“The paper defined an ideal power plan that can output 1 kW per kg of weight.
This is currently well outside the realm of possibility, with the best ion thruster power sources coming at something like 10 W per kg and even nuclear electric propulsion systems outputting 100 W per kg.”
- "ideal power plant that can output 1 kW per kg of weight"
I wished I remember where I read about this engineering problem, because it's an entertaining one. The main constraint on your [kW/kg] past a certain point is heat dissipation—the mass of the radiators rejecting waste heat into the vacuum of space. Thermal radiation scales like [temperature^4] (a very fast-growing function), so that parameter's obvious—you have to scale your exhaust temperature as high as you can engineer. That's how you shrink the radiator size. And you still need to run a heat engine—you need a significant temperature gradient, on top of the already-high exhaust temperature, to get useful work out of it. That's the temperature output of your primary heat source. So, the high-level design solution is: you have an array of infrared radiators glowing red, and a nuclear fission reactor glowing orange. That's the way to get a high power/mass ratio in space.
Also, everything's built around pipes of molten metals of different species (optimized heat transport), and the heat engine is like a steam turbine that spins on boiling molten potassium metal. (I think?) They're really exciting-looking machines. I wish someone would build one!
Is this where a Nuclear Thermal Rocket[1] is more efficient by mass as hydrogen running through the reactor might take some of the heat away, reducing the need for radiators? This is such a fun topic.
Yes, and even chemical rockets make use of the exhaust as effectively coolant. But the problem is that then you have to carry enough mass to carry away heat. And suddenly your rocket is 99% expendable coolant and only 1% payload (or less).
Or to put it another way, nuclear-thermal rockets might have an Isp of 3000 or so, which is amazing compared to a chemical rocket which might have 1/10 of that ... but an ion engine can have an Isp of 30000-70000, which is what makes a 550-AU trip even remotely practical.
I don't have the math right now to estimate how hot a conventional combustion chamber would have to be to get ion-engine-like exhaust velocities through a rocket nozzle/bell, but I think the answer is probably "ridiculously hot".
1 kW/kg is an ideal power plant. Does the paper define a minimally-viable one?
If we beef up the chemical stage, e.g. by launching on Starship and re-fuelling in LEO, can we make do with 100 or even 10 kW/kg?
(Also, to put 550 AU in perspective, Voyager 1 is 165 AU out [1]. At 38,000 mph Voyager 1 [2] travels about 3.6 AU/year [a]. Going straight out, it would reach the Solar gravitational lens in 2131 [b].
NASA has designs for thermoacoustic power plants that are much denser than this, but I note that they seem to be conveniently ignoring the weight of the fuel tanks in those calculations. If the generator were for instance 2% of the weight of the system, that means it has to be 50x as powerful as the goal.
With this unrealistic ion thruster method they estimate 13 years, whilst the realistic solar sail Turyshev method, which is actually planned, needs 17 years. Just the payload is much less.
This is pure science fiction, with no real practical benefit.
There is a very high quality video about how Solar Gravitational Lens could be used to map exoplanets, and full explanations about the images reconstruction and engineering challenges: https://youtu.be/NQFqDKRAROI
I've been aware of gravitational lensing before but this is the first time I've heard of it being used to resolve exoplanets.
This is super exciting and needs to be made a higher priority than it is considering our only other way around this is to build gigantic telescopes far beyond our current capabilities.
Idea: achieve massive power provision by transmitting energy from a base station to the spacecraft using space lightning.
Transmitting power is not a new idea: lasers are the go-to example for this. Powering the craft with solar energy is another theoretical way of doing it.
My idea on the other hand is different. Imagine spooling a long wire behind the space ship and just transmit electricity to it the same way you transmit power to your hoover. Except instead of sending power up a wire, send it up as bolts of lighting through the ionised gas trail your ship is trailing behind it.
Unless you have two beams, it would have to AC. You'd have to keep the series capacitance low enough that a significant amount of power is conducted (which might kill it). Remember that free space impedance is 377ohms sqrt(mu0/eps0) so non-photonic power likes to become Lambertian radiation.
Another way to get a spacecraft to 550AU would be to use solar sails and ground based lasers.
You have a small spacecraft (USB memory stick sized). Do a sub-orbital launch to get it out of the atmosphere. Then deploy the solar sails and fire ground based lasers at the craft to accelerate it. IIRC the theoretical maximum acceleration gets a you to 550AU in a few months.
While you could get a _platform_ there in those months, “what’s the point of it”? You would like to get a _payload_ there able to do science as well, so a USB Stick sizes spacecraft may not be suitable (for now) for it.
I can however imagine a proper ion-thrusted satellite with enough Antenna gain to communicate from 550AU back with real science.
I could be wrong, but these two things don’t seem like they’re really that comparable. Apollo was certainly a monumental engineering achievement, but did it require that we 100X the state of the art efficiency of some critical tech?
> Apollo was certainly a monumental engineering achievement, but did it require that we 100X the state of the art efficiency of some critical tech?
One order of magnitude in propulsion.
When Kennedy made his “We choose to go to the Moon” speech [1], our most powerful rocket was the Saturn I. Its H-1 engines thrusted at 200k lbf [2]. The Saturn V’s F-1s did 1.5mm lbf [3]. (The Saturn V, similarly, could lift an order of magnitude more mass to LEO than the Saturn I.)
It wouldn’t surprise me to find 100x increases in some material’s performance, et cetera.
So I specifically called out efficiency because 100X’ing your thrust is about scaling up technology. That generally involves solving engineering problems—figuring how to control vibration resonance would be one example in the case of the Saturn. 100X’ing efficiency, it seems to me, is another animal entirely—it’s often about legitimate scientific breakthroughs. Like going in, you don’t even know for sure of its possible.
> 100X’ing efficiency, it seems to me, is another animal entirely
Saturn V had a specific impulse of 263 s [1]. NSTAR did 3,000 s [2]. That’s an order of magnitude improvement in efficiency in 30 years of low-effort improvements.
Starship should demonstrate in-orbit refuelling next year. That’s another 10x technology. Add on a solar sail and you’re in the realm of two orders of magnitude of efficiency gains with known technologies. (Three from Apollo.)
The main reason the Apollo program wouldn't happen today is safety:
The first Apollo test (unfueled) had three astronauts die due to an electrical fire! At the time Apollo 10 launched the lunar module wasn't finished yet (they wouldn't have been able to get back from the moon because it was too heavy). Apollo 16 and 17 by chance missed the 1972 solar storm by months - if an astronaut had been outside the Earth's magnetic field during the storm they have received a potentially lethal dose of radiation. One reason why after Apollo 17 the rest of the flights were cancelled even though the rockets etc. were already ordered or even built and why it took until 1981 (apart from Apollo–Soyuz) to get humans into space again.
Common philosophical fallacy: just because we are able to do things in one stage does not mean we can do things in the next.
Of course, I am an optimist, but one cannot relate historical circumstances in the same way. I will be glad if it does happen of course, but I do not expect it to be so based on past performance.
>2.5% of the US GDP ($26 trillion in 2023) would be 600 billion. At its peak in 1967, the Apollo program budget was 3 billion, while the US GDP was about 850 billion. So 0.35 percent. US government spending in 1967 was about 112 billion, so closer to 2.5 percent of the federal budget, not the GDP. Converting to today’s 6,000 billion federal budget, about 150 billion today, or not quite 20% of the defense budget, the largest federal expenditure after Social Security (the defense budget is essentially tied with Medicare).
I’m not sure we want those sort of expenses anymore.
Because America was lagging behind the Soviet Union in space achievements, and the Soviet Union was parading their superior space program to promote Communism.
It's not a generational thing, it's that the Moon landing was a top priority Cold War effort to beat the Soviets and show that Capitalism is the best. This mission in the other hand would be neat but has limited political value. What money were are willing to spend on space will mostly be spent on having a permanent moon base before the Chinese.
The term 'capitalism' is often used as a smear by the adherents of 'opposing' ideologies like socialism and communism so let's agree on a definition using Britannica Money's example [1]: capitalism, economic system, dominant in the Western world since the breakup of feudalism, in which most means of production are privately owned and production is guided and income distributed largely through the operation of markets..
Given this definition and weighing the positives and negatives it still seems to be the best system, something which I do not see changing as long as humans remain in control of society.
Do you have any examples which show where another system has been proven to be superior at a large scale? That - scale - is an important factor here since there is a direct relation between the scale of the group and the applicability of economic systems.
The problem of capitalism is the instability of "the operation of markets".
Especially after 2000, there has been an extreme consolidation in almost all markets, so that they are controlled by quasi-monopolies and the operation of those markets resembles more and more every year with the economies of the former socialist countries.
The economies of the former socialist countries were pretty much identical with the end stage of evolution of the capitalist markets, when a true monopoly controls each market.
When the dominance of Russia has collapsed after 1990, for about a decade there was a huge hope of improvements that would lead to economies everywhere functioning according to the ideal "free markets" theories. However, to the dismay of those liberated from the Russian influences, the Western countries have evolved since then to resemble more and more the systems that they were formerly criticizing, not only in the monopolistic markets, but also in mass surveillance, whistleblower punishments, great discrepancies between what the politicians say and what they do, politically-controlled Supreme Courts of Justice and so on.
lindburg just shows up unanounced
at the cape ,security did not call ahead
just escorted him up to the main deck
everything stops,he hangs out for a bit,heads on his way
and then they get back to work
there might not even be a photo,and so you have to
trust that it happened,and in that is a large part
of how shit got done,on trust
I would have thought a direct fusion drive might be the way to go where you fuse something like helium-3 and deuterium and then just let it escape put the back to propel the rocket. (https://en.wikipedia.org/wiki/Direct_Fusion_Drive)
Admittedly it's hypothetical and no one's built one but it would be quite similar to the thing Helion is building which is theoretically supposed to make electricity this year. Although many are skeptical about that.
Take us -to- it, or take us past it? It’s one thing to go zooming past a point and another thing to get halfway there then turn around and slow down and park there. A telescope would almost certainly want to park there.
Do you honestly think this wouldn’t be taken into account? I’m sure even without reading the article this is exactly the sort of thing that will be accounted for in any serious proposal.
It’s not as if our species doesn’t have millennia of experience with sailing ships to compare to when evaluating how to manage navigating to a specific point using either a powered or sailing vessel, or decades of experience maneuvering spacecraft at high velocities with high precision using both thrusters and gravitational slingshots within the solar system.
Then you have to develop new comms that can travel that far with enough bandwith for photos. The lag could be a week or so. Or you do relays which is equally difficult in a different way
I think that's only a challenge at all because of how mass and power limited such a craft would be. The obvious solution is to use lasers.
It might even be possible to use a modulated retroreflector on the craft-- e.g. we fire a high power low divergence beam at the craft and it reflect it back. By modulating the reflection in can send back data with practically no power usage.
The angles accepted by the reflector could be fairly hard and so the difficult pointing problem would only exist on our end.
By far the most realistic engine for deep space travel is the Orion project [1]. You load a large rocket with lots of nukes, and detonate one at a time behind a pusher plate.
Of course, humanity being what it is, we'll never trust each other with the idea of building thousands of nuclear bombs with the "firm promise" that they'll only be used for space travel.
Nuclear explosions release mostly heat and radiation. Are you turning your pusher plate into high energy plasma with the explosion and using that to propel your spacecraft? My gut feeling is that your total ISP for this is disappointing compared to the amount of mass you add for the huge store of nuclear bombs.
> Are you turning your pusher plate into high energy plasma with the explosion and using that to propel your spacecraft?
I think the idea was for the bomb to vaporize a certain amount of propellant. According to wikipedia, the propellant was supposed to be tungsten, but I imagine that any substance would do. For example ice. The vaporized propellant hits the pusher plate and is reflected, resulting in an exhaust jet of very high velocity. The ISP was initially calculated to be between 4000 and 6000 seconds (so 10 times higher than the Space Shuttle), but later when they did the calculations with fusion bombs they concluded that an ISP of 75000 seconds is possible.
Not sure if this is the same idea as the one above, but the ones I’ve heard send the nukes up separately, so that the final launch craft is not impacted by the mass of the nukes. Instead as the craft passes by each nuke, they are detonated, and so it accelerates.
This plan is somehow even more insane than the original one. Now you have to launch bombs into space many years before launching your craft, then somehow fly in absolute precision with them to pass close enough to get an even push but not so close as to collide.
You need vastly different design parameters for Earth orbit, interplanetary and interstellar craft and masses corresponding to each.
Starship is all about a lot of mass to Earth orbit. This is a little mass somewhere between interplanetary and interstellar design parameters. Yes, Starship could put it into Earth orbit, when expendable probably interplanetary, sure.
And it may well help launch it, but that’s where its relevance ends, it’s at the other end of the spectrum of what you would design for. For example, at this corner of the design space, chemical propellants aren’t a thing.
>And it may well help launch it, but that’s where its relevance ends.
You can expand in parallel. If Starship works you can launch 100 spacecraft instead of 1, or in other words, you can expand your system by a factor of 100 "in parallel" and hence increase whatever you get from that small payload margin by a factor of 100.
Besides that there are other approaches like laser array + photon sail which more directly benefit from mass to LEO.
> But nuclear reactors aren't usually anywhere near 7 times heavier than their fuel.
On not on earth, but Soviet spatial reactors[1] weren't too far from that:
> The fuel core of the reactor was 0.24 m in diameter, 0.67 m long and weighed, as an assembly, 53 kg,[1][2] and contained 35–50 kg of enriched uranium. The entire reactor, including the radiation shielding, weighed 385 kg.
The problem is yield, out of 100kW of thermal power it was only able to generate up to 3kW of electricity due to lack of efficient cooling in space.
Hum... You can make things about 20 times better by enriching the uranium more.
But then you'll get into severe storage and control problems. And that thing has to work for 13 years, untouched. There's a maximum somewhere on the middle.
Anyway, I don't think reactors on earth are anywhere close to 140 times the mass of the fuel either. And they don't have to use radiative cooling.
Nuclear plants on Earth are way more than 140 times the mass of the fuel, but that's mostly concrete for the radiation shielding. If you're only worrying about the core and cooling infrastructure it's not nearly as bad. But of course as everyone has mentioned cooling things in space is hard and you'll want to minimize the number of moving parts because maintenance is impossible and you can't use convection to move heat around which makes it even more difficult.
Remember that on Earth nuclear reactors create electricity by boiling water to turn turbines. Such a system will be far more difficult to design for space.
Is there anything like a "space nuclear fission reactor" in the first place? My understanding was that all the so called "nuclear reactors" in space where jus powered by radioactive decay of short-lived material, not by fission.
