They're Building a REAL Nuclear Fusion Reactor! - Holy S#!T

Linus Tech Tips ·Linus Tech Tips ·2018-05-06 · 2,020 words · ~10 min read
Floatplane YouTube

Transcript

JSON SRT VTT 188
0:00 so guys we are here on location at general fusion where they are attempting
0:05 to build the world's first commercially
0:09 viable nuclear fusion power plant and uh
0:14 it might not look like much but the only way that the technology
0:20 inside could be any more space age would be if i was actually shooting on the
0:24 moon today so why don't we go take a look
0:28 after i tell you about tunnelbear tunnelbear makes really simple easy to
0:33 use vpn applications so you can enjoy a more open internet check them out at
0:37 tunnelbear.com LTT
0:49 so what exactly is nuclear fusion well
0:54 on an atomic level basically you take a couple of hydrogen
0:59 atoms get them up to a really high temperature
1:03 and then you like smash them together
1:06 once they're fused you'll be left with a helium atom
1:10 oh i'm steven today so because one proton plus one proton equals two
1:15 protons and you'll get some energy or
1:18 rather you'll get a lot of energy like
1:21 like ten times what you'd get from a nuclear fission reaction like the ones
1:26 that power traditional nuclear plants so
1:30 that all sounds super cool already and we haven't even gotten to the best part
1:34 yet nuclear fusion works around the clock unlike solar or wind and it's 100
1:40 clean and safe the reaction is actually
1:44 incapable of causing a meltdown or other radioactive incident and as much as this
1:49 might look like some kind of a you know death ray the only direct
1:55 byproducts are helium like for party balloons and energy so
2:01 no depleted uranium seeping into the environment for thousands of years no
2:04 greenhouse gas emissions and get this the fuel for it is practically free you
2:10 input deuterium a hydrogen isotope that you can extract from ordinary seawater
2:15 with a centrifuge
2:19 oh that's awful
2:22 and lithium and some critics might point out that
2:27 these materials particularly lithium are not technically renewable but
2:32 at least for heavy water the reaction uses so little of it and there's enough
2:38 sitting around that by the time we run out the sun will have enveloped the
2:42 earth anyway so uh yeah i think we are probably good there
2:48 well okay then Linus if it's so great
2:51 why is fusion power taking so long ah i'm glad you asked
2:56 while the fundamental principles have been reasonably well understood for
3:00 decades the engineering and science required to build something practical in
3:05 the real world are well in scientific terms a total to get right in order for
3:11 nuclear fusion to occur between two hydrogen atoms they need to be heated up
3:15 to 150 million degrees celsius that's about
3:19 10 times hotter than the center of the sun and you've got to do that efficiently
3:25 no one's going to buy a fusion power plant that consumes more power than it
3:29 outputs duh so then how would you get something
3:34 that hot ah well
3:38 you could use a device called a talkamac to make a plasma donut and then
3:44 manipulate it with liquid helium cooled magnets
3:48 sounds promising you could take a uh see here
3:54 peppercorn sized piece of frozen deuterium and then
3:59 bombard it from all sides at exactly the same time with what is quite literally
4:04 the world's most powerful laser effectively blowing up the outside
4:10 compressing the inside enough to start the reaction and
4:14 both of those methods have actually succeeded at using nuclear fusion to get
4:19 close to producing more power than they consumed
4:23 but the problem is that via both of those
4:26 methods it is extremely difficult to actually harness the power to generate
4:32 electricity ah
4:36 i'm not gonna do that again general fusion's approach is kind of
4:40 like a hybrid of the other two
4:43 and it operates kind of like a super high tech diesel engine so first they
4:48 use one of these guys to create a 5 million degree ball of plasma
4:53 then they rapidly compress it with
4:56 steam-powered pistons the pressure from which
5:00 heats it to 150 million degrees
5:03 at that point the deuterium kind of ignites releasing energy
5:08 and if all this sounds difficult don't worry in real life it is way
5:14 harder this is general fusion's last generation
5:18 large-scale plasma injector where they prototyped their process of creating a
5:24 plasma of sufficient quality that it can survive for long enough for the pistons
5:30 in the compression chamber to fully compress it
5:33 so during a cycle these valves right up here will inject
5:38 just two milligrams of deuterium evenly
5:42 into the chamber which goes all the way around in a ring then
5:46 one million amps at 15 000 volts so this is
5:51 about the same as your average lightning bolt goes across the electrical feed
5:55 right here and strips the electrons off
5:59 the deuterium creating a ball of plasma
6:02 these particles then spin around the ball of plasma
6:06 wrapping it in its own magnetic field creating an insulating layer around the
6:10 plasma to give it a longer lifespan from there
6:14 these big electromagnets here will push the ball of plasma out the front of the
6:19 injector into the compression chamber which would be under this part right
6:24 over here now the rest of this guy which is actually
6:29 about the front two thirds is just for data collection
6:35 every test that they run generates about a Gigabyte of data and even after 150
6:41 000 experiments there is still work to do i guess that's why general fusion is
6:47 so open about showing off their work
6:50 because even if you stole all the drawings and
6:53 built yourself one of these machines
6:56 there are literally hundreds of thousands of variables to tune for
7:01 plasma quality so you'd basically have built yourself a
7:06 multi-million dollar paperweight on the subject of millions of dollars much of
7:11 the research is actually done on smaller
7:14 less expensive little injector
7:17 prototypes like spector here these guys
7:20 only take a couple of months to create instead of years and they give the team
7:26 valuable insight in between larger projects mind you even this kind of
7:31 thing does not come easy
7:34 every aspect of this research is complex
7:37 and expensive like okay here how do you know
7:42 if your plasma is at five million degrees
7:45 you can't just stick a k probe in it
7:49 ah so it turns out you have to shine a
7:52 super strong laser through the plasma
7:56 and then measure the small bit of light that is deflected by the plasma and this
8:01 is only about one in every quadrillion
8:04 photons that you pump in that get turned into some kind of useful signal so
8:09 interpreting that data is quite literally the full-time job of bill here
8:15 who has a phd from the university of maryland
8:18 and that is just one example but
8:23 they are making progress inside this cage is the largest and most
8:29 powerful plasma injector in the world
8:33 it's the latest generation of the one that we looked at earlier and they're
8:37 running tests on it every day in hopes
8:40 of using a similar model to create a fusion demonstration plant in as little
8:46 as four to five years so
8:49 unlike the other one this guy is actually hooked up do you see all these
8:54 containers around me these are capacitor
8:57 banks so when all of them are operating
9:01 they'll be capable of 20 gigawatts of power that's enough to send
9:06 16.5 deloreans back in time
9:10 simultaneously the problem though is that when you're
9:14 unleashing a lightning bolt worth of power you throw out a ton of electronic
9:20 interference it's such a big problem that all of the sensor data has to be
9:25 transmitted through fiber optic cables to a nearby faraday cage full of general
9:31 fusion's own custom-made digitizer boards
9:34 they ended up designing their own because creating an electronics
9:38 engineering department was cheaper than buying gear like this off the shelf now
9:43 general fusion weren't the first ones to come up with the idea of
9:48 smashing plasma but the us naval research lab's
9:53 incredibly named project Linus ended up
9:56 getting abandoned back in the 1970s because there were a couple of problems
10:01 they just couldn't overcome behind me
10:05 is a couple of solutions this is general fusion's piston driven
10:10 compression chamber proof of concept now the real one would have many more
10:16 pistons but the operating principle would be similar
10:20 every piston would push down simultaneously with a tolerance of about
10:26 10 microseconds injecting liquid metal
10:30 stored in the pistons and in the 1000
10:33 RPM spinning grid on the inside
10:36 and that metal would squeeze the plasma
10:39 bubble that would be injected from the top
10:42 now the timing for all this has to be so precise because much like squeezing a
10:48 water balloon in your hands if you don't compress it evenly then it'll kind of
10:53 splurt out or break apart and that would
10:56 cause a misfire assuming everything goes right though and the simulations say it
11:00 should the fuel ignites fusion occurs
11:04 and that liquid metal gets heated to about 600 degrees celsius after which it
11:10 gets circulated through insulated pipes by a magnetohydrodynamic pump to a
11:16 traditional steam turbine power generator just like what you would find
11:20 at a coal or a natural gas power plant today which should make general fusion
11:25 solution sort of a drop-in upgrade for an
11:30 existing facility now
11:33 fun fact it is not a perpetual motion machine
11:37 because it does require a constant supply of fuel
11:41 but some of the steam that gets generated is actually going to be used
11:46 to actuate the pistons keeping the cycle
11:49 going general fusion hopes that once they
11:52 reach production they'll be able to run one of these reactions every second
11:57 generating about 100 to 200 megawatts of
12:01 power about the sweet spot for replacing existing coal power plants in developed
12:06 countries and not so much that it would
12:10 blow up the power distribution systems in developing countries a market where
12:15 the most new power generation will have to happen in the coming years now none
12:19 of this is a sure thing i mean nothing in life is but
12:23 if general fusion and its investors
12:26 which includes jeff bezos apparently by the way can stay on course
12:31 then the future looks really bright not just for their company but also for the
12:36 world speaking of companies with bright
12:40 futures are you finding as a freelancer or small business owner that the thing
12:44 holding you back is spending your time on complicated accounting software
12:48 instead of just doing your work well freshbooks is the solution you're
12:53 looking for it's a simple way to be more productive more organized and get paid
12:57 faster you can create and send professional looking invoices in less
13:00 than 30 seconds you can set up online payments with just a couple of clicks to
13:04 get paid up to four days faster you can see when your client has seen your
13:08 invoice and put an end to those guessing games and you can take the whole
13:12 experience with you on the go tracking your hours and expenses with their iOS
13:16 and Android apps so for an unrestricted 30-day free trial go to freshbooks.com
13:21 tech tips and enter Linus tech tips in the how did you hear about us section
13:25 we'll have that linked below so thanks for watching guys if this
13:29 video sucked you know what to do but if it was awesome get subscribed hit that
13:34 like button or check out the link to where to buy the stuff we've featured if
13:38 you are a large government and you're watching this
13:42 5 to 15 years in the future whatever don't worry about anyway also linked in
13:46 the description is our merch tour which has cool shirts like this one and our
13:49 community forum which you should totally join