Some questions on fusion technolgy

[DerpenWolf]

Ok, so recently in order to avert boredom I've been reading stuff about fusion and ideas on how we might create a reactor and I have a few questions.

-How can we use energy created by a reactor?

-From what I understand it seems that toroidal fusion reactors would work by accelerating mass around in a circle and then oscillate the mass so that it interacts with itself, Is my understanding correct? Its just that it seems a bit weird.

-So the desktop sized hobbyist device known as a Fusor seems to have the main problem that is the fact that most of the deuterium hits the inner cage. If this problem could be averted then would this actually function as an effective reactor or would many other problems plague the techniques it utilizes?

-ICF fusion, The national ignition facility is pretty crazy and has had very good results, what are your thoughts on this?

-What do you suppose Lockheed Martin is up to?

-What are major advantages and disadvantages to the various methods one could use to induce fusion? What do you think would ultimately work out the best?

-If its possible, what do you think would be the best way to use controlled fusion technology to propel a spacecraft? (use power to run a big Magnetoplasmadynamic thruster? Direct ejection of exhaust? Etc....???)

Edited January 16, 2015 by DerpenWolf

[Bill Phil]

Power from a fusion reactor would work much like power from a fission one. Steam. That's right, when you look at all methods of producing power, it's mostly Steam power in the end.

Oh and:

Lockheed martian

[UmbralRaptor]

-How can we use energy created by a reactor?

A simple but silly way would be by boiling water.

-From what I understand it seems that toroidal fusion reactors would work by accelerating mass around in a circle and then oscillate the mass so that it interacts with itself, Is my understanding correct? Its just that it seems a bit weird.

My understanding is more of a compression than oscillation, at least at the macro scale. Though oscillations more or less make sense at the particle scale. (More or less because the density near the center is a bit high for that)

-So the desktop sized hobbyist device known as a Fusor seems to have the main problem that is the fact that most of the deuterium hits the inner cage. If this problem could be averted then would this actually function as an effective reactor or would many other problems plague the techniques it utilizes?

polywell designs are supposed to solve this. It seems like work on it has stopped, though

-ICF fusion, The national ignition facility is pretty crazy and has had very good results, what are your thoughts on this?

My limited understanding is that they've been able to get some of the shots to output enough neutrons that the pellet probably produced more energy than impacted it. It's no longer a complete failure, but the value is ambiguous, and all sorts of details of the design are orders of magnitude too slow and expensive for commercial power generation.

-What do you suppose Lockheed Martin is up to?

Smoke and mirrors.

-What are major advantages and disadvantages to the various methods one could use to induce fusion? What do you think would ultimately work out the best?

I'm not optimistic about any of them, but given that Polywell and NIF seem to be failing, I guess my hopes lie with ITER now. (Nevermind the half century of tokamaks failing to produce enough power)

-If its possible, what do you think would be the best way to use controlled fusion technology to propel a spacecraft? (use power to run a big Magnetoplasmadynamic thruster? Direct ejection of exhaust? Etc....???)

My sarcastic response would be one of the city sized interstellar Orions. A more serious one will have to wait for a reactor that can be declared a success.

[GreeningGalaxy]

ible, what do you think would be the best way to use controlled fusion technology to propel a spacecraft? (use power to run a big Magnetoplasmadynamic thruster? Direct ejection of exhaust? Etc....???)

There's a lot of different ways to use fusion power as a propulsion system. If you use the KSP Interstellar mod, it does a pretty good job of representing many of them.

The most basic one doesn't even need the fusion to produce more energy than you put into it - you just use a fission-electric powered z-pinch or laser inertial confinement system, get enough fusion to make a nice high exhaust velocity, and you're golden. You won't get much thrust at all since you're losing energy to the fusion reaction, but you'll get a respectable Isp since the individual fusion events will launch charged particles away from you at very high speed.

Assuming you figure out how to actually achieve ignition with ICF, that system would work substantially better. You could make what essentially amounts to a miniature, constantly running Orion drive - toss the fuel pellets out the back, strike them with powerful lasers, and then funnel the high-speed fusion products into an exhaust beam with magnetic fields. According to our good friend Project Rho, the theoretical limits of that kind of engine are quite substantial! (See "FUSE: IC-Fusion (MAX)" towards the bottom of the table).

The easiest method for a tokamak-like design is probably a NERVA-like thermal rocket, pumping a hydrogen propellant through the hot part of the reactor. Deuterium-tritium fusion puts out a lot of neutron radiation which isn't too hard to thermalize with a nice big slab of tungsten, so you won't have much trouble designing an engine like that once you have a working reactor. A thermal rocket would give you a pretty respectable thrust and Isp, but nothing like the direct-drive external inertial-confinement system would when properly refined.

You could also use the reactor to generate electricity through some type of heat engine or thermoelectric system and then use that electricity to run a plasma thruster, but that system is going to be inefficient since you're converting your energy so many times - today's Brayton-cycle thermal generators only get a few percent efficiency, and then you'll lose energy from your power distribution system, and then the plasma engines will make heat as they run too, and eventually the energy that actually shows up as useful work will be minuscule compared to what you made in the reactor, and you'll have to find a way of dissipating all the heat you lose, presumably with large and heavy radiators.

For the foreseeable future, I'm with UmbralRaptor and the fusion-heavy Orion concept. Sure, it's dirty and complex and involves a lot of wildly terrifying engineering, but it would get the job done - the theoretical maximum Isp would be around a million seconds. If you could figure out a good way of smoothing out the acceleration (the Medusa concept looks good), you'd be pretty much golden.

Edited January 16, 2015 by GreeningGalaxy

[Ralathon]

Ok, so recently in order to avert boredom I've been reading stuff about fusion and ideas on how we might create a reactor and I have a few questions.

-How can we use energy created by a reactor?

Many ways to do this. If your reactor produces charged particles you can directly convert those to current. The fusion energy also shows up as heat, so you can power a good ol' steam engine with it, like we do for coal, gas and fission plants.

-From what I understand it seems that toroidal fusion reactors would work by accelerating mass around in a circle and then oscillate the mass so that it interacts with itself, Is my understanding correct? Its just that it seems a bit weird.

It's not that weird if you think about it. Your plasma is about 150 million degrees, if it hits the wall it'll cool instantly and stop fusing. So you need to contain the stuff with electromagnetic fields. A torus is a good shape for doing so because it also gives you a convenient loop that you can use to carry a current. A current produces an magnetic field that causes the charge carriers of the current to tend towards the center. So if you have a torus of plasma and you force a current through it the plasma will automatically contract in on itself.

-So the desktop sized hobbyist device known as a Fusor seems to have the main problem that is the fact that most of the deuterium hits the inner cage. If this problem could be averted then would this actually function as an effective reactor or would many other problems plague the techniques it utilizes?

Yea it would, and there are some ideas to do so. But there hasn't been much research because the maths is a bit dubious. Most people think you can never get more juice out of the reactor than you have to pump in.

-ICF fusion, The national ignition facility is pretty crazy and has had very good results, what are your thoughts on this?

I doubt ICF is ever going to be useful except as a research tool. Yes, they had breakeven, but thats if they ignore the inefficiency of their lasers. If you take that into account they're far from breakeven. It is also quite slow so it won't have much power output.

-What do you suppose Lockheed Martin is up to?

If something sounds too good to be true, it probably is.

-If its possible, what do you think would be the best way to use controlled fusion technology to propel a spacecraft? (use power to run a big Magnetoplasmadynamic thruster? Direct ejection of exhaust? Etc....???)

Probably direct ejection of the exhaust. That'll give you the highest ISP with the least energy conversions (And thus efficiency losses). It also saves on heat radiators, high power energy generation in spaceships is a ..... because you cant get rid of the heat. But before we get to fusion powered spaceships (ignoring solar panels) we first need to get it running on earth. And once we get it commercially viable we'll have to wait a looooong time for knowledge to build up, technological advances to make fusion power lighter and power densities to go up.

[Streetwind]

To extrapolate on the "breakeven" success of the ICF lab: they basically succeeded in igniting a fusion event that produced enough energy to theoretically ignite itself again. The problems with that? It was a single fuel pellet. Since it underwent fusion, it was consumed. The energy produced might have been theoretically enough to set off another fuel pellet, but the test setup did not allow for another one to be present - nor did it allow for refueling in mid-operation, or for actually shaping the burst of energy in such a way that it would compress and ignite another fuel pellet.

It was purely a mathematical success - they produced a record-breaking but ultimately meaningless number.

Oh, and by the way: they didn't count the energy consumed by the lasers to trigger the fusion event. Which was many thousands (if not millions) of times higher than what was produced.

However!

Inertial confinement fusion does have another application, which addresses your spacecraft propulsion question. It works pretty much like Project Orion that keeps getting quoted here. It's a fusion pulse rocket that uses solar power to trigger magneto-inertial containment fusion events on fuel pellets that are released out the back one by one. The fusion products are expelled through a magnetic nozzle, also solar powered. The drive has an estimated, constant Isp of about 3000s, and a variable thrust that is limited more or less by how much solar power you can create (more power -> faster pulse frequency). It should produce noticably more thrust for the same power at the same Isp as current generation ion thrusters do.

In contrast to Project Orion, however, this thing is actually safe, sane and consensual (hyuk hyuk hyuk!) and under active development by an US company with grant money from NASA. In contrast to Project Orion, it does not require the establishment of factories dedicated to building weapons of mass destruction. In contrast to Project Orion, it uses low-radiation, pure aneutronic fusion explosions that eliminate the need for shielding the rest of the spacecraft and eliminate dangers to nearby objects (unless you point the nozzle at it. See: The Kzinti Lesson). The downside compared to Project Orion is that it won't work as well in the outer solar system, where it won't be able to efficiently run off of solar power.

The thing is currently projected to fly as a demonstration prototype in space by 2020. So I'm personally expecting it to actually fly as a demonstration prototype by 2025, if it doesn't run out of funding

Read more about it here: Magneto-Inertial Fusion Driven Rocket

EDIT: Video animation demonstrating the operation principle:

Edited January 16, 2015 by Streetwind

[78stonewobble]

I'm not optimistic about any of them, but given that Polywell and NIF seem to be failing, I guess my hopes lie with ITER now. (Nevermind the half century of tokamaks failing to produce enough power)

Sorry, for cutting your post down, but I just wanted to comment on this part.

To be fair Polywell and the NIF, has also failed to produce enough power. As far as I can see, in the case of atleast Polywell's and Tokamaks, the problem seems to be or is claimed to be a matter of scaling the devices up.

ITER is supposed to do that, though I question the wisdom of not going straight to DEMO and not testing out a large size Polywell.

[Ralathon]

Sorry, for cutting your post down, but I just wanted to comment on this part.

To be fair Polywell and the NIF, has also failed to produce enough power. As far as I can see, in the case of atleast Polywell's and Tokamaks, the problem seems to be or is claimed to be a matter of scaling the devices up.

ITER is supposed to do that, though I question the wisdom of not going straight to DEMO and not testing out a large size Polywell.

High energy plasma physics is a fiendishly difficult field and there are still a lot of holes in our knowledge. If we knew how to go straight to DEMO we would, but the science and engineering just isn't there yet.

[Nuke]

k so here are my answers

-either direct conversion or a thermodynamic cycle. former is better, latter is easy and well understood.

-toroidal reactors lol. they pretty much just confine plasma. you heat it up and then you cross your fingers and hope the particles slam into eachother. there is a tiny chance of this happening and if they fail to fuse then the particles go flying into the walls of the reactor. because how far the particles jump is known and how many failed fusions need to happen before a successful one happens is also known, you can figure out how big the reactor needs to be to get a fusion rate that makes it self sustaining. this is why iter is so frickin huge (in the world of fusion, bigger is better. need citation? well then go look at the sun).

-the fuser will never work because of energy lost to the grids. its a great tool for education because it explains how to make fusion and the problems you need to solve, its also cheap. people have built them in their garages. however the solution to the grid problem is the thing that the polywell reactor is trying to get around (its essentially the same thing, but the grid is replaced with a magnetically contained electron cloud).

-laser based ignition is limited by the efficiency of laser technology, which sadly is very poor. you will never make a laser based fusion reactor breakeven.

-fusion research of the classified well funded kind, because murica.

-making fusion is actually easy, its making things break even thats the problem. i dont really like the iter/demo plan, because the end result is a reactor than nobody can afford (where fission is cheaper). i kinda think the smaller scale reactors like polywell and dpf might be the ones to look at.

-if a small reactor is possible, then it will revolutionize spacecraft power supply. you would then have the juice to power some really high isp engines with thrust that is actually respectable. a big reactor on the other hand wouldnt do much for propulsion, but it might be useful for offworld colonies. there is also direct fusion propulsion which would also make space travel a lot more fun.

[78stonewobble]

High energy plasma physics is a fiendishly difficult field and there are still a lot of holes in our knowledge. If we knew how to go straight to DEMO we would, but the science and engineering just isn't there yet.

Oh yeah I realise that, but what I meant, was more to make sure that the "testbed" could scale further of the way or all the way. Even if there are science and engineering kinks, to work out along the way.

I just quickly looked up the lockheed martin thing and it's kinda the opposite thing they are doing. Keeping the testbed small to work out kinks.

However then they might run into scaling issues for the last demonstration. Like polywell's and NIF.

With ITER, "we" are allready spending billions to get further in the way of scaling, but still not the entire way.

[Nuke]

i found some pics of the lm fusion machine here in this article:

http://aviationweek.com/technology/skunk-works-reveals-compact-fusion-reactor-details

and here, this one has a video:

http://aviationweek.com/blog/high-hopes-can-compact-fusion-unlock-new-power-space-and-air-transport

and thread from talk polywell which has some real nerdtastic posts about said reactor.

http://www.talk-polywell.org/bb/viewtopic.php?f=10&t=5643

[KerikBalm]

I'm pretty pessimistic about any torus design.

I'm optimistic about the polywell, and optimistic about Dense Plasma Focus reactors ("Focus Fusion") which nobody here has commented on.

That last one also would be most easily adapted to spacecraft propulsion.

For pretty much all of them, they *should* work if you scale them up enough.

[Nuke]

the lm machine seems to employ a lot of lessons learned with polywells. the coils are of similar construction. cusps are closed up better. its inline configuration seems like it would make direct conversion easier (though it would need to burn something other than d-t to do this). the weird thing is the coil diameters are not uniform. you still have a quasi spheroid plasma ball at the center. what i like about their machine is that it looks really expensive, all the coils are mounted on rails and it seems like they are trying a lot of different coil configurations to see which one is better at hearding charged particles.

one thing i noticed about test shots in tokamaks is that the plasma doesn't look like its being controlled very well at all. its always drifting about to the bottom of the toroid or jumping around wildly. then you look at a polywell and you have your plasma in a nice little ball of awesome. dpf is a different beast in that it really lets the plasma do its own thing and doesn't try to control it at all (if anyone knows a technical source for dpf that isnt 98% greeny propaganda and that isnt behind a paywall* let me know).

you cant really do fast test cycles on a tokamak either. its always 'we need a bigger one. this multi billion dollar rig that took 10 years to build wasnt big enough, we need another one' as opposed to 'we need 200 million and six months to build a new polywell'. i think much of that 50 years away nonsense is construction time.

* e

Edited January 19, 2015 by Nuke

[KerikBalm]

http://link.springer.com/article/10.1007%2Fs10894-012-9547-z

http://link.springer.com/article/10.1007%2Fs10894-011-9385-4

http://scitation.aip.org/content/aip/journal/pop/19/3/10.1063/1.3694746

http://link.springer.com/article/10.1007%2Fs10894-012-9567-8

http://scitation.aip.org/content/aip/journal/pop/19/11/10.1063/1.4764894