Why Not Space? [Op-Ed]

[K. A. Pital]

maybe I was not clear. Are they run with a profit or with a loss? By a private company or by the US/USSR government?

Bigger yet. World government. A real one.

Just below what you quoted I said:
-you use atmosphere as propellant
-the delta-v needed to get anywhere in a speed worth using Boom-Boom-Orion is 3-4 times that of reaching orbit.
Since delta-v to reach orbit is somewhere around 10-11 km/s and 3-4 times that is 30-40 km/s... Which is relatively crappy for a true high-speed transit, but should bring you to Mars in 3 months.

Anyway, you need somewhere like 2400-3200 bombs for a one-way trip to inner system destinations (Mars, venus, mercury, asteroid belt), or say 4800 bombs to go to Jupiter with something like 8 months of coasting. And you cannot do much better than that, my spreadsheet tells me you'd have a payload below 0 kg already.

The reference design was to be constructed of steel using submarine-style construction with a crew of more than 200 and a vehicle takeoff weight of several thousand tons. This low-tech single-stage reference design would reach Mars and back in four weeks from the Earth's surface (compared to 12 months for NASA's current chemically-powered reference mission). The same craft could visit Saturn's moons in a seven-month mission (compared to chemically-powered missions of about nine years).

http://en.wikipedia.org/wiki/Nuclear_pulse_propulsion
http://hem.bredband.net/ulfn/checklists ... rbiter.htm
http://www.oriondrive.com/

Something's not right here.

That is with the relatively realistic Orion designs, before they pull really weird shit out of nowhere.

Explain what "weird shit" is pulled out in these studies.

The kinetic impactors I talked of above can keep accelerating your ass for as long as you want them to (you can pull brachistochrones! Yay!), with a Ve which is a little less than double that of an Orion at optimal conditions, which means they basically beat the crap out of Orion performance-wise for a fraction of the cost (especially if you can make them with ISRU and just ship the electronics from Earth). The only difference is that you have to book the flight a year or so of time ahead of the departure, not even Orions can dramatically change their course.

Is it like the system suggested by the guy here?
http://www.bautforum.com/showthread.php ... e-NEA-ISRU

Also, the swarm of kinetic impactors can have military uses as well, since it can basically sweep away any orbital asset, and any asset on the ground not protected by an atmosphere. Also do a good job at intercepting whatever wants to hit the vessel that is using them as propulsion.

Takes a damn while to get them to the target, but it's something that can be used as space-equivalent of ICBMs for space assets that don't move much faster either.

I have no desire to fill my universe with weaponized space. Earth is united, humans live in peace and the next war is thousands of years away. But in any case the impactor system would rely on external control of the impactor swarm, which is problematic if we'd face interference of some sort. On the other hand, the Orion controls its own propulsion.

To solve the radioactives problem I'd either start fucking with thorium (it's relatively abundant in the solar system and on Earth, you can use it to breed U233 which can be used to make nukes or fissioned to produce power in a reactor, not so easy as it sounds but doable) or go to fully-fledged fusion bombs.

Good ideas, thanks.

If magnetically-confined fusion (which is a major pain in the ass) is a reality, why not using that to build a fusion torch instead of Orion?

Would it allow for faster transit between planets in the Solar system?

In 1890 no one knew how to build a rocket that would take a man into space. We can say that it was beyond their capability. No amount of time, with their fixed 1890 scientific knowledge and industrial base, would fix this. In 1960, it was done. In 2010, we are in at worst the same position we were when Saturn V program started. In fact we are in a better position, because we have in addition all the technical knowledge and experience from the Saturn V program, and also subsequent scientific, technological and industrial advances. It therefore would take us less time, be cheaper, better, or all of those. That is what is generally considered to be meant by progress.

Yeah, I didn't mean to say that we're materially in a worse position than in the 1960. Computing advances, material science and so on and so forth have actually improved our potential. Our real abilities hinge on the goals and targets of space exploration programs and the methods used thereby, however.

The guys who work in the field are reasonably optimistic now- they have the physics nailed down well enough to make the reactor work, and the engineering is just expensive, now that the computer technology is up to making it all designable.

I saw a documentary on ITER. Even though the fusion isn't completely clean, there are few if any fundamental challenges to operation. Removing the contaminated plates from inside the reactor will be done by robots much like we handle nuclear waste now. Yup, not 100% clean, but you don't get a chance of it going Fukushima, which is in my view a huge improvement. The ITER pessimism stems from the early 2000s and the 1990s, right now I'm not sure someone would seriously argue the thing wouldn't run. It might be a little tricky and there are engineering challenges, but not theoretical physics ones.

[HMS Conqueror]

Really?

Well there are a couple of issues. It's not just a matter of hooking it to a turbine.

It kind of is- there's a cooling jacket around ITER that draws heat straight off. Since ITER is a research reactor, they just bleed the heat off into radiators, but you could totally hook a turbine to that.[/quote]
But that still wouldn't make it a commercial power reactor, for the reasons I went on to talk about...

ITER will only run for a few minutes (albeit at 500MW, comparable to a real power plant), which is a lot longer than other fusion experiments, but insufficient for a power reactor.

Since the reaction produces heat, I suspect that if you can make it run in long enough bursts you don't have to run the fusion reaction continuously- just keep hitting it in bursts and keep the reactor temperature topped off.

Possibly so, but the recovery time is important. The reason, I think, is that the components cannot yet withstand the heat flux continuously. In older reactors it was because the magnets started to melt, but ITER uses cooled superconducting magnets, so I am not sure if this is still the limitation.

Second, there is the issue of how to sustain neutron bombardment for long periods.

If you mean what I think you mean, I know ways to bathe a target in neutron beams indefinitely for a relatively limited cost of power input. I once saw an interesting poster on using a method like that as a turbocharger-analogue for a molten salt reactor.

That was badly phrased, but by the time I re-read the edit button was gone.

I mean that the materials need to survive sustained neutron bombardment. The reaction creates a sustained neutron bombardment, which is useful for breeding fuel, but also the soruce of this problem.



Really ITER is a sort of engineering test-bed. It scales up the size of the reactor which reduces the turbulence issues and should reach the physics milestones of breakeven and possibly even ignition. But it will also be the first time things like the breeder blanket, sustained(ish) neutron-resistant materials, &c. have ever been fully tested in a realistic environment.

[Winston Blake]

Stas Bush: I feel like cut'n'rebutt is taking this off into a million unrelated tangents, so instead I will just rephrase my argument in more concise and hopefully more convincing way:

In 1890 no one knew how to build a rocket that would take a man into space. We can say that it was beyond their capability. No amount of time, with their fixed 1890 scientific knowledge and industrial base, would fix this. In 1960, it was done. In 2010, we are in at worst the same position we were when Saturn V program started. In fact we are in a better position, because we have in addition all the technical knowledge and experience from the Saturn V program, and also subsequent scientific, technological and industrial advances. It therefore would take us less time, be cheaper, better, or all of those.

That is what is generally considered to be meant by progress.

This is all absolutely true... if capability is defined as a matter of 'fundamental scientific and industrial capability'. The earlier tangents in the thread were due to a failure to clearly define 'capability', I think. There are different levels of capability. Just prior to when Skylab was lifted in 1973, if someone asked 'Can the U.S. put a 77t space station into orbit?', then the 'yes' that would have been given is different to the 'yes' of, say, 1985, or the 'yes' of 2012.

The question 'Can the U.S. put stuff into space?', is a bit like someone asking 'I'm moving house, can you help me move my stuff?'. There's a difference between:

• 1. 'Yes' (But only by carrying small things in my arms).
  2. 'Yes' (I have a small car with a small boot, and it costs a lot to run, so I don't use it as much as I had hoped).
  3. 'Yes' (I operate a big moving truck that I can rent to you for cheap).
  4. 'Yes' (I'm saving up for a house and I can't spare any money. I once helped someone move, but I sold my car, and I take the train now so I don't need a car. Theoretically if I really wanted to I could buy a new car and help you move. I won't though.).

It's not surprising that many people would interpret #4 as really meaning 'No'.

It wouldn't help to say 'Oh new cars are cheaper nowadays, so I'm actually even more capable of helping you than I was when I bought my old car. I still won't though.'.

If you sell your car because you don't need it, then despite the fact that you could buy a new one, if you really wanted to move something, the fact is: you don't actually have a car. I guess the idea of hiring a mover or using a friend's car is a bit like NASA using Soyuz but I don't want to drive this analogy too far.

[Sky Captain]

If magnetically-confined fusion (which is a major pain in the ass) is a reality, why not using that to build a fusion torch instead of Orion?

Would it allow for faster transit between planets in the Solar system?

Transit times probably would be quite similar. Only advantage nuclear pulse drive would have it would provide faster acceleration. Also fusion torch drive may be able to scale down without loosing performance better than Orion and would be able to refuel from local water sources. To get really high performance from Orion you would have to use thermonuclear bombs and that would require very large and massive ship, possibly supertanker sized.

I have no desire to fill my universe with weaponized space.

Well, any kind of propulsion that allow easy interplanetary travel could easily be used as weapon of mass destruction, just stick it to something massive, accelerate it to really high speed and you have impactor from hell.

[Dass.Kapital]

I have no desire to fill my universe with weaponized space.

And to remind people of the old adage,

"Any engine powerful enough to be interesting as a from of quick travel any where...Is an engine powerful enough to be used AS a weapon."

Or, in other words, 'The Kzinti lesson'.

No need for accelerating the entire ship at something. Just turn it around and wave the engine exhaust beam at something.

[Simon_Jester]

That is vastly over-simplistic. Weaponized rocket exhausts don't have to be dangerous at long range, and at short range they become targets for things like cheap-shit surface to air missiles (if you're attacking a ground target) or... pretty much any conceivable weapon (in space).

A rocket engine will always be inferior as a weapon to any custom-built device intended as a weapon and built to the same scale of power output.

[PeZook]

Yeah, it's like using mining charges vs. proper artillery. You can technically make some MEAN IEDs with mining charges, but artillery is superior in every single way.

[Sky Captain]

It also depends on engine type. A multi terawatt fusion torch if the individual exhaust particle energies were high enough probably could cause radiation damage at fairly long range kinda like particles from solar storm although it would be inferior compared to dedicated long range beam weapon.
Only exception would be laser stations designed to propel laser sail spacecraft. Those probably would require only some minor changes in focusing optics to turn them into deadly long range laser weapons.

[Skgoa]

The problem with that is the words "if", "were high enough" and "could": yes, of course such a weapon is conceivable but that doesn't make it practical. Not just because even just using that kind of power to accelerate even a "dumb" block of lead would cause far greater damage - but also because space engines that are based on accelerating (not force-carrying) particles all have the characteristic that the exhaust stream tends to dissipate relatively fast. This is due to the particles 1) not being emitted perfectly straight and b) imparting forces on each other. I.e. the reason why particle cannons are hogwash. So you would want incredibly cold uncharged and non-magnetic particles... why not just use a conventional mass driver?
Though even lasers aren't perfect and thus from a certain distance any energy weapon becomes pointless.

If magnetically-confined fusion (which is a major pain in the ass) is a reality, why not using that to build a fusion torch instead of Orion?

Would it allow for faster transit between planets in the Solar system?

Transit times probably would be quite similar. Only advantage nuclear pulse drive would have it would provide faster acceleration. Also fusion torch drive may be able to scale down without loosing performance better than Orion and would be able to refuel from local water sources. To get really high performance from Orion you would have to use thermonuclear bombs and that would require very large and massive ship, possibly supertanker sized.

You don't always want higher acceleration. (E.g. when you have achieved 10g with one type of drive, there is no point in switching to a drive that can do 30g - either would kill anyone on board.) What you want is efficiency. Or more precisely: you want as high a sustained impulse as you can have for a given amount of weight. This concept is called (weight) specific impulse and is represented in formulas as Isp. (a capital I with "sp" annotated at the lower right.)
A fusion reactor entirely contained by magnetic fields might be incredibly light. Orion will definitely be extremely heavy. I am not claiming that one of the two will be better in reality (neither is even remotely close to feasible right now) but one of the two seems sensible and has real world research going into it while the other is shown to first semester rocket science students as a joke.

[K. A. Pital]

A fusion reactor entirely contained by magnetic fields might be incredibly light.

And wouldn't run into scalability issues?

Orion will definitely be extremely heavy. I am not claiming that one of the two will be better in reality (neither is even remotely close to feasible right now) but one of the two seems sensible and has real world research going into it while the other is shown to first semester rocket science students as a joke.

Heavy, sure, though considering the thing will be assembled in space, why the problem? And yeah, nuclear pulse propulsion is slightly out of fashion, which doesn't impact the worth of the idea at all. I'm talking about a system of orbit-to-orbit transit for a hypothetical future of civilization; not "give me nukes in space right now".

[Simon_Jester]

Stas, I think you're missing important points. For orbit to orbit transit, many engine options other than Orion drives become practical. The great advantage of Orion- spectacularly high thrust- is reduced. Moreover, fissiles are a valuable and limited resource, and Orion expends them in large quantities- the kind of subkiloton bombs used for propulsion in Orion are very inefficient at converting plutonium into energy.

Basically, any society disciplined enough to unite the world should be disciplined enough to forgo Orion for interplanetary transport and use more economical options.

Hell, I suspect the nuclear salt water rocket is more plausible.

[Skgoa]

Stas: The point is that heavy mass is the same as inertial mass. (I hope I haven't misstranslated the terms...) I.e. if your ship is heavier, it will take a higher impulse to get it to the same speed. Which basicly means that you need to burn through more fuel (whatever type) and thus you have to carry more of it with you. This either eats into your payload mass or your achievable deltaV. It gets worse when you are accelerating near a source of significant gravity.
Now, for your setting it is easy to justify it via whatever you want. E.g. nothing else can be scaled as huge and thus it's the only type of drive that can move the kinds of interplanetary bulk cargo you have in your setting. I just wanted to point out some more Real Rocket Science(tm), because I'm a huge nerd and because the concept of effectivity v. raw thrust seems to be something most people don't know/get about space.

[K. A. Pital]

Um... any calculations? Give me the fastest passenger transit between Earth and say Mars, feasible with current physics. What would that be? I'm not dead set on "Orions" as intra-system vessels.

[Skgoa]

Absolute fastest? Some kind of fixed external accelerator, e.g. a huge railgun. You would simply launch the interplanetary shuttle from the orbit (or an airless moon) of the planet you start on and right into the railgun in orbit around the destination. Since none of the critical parts have to move, you can build them as big as you want to. It would start to make economic sense from a certain volume of travel onwards.

Other than that... my opinion remains unchanged. Some form of electromagnetic drive would be the most sensible, IMHO. Electric thrusters are relatively small and light, as is their reactant mass. They can achieve ungodly high Isp, e.g. ESA's Dual-Stage 4-Grid has achieved 19300(!) seconds of Isp in '06. This - and the fact that most drives of this type produce only little waste heat - means you can put many of them on the ship and have them run for very long burns to achieve huge velocities over time. In the inner solar system you could even get away with powering them with solar cells exclusively but some form of fusion or matter/anti-matter reactor seems prudent. Since the ship operates out in space, not nearly as much shielding is needed, which means we can get away with much less weight in comparison to reactors on the ground.
The ship in Avatar is a good example of a realistic long distance space ship. (Their version of a ground-to-orbit shuttle OTOH...)

[K. A. Pital]

Um... I'm not talking about interstellar transit yet (the one in Avatar was interstellar). I'm interested in how a hypothetical rapid passenger transit system would operate.

Say, we place a mass driver for interplanetary shuttles on the Moon. What next? The absolute limit on how fast this shit can deliver us to Mars orbit without (a) killing everyone on board (b) requiring space-based structures infeasible even with, say, graphene?

A small Orion craft can turn around using smaller maneuever engines and decelerate when coming into orbit. How is the interplanetary shuttle going to brake upon reaching Mars? What if it can't "get right onto the railgun" in orbit of Mars, since docking is a complicated process? Then what, the shuttle misses by a few centimeters and becomes space dust? Seriously, I need a ship which can also brake upon arrival. Alternatively, tell me just how advanced the docking tech should be to perfectly catch and decelrate a shuttle coming from the Moon into Martian orbit.

some form of fusion or matter/anti-matter reactor seems prudent.

What if we still suck at fusion and it is poorly scalable, running in hugeish reactors on Earth but not so good on small spaceships? And let's not get carried away with antimatter. If uranium's expensive, antimatter is just way out there.

[PeZook]

Absolute fastest? Some kind of fixed external accelerator, e.g. a huge railgun. You would simply launch the interplanetary shuttle from the orbit (or an airless moon) of the planet you start on and right into the railgun in orbit around the destination. Since none of the critical parts have to move, you can build them as big as you want to. It would start to make economic sense from a certain volume of travel onwards.

Uh...the critical parts would ABSOLUTELY have to move. First, since geometry of the transit would keep changing due to planets going around their orbits, at the very least you'll need a way to orient the decelerating railgun to catch the incoming ships. Second, it would need the ability to alter its own orbit for a variety of reasons, like avoiding debris impacts or again orienting the decelerator towards incoming traffic (it has to be perfectly synchronized with the target's arrival to catch it, including position in orbit) - and since it's virtually certain that the ship will need to do correction maneuvers duirng transit (because you'd need to hit a relatively tiny target in orbit around the destination), that's gonna be a huge concern.

And then, of course, there's the problem of heat management which becomes seriously problematic in a structure "as big as you want it to be", especially one that accelerates ships to absurd speeds over the length of the barrel. You really don't want uneven heating from the Sun to deform the barrel ever so slightly while a ship is rushing through at 30 km/s

But storywise, it's a really cool concept. Hell, the engineering difficulties might be mentioned as a way to make your hypothetical civilization look more awesome for having solved them

[K. A. Pital]

The Commune is more centered on efficiency than overengineering, though

However, a Lunar railgun or maglev-assisted launch is quite feasible (I don't understand though why not go all the way and if you can make space elevators on smaller bodies, why not do an elevator-launcher combo). The problem is the deceleration and precise docking in orbit over any other body, Venus, Mars or the Jovian moons.

If the "interstellar shuttle" is really a small shuttle which is moving way beyond the speeds that are feasible with it's own engines, deceleration might be a bitch.

Okay, let's say people on Mars are gods at timing (far superior to our pathetic docking procedures which even the chief of the Russian space program called "walking on a thread every time") and despite the 10 to 30 minute delay with Earth-Mars communications and any unforeseen in-transit course changes by the shuttle, they can precisely orient their railgun/space elevator/whatever to catch the incoming shuttle and decelerate it. What with Venus, Jovian moons, farther planets? Pluto, Eris, asteroids - where infrastructure construction is going to be limited primarily by the huge distance itself, and will most likely involve pre-fab automatic stations, and no humans?

How will we "place a railgun" there and coordinate it precisely enough so that a "space bullet" with a human inspector spending 3-4 months in transit to Pluto doesn't end up in a cloud of debris because the delays disallowed perfect coordination of the decelerator?

I am deeply skeptical about this, much more so than about Orion. I need a reliable ship which can speed up and decelerate at will, not requiring the orbital infrastructure operating with a zero-failure tolerance! Dual-Stage 4-Grid is interesting, but how scalable is such an ion thruster and what would be the typical transit times if we build a man-rated ship with them in orbit (equipped with a fissile source of power, for example)?

[Simon_Jester]

For one, it's actually a huge advantage if your spaceship only needs to provide braking thrust, not acceleration: halving the delta-v required to make the transit at a given speed means you can travel a lot faster.

For another, I think Skgoa's point was that this might be theoretically optimum, that doesn't make it practical. Orion drives aren't really that practical either under the circumstances. The unsolved engineering problems are significant, the cost in fissiles is significant to a civilization that cares about peak uranium, and the speed is at most somewhat desirable for a narrow subset of travel applications: people who absolutely have to be on Mars in one month instead of (say) two.

Since you've already gotten around (somehow) the need for high-thrust super-engines to lift cargo off of planets, Orion just... doesn't make as much sense as using huge batteries of engines that provide a 0.01g-0.1g acceleration over very long periods.

Many of the same arguments apply here as would apply to individual autos versus mass transit, in my opinion.

[Sky Captain]

A fusion reactor entirely contained by magnetic fields might be incredibly light.

And wouldn't run into scalability issues?

Orion would have similar or even worse scalability problems. A small Orion craft would have worse ISP than ion craft. There was a NASA study back in the sixties about small 10 m diameter Orion craft that would be assambled by 4 or 5 Saturn V lauches for Mars mission and it had specific impulse only little better than nuclear thermal rocket. Not really worth messing with nukes on such small scales. If you want Orion you want it so big that it can use H bombs so you could get really high specific impulse. Such drive system may make sense for gigantic interplanetary bulk carrier.
Also nukes need certain minimum amount of fissionable material to work and small Orion craft requiring sub kt explosions would be extremely wasteful because most of plutonium would not fission.
For passenger transport unless the ships are as big and carry as many people as largest cruise ships Orion would be suboptimal.

For small passenger transport a VASIMR type engines powered by lightweight fission reactor may work better and there already are working exapmles of such engines that soon will be tested on ISS. Another advantage - redundancy. Orion craft would have only one engine and if something goes wrong then you are stranded maybe on solar escape trajectory. VASIMR craft could easily have several independent reactors and engines making total propulsion system failure unlikely.

[K. A. Pital]

For one, it's actually a huge advantage if your spaceship only needs to provide braking thrust, not acceleration: halving the delta-v required to make the transit at a given speed means you can travel a lot faster.

Yup. That means you can hurl efficient ion-engine equipped ships or Orions into space from the Moon, right? Spares you the problem of building railguns and docking on faraway planets.

For another, I think Skgoa's point was that this might be theoretically optimum, that doesn't make it practical. Orion drives aren't really that practical either under the circumstances. The unsolved engineering problems are significant, the cost in fissiles is significant to a civilization that cares about peak uranium, and the speed is at most somewhat desirable for a narrow subset of travel applications: people who absolutely have to be on Mars in one month instead of (say) two.

That is if you use pure fission bombs. What if you use almost pure fusion bombs? What if your Earth power infrastructure already operates on pure fission+renewables and thus does not require uranium par se?

Since you've already gotten around (somehow) the need for high-thrust super-engines to lift cargo off of planets, Orion just... doesn't make as much sense as using huge batteries of engines that provide a 0.01g-0.1g acceleration over very long periods.

I didn't "get around it" for Earth. And Orions are not used for anything but passenger transit, since human time is the ultimate non-renewable resource (humans are still mortal). The proposal is the following: there's an SSTO which lifts you up to an orbital dock (Earth). Space elevator and human permanent colonies operates on the Moon and Mars. Possibly there's a skyhook on Venus to a yet-small floating station (undecided on whether the Commune is batshit crazy about Venusian exploration; with their current capabilities terraforming is out of the question).

That's about all the human-inhabited infrastructure I have. The year is ca.2090.

Many of the same arguments apply here as would apply to individual autos versus mass transit, in my opinion.

I understand your points but I fail to see how huge batteries of engines accelerating ships for long periods are useful for rapid intra-system transit. The ion thruster is neat, but I haven't yet gone into the theme deep enough. I researched Hall thrusters early on when writing, but they had scalability and other issues - no good for passengers.

Orion would have similar or even worse scalability problems. A small Orion craft would have worse ISP than ion craft. There was a NASA study back in the sixties about small 10 m diameter Orion craft that would be assambled by 4 or 5 Saturn V lauches for Mars mission and it had specific impulse only little better than nuclear thermal rocket.

I'm thinking more in terms of space liners (40-50 m diameter), would that also have crap Isp?

Not really worth messing with nukes on such small scales. If you want Orion you want it so big that it can use H bombs so you could get really high specific impulse. Such drive system may make sense for gigantic interplanetary bulk carrier.

Yeah, but can't you use H-bombs on a 50m pusher plate? What about 100 m or so?

For passenger transport unless the ships are as big and carry as many people as largest cruise ships Orion would be suboptimal.

There are excessively large interstellar Orions in my story (which set the stage for a later interstellar war when FTL is discovered), but I thought that mid-scale ones would be a good pick for intrasystem hauls.

For small passenger transport a VASIMR type engines powered by lightweight fission reactor may work better and there already are working exapmles of such engines that soon will be tested on ISS. Another advantage - redundancy. Orion craft would have only one engine and if something goes wrong then you are stranded maybe on solar escape trajectory. VASIMR craft could easily have several independent reactors and engines making total propulsion system failure unlikely.

Hmm... So can a VASIMR-equipped ship make it to Mars orbit in a few weeks, like the Orion study proposed? Or would it still take months? The issue I have is that humans (especially future humans) treat time as the sole scarce resource and want to reduce their own transit times to an absolute minimum. Cargo function is thus irrelevant and probably served by massive ships with ion thrusters.

[PeZook]

The Commune is more centered on efficiency than overengineering, though

However, a Lunar railgun or maglev-assisted launch is quite feasible (I don't understand though why not go all the way and if you can make space elevators on smaller bodies, why not do an elevator-launcher combo). The problem is the deceleration and precise docking in orbit over any other body, Venus, Mars or the Jovian moons.

Well, if timing is too much of a hassle and represents a too much of a single point of failure (not an unreasonable fear at all!), then you can always accelerate the ship to a reasonable transfer speed and then have it use its own engines/aerobraking/laser-blasted solar sail to brake at the destination.

It is still a great arrangement, even for HEUGE vessels like Orion, because it saves piles of fuel.

Okay, let's say people on Mars are gods at timing (far superior to our pathetic docking procedures which even the chief of the Russian space program called "walking on a thread every time") and despite the 10 to 30 minute delay with Earth-Mars communications and any unforeseen in-transit course changes by the shuttle, they can precisely orient their railgun/space elevator/whatever to catch the incoming shuttle and decelerate it. What with Venus, Jovian moons, farther planets? Pluto, Eris, asteroids - where infrastructure construction is going to be limited primarily by the huge distance itself, and will most likely involve pre-fab automatic stations, and no humans?

Frankly, I think it's not unreasonable to expect a great many ship classes optimized by their destination. For example, ships moving between Mars and Earth would have provisions for aerobraking and solar sails, because Earth and Mars would have well-developed infrastructure, solar sail lasers powered by vast arrays of solar panels on the Moon etc etc etc.

While ships going to outer planets would be vast, self-sustaining things capable of independent maneuvering - for obvious reasons.

I am deeply skeptical about this, much more so than about Orion. I need a reliable ship which can speed up and decelerate at will, not requiring the orbital infrastructure operating with a zero-failure tolerance! Dual-Stage 4-Grid is interesting, but how scalable is such an ion thruster and what would be the typical transit times if we build a man-rated ship with them in orbit (equipped with a fissile source of power, for example)?

Ion thrusters and the like are cool because all they really need is electrical power. There's few moving parts, they don't take a huge amount of space etc.

So just circumvent the low, low thrust by using huge arrays of them and thrusting all the freakin' time.

For small passenger transport a VASIMR type engines powered by lightweight fission reactor may work better and there already are working exapmles of such engines that soon will be tested on ISS. Another advantage - redundancy. Orion craft would have only one engine and if something goes wrong then you are stranded maybe on solar escape trajectory. VASIMR craft could easily have several independent reactors and engines making total propulsion system failure unlikely.

Having an engine which uses a non-disposable power source is also good from a conservation standpoint: instead of throwing fissile materials out with every pulse unit, you can reprocess the onboard fuel and reuse it, GREATLY extending your available fissiles supply.

[K. A. Pital]

I guess my characters will have to take their time to visit the Moon and be launched from there by means of a railgun or teh space elevator itself, if it means saving heaps of fuel. And I guess the Moon launch complex is going to be quite massive, too.

For example, ships moving between Mars and Earth would have provisions for aerobraking and solar sails

Solar sails (or rather laser sails) are cool (haven't really thought about that), but I don't see a need for a purely orbit-to-orbit vessel to have aerobraking features. Instead the ship would be better off with carrying a bunch of escape pods, I guess. After all, there's a space elevator on Mars extending from Olympus Mons (I shamelessly took that from the nice book "Leaving the planet by a space elevator") and thus shuttles are redundant. This elevator services almost the entirety of Mars' inbound and outbound passenger and cargo transit.

If you can get a high enough speed with ion engines by accelerating all the time, I might just forego the Orions and only leave the interstellar ones as artifacts. After all, I'm aiming for a very hard sci-fi start

[PeZook]

You don't have to aerobrake all the way ; Just calculate it in a way that will let you bleed off enough speed to get into orbit, and then maneuver to docking with traditional propulsion. Russian Zond flights used a somewhat similar technique called "skip re-entry", where the spacecraft would progressively bleed off speed by skipping off the atmosphere. Just forego the "re-entry" phase by, say, using some sort of disposable shield that you'd jettison once you bled off enough speed.

It doesn't have to be an exclusive method, either - since you mentioned energy is not much of a problem, some ships might use it to save fuel, while others just go full retard and use their main engines

[someone_else]

Explain what "weird shit" is pulled out in these studies.

It's the interaction between the nuke and the plate. It's unclear how much of the energy of the nuke can go into thrust and how much must be wasted to avoid blowing away the plate or due to other issues. All tests were made with nuclear fireballs (underground), not with a nuclear shaped charge.

If you start pulling efficiencies from your ass, you end up with shit like this or this.

To get the numbers I told you in my last post I used the numbers from Atomic Rockets (which are also similar to the ones I've seen thrown around in other places), that is the Orion's relatively realistic performance.

Is it like the system suggested by the guy here?

Yes. It uses another way to accelerate the kinetics, but it's more or less the same in practice.
Solar kinetic uses solar sails and a relatively close pass to the sun to make a u-turn and place them into retrograde orbit (I posted links to papers about this in my posts above). Given that they are launched from Earth/Moon they were in orbit around the sun with speeds around 30 km/s, an impact between a vehicle and the kinetic stuff is at 50-60 km/s, or more if you pull other tricks.

But in any case the impactor system would rely on external control of the impactor swarm, which is problematic if we'd face interference of some sort. On the other hand, the Orion controls its own propulsion.

Heh, that's a swarm orbiting at interplanetary distances from anything. It's hard to jam them all due to sheer distance involved, and they can easily overcome any long-range jamming if the ones not jammed relay the message to others with their own short-range comms (likely much stronger than any realistic long-range jamming). What interferences you were thinking of?

If you fear about people fucking with the craft itself to screw the terminal guidance signals somehow, heh, any craft is easy to screw up as well.

As for solar weather (the only interference not requiring saboteurs on the vessel, imho), you're more or less forced to stop and wait the storm to go away for both Orion and kinetics. Or risk a launch system/bomb timer hiccup and blow up yourself with a misfired nuke. For kinetics the rad shielding of every single sail will likely be enough to survive in space, but surviving a direct hit from a CME is a bit too much weight to be worth it.

The point is that you can send much more solar kinetics than needed, and the ones that aren't used can simply steer away and return in their parking orbit when their job is over. Or be sold with a discount to boost something else.

Would it allow for faster transit between planets in the Solar system?

Probably too early to say for sure. Anyway, thispaper gives some eyeballed estimates which don't look that bad (magnetic confined fusion is slightly crappier than Orion as imagined, inertially confined fusion has more or less the performance of fusion-bomb orion, antimatter is nonsense, there is a neat talk of fission rockets as well, but no word on Orion-like systems).

All in all, a very good read even if you don't understand most of the physics like me.

Magnetic fusion engines with specific powers in the range of 2.5–10 kW/kg and Isp of 20,000–50,000 s could enable round-trip missions to Jupiter in less than a year. Inertial fusion rockets with ap >100 kW/kg and Isp > 105 s offer outstandingly good performance over a wide range of interplanetary destinations and round-trip times. Even Pluto is accessible with round-trip travel times of less than 2 years.

Hmm... So can a VASIMR-equipped ship make it to Mars orbit in a few weeks, like the Orion study proposed? Or would it still take months?

To put it simply, the main difference between a fusion torch and an ion engine is only that the former does not need outrageous amounts of power to accelerate the propellant (because it generates its own power by fusing the fuel).
The issue with ion-engines in general is simple: what gives you the hellacious amounts of power to operate them?
Rememeber the formula of the thrust power from Atomic rockets. What engine gives you X performance is irrelevant. That energy for that must come from somewhere.

VASIMR is a bit of a fraud, to reach Mars in so little time it requires fission reactors with a power-to-weight ratio that makes no sense whatsoever as discussed here.
And once you have fusion generators you have already pretty damn fast plasma so why not shoot that out of the tailpipe?

Realistic fusion engines (as described in the paper I linked above) should work more like an ion drive than a rocket:
-relatively crappy acceleration, but MUST be more than 1 milligee, so you can do brachistochrones
-has to accelerate continuously for months.

[someone_else]

You don't have to aerobrake all the way; Just calculate it in a way that will let you bleed off enough speed to get into orbit

The problem with aerobraking is that the higher the transit speed the less useful this deceleration gets. I.e. the fuel needed to decelerate the same amount of speed would mass less than the mass of an aerobraking shield.

Albeit for Orion that's not really an issue.