[Kosh]

A better solution - EVEN with losses in thermal exchange - would be to contain the fusion-plasma in the reactor, trap most of the gammas into radiation shielding, which causes the reactor - including the shielding - to heat up and the thermal energy can be used to expand suitabe propellant to large volume and to be propelled backwards that way. This has the advantage of taking care some of the cooling of the reactor as well - at least while there's any propellant left. The propellant used could be liquid hydrogen, nitrogen or perhaps some inert gas like Xenon, Neon etc. In this scenario, certainly the fusion waste helium can be added into the mix, but you would still need a separate mass of propellant, the stuff coming from fusion is not really sufficient to offer a long term propellant source.

So about how much thrust could something like that (or any of the other suggestioned designs) make?

[Wanderer]

As Herra Tohtori said... That depends on the amount of fuel ejected and on the exit velocity of the said ejected fuel. As neither are known it is rather difficult to estimate it.

EDIT... With quick google search...

http://www.projectrho.com/rocket/index.html
http://www.projectrho.com/rocket/rocket3c.html

Dunno how 'good source' that is but at least but...

[Herra Tohtori]

So about how much thrust could something like that (or any of the other suggestioned designs) make?

Oh brother...

I'm gonna mostly ignore the neutron flux, because I assume you don't have a way to redirect them neutrons towards the rear of your ship - an easy task with alpha particles because of their +2e charge (damn, this starts to sound like an RPG ), all it takes is a magnetic field... neutral particles won't change their direction in mag field, so the neutrinos from the reaction would travel on pretty much linear trajectories, and assuming your fusion chamber is still mostly concealed by the ship, the ship's structure would be equally or almost equally hit by them -> no notable change of momentum from them.

Okay then... Assuming you're using deuterium-tritium fusion fuel mix, the resulting helium nuclei (alpha particles) will have roughly 3.5 MeV energy (source, about what kind of nuclei were observed coming from fusion going on in JET tokamak reactor running a 1.5 MW fusion reaction).

3.5 MeV is 5.60761762e10^-13 Joules of kinetic energy on each helium nuclei produced. From this it's easy to calculate the velocity of the nuclei - that is

E(k) = ½ m v^2

v^2 = 2 E(k) / m

v = Sqrt (2 E(k) / m)

since the mass of alpha particle is 6.644656e10^-27 kg, just throw them into the equation to see that the velocity is

v =~12991758.6 m/s, which is a heck of a lot of speed. Just to check if it's accurate, compare it to speed of light - approximately 300,000,000 m/s. The velocity of the alpha nuclei seems to be about 4% of speed of light, which is thankful because now it's unnecessary to use relativistic equations for defining momentums and stuff - for the accuracy we require, anyway.

Okay, now we have the (average) velocity of alpha nuclei, and we're going to divert their vector towards the rear of the ship, so that effectively the ship gains the same momentum as each alpha nuclei, but forwards.

Now the momentum of a single alpha nuclei seems to be about

p = m*v =~ 8.633e-20 kgm/s

And every reaction releases one of these critters. Okay then, using chemistry terms, if we have one mole of deuterium nuclei (not molecules!) and one mole of tritium nuclei, we end up with one mole of helium nuclei, assuming the fusion is perfect (which isn't gonna happen but since we're looking for upper limits, best case scenario, it's OK).

Deuterium's atomic weight is 2u, tritium's is 3u, helium's is 4u (and the neutron that goes away is 1u, but nevermind that now). Roughly. There are some mass loss, but on this scale, it's not all that notable anyway, so I'm gonna blatantly ignore it aside from saying that actually helium+neutron weighs a little less than deuterium+tritium, and that's it. Helpfully, the atomic mass unit is configured so that when the atomic weight is 1 u, one mole contains one gram of the matter.

So, when the engine consumes grand total of 5 grams of deuterium-tritium mix (in proper ratios), it gives us 4 grams of helium to use as propellant.

Now then... the one mole of helium weighs (about) 4 grams, as was determined. One mole contains 6.0221415e23 particles (Avogadro's number), which means we're gonna multiply the momentum with this... multiplier number... to determine the momentum that the ship can at best gain from using grand total of 5 grams of fusion fuel:


p = 6.0221415e23 * 8.633e-20 kgm/s = 51989.1475695 kgm/s


which feels like a lot of momentum, but what it essentially means is that with 5 grams of fuel, you will manage to make a 52 ton ship travel at a whopping one metre per second, in the time it takes for the reactor to consume the 5 grams of detrimix fuel...


To put it in thrust terms (or force);

F = dp/dt

which means that if the fusion reactor can consume 5 grams of fuel in, say, 60 seconds, the thrust from it would be

F = 52000 kgm/s  / 60 s = 866.7 N


If the reactor can use the mentioned 5 grams of fuel in 1 second, the maximum thrust would be 52 kN.


And seeing how utterly incredible amounts of energy are released, the achieved momentum/thrust/acceleration ratio is not really mind-bogglingly good. Also when you think of 52 ton space ship, you should be looking at something about the size of a good-sized airliner (like Airbus A320, or perhaps Boeing 757) as a reference to size, and compare to existing chemical rocket engines - Apollo-type service module engine produced 98kN thrust.


And I reserve the very viable possibility that something went wrong in the calculations.


As a final analysis, it's obvious that as far as propellant consumption is considered, this kind of drive is rather efficient, but propellant consumption is not all in all. In this case, fusionable isotopes are not as abundant as normal hydrogen in space, so where you gonna get all those precious grams of detrimix fuel?

[Kosh]

As a final analysis, it's obvious that as far as propellant consumption is considered, this kind of drive is rather efficient, but propellant consumption is not all in all. In this case, fusionable isotopes are not as abundant as normal hydrogen in space, so where you gonna get all those precious grams of detrimix fuel?

So the only reason that this hasn't been done is that there isn't enough fuel? I'm assuming all these ideas have been around for a long time so I would hope fuel shortages would be the only thing holding this up........

[Herra Tohtori]

No, the reason is that it isn't technically or energetically feasible.


I once did some calculations on how energy consumption (ie. power requirement), propellant consumption, ejection velocity and momentum are connected. In the end the equation is really simple - the faster you eject the propellant, the more energy it consumes in the ship's reference frame - remember, double the ejection velocity, double the momentum, BUT quadruple the amount of energy consumed to accomplish the relative velocity between ejected propellant and the ship.

That's also why it takes obscene amounts of energy to propel anything to 4% speed of light. And energy consumption is also important when space ships are considered... or rather, if there's sufficient amount of fuel/propellant, it's the cooling that becomes problem rather than power levels.


Also, the main reason no one even thinks of doing this in their right mind is that right now, we don't even have self-sustaining, energy-outputting fusion reactors. They need to be maintained with constant energy input that is more than the reaction itself yields. Probably situation will change as ITER and further in future DEMO become active, but right now fusion hasn't even been accomplished in huge research facilities - much less in any kind of vehicle, not to mention a space ship in which light weight is kinda priority. Also, radiation shielding can't be accomplished without a lot of mass aboard, so it's kinda so-so whether or not a nuclear reactor on board a manned space ship would be a blessing or curse.

[Stealth]

I never played Descent and I'am still consider myself a FreeSpace fan.

then i've got news for you.......... you're not a fan.

[Koth]

I never played Descent and I'am still consider myself a FreeSpace fan.

then i've got news for you.......... you're not a fan.

Who are you to decide that?

[Kosh]

Probably situation will change as ITER and further in future DEMO become active,

What is so different about their designs to change this?

[Herra Tohtori]

Mostly the fact that they will be the first two fusion reactors designed to offer positive energy output from the get-go. Thus far, the research has been pretty much concentrated on getting the fusion happen in the first place, in a controlled manner, then increase the energy output and reduce the energy consumption.

JET-tokamak reactor has in some tests managed to produce more energy output than input, but only for really short periods of time. ITER will be bigger, stronger and better and hopefully will be able to keep the energy output consistently and meaningfully bigger than the energy used tto keep the thing going on. DEMO will be a, well, demo of a commercially viable tokamak fusion reactor.

Basically it's pretty much like comparing previous attempts on powered flight to Wright brothers' first Flyer. The biggest difference between them are the fact that Flyer had a sufficiently lightweight and powerful engine, whereas the previous ones didn't. Also, Lilienthal, Bernoulli et al had done a lot of work that the Wrights used to make their flyer aerodynamically viable flyer... and the result was that the Flyer flied, when previous attempts had failed.

Similarly, the basic idea on JET, ITER and DEMO is the same - toroidal chamber with magnetic coils keeping the plasma in the middle. Now it's just making the process more streamlined, less energy-consuming while increasing the volume of the reactors to increase power capacity.

[Flaser]

I don't think the JET or other tokamak designs are a good source to estimate how a fusion engine would function.

Namely, the engine has a huge advantage, namely, that it can simply eject the spent plasma; so you don't have to hold the whole thing in place, you can have a strong current/stream (movement) in the plasma that could also help its handling.

However, that's the lesser source of miscalculation: in the Tokamak designs, it's not an aim to further energize the particles in the reactor...quite the opposite. The missing mass has become a huge load of energy - unlike the power plant designs, here we don't just absorb/radiate it all over the place if we can help it. We want to convert that energy to kinetic energy of the propellant.

Other calculation - namely the ones on project Rho - have shown a lot bigger specific impulse with these kind of engines:

"Deuterium-Tritium Fusion rockets use the fusion reaction D + T ⇒ 4He + n. If you add up the mass of the particles you start with, and subtract the mass of the particles you end with, you can easily calculate the mass that was converted into energy. In this case, we start with one Deuteron with a mass of 2.013553 and one atom of Tritium with a mass of 3.015500, giving us a starting mass of 5.029053. We end with one atom of Helium-4 with a mass of 4.001506 and one neutron with a mass of 1.008665, giving us an ending mass of 5.010171. Subtracting the two, we discover that a mass of 0.018882 has been coverted into energy. We convert that into the fraction of fuel that is transformed into energy by dividing it by the starting mass: Ep = 0.018882 / 5.029053 = 0.00375.

Plugging that into our equation Ve = sqrt(2 * 0.00375) = 0.0866 = 8.7% c. "




That's almost 3 times your specific impulse, but that's also the theoretical limit.
I also find the couple of gramms / second consuming engines somewhat unlikely, though your hypothesis based on the Tokamaks and therefore the current limits of magnetic engineering may more than apply. (I think the open bottle nature of a rocket is what would make a big difference here).

[Herra Tohtori]

I don't think the JET or other tokamak designs are a good source to estimate how a fusion engine would function.

Namely, the engine has a huge advantage, namely, that it can simply eject the spent plasma; so you don't have to hold the whole thing in place, you can have a strong current/stream (movement) in the plasma that could also help its handling.

Ah, but then problem becomes - how to contain the deuterium-tritium plasma that is yet to be fused, and how to let the fast charged alpha particles escape to generate thrust?

The problem with fusion is that in an uniform temperature/pressure plasma cloud (with sufficient energy levels to cross the Coulomb barrier), fusion reactions happen pretty much uniformly as well, which increases the temperature of the plasma, propelling newly formed alpha particles everywhere inside the plasma, trying to expand the plasma cloud's volume, but while there still are also deuterium and tritium in the mix, you don't really want them to escape as well, as it would seriously weaken the efficiency of the engine.

The answer could possibly be to use some kind of pulse fusion technology - not using a continuous fusion, but instead alternating between confining magnetic field and "guiding" magnetic field to guide the expanding cloud of positive ions backwards. Pour a few micrograms of plasma into confining magnetic field, light it up with lasers, let it fuse (which increases it's temperature), then switch to guiding magnetic fields, which releases the ions from confinement and turns the expansion of plasma into propulsion.

With sufficiently large frequency, the pulses would feel more like steady humming on the ship.

However, that's the lesser source of miscalculation: in the Tokamak designs, it's not an aim to further energize the particles in the reactor...quite the opposite. The missing mass has become a huge load of energy - unlike the power plant designs, here we don't just absorb/radiate it all over the place if we can help it. We want to convert that energy to kinetic energy of the propellant.

Yes, but since the reaction itself that I used in the example is the same, it means that the alpha particles from similar reaction would have (approximately) the same kinetic energy from the reaction.

Other calculation - namely the ones on project Rho - have shown a lot bigger specific impulse with these kind of engines:

"Deuterium-Tritium Fusion rockets use the fusion reaction D + T ⇒ 4He + n. If you add up the mass of the particles you start with, and subtract the mass of the particles you end with, you can easily calculate the mass that was converted into energy. In this case, we start with one Deuteron with a mass of 2.013553 and one atom of Tritium with a mass of 3.015500, giving us a starting mass of 5.029053. We end with one atom of Helium-4 with a mass of 4.001506 and one neutron with a mass of 1.008665, giving us an ending mass of 5.010171. Subtracting the two, we discover that a mass of 0.018882 has been coverted into energy. We convert that into the fraction of fuel that is transformed into energy by dividing it by the starting mass: Ep = 0.018882 / 5.029053 = 0.00375.

Plugging that into our equation Ve = sqrt(2 * 0.00375) = 0.0866 = 8.7% c. "




That's almost 3 times your specific impulse, but that's also the theoretical limit.
I also find the couple of gramms / second consuming engines somewhat unlikely, though your hypothesis based on the Tokamaks and therefore the current limits of magnetic engineering may more than apply. (I think the open bottle nature of a rocket is what would make a big difference here).

Now that is interesting, my thanks for more accurate data.

To be honest I'm a bit surprized that I got to at least approximately same magnitude of achieved impulse compared with the calculations of that mentioned group, which are much more likely way more accurate. The relative accuracy of my calculations surprizes me, especially since I used very crude approximations and mostly just picked an estimate of the kinetic energy/velocity of alpha particles from fusion, based on the first source I could find that seemed reliable.

In guesstimation physics, I consider it a relatively good achievement if the results are on the same magnitude (in this context, same exponent of ten) with more accurately calculated, actual physics (either theoretical or experimentally measured values)...


And, as said, even with three times the specific impulse (and three times the thrust), it would still mean consuming 5 grams of deuterium-tritium mix in a second to achieve thrust of 150 kN (or something like that anyway). As a theoretical maximum. A bit more than the amount of thrust from Pratt&Whitney J58 with it's 142 kN thrust (the engines used in SR-71 Blackbird).

Now, as I stated before, as you increase the ejection velocity and reduce the mass of propellant, the amount of work increases exponentially. Seeing how the mentioned J58 engine looks like this with afterburners on...



Glows pretty nice and warm, right? Well, the energy released in the chemical reactions is much much less than what is released in fusion of five grams deuterium-tritium, and releasing that amount of energy in a second would result in pretty brightly glowing end point of the rocket. Of course a lot of the energy would go into the plasma expansion thrusters as kinetic energy, but even so, it would make the plasma itself glow terribly brightly, not only on gamma ray wave lengths but also infra-red and visible light - which could have serious adverse effects on the rear end of the space ship.

[Hellstryker]

If you're talking about the last thing, you can't call yourself a real gamer if you have never seen that.

No i've seen it, i just recognized the similarity now tho

[Nuke]

i figure the hardest part of rocket science is keeping your engine from melting.

i think fission + ion is the way to go for now. by the time our magnetic technology reaches a level to make fusion viable, other propulsion technologies may become feasible, such as accelerating the interstellar medium. you could probably also use magnetics to contain hotter chemical reactions than could be achieved by most materials and cooling approaches used today.

forcefields are good for technology

[Admiral_Stones]

Holy ****. Can someone translate all this technobabble for your average sub-gymnasium scholar

[Herra Tohtori]

Fusion = hawt.

[Kosh]

Holy ****. Can someone translate all this technobabble for your average sub-gymnasium scholar

Yeah, there's so much technobabble in this thread you'd think it was dialog from a star trek episode.

[perihelion]

Actually, Herra and Flaser have been doing a very good job of condensing extremely complicated material into layman's terms.  It is very difficult to appreciate just how hard it is to achieve a controlled fusion reaction without going into obscene detail.  You haven't even begun to see the technobabble yet.  Try taking a course in plasma engineering.  Just one.  For a semester.  It'll make your head spin, you'll never use anything you learn in it, but man, what a rush!

[Nuke]

Holy ****. Can someone translate all this technobabble for your average sub-gymnasium scholar

Yeah, there's so much technobabble in this thread you'd think it was dialog from a star trek episode.

unlike in startrek, theres actually math in the technobabble

[Stealth]

I never played Descent and I'am still consider myself a FreeSpace fan.

then i've got news for you.......... you're not a fan.

Who are you to decide that?

... a better and more experienced and respected fan than you...

[redsniper]