Hacker News Reader

Comment by lutusp | Hacker News Reader

Something is rotten in Denmark. If this project achieves a sustained fusion reaction as claimed, that represents a huge breakthrough, and not just for space propulsion, but for the general topic of fusion power.

If it succeeds, it would sweep away all present fusion reaction prototype methods, including the well-tested but so far ineffective( * ) tokamak and laser-fusion approaches.

* By "ineffective" I mean none of them has reached the break-even point, that point where more energy is released than is required to initiate the reaction in the first place.

The article doesn't say whether the prototype device has actually succeeded in igniting a sustained fusion reaction. If it does, it would quickly move beyond its present goal of producing a more effective source of space acceleration and would answer some longstanding questions about fusion power itself.

The fact that this isn't being discussed leads to my statement above -- something is rotten in Denmark. Either the project is overselling its possibilities, and/or it can't really achieve fusion break-even.

Speaking hypothetically, if the device could produce a sustained fusion reaction with substantial power, it could be scaled up and used to propel a spacecraft to Mars in much less than 30 days. Assuming a sustained acceleration of 1 g, the hypothetical craft could accelerate for 1/2 the trip, turn around and decelerate for the other 1/2 of the distance, arriving at Mars with zero velocity. Apart from minimizing travel time, this hypothetical profile would prevent the bone loss that accompanies sustained time at zero-g.

Making the above assumptions, and assuming that Mars is at a close approach point in its orbit, the travel time could be as little as ... wait for it .. 50 hours.

Derivation:

1. Distance d (meters) for acceleration a (m/s^2) and time t (seconds): d = 1/2 a t^2

2. Time t for distance d and acceleration a, assuming 1/2 acceleration and 1/2 deceleration: t = 2 sqrt(d/a)

3. Result for Mars close approach (7.834e10 meters) and acceleration of 1 g: 49.64 hours.

Again, speaking very hypothetically. I still think something is rotten in Denmark.

View on HN

> Something is rotten in Denmark. If this project achieves a sustained fusion reaction as claimed, that represents a huge breakthrough, and not just for space propulsion, but for the general topic of fusion power.

Achieving fusion in a tokamak or with lasers has been a solved problem for a couple of decades.

The problem with fusion power is generating more energy from the fusion than consumed to power the lasers or tokamak.

There is no such problem here as I understand it; the system isn't supposed to sustain itself from fusion, but instead will rely on solar panels or some other energy source to power the fusion reactor.

> Achieving fusion in a tokamak or with lasers has been a solved problem for a couple of decades.

I should have been more clear -- I mean sustained fusion generation, meaning a net energy gain over that required to start the reaction in the first place. That hasn't been achieved.

> There is no such problem here as I understand it; the system isn't supposed to sustain itself from fusion ...

Not according to the NASA documents (see below). If that were true, there would be no point in using the system. If the fusion reaction produces less power than it requires, the designers would be better off using the source electrical power to drive an ion thruster.

The claim being made is that the fusion scheme creates more power than is required to start it -- by using electrical power to initiate a fusion reaction that produces more power than it requires. If this were not true, the amount of converted solar energy described in the project (i.e. 200 KW) is not enough to propel the relatively massive spacecraft to Mars in 30 days using other methods.

Here's the evidence:

Link: http://www.nasa.gov/directorates/spacetech/niac/2012_phaseII...

Title: "The Fusion Driven Rocket: Nuclear Propulsion through Direct Conversion of Fusion Energy"

Quote: "an in-depth analysis of the rocket design and spacecraft integration as well as mission architectures enabled by the FDR need to be performed. Fulfilling these three elements form the major tasks to be completed in the proposed Phase II study. A subscale, laboratory liner compression test facility will be assembled with sufficient liner kinetic energy (~ 0.5 MJ) to reach fusion breakeven conditions." [emphasis added]

Which brings us back to square one. If this project were to succeed, it would instantly replace the existing approaches (i.e. tokamak and laser inertial confinement) as the most promising candidate for large-scale fusion power generation.

As I said before, something is rotten in Denmark -- why isn't this project being described as a candidate for earthly fusion power generation, given that it must achieve break-even to accomplish its mission?

The thing about a fusion rocket engine is that the energy is used to throw stuff out the back. In a fusion generator this would not be considered a feature - if you want a stationary generator that doesn't melt your city you have to contain that plasma, which is very, very energy-intensive.

Yes, true, but neither scenario has a working model of a fusion generator with power gain > 1.

> if you want a stationary generator that doesn't melt your city you have to contain that plasma, which is very, very energy-intensive.

Actually, if you think about it, a rocket engine that can create a fusion reaction and direct the energy out "the back" as you put it, and a power generator that also directs fusion energy to a secondary process, are very similar. In the extreme case, you could take the space device, put it in a vacuum chamber, and direct the thrust into a steam generator.

  I should have been more clear -- I mean sustained fusion generation, meaning a net energy gain over that required to start the reaction in the first place. That hasn't been achieved.

Yes, it has. Fusion produces energy; the energy you have to put in to initiate fusion does not magically disappear. Thus, if you can do fusion at all, you will end up with more total energy than you started with.

What has not been achieved is capturing enough of that resultant energy in a form (such as electricity) in which it can be used to keep the reaction going. That step, however, is completely irrelevant to a rocket engine. The energy released by fusion is largely kinetic energy, and that is exactly what you want in rocket exhaust, no need to go through a lossy conversion step.

  If the fusion reaction produces less power than it requires, the designers would be better off using the source electrical power to drive an ion thruster.

Not so. If the fusion reaction generates any energy at all, then that is better than just using your electrical power source by itself, for exactly the same reason that it's better to use a starter motor to initiate ignition of gasoline in your car than it is to try to run your car off of a weak starter motor. This is true even if you have no way of using the engine to recharge your battery.

  If this project were to succeed, it would instantly replace the existing approaches (i.e. tokamak and laser inertial confinement) as the most promising candidate for large-scale fusion power generation.

No, it wouldn't. That would only happen if the additional step were taken of discovering a highly efficient means of capturing the kinetic and radiant energy of the engine exhaust and converting it back to electricity. And research into fusion generators that work pretty much like that has been and is being done.

Edit: The slides at http://www.nasa.gov/pdf/636883main_FDR_talk_NIAC_2012_final.... even say:

  It is assumed that initially FDR employs solar panels for house keeping power
  Eventually it would be derived directly from nozzle flux compression

I.e., "once we do figure out how to turn this into a generator, then you don't need solar panels anymore." But they don't know how to derive energy from the exhaust stream through the nozzle yet, so they have to keep it going with solar panels. And extracting energy from the exhaust would necessarily reduce the ISP of the engine somewhat.

>> I should have been more clear -- I mean sustained fusion generation, meaning a net energy gain over that required to start the reaction in the first place. That hasn't been achieved.

> Yes, it has.

No, it has not. Apart from stars and thermonuclear weapons, there are no fusion reactions that yield more energy than they require, i.e. the achieve break-even. It has not happened.

> Thus, if you can do fusion at all, you will end up with more total energy than you started with.

You just changed the subject. Apart from stars and weapons, existing experimental fusion reactions produce much less energy than is required to create them. For example, all the laboratory experiments to date.

Also, very important, in a fusion reactor at less than break-even, the input power must be used to perpetually sustain the reaction (the state of the plasma), so that power is unavailable for any other purpose. Only the fusion reaction's power can be exploited.

So in a hypothetical reactor that requires 1000 watts to sustain fusion but produces 250 watts of fusion power, only the 250 watt fraction is expoitable -- the original power must be reserved for heating the plasma. That's why break-even is essential.

>> If this project were to succeed, it would instantly replace the existing approaches (i.e. tokamak and laser inertial confinement) as the most promising candidate for large-scale fusion power generation.

> No, it wouldn't.

Yes, it would. Given a power gain > 1, it would be child's play to generate steam and spin a turbine, as just one example.

> And research into fusion generators that work pretty much like that has been and is being done.

To date, there have been no -- that's NO -- laboratory fusion generators that produce more power than they require for initiation.

  No, it has not. Apart from stars and thermonuclear weapons, there are no fusion reactions that yield more energy than they require, i.e. the achieve break-even. It has not happened.

That's true, but irrelevant. Producing more energy from the reaction than was required to start the reaction is a different thing from ending up with more total energy in the system than you started with. Even stars don't do that- they just have sufficiently good containment that they don't lose all of the initial ignition energy, and thus don't need additional power inputs to replace non-existent losses.

You seem to be confusing the total power available in a fusion system with the total power available at the output terminals of a generator. The first is relevant to a rocket. The second is not.

To date, energy recovery inefficiencies for fusion reactors have always been high enough that the energy lost to neutrinos / waste heat / etc. is large than the amount produced by the reaction, meaning that the power available at the output terminals is less than the input power. Break-even does not necessarily mean that the reaction itself produces more power than the ignition apparatus- it means that the total useful power you can extract from the system, whether you put it there to begin with or not, is larger than the power required by the ignition apparatus. But a rocket doesn't care about recovery. Any power produced by the fusion reaction counts as a gain.

  So in a hypothetical reactor that requires 1000 watts to sustain fusion but produces 250 watts of fusion power, only the 250 watt fraction is exploitable -- the original power must be reserved for heating the plasma. That's why break-even is essential.

You are implicitly assuming that some of the 250W surplus can be extracted, but that none of the original 1000W can. That's a physically indefensible assumption. If you put in 1000W and the reaction generates 250W, then there's a total of 1250 indistinguishable watts running through the system, and you need to be able to harness at least 1000W to keep the system in steady state. If you can extract more than that, you've got a generator.

  To date, there have been no -- that's NO -- laboratory fusion generators that produce more power than they require for initiation.

I made no claim that there were. I said that research into this kind of fusion generator has been done, not that it has resulted in a working generator yet. See, e.g., focus fusion, or magnetoplasmadynamic generators.

>> No, it has not. Apart from stars and thermonuclear weapons, there are no fusion reactions that yield more energy than they require, i.e. the achieve break-even. It has not happened.

> That's true, but irrelevant.

That's the topic of discussion. Therefore it is relevant.

> Break-even does not necessarily mean that the reaction itself produces more power than the ignition apparatus ...

That is exactly, precisely what it means. That is how break-even is defined.

http://en.wikipedia.org/wiki/Fusion_energy_gain_factor

Quote: "The fusion energy gain factor, usually expressed with the symbol Q, is the ratio of fusion power produced in a nuclear fusion reactor to the power required to maintain the plasma in steady state. The condition of Q = 1 is referred to as breakeven."

Any questions?

> You are implicitly assuming that some of the 250W surplus can be extracted, but that none of the original 1000W can.

I am not implicitly assuming anything. In the example, because 1000 Watts is required to sustain the plasma in a fusing state, none of that power is available for any other purpose -- it might as well not exist. An attempt to harvest any part of that power will extinguish the fusion reaction. This leaves 250 watts. Those are the facts.

> That's a physically indefensible assumption.

Okay, you need to learn basic physics. One cannot harvest more than 250 watts from the hypothetical system without extinguishing the reaction. The original 1000 watts is unavailable -- it might as well not exist.

>> To date, there have been no -- that's NO -- laboratory fusion generators that produce more power than they require for initiation.

> I made no claim that there were.

Yes, you did. That was your claim -- that fusion reactors produced more power than they required for initiation. Here's what you said:

> Thus, if you can do fusion at all, you will end up with more total energy than you started with.

It is false. While the reaction is underway, you do not have more total energy than you started with, you have less. I have explained why this is so, very clearly.

You simply are not even reading. Nobody said anything about maintaining any steady state. That's needed for a generator. This project is talking about doing one off detonations.

Have you heard of a hydrogen bomb? If they lost energy to creating the fusion in a hydrogen bomb, what would be the point? If you removed the fusion part of a hydrogen bomb, would you get a bigger explosion? No. You get MORE ENERGY using fusion. However, we don't have a way to harness that to make electricity yet to maintain a steady reaction. But maintaining a steady reaction is not what this project proposes doing. Do you get it yet?

Edit: smaller explosion -> bigger explosion. So many explosions.

> You simply are not even reading.

And your evidence is that I quoted everything that I replied to, word for word, and quoted from the original NASA project documents as well?

> Nobody said anything about maintaining any steady state.

I did, and so did NASA. You missed its significance. Pulsed systems have an average power level, and a peak power level. Both need to be analyzed.

> That's needed for a generator.

Yes, and NASA wants a net generator of energy, something better than break-even, otherwise it's not worth doing. And they say this in their documents about the project.

> This project is talking about doing one off detonations.

I can't believe you missed the significance of NASA'a remarks about break-even. Don't you understand that, pulsed or not, break-even still has a meaning, and if they can't get to break-even, the project makes no sense?

A steady state generator either does or doesn't achieve break-even. A generator that consists of a series of pulses also does or doesn't achieve break-even. Here's what NASA has to say about this:

http://www.nasa.gov/directorates/spacetech/niac/2012_phaseII...

Quote: "A subscale, laboratory liner compression test facility will be assembled with sufficient liner kinetic energy (~ 0.5 MJ) to reach fusion breakeven conditions."

I can't believe NASA thinks break-even is an essential program goal. Maybe they should hire you as a consultant, so they won't waste taxpayer dollars trying for break-even, after all, according to you, because the output is pulsed, break-even has no meaning.

> But maintaining a steady reaction is not what this project proposes doing.

You very clearly do not understand the relationship between peak and average power. The device being described generates a series of pulses, but for there to be a point to the exercise, the average output power must exceed break-even.

A radar has a peak output power of two megawatts and a steady-state input power of ten watts. Does the radar violate the principle of energy conservation? Yes or no?

A fusion reactor has a peak output power of two megawatts and requires an average plasma sustaining power of 200 KW. Such a generator either does or does not achieve break-even over time, and as quoted above, NASA cares very much which is so.

> But maintaining a steady reaction is not what this project proposes doing.

You need to learn the relationship between peak and average power. Stop embarrassing yourself.

http://www.radartutorial.eu/18.explanations/ex28.en.html

http://hank.uoregon.edu/experiments/modelocked-fiberlaser/20...

> Do you get it yet?

You have managed to miss every key fact about this project.

I guess the point is less about the fusion method being efficient (and getting past break-even) but more about propulsion. ie. for 200KW of input do you get more propulsion from initiating a fusion reaction than using an ion ejection method.

Their aim is to move a heavy object with as little fuel as possible not necessarily produce perfect fusion. (although obviously that would be nice for everyone!)

> I guess the point is less about the fusion method being efficient (and getting past break-even) but more about propulsion.

But "break-even" means yielding more energy than is required to start the reaction. Obviously anything less than break-even is not worth having -- a chemical rocket would be more efficient.

> Their aim is to move a heavy object with as little fuel as possible not necessarily produce perfect fusion.

This misses the point that present laboratory fusion reactions require more power than they release. It's not a question of "perfect fusion" but any fusion that has a net positive energy yield.

> Obviously anything less than break-even is not worth having

No. Not at all. We are talking propulsion, using fusion reactions to pass on energy to the propellant and convert that into thrust. Although it would be desirable, it's not required to make it a net-positive reaction - just giving the propellant more energy than a chemical reaction is enough to be more efficient than a chemical rocket.

As the article mentioned, you can power this rocket with an ISS worth of solar panels (which is quite a lot of mass). Or, as it didn't mention, a very small fission reactor (provided you could negotiate putting a 200 KW reactor in space).

>> Obviously anything less than break-even is not worth having

> No. Not at all.

Yes -- if the fusion reactor didn't achieve break-even, the designers would be better off using an ion thruster. Also, the NASA documents that describe the project assume that break-even must be achieved:

http://www.nasa.gov/directorates/spacetech/niac/2012_phaseII...

They are talking about doing away with the present lines of fusion research, which haven't achieved break-even, and using a different method. But they don't suggest that this, or some variation, might be used for conventional power generation.

> Although it would be desirable, it's not required to make it a net-positive reaction

Yes, it is -- that is required. Were this not true, the designers would be better off using an ion thruster, which already exists and is quite efficient.

> As the article mentioned, you can power this rocket with an ISS worth of solar panels ...

Yes, that's in the description, but the power available (200KW near earth, 100KW near Mars) is not enough to propel the relatively heavy craft to the mission profile (i.e. 30 days to Mars) without some other source of energy, like from a net fusion power gain > 1.

A rocket only needs to achieve thermal break-even. A power plant needs to achieve electrical break-even, which includes the unavoidable loss from converting the heat to electricity in a heat engine.

In this way a system can be simultaneously useful for space travel but useless for power generation.

> A rocket only needs to achieve thermal break-even. A power plant needs to achieve electrical break-even ...

Yes, all true. But if a fusion reactor ever achieved break-even, that would be such a breakthrough that the specifics would be reduced to footnotes, and both thermal and electrical applications would soon follow.

> In this way a system can be simultaneously useful for space travel but useless for power generation.

My point is that if break-even were to be achieved, it would be break-even for both applications. The reason is that the plasma conditions for fusion break-even would have much more in common in the two cases than the differences.

If fusion doesn't provide a substantial energy gain, it's not worth the effort. Existing ion thrusters are quite efficient (60+%) at translating electrical power to exhaust kinetic energy.

Indeed, but they offer very low thrust. For manned space travel, you may need something that can give you a higher acceleration, even if at the expense of efficiency.

Also, fusion reactors designed for power generation have very different goals than fusion rockets. With power generation, particles leaving the reactor may be considered wasted energy. With rockets, the whole idea is to have particles leaving the reactor in a certain direction taking as much energy as possible with them. You just point the jet at the direction opposite to the one you want to go.

Well, controlled fusion hasn't reached energy breakeven at all, even taking into account the kinetic energy of the reaction products. But the low thrust of current ion engines is mostly due to power restrictions (https://groups.google.com/group/sci.space.science/msg/0cb332...). If your spaceship has a mass of one ton and you want an acceleration of 1 centigee with an exhaust velocity of 30 km/s, you will need at least:

0.5 * thrust * velocity = 0.5 * (0.1 m/s^2 * 1000 kg) * 30 km/s = 0.5 * 100 N * 3E4 m/s = 150E4 W = 1.5 MW

of power.

"kinetic energy" means nothing alone, thrust and specific impulse are better metrics to look at. Doing a lot better than ion thrusters at either without being too much worse in the other and energy efficiency would be interesting.

However, the article quotes an exhaust velocity of 30km/s (or ISP of 3000s), and using 200kW doesn't leave much room to beat http://en.wikipedia.org/wiki/HiPEP on thrust without counting on net energy gain. The linked slides also claim net energy gain.

Yes, I agree. My point was that you cannot do substantially better than ion engines without having energy gain and, in fact, you are probably going to do much worse due to the weight of the "fusion hardware".

I'm quite optimistic about magnetized inertial fusion. But the idea of doing the job much better than the Z-machine, with something lightweight enough to carry into space and in less than 10 years seems to me... unlikely, to put it mildly.

>in fact, you are probably going to do much worse due to the weight of the "fusion hardware".

Ion engines have weight, too. What leads you to believe that the fusion hardware will be heavier?

In the short run:

- Ion thrusters are way simpler and they are trying to do something that is much easier.

- Ion thrusters are relatively mature, working "fusors" don't exist.

- Existing machines that are trying to get fusion breakeven are building-sized.

- Trying to start a fusion reaction is hard, it only makes sense if you get a commensurate result (energy gain).

In the long run, I'm very optimistic about fusion.

Ion drives are highly efficient, but have ridiculously low thrust. Which doesn't help at all when you are trying to reach some point quickly.

If you can construct an ion drive with enough thrust to match this proposed fusion drive (or even a NERVA: http://en.wikipedia.org/wiki/Nuclear_thermal_rocket), talk to NASA, I am sure they will be interested in buying several from you.

> Which doesn't help at all when you are trying to reach some point quickly.

Sure it does. You just leave it turned on. This thruster assumes 6 days of thrust then 24 days of coasting. With an ion thruster you leave it on for all 30 days.

There are no high thrust systems that can just be left on - even nuclear ones are used for a short period then turned off. The idea of an ion thruster is that you leave them on, and achieve the same total thrust, over the same time.

> With an ion thruster you leave it on for all 30 days.

With a thrust of - at best - 5 newtons, you won't achieve the goal of getting to Mars faster. You may get there cheaper and using less fuel, but for pure speed you lose. Ion thrusters are good for very long trips when you're going to leave the engine on for months, or for trips where the total time doesn't matter much, only fuel economy (e.g. for cargo shipments or probes).

5 newtons won't get you there in 1 month, but according to http://en.wikipedia.org/wiki/VASIMR it will in 5 months which is still pretty good, and much better than a standard rocket.

If you don't mind waiting a year.

Existing ion thrusters don't give enough thrust to get to Mars in 30 days.

You are completely missing a point of this. It is a propulsion system, not a power generating one. Chemical rocket engine doesn't generate any electrical power either. Main factors in propulsion is specific impulse and reaction mass and this idea promises a huge impulse - basically once in orbit, it could ferry between Earth and anywhere-where-solar-power-is-sufficient because it doesn't need tonnes of fuel, only relatively small load of pellets would get the job done. Of course there is the issue of getting it and it's payload into orbit, but I'd say Skylon will take care of that

> You are completely missing a point of this. It is a propulsion system, not a power generating one.

To represent an advantage over an ion thruster, it must be both. And the NASA documents that describe the project specify a fusion power gain > 1:

http://www.nasa.gov/directorates/spacetech/niac/2012_phaseII...

The phrase "fusion breakeven conditions" means the reaction must produce more power than is requires to start it. This is a requirement to justify the project over other approaches.

It is not even supposed to be used to generate electrical power. Citing from the site itself: "Virtually all of the radiant, neutron and particle energy from the plasma is absorbed by the encapsulating, metal blanket thereby isolating the spacecraft from the fusion process and eliminating the need for large radiator mass"

No radiators implicitly mean no electricity generation, because it creates waste heat that has to be utilised.

http://en.wikipedia.org/wiki/Fusion_rocket

Proposed concept is a direct drive, and it actually claims they fixed the problem with neutrons (they are caught by the compressing metal)

As for breakeven conditions they merely state that the power that the fusion generates must be at least equal to power used to start/sustain it. http://en.wikipedia.org/wiki/Fusion_energy_gain_factor Theoretical limits of efficiency of heat engine is around 70-80% (roughly) http://en.wikipedia.org/wiki/Thermal_efficiency and you have to use those to convert heat generated in controlled fusion into electricity. To actually achieve net power generation in a power plant, you have to generate a lot more heat that electricity, while for propulsion you merely care about actual raw (so to speak) power.

Ion thrusters have their own issues, while their ISP is considerably better than chemical rockets, maximum thrust is abysmal, but it is hard to argue about that due to the lack of data on FDR. However, with ion thrusters there are other issues (for example:wear, charge imbalance)

> As for breakeven conditions they merely state that the power that the fusion generates must be at least equal to power used to start/sustain it.

Yes, but (apart from stars and weapons) this has never been achieved anywhere, ever. This is a big issue, bigger than most people realize.

https://en.wikipedia.org/wiki/Fusion_power

Quote: "As of July 2010, the largest experiment by means of magnetic confinement has been the Joint European Torus (JET). In 1997, JET produced a peak of 16.1 megawatts (21,600 hp) of fusion power (65% of input power) ... Fusion powered electricity generation was initially believed to be readily achievable, as fission power had been. However, the extreme requirements for continuous reactions and plasma containment led to projections being extended by several decades. In 2010, more than 60 years after the first attempts, commercial power production was still believed to be unlikely before 2050."

My point is if this project creates a net gain > 1, it will address a lot more than the problem of getting to Mars.

As other posters have noted, ion thrusters don't generate very much thrust.

s/Skylon/SpaceX/

That's what I was thinking. They're actually already doing something. On the other hand, their not-yet-developed Falcon Heavy can "only" lift 53 tons, about a third of the weight of FDR. Hopefully they can come up with a way to lift it in stages, or we still need a bigger rocket.

> Obviously anything less than break-even is not worth having -- a chemical rocket would be more efficient.

Why not? The energy to start the reactor comes from solar power, so they didn't have to carry it up with them.

It's not a sustained reaction. It's just a bunch of short fusion reactions, one after another. We've been capable of short, uncontrolled fusion reactions for decades--it's just that, being bombs, we didn't have much luck generating electricity from them. A hydrogen bomb uses a relatively small quantity of conventional explosives to start a reaction which ultimately releases orders of magnitude more energy than the original input. This engine claims to use a small input of electrical energy to trigger a reaction which releases a lot of energy. Then, much like a car getting blown off the road by a nuclear test, they just position their ship near this reaction and get shoved away.

The proposal specifically says they will not convert the energy to electricity, but will use the reaction directly for thrust.

Actually hydrogen bombs don't use conventional explosives. They use fission. See http://en.wikipedia.org/wiki/Pure_fusion_weapon (a hypothetical bomb).

> Actually hydrogen bombs don't use conventional explosives. They use fission.

Sort of, but because the fission trigger requires conventional explosives, one could argue that conventional explosives play a key role.

In temporal sequence:

1. Conventional explosives compress the fission core.

2. At maximum compression, a special neutron source releases a burst of neutrons to start nuclear fission.

3. The fission trigger delivers a huge pulse of energy to the mixed hydrogen isotope / fission secondary, compressing and heating it.

4. The secondary begins to fuse, releasing much more energy than the fission trigger could produce on its own.

The above is somewhat oversimplified:

http://en.wikipedia.org/wiki/Thermonuclear_weapon

It's our weak imitation of a star.

The same guy (John Slough) started Helion Energy, which is attempting to build a power-producing fusion reactor. Last I heard they had a 1/3 scale device with promising results, and needed $20 million for a full-scale test. But he got funding for the rocket first.

The rocket project may actually be easier. It only pulses once per minute, just runs for three days, and doesn't have to deal with the economic considerations of a practical power plant.

Incidentally, Helion is not the only alternative fusion project in the works. Others include Sandia's MagLIF, Lockheed's recently-announced project, Tri-Alpha, General Fusion, Focus Fusion, and IEC/polywell.

> Last I heard they had a 1/3 scale device with promising results, and needed $20 million for a full-scale test.

Research takes time, the potential payoff is huge, but in the final analysis, the fusion reaction has to produce more power than it requires. It has to achieve what's called "break-even".

A fusion reaction with a power gain < 1 is endothermic -- its temperature is less than that of its power source. A fusion generator by definition produces more power than it requires for initiation -- it has a power gain > 1, and it is exothermic.

> Incidentally, Helion is not the only alternative fusion project in the works. Others include Sandia's MagLIF, Lockheed's recently-announced project, Tri-Alpha, General Fusion, Focus Fusion, and IEC/polywell.

This is a good example of long-shot science, applied research. None of these ideas show very much promise, but the potential payoff is so big that (IMHO) the research money is well-spent.

>A fusion reaction with a power gain < 1 is endothermic -- its temperature is less than that of its power source.

You are trolling! Or you just don't get it. The energy does not disappear. If you put in 1, and the reaction gives you .5, you have 1.5. You do not have to maintain a stable reaction, in the same way a hydrogen bomb doesn't have to produce however many terajoules of electrical energy to be more useful than TNT. It just has to be more total.

They are not making an electric generator, they are making a rocket. They are not the same thing.

> The energy does not disappear.

Below break-even, the source power is requires to sustain the plasma in a high-energy state to maintain fusion. Therefore it's unavailable. Any effort to extract more power than the fusion reaction produces will extinguish the reaction.

Example -- let's say we have a fusion reactor that doesn't achieve break-even (example: all existing laboratory reactors). Let's say very hypothetically that heating the plasma sufficiently requires 1000 watts (sometime in the future) but the fusion reaction itself only produces 250 watts.

Once the fusion reaction has been established, we can only extract 250 watts from the system. Any effort to extract more will extinguish the reaction. Therefore more power must flow into the system than can be taken out. It's endothermic.

Which word didn't you understand?

> If you put in 1, and the reaction gives you .5, you have 1.5.

You, too, can learn the physics required to understand this system. You just have to think a bit harder.

> You do not have to maintain a stable reaction

On average, yes, you do. This project has a peak and an average power that differ greatly, but terms like break-even still have their normal meaning, which is why NASA mentions break-even in the project's documents as a required project goal.

http://www.nasa.gov/directorates/spacetech/niac/2012_phaseII...

Quote: "A subscale, laboratory liner compression test facility will be assembled with sufficient liner kinetic energy (~ 0.5 MJ) to reach fusion breakeven conditions."

What does break-even mean in a pulsed fusion generator? It means the average output power is greater than the average input power.

> They are not making an electric generator, they are making a rocket. They are not the same thing.

Yes, they are the same thing. They both require break-even, and they both rely on a fusion reaction.

> You are trolling!

I know this subject, you do not. Therefore it is you who are trolling when you should be reading.

> Which word didn't you understand?

I understood all of it, I just think you're wrong. And unless you ever address what anyone is saying, all I can do is keep on thinking you're wrong.

Nobody - nobody here, nobody on TFA - said anything about sustaining a reaction. It is a one-off explosion much like a weapon. An explosion like that is hard to use for generating electricity, just like a bomb, but great for pushing things around. Why isn't this feasible? Please explain. Why can we detonate a hydrogen fusion bomb, which generates a lot of energy (but no electricity), but we cannot initiate a tiny short term burst with a carefully designed fuel pellet?

I'm not saying they have everything figured out. I'm just saying I wouldn't be so skeptical to think it would be some amazing breakthrough if they do it. If this rocket works, we still don't have usable fusion electricity. That problem still needs to be solved. But it is not the same problem.

> Yes, they are the same thing. They both require break-even

No, electrical generation means you need to need to somehow use your energy to heat up something and turn a turbine, or something like that. Which means you need some kind of ongoing stable and contained reaction. There are so many difficulties with that, it really WOULD be a big deal if they solved that problem. Just making a small explosion? Not as far fetched.

> I know this subject, you do not.

And don't try to argument-from-authority me. You're a random account on the internet just like me. As far as I know you're some crackpot pseudoscience crazy. Not that I believe that, but I don't know better.

It's not continuous, it's pulses. But that doesn't actually matter, attach this thing to a big wheel and pulse it to run a generator. Or a few of them pulsing in sequence.

But I wonder if the radiation exhaust might be to high to be practical on earth.

The fusion products are presumably fairly hot - so why not use them to heat water and drive a steam turbine generator combination?

[NB This is in reply to the idea of using whirling fusion generators as a means of generating power, not part of the discussion on using it as spacecraft propulsion!]

I think the suggestion is that the fusion is achieved only around the tiny pellet, and that the fusion acts against the magnetic containment to actually propel the pellet out the back. That's my guess - which slightly beats the idea they have achieved sustainable net positive energy fusion but cannot be bothered to solve all the worlds energy problems and become trillionaires.

If this idea had been reduced to practice, it would mean the end of the National Ignition Facility, which after decades of effort, has yet to produce more power than is required to start the reaction:

http://en.wikipedia.org/wiki/National_Ignition_Facility

If the described method actually worked, it would force a complete reevaluation of the other approaches to fusion, none of which have actually worked (in a practical sense) after decades of effort.

That's why I doubt that the described method has moved beyond the theoretical phase. If this isn't true, then people would try to create a continuous-power version of the technology, in which prior energy releases and high temperatures would be used to sustain new fusion energy releases.

I've been reading about fusion power research for decades, I know the problems, and I remain skeptical.

It's actually surprisingly difficult to capture the heat energy.

The energy is released as neutrons, neutrinos and gamma rays, all of which tend to escape rather than heat the surrounding area.

But all of them are magnetically neutral, so I wonder how they propose to capture their momentum.

Also speaking very hypothetically, why do you suppose that all the input electrical power is wasted for the propulsion goal? Each fusion mini-explosion will have a total energy that is the sum of the input (ignition) energy and the fusion itself, not just that of the fusion.

Edit: This would mean that the thrust could be higher than the electrical input power even below break-even for the controlled fusion. If you only consider the electrical input, you could have a >1 ratio for the electrical energy/kinetic energy conversion (thanks to the fusion).

> Also speaking very hypothetically, why do you suppose that all the input electrical power is wasted for the propulsion goal? Each fusion mini-explosion will have a total energy that is the sum of the input (ignition) energy and the fusion itself, not just that of the fusion.

Yes, technically true, but consider this hypothetical scenario:

1. A power source, let's say 1000 watts, can be used for any purpose, or can be used to heat a fusion plasma.

2. A fusion reactor requires 1000 watts to sustain fusion but delivers 250 watts of fusion power.

3. The total power in the system is now 1,250 watts.

4. But that total power is not available for external uses. 1000 watts of it must perpetually be applied only to the task of heating the plasma.

5. That leaves 250 watts for other uses.

That means the fusion reaction uses more power than it produces, only 250 watts is available for other purposes, and there is no remedy except to not use the fusion reactor.

That's why break-even is an essential precondition for this project's viability.

"It's very scalable; we can achieve fusion at a much smaller scale," he said. "We could run the designed engine off 200KW of solar panels, which is about the same power as generated by the panels around the International Space Station"

So this is hot fusion, not cold. Hot fusion has been around for a long time. In fact I remember a couple of high school kids made a hot fusion device once.

My understanding is that "hot" fusion is actually fairly easy - it's generating a surplus of energy from fusion[*] outside of a nuclear weapon that is the tricky part.

http://en.wikipedia.org/wiki/Fusor

[Note: Typically in most H-bomb designs most of the energy still comes from fission - the fast neutrons from the fusion in the secondary igniting fission of the DU tamper round the secondary].

> My understanding is that "hot" fusion is actually fairly easy ...

Unless you're talking about a thermonuclear weapon, it's not easy at all. In fact, decades of research have yet to produce a break-even reaction in the laboratory.

I think you needed to read the second half of that sentence before replying.

> Hot fusion has been around for a long time.

Only in the stars and in thermonuclear weapons. It has not been achieved in a break-even sense anywhere else, in spite of decades of research. Many very large projects are dedicated to achieving it, including the National Ignotion Facility:

http://en.wikipedia.org/wiki/National_Ignition_Facility

The linked article shows that, in spite of vast amounts of money being spent, the project has yet to achieve break-even (i.e. produce more power than required to start the reaction).

> In fact I remember a couple of high school kids made a hot fusion device once.

No, that did not happen. It's an urban myth.

http://www.ncsu.edu/park_scholarships/news/article.php?id=24...

Nowhere near break even though.

This is just a plasma generator, not a fusion reactor. Not to dismiss this student's achievements, but any fluorescent lamp is a plasma source. Plasma is the "fourth state of matter", essentially separate electrons and nuclei in a relatively high energy, electrically conductive medium:

https://en.wikipedia.org/wiki/Plasma_(physics)

It's a fusor - which does (as its name suggests) do fusion - which is why they are used as neutron sources.

http://en.wikipedia.org/wiki/Fusor

A quote from your linked article: "Hopes at the time were high that it could be quickly developed into a practical source of fusion power. However, as with other fusion experiments, development into a generator has proven to be difficult."

Translation: no break-even fusion reaction. If this approach held promise for fusion power research, it would be being explored instead of the millions of dollars in the much more common laser-confinement and tokamak approaches.

Strictly speaking and from a technical standpoint, if it doesn't produce more power than is present for conventional reasons (like electronic current flow), and in spite of its name, it's not a fusion reactor as that term is understood in physics.

It's a nice plasma source, and it produces neutrons -- very useful -- but it's not a fusion power source.

I never said it was a "reactor", or that it was close to break even - merely that creating a device where fusion reactions do occur is not that difficult.

Read my comment above I say "Nowhere near break even though."

Incidentally, there has been a lot of work in inertial electrostatic confinement - it doesn't look terribly promising and no it is not break even but people do appear to be working in this area:

http://en.wikipedia.org/wiki/Polywell

As established by arethuza, hot fusion has been accomplished for some time, relatively easily, outside of thermonuclear weapons and stars. That was the claim in dispute. Nowhere in this thread was there a claim about power generation.

> As established by arethuza, hot fusion has been accomplished for some time ...

Let's be clear about what we're talking about. A fusion generator by definition produces more power than it requires. Apart from stars and thermonuclear devices, this has not been achieved anywhere. Without clear terminology, we will go in circles.

Also, the NASA project documents specify and require a net power gain in the fusion reaction:

http://www.nasa.gov/directorates/spacetech/niac/2012_phaseII...

> Nowhere in this thread was there a claim about power generation.

Except in the NASA documents that describe the project under discussion.

Yea. That is what is called hot fusion - where you have to put in energy to keep the fusion going.

And yes I was referring to the Fusor linked below. That is hot fusion.

> That is what is called hot fusion - where you have to put in energy to keep the fusion going.

No -- in general, "hot fusion" means producing more power than is required to start the reaction. Otherwise the term "hot fusion" makes no sense, since a reaction with a power gain < 1 is endothermic.

I emphasize that the term "hot fusion" is used in a lot of different ways, by people who aren't using it in its strict physical meaning. But in physics, "hot fusion" should mean an exothermic fusion reaction.

> And yes I was referring to the Fusor linked below. That is hot fusion.

It's not a fusion generator, which by definition produces more power than it requires. And the NASA documents that describe the project under discussion specify a fusion generator, a device with a power gain > 1.

http://www.nasa.gov/directorates/spacetech/niac/2012_phaseII...

"an in-depth analysis of the rocket design and spacecraft integration as well as mission architectures enabled by the FDR need to be performed. Fulfilling these three elements form the major tasks to be completed in the proposed Phase II study. A subscale, laboratory liner compression test facility will be assembled with sufficient liner kinetic energy (~ 0.5 MJ) to reach fusion breakeven conditions."

Note that last few words.

it's not about having net energy gain, it's about converting electrical and nuclear energy into kinetic energy.

> it's not about having net energy gain ...

That's not what the NASA documents say, the documents that describe the project:

http://www.nasa.gov/directorates/spacetech/niac/2012_phaseII...

Without a net energy gain, the project would be better off using an ion thruster, which has the advantage of already existing and being quite efficient (but with a power gain < 1).

To repeat what others have told you several times in this thread:

Why do we still use chemical rockets instead of ion thrusters for boosters? Answer: thrust vs. specific impulse. Ion-thrusters are efficient in use of (high specific impulse), but are weak in how much thrust they create. Your ship would have to be mostly ion-thruster to get it anywhere, thus ruining the efficiency advantage with respect to payload of the engine.

http://en.wikipedia.org/wiki/Ion_thruster "Given the practical weight of suitable power sources, the accelerations given by ion thrusters are frequently less than one thousandth of standard gravity."

What does linear kinetic energy have to do with fission reactors? It's not as simple as "attach a turbine to the kinetic energy conversion process." Or rather, it is that simple, and that's why it doesn't work, because it's another point at which one loses power to inefficiency. But if all you want is kinetic energy in the first place, then you don't lose efficiency to that extra step.

There is something you are not getting. Assuming a set of solar panels like those on the ISS, able to deliver 200KW near Earth and 100KW near Mars, one must still convert that power to useful thrust. That means achieving fusion break-even, a requirement that the NASA documents explain.

Obviously ion thrusters aren't practical for this kind of heavy lifting and short mission time, but that doesn't usher in fusion power as an obvious substitute -- unless the project creates a useful reaction, meaning a gain > 1.

Reply on news.ycombinator.com