Among the US fusion community, many researchers were worried that the US's stake in ITER would eat up the entire fusion research budget, leaving nothing for domestic programs.
NIF's main purpose, really, is weapons stewardship. France is building (built?) a similar facility too.
The thing about the X-Prize: governments have already gotten to space, and private companies are just replicating what was done in the 1960s. Fusion: governments haven't gotten a viable reactor. The technology simply isn't there yet.
My school also has an EECS department (Berkeley), but I would argue that it makes sense to combine the department. You can't run software without hardware.
Yeah, the Sun stuff is amusingly out of date. Should note that most of the EE computer labs are Windows machines, though they're locked up in terms of account permissions. Also, the inst.eecs webspace gets ddos-ed every so often.
I've never bothered to try a non-EECS domain computer... maybe I should.
I had to buy a signals & systems book (Oppenheim). I didn't want to pay $120 for a new book or slightly less for a used one, so I got an international edition off eBay for $20ish. It's actually pretty good quality. But the kicker is the list price, which the seller of my book covered with nail polish. I scratched it off to find "300 Rs." Currently, $1 USD = 45 Rs.
I agree totally. I built an AM/FM radio kit a long time ago, the Elenco one. The manual assumed that I had a bunch of test equipment (oscilloscope etc) handy, and gave vague explanations about the circuits.
I'm now in my third year of EE, and I still don't have the knowledge to fully analyze the circuits in the kit. Just last week I finally found a book that talks about discrete radio design, and it's an advanced undergrad/grad text! So some topics definitely require a serious amount of education.
I've never heard of deuterium fission. At the low end of the periodic table, it's far more energetically favorable for thing to fuse. I'm no expert, but D fission might even be endothermic.
You can think of their experiment like the classic Rutherford experiment, except they've got D+ ions being shot at a sheet of D. The two D+D fusion reactions happen with equal probability:
D + D -> T (1 MeV) + p (3 MeV) D + D -> He3 (0.8 MeV) + n (2.5 MeV)
What they did in the experiment was to look for 2.5 MeV neutrons, because that reaction will _always_ produce a 2.5 MeV neutron. They also looked at associated X-rays. (If it was a fission neutron, it would probably be a different energy, and again, I can't find any reference to such a phenomenon.) Then they correlated their results to a computer simulation. I don't know why they didn't bother to look for the H and T. They may not have had the equipment, or they considered it outside the scope of the experiment's purpose as a neutron source.
The caveat to this experiment is that neutron and x-ray detection is something of an art, and must be done correctly; I'm not qualified to comment on their setup. This experiment makes sense according to normal physics, though.
People have indeed looked into fusion/fission hybrids. Unfortunately, they seem to compound the dangers of fission with the (current) impracticality of fusion.
We _can_ burn our waste with breeder reactors, using a different fuel cycle. The problem there is that these particular fuels are the kinds you use in nukes, so politically (and security-wise) it's not practical.
There are lots of relativistic particles in cosmic rays, particle accelerators, etc, that approach c. Do two colliding particles at the Tevatron repulse each other? Or do the anti-gravities cancel out? You'd think that somebody would have noticed something like this.
They indeed process it on the fly, which is the point of the trigger: it "triggers" on the most interesting events, which get stored. Even after triggering, the data set is massive.
I'll add that according to current theory, extremely high energy cosmic rays create mini black holes all the time. We've seen these extreme cosmic rays, and we haven't been swallowed up yet, so it's reasonable to conclude that the danger is nonexistent.
Well, here at Berkeley, you get free Microsoft OSes and dev programs if you're an EECS major. I don't think very many people take advantage of it (probably mostly the CS people), but MS definitely has an ulterior motive in pushing their software.
"Sometimes it's over here, sometimes it's over there. In a thousand measurements the average is () and the variation is (), so we can say with x% confidence that it's between () and ()."
And as for the free content for UC Berkeley courses, we have only 100-level (or lower) classes which are basically prerequisites for a UC Berkeley education. I'm sorry to say that if you were looking for course content, you'll need to look elsewhere.
Somewhere like http://inst.eecs.berkeley.edu/classes-eecs.html. Not all classes are active, but there's archives. Berkeley doesn't have an official OpenCourseWhatever system, so you have to look around a bit.
My fusion professor had an essay question that went like this: The devil comes to see you, offering a choice of three things in exchange for your soul: a room-temperature superconductor, a full understanding of plasma transport phenomena, and a resilient first wall material. Which do you choose and why?
The concept of the question is that any one of those three things would give us a viable fusion reactor. They're all technical issues.
1) Superconducting magnets. Tokamaks rely on SC magnets to create their strong fields. With current technology, these magnets have to be cooled with liquid helium. One obvious problem is that a coolant failure would be bad, and it's expensive to keep a cold thing cold while it's next to a very hot thing. Another is that the magnet coils are in a position where they experience a lot of neutrons, damaging the material. So why would a room temperature superconductor be good? It would eliminate the cost of coolant and alleviate concerns about coolant failure, for one. Removing all the cooling facilities would also allow the tokamak to be shaped for better efficiency.
2) Plasma transport. This refers to heat conduction through a plasma. It surprised me to learn that we still do not understand plasma transport. The calculated heat flux out of the plasma via electron motion is off by orders of magnitude. And of course, the error goes the 'wrong way'; we calculate much less heat loss than actually occurs, and we want to keep heat in the plasma to maintain the reaction. Furthermore, we don't fully understand plasma instabilities, as confined plasma likes to do all sorts of wacky things.
3) The first wall. The 'first wall' is the wall right next to the plasma. In DT fusion (the only variety considered to be commercially viable) you produce 'fast neutrons' which really mess up most materials. The canonical example is stainless steel, which will deteriorate rather badly and produce Cobalt-60 in the bargain. So there has been interest in more exotic materials like Vanadium and Molybdenum. The problem here is that at fusion temperatures, materials tend to bleed off particles, and anything that's not D or T in the reactor poisons the reaction and reduces the yield significantly.
Point one: tokamaks run their plasmas at about 1 millionth of atmospheric density; the rule of thumb is 10^20 particles/meter^3. This means the plasma is in a vacuum vessel.
Point two: for DT fusion, you've always got neutrons coming from the reactions. And they're fast neutrons, which means they'll react with the Nickel in stainless steel to form Cobalt-60, which is a gamma emitter. But that's stuck in the wall, and you'd want to use a different material for your walls anyway.
Point three: if magnetic containment fails and the plasma hits the wall, the plasma just dumps its thermal energy into the wall, and fusion can no longer be sustained. This happens in experiments all the time, though they try to avoid it. At worst, this could rupture the wall.
Point four: I haven't studied this in detail, but if the wall ruptures, then there will be air sucked _into_ the reactor to equalize pressure. In a real plant design, you'd probably have separate air circulation for this region of the plant, but for disaster analysis you'd assume a small amount of what's inside the reactor gets outside into the world. The only radioactive stuff would be tritium, which is relatively harmless, but still a problem.
So if a fusion reactor fails, nothing catastrophic happens. You need extreme extreme density to have an H-bomb. This is what they do with in Inertial Confinement Fusion, compact DT ice with lasers. I don't have my notes right now, but that resultant density is a whole fricking lot more than 10^20 per meter^3.
I'll add that the fusion plasma density in a tokamak is very low -- about 10^20 particles/m^3, or 1 millionth the density of atmospheric pressure air. In other words, it's in a vacuum vessel, and will implode, not explode.
The magnetic field here is _not_ for squeezing the plasma, it's for controlling it, and even strong fields don't do a particularly good job of it. A greater danger is the energy contained in the superconducting magnets. Mega-amps or more are going through those things, and there's also a significant physical tension due to the magnetic field. Coolant loss or structural failure would be bad.
As long as the fusion is happening within water, you'll deposit some energy inside the core. But let's say you've found a way to minimize that, and you've got a surrounding chamber that can go superhot. A few issues to contend with are
1) Heat flux. How do you ensure that the sono chamber stays cool while right next to a much hotter system?
2) Neutronics/materials. The wall(s) keeping the systems from mixing are going to see a whole lot of fast neutrons, which is a big problem. You have to pick a material that holds up well under fast neutron flux + heat.
These two issues are, ironically/unsurprisingly, two of the issues "conventional" magnetic fusion faces. In such a device, you've got vacuum pumps that run at cryogenic temperatures (1), and a so-called 'first wall' that sees a whole lotta neutrons over its operating lifetime (2). Needless to say, we don't have good solutions to these problems yet.
Slashdot Mirror
Among the US fusion community, many researchers were worried that the US's stake in ITER would eat up the entire fusion research budget, leaving nothing for domestic programs.
NIF's main purpose, really, is weapons stewardship. France is building (built?) a similar facility too.
The thing about the X-Prize: governments have already gotten to space, and private companies are just replicating what was done in the 1960s. Fusion: governments haven't gotten a viable reactor. The technology simply isn't there yet.
People have looked into this. It doesn't seem very promising for energy generation.
I took a fusion class, and I seem to remember my professor saying that even V has problems, due to sputtering -> impurities -> bremsstraluhng.
I have to agree. Even in the original thread at SA, plenty of people have no idea of how much thermal paste is enough.
My school also has an EECS department (Berkeley), but I would argue that it makes sense to combine the department. You can't run software without hardware.
Yeah, the Sun stuff is amusingly out of date. Should note that most of the EE computer labs are Windows machines, though they're locked up in terms of account permissions. Also, the inst.eecs webspace gets ddos-ed every so often.
I've never bothered to try a non-EECS domain computer... maybe I should.
I had to buy a signals & systems book (Oppenheim). I didn't want to pay $120 for a new book or slightly less for a used one, so I got an international edition off eBay for $20ish. It's actually pretty good quality. But the kicker is the list price, which the seller of my book covered with nail polish. I scratched it off to find "300 Rs." Currently, $1 USD = 45 Rs.
I agree totally. I built an AM/FM radio kit a long time ago, the Elenco one. The manual assumed that I had a bunch of test equipment (oscilloscope etc) handy, and gave vague explanations about the circuits.
I'm now in my third year of EE, and I still don't have the knowledge to fully analyze the circuits in the kit. Just last week I finally found a book that talks about discrete radio design, and it's an advanced undergrad/grad text! So some topics definitely require a serious amount of education.
I've never heard of deuterium fission. At the low end of the periodic table, it's far more energetically favorable for thing to fuse. I'm no expert, but D fission might even be endothermic.
You can think of their experiment like the classic Rutherford experiment, except they've got D+ ions being shot at a sheet of D. The two D+D fusion reactions happen with equal probability:
D + D -> T (1 MeV) + p (3 MeV)
D + D -> He3 (0.8 MeV) + n (2.5 MeV)
What they did in the experiment was to look for 2.5 MeV neutrons, because that reaction will _always_ produce a 2.5 MeV neutron. They also looked at associated X-rays. (If it was a fission neutron, it would probably be a different energy, and again, I can't find any reference to such a phenomenon.) Then they correlated their results to a computer simulation. I don't know why they didn't bother to look for the H and T. They may not have had the equipment, or they considered it outside the scope of the experiment's purpose as a neutron source.
The caveat to this experiment is that neutron and x-ray detection is something of an art, and must be done correctly; I'm not qualified to comment on their setup. This experiment makes sense according to normal physics, though.
People have indeed looked into fusion/fission hybrids. Unfortunately, they seem to compound the dangers of fission with the (current) impracticality of fusion.
We _can_ burn our waste with breeder reactors, using a different fuel cycle. The problem there is that these particular fuels are the kinds you use in nukes, so politically (and security-wise) it's not practical.
There are lots of relativistic particles in cosmic rays, particle accelerators, etc, that approach c. Do two colliding particles at the Tevatron repulse each other? Or do the anti-gravities cancel out? You'd think that somebody would have noticed something like this.
He didn't say just how he was ranking them. He might be ranking them in terms of suckiness, which is why he's going to buy a Mac. /ducks
> Chick Publications
Those things are funny. And scary.
They indeed process it on the fly, which is the point of the trigger: it "triggers" on the most interesting events, which get stored. Even after triggering, the data set is massive.
I'll add that according to current theory, extremely high energy cosmic rays create mini black holes all the time. We've seen these extreme cosmic rays, and we haven't been swallowed up yet, so it's reasonable to conclude that the danger is nonexistent.
Heh, I'll raise you one. I'm half your age and prefer the radio.
Well, here at Berkeley, you get free Microsoft OSes and dev programs if you're an EECS major. I don't think very many people take advantage of it (probably mostly the CS people), but MS definitely has an ulterior motive in pushing their software.
No, that's your mind on drugs.
A quantum physicist would say:
"Sometimes it's over here, sometimes it's over there. In a thousand measurements the average is () and the variation is (), so we can say with x% confidence that it's between () and ()."
Yeah, it's not funny. Sorry.
And as for the free content for UC Berkeley courses, we have only 100-level (or lower) classes which are basically prerequisites for a UC Berkeley education. I'm sorry to say that if you were looking for course content, you'll need to look elsewhere.
Somewhere like http://inst.eecs.berkeley.edu/classes-eecs.html. Not all classes are active, but there's archives. Berkeley doesn't have an official OpenCourseWhatever system, so you have to look around a bit.
I didn't RTFA, but there's a book on stress called "Why Zebras Don't Get Ulcers". I wouldn't be surprised if the author was the lecturer.
My fusion professor had an essay question that went like this: The devil comes to see you, offering a choice of three things in exchange for your soul: a room-temperature superconductor, a full understanding of plasma transport phenomena, and a resilient first wall material. Which do you choose and why?
The concept of the question is that any one of those three things would give us a viable fusion reactor. They're all technical issues.
1) Superconducting magnets. Tokamaks rely on SC magnets to create their strong fields. With current technology, these magnets have to be cooled with liquid helium. One obvious problem is that a coolant failure would be bad, and it's expensive to keep a cold thing cold while it's next to a very hot thing. Another is that the magnet coils are in a position where they experience a lot of neutrons, damaging the material. So why would a room temperature superconductor be good? It would eliminate the cost of coolant and alleviate concerns about coolant failure, for one. Removing all the cooling facilities would also allow the tokamak to be shaped for better efficiency.
2) Plasma transport. This refers to heat conduction through a plasma. It surprised me to learn that we still do not understand plasma transport. The calculated heat flux out of the plasma via electron motion is off by orders of magnitude. And of course, the error goes the 'wrong way'; we calculate much less heat loss than actually occurs, and we want to keep heat in the plasma to maintain the reaction. Furthermore, we don't fully understand plasma instabilities, as confined plasma likes to do all sorts of wacky things.
3) The first wall. The 'first wall' is the wall right next to the plasma. In DT fusion (the only variety considered to be commercially viable) you produce 'fast neutrons' which really mess up most materials. The canonical example is stainless steel, which will deteriorate rather badly and produce Cobalt-60 in the bargain. So there has been interest in more exotic materials like Vanadium and Molybdenum. The problem here is that at fusion temperatures, materials tend to bleed off particles, and anything that's not D or T in the reactor poisons the reaction and reduces the yield significantly.
Mm, I'm late to this discussion, but...
Point one: tokamaks run their plasmas at about 1 millionth of atmospheric density; the rule of thumb is 10^20 particles/meter^3. This means the plasma is in a vacuum vessel.
Point two: for DT fusion, you've always got neutrons coming from the reactions. And they're fast neutrons, which means they'll react with the Nickel in stainless steel to form Cobalt-60, which is a gamma emitter. But that's stuck in the wall, and you'd want to use a different material for your walls anyway.
Point three: if magnetic containment fails and the plasma hits the wall, the plasma just dumps its thermal energy into the wall, and fusion can no longer be sustained. This happens in experiments all the time, though they try to avoid it. At worst, this could rupture the wall.
Point four: I haven't studied this in detail, but if the wall ruptures, then there will be air sucked _into_ the reactor to equalize pressure. In a real plant design, you'd probably have separate air circulation for this region of the plant, but for disaster analysis you'd assume a small amount of what's inside the reactor gets outside into the world. The only radioactive stuff would be tritium, which is relatively harmless, but still a problem.
So if a fusion reactor fails, nothing catastrophic happens. You need extreme extreme density to have an H-bomb. This is what they do with in Inertial Confinement Fusion, compact DT ice with lasers. I don't have my notes right now, but that resultant density is a whole fricking lot more than 10^20 per meter^3.
I'll add that the fusion plasma density in a tokamak is very low -- about 10^20 particles/m^3, or 1 millionth the density of atmospheric pressure air. In other words, it's in a vacuum vessel, and will implode, not explode.
The magnetic field here is _not_ for squeezing the plasma, it's for controlling it, and even strong fields don't do a particularly good job of it. A greater danger is the energy contained in the superconducting magnets. Mega-amps or more are going through those things, and there's also a significant physical tension due to the magnetic field. Coolant loss or structural failure would be bad.
As long as the fusion is happening within water, you'll deposit some energy inside the core. But let's say you've found a way to minimize that, and you've got a surrounding chamber that can go superhot. A few issues to contend with are
1) Heat flux. How do you ensure that the sono chamber stays cool while right next to a much hotter system?
2) Neutronics/materials. The wall(s) keeping the systems from mixing are going to see a whole lot of fast neutrons, which is a big problem. You have to pick a material that holds up well under fast neutron flux + heat.
These two issues are, ironically/unsurprisingly, two of the issues "conventional" magnetic fusion faces. In such a device, you've got vacuum pumps that run at cryogenic temperatures (1), and a so-called 'first wall' that sees a whole lotta neutrons over its operating lifetime (2). Needless to say, we don't have good solutions to these problems yet.