and it allows to get rid of the issue.
I'm not trying to pick a fight, but how is that in any way honorable?
The former creates no incentive to be a whistleblower - it just takes away one of the risks of doing so. The latter creates a strong incentive to be the first to report something.
People should be encouraged to come forward _before_ they do something illegal.
If you reward lower level employees for doing it after the fact, the rational thing for them to do is to continue acting illegally, as if the thing blows up they will be bailed out, and if they don't they might still suffer the consequences of it (i.e. get in bad relations with their boss).
I am not saying there is the same level of responsibility, but there should be _some_.
http://www.reuters.com/article/2015/10/08/volkswagen-emissio...
Those "engineers" were the leading engineer of R&D and the leading engineer of Engine Design. Both board members.
Sometimes, but more often they're more propaganda than truth.
Or at least its responsible investment behavior by its government.
USA would not build such a thing in today's political climate of nonsense unless it had weapons research possibilities.
Just because the US invests rarely in civilian research doesn’t mean they never do (compared to military research; compared to other countries they still invest a lot in civilian research)
Congress won't fund a civilian space shuttle anymore but we spend billions for the military to have their own.
Like I said, if it's not about killing people, destroying things and spying on the world (including our own people) we pretty much won't spend money on it anymore here.
Germany did the investment beyond war.
Deleted comment
Besides betraying a rather shallow understanding of history (e.g. implying that "round 1" had the same causes as "round 2"), I do think at this point an American accusing Germans of fascist tendencies and war-mongering is pretty rich, what with the drone wars and the global surveillance and all.
I can assure you Germany is not out of trash.
Germany is still years away from 100% renewable. But is actually making progress to there. There's a lot of coal still dug up and burnt for example.
Germany has had lots of immigrants for decades, and things have been mostly peaceful.
The current leader (Angela Merkel) is considered to be from the right, leading a conservative party.
The democracy is one of the most stable around, and many people are anti war.
There are serious issues with the German economy, and the state of the country in general. It's not highly ranked internationally for things like GDP or internet speed.
I agree that internet should be much better than it currently is though, yet still far from bad. Australias internet made me really sad when lived there... maybe everything feels good in comparison to that.
And with them saying things like "Yes, there is a problem with the refugees from Syria, sadly, the concentration camps are currently closed" it’s getting a bit scary.
It does seem much of that is a protest against the policies of CDU. These people would rather immigrants freeze it seems.
If an election was called now, then a left coalition would probably get in, since CDU and AfD are unlikely to team up.
But the right wing terror is getting scary, with dozens of asylum homes being burnt down just this year.
Not that the single trite explanation could address the whole of their success, but certainly could account for a 'bump'