Princeton Satellite Systems – $100 Million Plan for a Fusion Space Drive by 2027

Direct Fusion Drive (DFD) FD is a conceptual fusion-powered spacecraft engine. It should produce between 5-10 Newtons thrust per each MW of generated fusion power, with a specific impulse (Isp) of about 10,000-20,000 seconds. They would have 200 kW – 2MW available as electrical power. Approximately 35% of the fusion power goes to thrust, 30% to electric power, 25% lost to heat, and 10% is recirculated for the RF heating. The design uses a specially shaped radio waves (RF) “antenna” to heat the plasma.

The ISP is about 20,000. 5 to 10 newtons are generated per megawatt of fusion power. They are looking at 1 to 2 megawatts for the initial space system. This would thus produce 10 to 20 newton using a 2-megawatt system. They want to get their machine cycle down to 3 years. They want to complete the system for less than $100 million.

They believe they can also make a pure energy generation system. They refer to this as closed loop mode.

It would produce thrust from fusion without going through an intermediary electricity-generating step. It will use magnetic confinement and heating system, fueled with a mixture of helium-3 (He-3) and deuterium (D), to produce high specific power, variable thrust and specific impulse, and a low-radiation spacecraft propulsion system. They will heat plasma to 100 keV (which is 1,120,000,000 degrees K)

The plasma is confined in a torus-like magnetic field inside of a linear solenoidal coil and is heated by a rotating magnetic field to fusion temperatures. Bremsstrahlung and synchrotron radiation emitted from the plasma are captured and converted to electricity for communications, spacecraft station-keeping, and maintaining the plasma’s temperature.

Princeton Plasma Physics Laboratory has licensed the technology for Direct Fusion Drive (DFD) a fusion-powered rocket engine that could take people on a mission to orbit Mars for 30 days with total trip duration of 310 days.

They would create a cigar-shaped plasma—the superhot, electrically charged gas that fuels fusion reactions—inside a cylinder that is some 20 feet long and could produce up to 10 million watts of power. Propulsion would come from the stream of high-speed fusion exhaust that would blast into space through a magnetic nozzle.

A test facility, based on a concept known as magnetic field reversed confinement, could be completed by 2022.

Princeton technology can cut costs of RF equipment and reduce power consumption in NASA’s existing RF-based laboratory research and aerospace manufacturing operations. The improved RF gear can be used for microwave-heated thrusters and communication systems up to S-band. The switching amplifiers are also said to be more efficient than current RF systems using tubes, such as Traveling Wave Tube Amplifiers (TWTAs) and Solid State Power Amplifiers (SSPAs).

Princeton’s novel amplifier will have commercial applications:

HF band radars for coastal and over horizon systems (OTH)
Medium (MF) and High Frequency (HF) Radio (ITU Bands 5 and 6)
Communications channels up to S-band
Materials processing
Plasma heating for terrestrial fusion reactors
RF heating for manufacturing

SOURCES- Princeton Satellite Systems, NASA, Tennessee Valley Interstellar Workshop, FISO Future In-Space Operations working group, SBIR grants, NASA NIAC
Written By Brian Wang,

41 thoughts on “Princeton Satellite Systems – $100 Million Plan for a Fusion Space Drive by 2027”

  1. Rather late, but still want to say this: you need to input X energy to get the fusion reaction to happen. You can’t invest that in advance. If your fusion runs at less than unity, then your exhaust carries less than X energy. So theoretically, you can get more Isp by investing the same X energy in directly accelerating the exhaust.

    BUT, that depends on what acceleration technology you have. If it’s efficiency is lower than the fusion drive, you’re better off going with fusion. Or there may be other practical issues, like size constraints.

    But if the efficiency of your accelerator is better than that of your fusion reactor – which is possible with a sub-unity reactor – and the other factors are acceptable, then you’re better off going with the accelerator.

  2. ok, so my original question was how is the exhaust velocity achieved ? And why on earth they include slides about using the thermal output….

  3. I’m having so much fun ⊕1’ing comments, instead of blithering numerically this very early morning…

  4. No, that makes it the perfect analogy.
    Because, (as stated by other commenters) if we can do fusion we can make our own He3, so that isn’t a limitation anyway.

  5. A great example of how bad presentations can be.
    Something old from the internet archive i guess 🙂

  6. I know that in my case, all errors of spelling, grammar, fact and logic can all be blamed on the autocorrect function.

  7. I give slightly more credence to Leif Homlid’s annihilation reactor based on hyper-dense protium and deuterium. Basically he and his team say they have found a way to make hyper-dense hydrogen with a catalyst in minute amounts, and that such hyper-dense protium or deuterium can produce very strong nuclear reactions when illuminated by a laser, akin to those found in matter/antimatter reactions producing pions, kaons and muons.

    It’s fringe science but very interesting and seems closer to have replications (they have shared their setup with others, with positive results) and practical applications.

    The topic has been brought here before, but Leif and his group have been very active producing more papers and additional replications of their work.

    For example, this analysis of the kind of Isp and speeds a rocket using this presumed reaction (it would be truly relativistic):

    Of course, take it with a grain of salt, but I think it’s worth it to keep an eye on it, to see if anything comes out.

  8. But the big problem is getting enough diamonds

    You need a different analogy, there is no shortage of diamonds as billions of carats are produced each year.

  9. But if the fusion reactor can’t operate at a positive net gain then the effective energy density of the fusion fuel is actually less than that of a stretched rubber band.

  10. the big problem is fuelling that baby up…

    No it isn’t.

    You’re watching the magician waving his wand and ignoring that he’s pushing something with his foot under the table.

    The problem is not getting He3. I know they say that this the problem, but they are misdirecting you away from the real problem.

    The real problem is that they, just like everyone else, haven’t got (controlled) nuclear fusion working. It’s all just science fiction at this point.

    This is like someone saying that they can magically transform a kg of diamonds into a magical, flying, free energy pooping unicorn. But the big problem is getting enough diamonds.

    NO, the big problem is that they can’t transform anything (except transforming research funding into nice salaries and a company car.)

  11. Well Princeton has never solved the micro instability problem in 30 years. This is all about more hype to get funding. Make the problem even harder than the one they haven’t solved. Sad. Scientists who don’t at least describe the scale of what they don’t know aren’t really scientists.

  12. Rossi’s intellectual property is a trade secret because even if the USPTO allowed patents in the “cold fusion” class, his trade secret was published long ago as Mills’s potassium electrocatalysis.

  13. This is one of the few situations where you need He3 rather than the much more common He4, so I don’t see that as a problem.

  14. My understanding is that *IF* you can get the d-He3 reaction to go you can get the d-d reaction to go. So you would have d-d reactors producing energy & both T & He3 in situations where a blanket absorbing the neutrons is fine to have. Then you have the He3 to use in the d-He3 (or even He3-He3) reaction in situations like this space drive.

  15. If you can get net power from D-He3 then fuel isn’t a problem. The D-D reaction is easier and its waste product is He3 (half directly, half as tritium which decays to He3 with a 12-year half-life). D-D emits neutrons but they’re at fission energies, not D-T fusion energies.

    Fusion startup Helium is working on a hybrid D-D/D-He3 reactor, saying it would only emit 6% of its energy as neutron radiation.

  16. Penning traps really are not capable of storing antimatter at densities high enough for interstellar travel. Maybe for interplanetary travel if the antimatter is used to catalyze fusion.

  17. It beats ion drives in terms of thrust (hundreds of Newtons) and ISP of 20k.

    So it is basically not like an ion drive.

  18. Paying the energy up front to generate antimatter could work, because the resulting store of antimatter (and matching matter) will be small and light and so can be stored on board your spaceship.
    If you want to pay the energy up front for your below unity fusion drive… how do you store that energy on the spaceship?

  19. Getting enough He3 is the easy part of this scheme.

    Well, let’s call it the least difficult.

  20. And here is a 15 minute power point talk on the fantastic meal that would result.

    We show multiple artists impressions of the elaborate shapes we could carve the ham into. We show the various options of fried eggs, boiled eggs, scrambled, poaches and baked.

    Here are some extra slides showing the high levels of protein the meal will provide.

    With slides showing that it is far cheaper than some rival schemes where they foolishly assumed higher prices for their steak and chips (Once they get steak. And Chips.)

  21. Doomarz

    That is quite a hand waving. The analogy doesn’t hold . For an antimatter engine the energy comes from the matter-antimatter annihilation. So the antimatter could be there energy store and the annihilation is how you release it. But when you claim a fusion reactor which is supposed to generate 2 MW thermal that energy had to come from somewhere. Forget the scarcity of helium 3. Forget all the pretty pictures and isp calculations. If you are under unity to have to constantly hat the reactors by microwaves. The helim3 + deuterium of your energy store. But you cannot release is because the imagined reactor didn’t exist. 

    Why yes about Spencer of you have zero proof it cold work here on Earth. And if it works why even bother with the ITER monstrosity. Just use this magic mini rescue that fits in a truck…

  22. I’m no expert, just interested in space travel. However it seems to me that you are correct about antimatter being the way to go. If we’re going to reach for the stars why mess around? If we can get production and confinement (say from super cooled Penning traps) working then we should just go with the most powerful systems we are technologically capable of to nullify the extreme distances involved in deep space travel. Oh, and engine designs capable of working with antimatter of course; we don’t want the ship to go boom.

  23. low thrust high isp, so its more like an ion or plasma thruster with better performance that also generates electricity on the side, pretty good stuff

    of course, gasdynamic mirror, inertial confinement and finally antimatter catalysed fusion drives will all leave this one in the dust, but this is a good starting place for the technology which will allow us to conquer the stars

    the big problem is fuelling that baby up, He3 is in scarce supply this close to Sol, we could get some from the Moon but i dont think this would last very long as solar wind only deposits small quantities into the Lunar regolith

    in the long run we will need either some kind of net that can catch He3 on the solar wind, or a high-thrust engine that can drive a ship through the upper atmosphere of Saturn or Jupiter scooping up He3 before returning to orbit (i believe a gas-core fission rocket would be best for that) and then somebway to send He3 back to Earth (i suggest laser-propelled sail-freighters, the lasers could be energised either from fusion powerplants or from power harvested from Jupiter’s magnetosphere via electrodynamic tethers in the Io plasma torus)

  24. Fusion-Enabled Pluto Orbiter and Lander

    Presentations: 2.5_M.Paluszek_CT2016.pdf

    Travel time to Pluto, New Horizons vs. DFD Mission

    9.5 years vs. 4 years

    Delta-V 290 m/s vs. 70,000 m/s

    The additional mass takes a large toll even with the larger diff in deltav.

  25. so if have underunity reactor how is an y better than simply heating and accelerating ions ? are fusion products accelerated more? can we even use some of the fusion yield?

  26. Let’s build a rocket that uses a fuel that (essentially) doesn’t exist. If He3 was extracted from all the natural gas produced in the U.S. (largest producer in the world), that makes about 5 kg per year. And it won’t be cheap.

  27. Continually, for 3 decades, promising but not delivering on endless clean energy paid for by wealthy investors.

    Oh, wait, that’s Randell Mills’s SunCell.

    The Princeton fusion program is so much more credible since it politicians investing our money for, not 3 decades, but 6 decades — without delivering a working system. But at least the politician-funded guys have told us how bad, evil and/or deluded Randell Mills and his investors are since they aren’t producing anything that comes close to being a commercial product.

  28. Yep. This kind of thrusters will eventually become the protagonists of Outer Solar System crewed visits and settlement. We haven’t many choices, if we want to reduce the time people spends in microgravity and bathing in solar storms and cosmic rays. And still, it looks as if the trips will take 1 or 2 years, if you want to go to Saturn.

    But there still is plenty to do before that. Like having a viable demonstrator that looks as if it can be scaled up.

    I believe more in this kind of spaceships (as far as they may still be) than in the mythical gigawatt laser light sails.

    But again, I might be wrong and Solar Power Satellites may scale up faster than fusion and then we could start pushing sails at ludicrous speeds. Fusion has a habit of always being 10 years in the future after all…

Comments are closed.