Nuclear power gives you more energy, but you're still limited by how much reaction mass you can carry.
The plans that work look like early 1960s NASA plans - build infrastructure in orbit, assemble nuclear power interplanetary craft in orbit, nuclear power is from planetary orbit to planetary orbit only. That was Wernher von Braun's "Man Will Conquer Space Soon" plan.[1]
The Apollo program, at US$20 billion, was the economy version. To the moon and back, no space station, no permanent infrastructure in space.
You couldnt do it without publicizing. Space stuff is big and loud. Space itself is transparent. Doing big secret stuff in space is extrodinarily difficult.
You can't hide that your about to do something in space because launches are loud and obvious. However, once in space, you can absolutely hide what you are actually doing. We still don't really know what the Air Force Space Plane does while in orbit.
There are spy satellites that we know nothing about in orbit, only until someone like Trump leaks an image on twitter. Who is to say where that line of development ended, if ever, given so little is spoken publicly of these efforts? Perhaps others could detect some of your efforts, but you hold them to secrecy like you do other state secrets among allies out of mutual benefit. I'd expect there to at least be a well worked out plan for such an infrastructure along these lines by now, if not implementation. It's one of those things where the losing move is not to engage with it, for fear someone else is and making strides.
Except we know when they were launched, thier general orbital parameters, and what they are generally doing. We dont know the specifics but their existance as spy sats is well understood. There is even an online community of enthusiasts looking for and tracking them. Wikipedia is full of data.
I think advancing science sometimes comes with risks. I would welcome a friendly arms race for better access, but who knows what would actually happen.
Nuclear doesn't work with Elon's approach to development. You can't blow up multiple nuclear rockets within the atmosphere just to learn what needs fixing for the next attempt.
Then again, it's looking like the whole 'fail fast' approach is over anyway, with Starship grounded by red tape.
A full-stack Starship failure is seemingly too big/dramatic/risky to get away with doing repeatedly, especially admidst the politics of Twitter/X.
- "You can't blow up multiple nuclear rockets within the atmosphere just to learn what needs fixing for the next attempt."
Sure you can. Russia has blown up multiple experimental nuclear-powered cruise missiles in the last few years [0,1]. It's a political question of tradeoffs, of how much you value technological progress on a particular front.
America hasn't blown up any atmospheric nuclear reactors in this century. America is also unlikely to learn how to build aerospace nuclear reactors in this century.
Nuclear rockets are good in space, not in atmosphere and they aren't radioactive until you turn them on. You test in space. Also that isn't "Elon's approach to development" that's just how everything good has ever been made. Also if you plan to fail then you usually wind up failing more safely than if you don't plan to fail.
A reactor that has never been turned on isn't a significant radiation hazard. It's the fission products that are hazardous, not the fuel, if it's never gone critical there are no fission products yet.
How much under? I’m guessing the real fear is contact, which then begs the question of dose over time.
Because as I’m sure you know and can see where I’m going with this, you’re already living under an enormous amount of lethal radiation, you and everyone else has been for their entire lives… its called the van allen belts.
Assuming Elon gets Starship working reliably, which will probably happen, you could use it to lift a nuclear rocket to orbit. You could then fire the nuclear bit up away from the earth and use it to shuttle around the solar system.
Starship is planning to launch soon according to ars technica, it was also posted to hacker news a few hours ago. So not too bad considering the red tape and damage to the launch pad.
You can. You just can't legally do it in Texas. Maybe SpaceY will be based in North Korea or somewhere to escape the burdensome regulatory regime in the USA.
He could fund a new company elsewhere in the world and start from scratch, perhaps.
But even if he was super-careful about ensuring that restricted information wasn't shared with the new company, I'm sure his political enemies would find ways to have him locked up if he tried that...
AFAIK, the only nuclear rocket that makes sense is landing a reactor on an icy comet and using it to shoot super heated steam wherever you want thrust.
Project orion was super cool and I warmly recommend the book by similar name by George Dyson (son of Freeman Dyson!) who did a tremendous job of putting everything back together as well as humanly possible for an ops with very top secret components.
Nukes can totally function as the energy for rocket’s impulse. Very small and very directed nukes. The directed explosion plasma wave hits a pusher plate, where very specific plasma physics play out - the plate is not vaporized but merely receives an impulse - and then there are dampers to smoothen the blow.
Half of the interviewed experts were positive it would work and half were skeptical.
So, it’s sort of sad they had to wrap it up due to nuclear test ban before actually being able to run actual tests.
The best part? The bigger it is, the better. We could totally have gigantic nuclear space cruisers that go to Jupiter in like a week (I exaggerate because I forget the exact orbits but with ISP from nukes that bastard goes fast)
In the interviews some of the team members were concerned with proliferation of shaped-charge mini-nukes. IMO that's not a good reason not to do anything but not an expert.
The Project Orion design (shoot hundreds of fission bombs out the back and ride the shock wave on a reaction plate) is surprisingly practical, and probably our only way to get humans past Mars. It's just kind of terrifying.
The energy required to accelerate a useful amount of mass to a high enough velocity to travel interplanetary (or interstellar) distances in a timely manner is always going to be terrifying. Even if it runs on rainbows and hugs the kinetic energy imparted to the payload would turn it into a planet killer if you ever rammed it into one. Managing that risk so you get the payoff without any accidents along the way is the only sensible thing to do.
> The Project Orion design [...] is surprisingly practical, and probably our only way to get humans past Mars.
I don't think this is the general consensus. It is seen as one of the only ways to get humans out of the solar system, but other nuclear rockets[1], and even chemical options[2], could get humans to eg the Jovian moons[3] with TRL's much higher than Project Orion.
It's not just about propulsion, but radiation. The radiation environment in space is awful and you need thousands of tons of shielding to effectively screen a crew from cosmic rays. Moreover, because of secondary radiation, partial shielding from energetic cosmic rays is worse than none.
A unique property of nuclear pulse propulsion is that the engineering gets easier the more massive the spacecraft. That leaves lots of room for radiation shielding and consumables, and makes it the only halfway practical choice for multi-year missions into the outer solar system.
> That leaves lots of room for radiation shielding and consumables, and makes it the only halfway practical choice for multi-year missions into the outer solar system.
I disagree, one of the types I referenced is cyclers, which are specifically good for the same kind of reason. NERVA is another alternative that becomes similarly easier as you scale it up. All three types would need in-space assembly, so no big issues with any of them that way.
Don't get me wrong, Orion is cool (and the only way to conceivably run an interstellar mission with known technology/physics), but it isn't necessary (or even desirable imo) for interplanetary missions, at least as far as saturn anyway.
It requires something like several hundred g/cm^2 of polyethylene for adequate GCR shielding. So back of the envelope, if the crew compartment is a cube 10 meters on a side, it requires 500g x 6,000,000 cm^2 = 3,306 tons of plastic to shield. I'm not aware of any non-nuclear-pulse design that can carry around that kind of heft. Just how big are the cyclers you mention allowed to get?
Should also note that the Orion design most people are familiar with is from the 60s, when the latest in engineering used vacuum tubes. Modern innovations in material science, computer assisted design, and just technology that just simply didn't exist yet such as laser ignition would permit ditching the ridiculous pogo stick pusher plate arrangement with something that's more efficient, runs smoother, and looks a lot more traditional too. I wouldn't be surprised if a modern variant looks like that one Sea Dragon concept with beefier nozzle suspension
If someone siphons your gas tank, they're not going to be able to nuke Paris. But a nuclear pulse rocket needs thousands of bomblets the size of coke cans. That's the scary part.
The summary of this methodical and enlightening article from an excellent source for mathematically backed science fiction / space travel is:
No. Nuclear is not worth it for starship, as the added engine weight limits the additional delta V and the additional challenges of handling nuclear material.
The no accompanies the short explanation of a nuclear planetary transfer vehicle that never lands on earth. A lot of conceptual work has been done on these designs, and Andy Weir’s The Martian features one.
Nuclear powered space tugs is a particular rabbit hole I’ve spent too much time on.
This is seriously underselling what US nuclear rocket program of the '60s [1], by calling everything that came after "modern". That program was the only one that built something, everything else was speculation on paper.
The obstacles faced by the NERVA project were immense, and the iterations achieved spectacular improvements in record time. That project that lasted for less than one decade and was done on a shoestring budget is not quite the equal of the Manhattan project, but it's not much below it. It achieved the highest density of power generation by maybe a factor of 100 compared to any other reactor in history.
All that stuff was pretty much lost. NASA is trying to revive this, but it will not have legs. We simply don't have the same pool of talent to recruit from. In the 60's there were lots of scientists and engineers that had first hand experience building nukes and nuclear reactors. Now this country has not built a reactor in 3 decades, with the exception of Vogtle.
First I am reading about nuclear engines applied to Starship. I wonder if it would be more practical to launch the reactor, fuel, and propellant into orbit using conventional rockets. Later a mission could pick up this Nuclear rocket engine and use it as a sort of booster while not inside a strong gravitational field. Ofc that doesn’t sound like Starship at all.
Nuclear rockets are useless for launching things into orbit. They don't have the thrust-to-weight ratio to get off the ground. Nuclear rockets are useful because they have high specific impulse which means using less fuel for thrust.
However, in KSPI [1]: Consider the combination of a nuclear lightbulb with compressed air as the reaction mass: Surprisingly good midpoint qua TWR and ISP! If you install a compressor you can get some pretty interesting SSTOs out of it. Whether these would work IRL is a different question, of course. [2]
Nuclear lightbulb is of course highly theoretical, but scientifically/engineeringly plausible at the least. [3]
Did you read to the end? That's the idea he ends up with: only use nuclear propulsion for space-to-space travel where you don't care as much about shielding.
I make no claims about my reading comprehension. That last bit seemed too narrow to me. Once you have all the bits in orbit, shielding doesn’t seem too onerous. Is the claim that even space-space shielding + reactor is too heavy to be worth the advantage of the nuclear drive?
No, they are saying that the nuclear drive shall be reserved to a shuttle that remains in space. At that point, it makes much more sense as you can really use the nuclear drive to its full capabilities.
> If we need the full performance advantage of nuclear propulsion, we should design a spaceship that is intended for it from the get-go. It never lands, only going from orbit to orbit, so there is no need for heat shielding, flaps, high thrust engines, thick steel structure or aerodynamic shaping requirements.
>If we need the full performance advantage of nuclear propulsion, we should design a spaceship that is intended for it from the get-go. It never lands, only going from orbit to orbit, so there is no need for heat shielding, flaps, high thrust engines, thick steel structure or aerodynamic shaping requirements.
I think we'll see nuclear propulsion for Starship variants that receive fuel rods in orbit and are not designed for reentry. Interplanetary logistics between Earth/Moon/Mars orbits.
If he really wanted to, he's rich enough to build the enrichment facilities himself. Could hide it underground in one of those tunnels, claiming the power consumption is a cryptocurrency mining rig or an AI server farm.
I suspect at least ten government agencies from various nations all have him/his assets under 24/7 surveillance just to make sure he doesn't do something at least this unwise.
If SpaceX was willing to do the work and put in the investment they could absolutely get access to fissile material. The US government evaluates things based on regulations and not who is popular on twitter.
Starship and the booster already collectively carry 70 kiloton-tnt-equivalent masses of liquid methane and oxygen. As a weapon of mass destruction, it's hard to beat relativistic impact.
How do you calculate this? Starship has a propellant capacity of 1200 metric tons - 1.2 kilotons. Super heavy booster has 3400 tons, total of 4.6 kilotons, per wikipedia [1]. By weight, you need about a 4:1 ratio of O2:Methane for complete combustion (Starship is rumored to carry 22% methane). Which means you're getting about 0.92 kilotons of methane completely combusted. At 50-55MJ/kg, that gives you - roughly - 50 thousand gigajoules. A ton of TNT is 4.184 thousand gigajoules[2], which means that Starship+Super Heavy carries the energy-equivalent of about 12 kilotons of TNT [3][4].
[3] It can't explode all at once unless pre-mixed in perfect proportion, of course. What you'd really expect is an awfully big fireball + fire but not the same massive force all at once of a nuclear detonation.
[4] Or 50 terajoules (the ship) and .184 terajoules (TNT), which come out to the same thing, but I find gigajoules both more intuitive and more a more fun BTTF reference.
> As a weapon of mass destruction, it's hard to beat relativistic impact.
> relativistic impact
You... er, realise they're not actually starships, right? At the speeds they go at (or would go at if they could take off without exploding) there is no significant relativistic effect.
Surprisingly little, as the thing is mostly fuel — by the time it hits anything[0] it's almost empty and therefore very light.
It wouldn't even be like the effect of the planes on the WTC in 9/11, as planes use fuel continuously during flight, whereas spacecraft use almost all of it just to take off.
[0] which implicitly assumes the self-destruct still isn't working right
Since this is presumably for interplanetary, and not heavy lift, I'm not sure why you'd use this over just having thin-film solar + pulsed plasma thruster (which can do anything from 1km/s to 3000km/s exhaust velocity).
I only skimmed the article; but as I understood it the liquid hydrogen here is being used strictly as ejecta. It is not being combusted, the way that it is used in the space shuttle boosters.
If that is the case, liquid hydrogen seems like an awful substance -- hard to produce and transport, hard to store and the mass required to support keeping it is not insubstantialy, all for a mass density of ~70g/L.
Why not just use water? Water is 1000g/L and can be stored without any real effort at all. If all you're doing is shooting it out the rear end of the rocket with as much energy as possible, it's mysterious why you would even consider liquid H2.
The article covers this. Virtually all nuclear engine proposals have used H2 as propellant because of the specific impulse, but this article mentions methane as an alternative. Of course, Starship today uses methane and Elon plans to create methane on mars.
> Most significant for our purposes is methane as propellant. It is six times denser than liquid hydrogen, can be stored at 100K, which is compatible with liquid oxygen, and it can be produced using water and carbon dioxide. At high temperatures, it breaks down into hydrogen and carbon, turning it from a 16 g/mol molecule into a 3.25 g/mol plasma. That is how it achieves a specific impulse only mildly lower than what is achievable using liquid hydrogen. Zubrin lists its specific impulse as
>Previous calculations using hydrogen propellant revealed how volume-limited the Starship design was. There was no room for the bulky liquid hydrogen, and getting to orbit meant sacrificing the payload mass and volume advantages that the Starship is built around.
>These could be solved by using denser liquid methane as propellant for the nuclear propulsion system. The Isp will be lower, but the mass ratios become so much better that more deltaV is available overall.
Shooting just electrons out the back of your ship is fine for the brief period before your ship gains a positive electric potential comparable to the net potential those electrons experience due to the inside of the particle accelerator you were using, after which your ship's own field is a major source of drag.
If your ship has a positive electric potential of over 1.044 MeV[0], you also start getting positron-electron pairs forming on your hull.
[0] less in practice, because any electrons you're going to encounter were already moving
Don't ionic thrusters typically solve this problem by also emitting oppositely charged particles from a thin rod behind the motor? Maybe that still causes drag, but overall the main problem is that while the ISP is high, the thrust is almost non-existent.
They do indeed; the oppositely charged particles being electrons. But there's no charged particles lighter than an electron, and the only one with the same mass is the positron. If you try to balance it with protons, then what you've really got is a slightly over-complicated hydrogen-based ion drive.
First, you have a way to store an absurd quantity of positrons. For a sense of scale, without shielding the electric fields, 1 picogram of positrons (or electrons) confined within a 10 cm radius is going to trigger positron-electron pairs formation, thanks to free electrons in the area responding to the surface potential.
Second, AFAICT if you can do that then you're either going to want to use them as an energy source to propel your cheap reaction mass even harder than a fission rocket would, or you're going to want to react them with electrons to make a photon rocket.
- You'd need to account the mass of the copper-63 waste, in addition to the electron
- This math is specifically about thermal propulsion; your 17 keV electron wasn't in thermodynamic equilibrium with anything, it's a different sort of problem
That's not quite what I'm picturing (and let's be honest, the idea is silly). That page describes capturing the heat of a nuclear reactor, converting it to electricity, and then using that electricity to force ions out of a thruster. I'm saying, the radiation itself has momentum; don't bother with the contraption.
I need to play Kerbel Space Program. I have absolutely no intuition about this sort of thing. Some intuition around orbital mechanics from my physics days but rocketry is a bit of a mystery.
For me it helped to remember that the term that matters in the rocket equation is exhaust velocity, not momentum. You want to be shooting the lightest things you can out the back of the rocket, at the highest speed.
Unfortunately with chemical rockets, the energy source and reaction mass are the same thing, so you're kind of stuck with whatever burning your fuel gives you. But when you can separate reaction mass from energy source (like in nuclear or electric rockets), hydrogen is always the best bet.
Mostly true. However, there's tradeoffs qua thrust and specific impulse for different propellants. For pure vacuum work hydrogen is a good candidate. For operating in a gravity field (eg. the moon), you might be able to use a smaller nuclear engine with a denser propellant. Obviously you ARE trading in some range when doing this. But if the question is whether you can get off the ground in the first place, you may have less of a choice. It might also depend on what you can actually lay your hands on qua ISRU (if/when relevant)
I think all real-life actually-tested nuclear thermal rockets have used hydrogen as propellant, though use of other substances has definitely been proposed.
Water is annoying: when you heat it, some of the energy goes into dissociating it rather than accelerating it. In the end you apparently get pretty bad performance out of it.
In deep space you don't really care about the volume of your fuel tank, so then it does make sense to use something light like hydrogen.
It's not a stupid question. Space is already highly radioactive, with lots of protons and bigger nuclei from outside the solar system flying around in all directions, along with a ton of proton radiation from the Sun.
Most nuclear spaceship designs have a long stick with the reactor at one end and the crew on the other, and some kind of shielding in between. Since you only need to shield a cone, it can be quite effective.
Yes, but none of that changes the fact that when we talk about fission rockets, we have to think about nuclear safety on top of everything else. (All of your comment is thinking about nuclear safety.)
And if we use them inside of a planet's magnetosphere, we have to think about environmental contamination too.
Yes, but space is already full of radioactivity. An incredibly large amount of it. The average range (outside of the upper Earth's atmosphere) would be equivalent to getting somewhere around 10 chest x-rays per day, every day.
More or less, yes. Space is so large and so rarified and so riddled with energetic particles and photons that the contribution of a few measly rockets is of no concern.
I don't know why you're downvoted. I think it's a legitimate question. When I was in high school I asked "Why don't we just shoot our nuclear waste into the sun?".
As an adult, Kerbal Space Program has taught me that it's incredibly difficult to lose the orbital velocity and get things into the Sun!
But the answer my teachers told me was "Well, even if rockets have a 99.5% success rate and a 0.5% failure rate, and only 1 out of every 200 rockets explodes during launch... then eventually a rocket carrying nuclear waste is going to explode and spread it everywhere"
As others have pointed out: Nuclear rockets don't have the thrust ratio necessary for take off, so presumably they'd be assembled and used in space. My question is: What level of risk does launching nuclear rocket parts/fuel into space pose?
A rocket explosion is unlikely to pulverize solid nuclear fuel or vitrified nuclear waste. It would most likely fall, intact or in large pieces, into the ocean which is a pretty good place for it anyway. If it blows up over Florida that's worse, but the worst case scenario is that it reenters earth's atmosphere at near-orbital speeds over a populated area and burns/breaks up. Then you have a situation like Kosmos 954, where the radioactive material is spread over a large area and is consequently much harder to clean up.
The biggest issue for the nuclear waste disposal idea is that it doesn't make any economic sense. It's better to dump it into the ocean, or down a mineshaft, or even just to let it sit in storage near the power plant for decades. But with nuclear fuel for rovers, satellites or rockets the rewards could outweigh the risks, and so this is sometimes done.
> The General Purpose Heat Sources (GPHS) inside the
MMRTG is designed specifically to prevent such an occurrence. The fuel inside each GPHS is surrounded by
several layers of protective materials, including the type
of tough material used in the nose cones of missiles
designed to survive fiery conditions during re-entry into
Earth’s atmosphere. In addition, the radioisotope fuel
is manufactured in a ceramic form (similar to the mate-
rial in a coffee mug) that resists being broken into fine
pieces, reducing the chance that hazardous material
could become airborne or ingested.
> The biggest issue for the nuclear waste disposal idea is that it doesn't make any economic sense.
No the biggest issue is that its a stupid idea in the first place. Nuclear waste is fine and perfectly reasonable to handle here on earth with close to zero chance of actual danger.
Yes, but nuclear reactors carried up to space for rockets or electricity can still make sense. Before you turn on your reactor it's not very radioactive, to survive to the present day uranium has to have a very long half-life which means low radioactivity. But as soon as you turn it on you start transmuting the uranium into all sorts of things with much shorter half-lives and so much higher radioactivity.
So its safe to fly a new reactor into a stable orbit, if something bad happens you'll scatter some uranium but that's just toxic heavy metal bad, not radiation hazard bad. Just make sure the orbit is high enough to be stable, not like Kosmos 1402.
Why would you dispose of such a useful material by attempting to shoot it into the sun? Sure, there is some danger associated with storing it, but it's not as if nuclear material is the only hazardous substance on the planet, and with the rest of it we seem perfectly ok with just sticking it in a hole somewhere. Nor does nuclear material, by itself, represent a danger of civilisation-ending proportions (only if you build a bomb out of it, and that is quite a significant operation that can only be conducted by people that know what they are doing. In other words, it won't explode if left by itself.).
Amazingly, this project (nuclear third stage for Saturn V) was actually under serious development. This to be used on a rocket stack which had had two major incidents in 13 launches; it really was a very different safety culture back then.
This article addresses that. They claim that, like you said, they are only used at very high altitude or in space, and the raw materials, before ignition are not dangerous. They say that the low enrichment uranium is even safe to touch with your bare hands.
The sun is a giant fusion reactor known to cause cancer on a good day, and emit radiation bursts that can damage or disrupt power grids often enough that people can reasonably expect to read reports of them more than once in their lifetimes.
Even then. The sun is always a big ball of dangerous radiation, even despite things like the ozone layer. A nuclear rocket taking off isn't something you want to be near, but even literal Chernobyl wasn't as deadly in total as skin cancer was just in the year the reactor exploded.
So here is my science fiction theory: fusion will never be possible, but fission already is. So in the galactic civilization, rare uranium is the most valuable resource since it enables interstellar travel. Uranium is only produced in the collisions of neutron stars, so we are lucky to have any.
We should not squander our uranium for power generation when we can use solar for that. It's like not using oil or gasoline for heating.. we need it for transportation.
> We should not squander our uranium for power generation.
I think we might have enough.
"He noted that fast breeder reactors, fueled by naturally-replenished uranium extracted from seawater, could supply energy at least as long as the sun's expected remaining lifespan of five billion years"
Nuclear power gives you more energy, but you're still limited by how much reaction mass you can carry.
The plans that work look like early 1960s NASA plans - build infrastructure in orbit, assemble nuclear power interplanetary craft in orbit, nuclear power is from planetary orbit to planetary orbit only. That was Wernher von Braun's "Man Will Conquer Space Soon" plan.[1]
The Apollo program, at US$20 billion, was the economy version. To the moon and back, no space station, no permanent infrastructure in space.
[1] https://www.youtube.com/watch?v=Sz7njI0wEIw