• This forum is specifically for the discussion of factual science and technology. When the topic moves to speculation, then it needs to also move to the parent forum, Science Fiction and Fantasy (SF/F).

    If the topic of a discussion becomes political, even remotely so, then it immediately does no longer belong here. Failure to comply with these simple and reasonable guidelines will result in one of the following.
    1. the thread will be moved to the appropriate forum
    2. the thread will be closed to further posts.
    3. the thread will remain, but the posts that deviate from the topic will be relocated or deleted.
    Thank you for understanding.​

Spaceship Propulsion

Joined
Jun 29, 2008
Messages
11,042
Reaction score
841
Location
Second star on the right and on 'til morning.
Website
atsiko.wordpress.com
If a similar thread shows up under my name, blame it on my crappy internet connection.

I'm trying to come up with a somewhat scientifically accurate "sub-light drive" for use in various sizes of spaceships. Most importantly, a sub-generation survey/exploratory vessel big enough to handle a few thousand hands for a few months. I've caved in and used a blackbox FTL drive, but I want as many other systems as possible to be close to "real". From what I understand, the main three propulsion methods for use on-board a ship are chemical rockets, ion-drives, and possibly solid-state propellant mass drivers. Solar sails would not be feasible in terms of the purpose of the ship. Ion-drives have a high specific impulse, which is great, but they have relatively low acceleration. Chemical rockets have better acceleration as far as I know, but I'd prefer to avoid large quantities of rocket fuel on the ship, for obvious reasons--it can explode. Mass drivgers require large throwable objects... which strikes me as a bit worrisome--what if you run out of them?

I'd love to get some comments and suggestions on the most efficient system.
 

Albedo

Alex
Super Member
Registered
Joined
Dec 17, 2007
Messages
7,376
Reaction score
2,958
Location
A dimension of pure BEES
In my opinion any civilisation that's invented faster-than-light travel will probably already have 'conventional' rockets powered by direct fusion or even antimatter. Hell, the engineering behind (fission) nuclear rockets has basically been feasible since the 1960s, they've just always been a bit too, erm, politically incorrect to use.

Project Rho is an absolutely invaluable resource on realistic rocketry in science fiction. They have a page full of drive types here.
 

Dommo

On Mac's double secret probation.
Super Member
Registered
Joined
Apr 18, 2008
Messages
1,917
Reaction score
203
Location
Oklahoma City, OK
Easy. Nuclear Pulse Propulsion.

Allows for ships of up to into the multiple million ton range to be ground launched from earth(assuming you can live with the fallout, or perfectly doable if you had clean nukes). Doable with current technology, and allows for a top speed of somewhere between 5-10% light speed depending on the type of nukes used.

If I were ever going to build a colony on mars, this is how I'd do it. I could take a ship that's larger than an aircraft carrier, and with that much space and cargo ability, it would allow for a viable founding population to start on a colony(say a thousand people), with all of the needed equipment and tools to build a community around the ship and start mining the planet for resources.

http://en.wikipedia.org/wiki/Antimatter_catalyzed_nuclear_pulse_propulsion

http://en.wikipedia.org/wiki/Project_Orion_(nuclear_propulsion)
 

Lhun

New kid, be gentle!
Super Member
Registered
Joined
Jan 30, 2007
Messages
1,956
Reaction score
137
Personal favorite: go with an ion drive.
There is no specific reason why an ion drive has to have a very low thrust, that's just current technological limits. If you have a whole quantum-bolognium-singularity power generator on the ship, you have more than enough energy to power a really large ion accelerator. Heck, a big fusion reactor should do the trick.

The upsides are:
Very high thrust-to mass ratio. So you don't need to carry around humongous amounts of reaction mass.

Draws it's power from the ships reactor, not from a fuel/reaction mass combination like a rocket drive, so it's not volatile, and you're a lot more mass efficient. Otherwise your rocket's dead weight while in FTL and the popwer generator for FTL is dead weight while in STL.

Clean. You don't produce radioactive fallout or highly energetic EM radiation like you would with a fusion drive, a nuclear drive, or an antimatter drive. So you could use it on shuttles to land on a planet.
Landing a starship on a planet is a really bad idea anyway. The differences in designs that are optimal for deepspace flight and athmospheric takeoffs are so huge, you'd have to design something like a submacopter. Spaceships should stay in orbit while shuttles do the transfer. On the other hand, if you have a universe where the spaceships carry very thick material armor anyway, they're probably good for planetary landings without being designed for them. If it's armored enough to survive missile hits, it's probably hard enough to survive athmospheric pressure, it's own weight on the ground, and air friction as well.

Improvised weapon. Other than something with a pretty low exhaust velocity, such as a chemical rocket, or something with an unfocused blast such as an Orion drive, an ion drive is basically a big-ass particle accelerator cannon. Might not have the range of a weaponized laser, but it will definitly be the most powerful energy projector on any ship.

In terms of coolness though, the nuclear saltwater drive tops everything imo. The idea is just awesome. :D Though they are dirty as hell, and have the same efficiency disadvantages as chemical rockets.

Addendum:
Or use a photon drive if you have a power generator for your FTL that can produce ridiculous amounts of energy. All the advantages of an ion drive without even needing physical reaction mass.
 
Last edited:
Joined
Jun 29, 2008
Messages
11,042
Reaction score
841
Location
Second star on the right and on 'til morning.
Website
atsiko.wordpress.com
Well, I'm against nuclear pulse propulsion, but nuclear reactors are perfectly okay for the power source. I've been considering using a NIF-like method of nuclear fusion to generate the power, probably with the direct-drive approach.

The FTL drive doesn't rely completely on enormous power generation, but it might manage to power a photon drive. But I'm still leanign towards ion. Its perfectly reasonable for the ship, which would be pretty big, to employ a very large ion drive.

Another question this brings up is what type of drive you would use for something like, say, attitude adjustments. Would small ion drives be effective, or would you be better off with chemical rockets?
 

Lhun

New kid, be gentle!
Super Member
Registered
Joined
Jan 30, 2007
Messages
1,956
Reaction score
137
Well, I'm against nuclear pulse propulsion, but nuclear reactors are perfectly okay for the power source. I've been considering using a NIF-like method of nuclear fusion to generate the power, probably with the direct-drive approach.
NIF uses inertial containment fusion iirc. Check out the polywell fusion reactor design, that would scale a lot better for really big reactors.
The FTL drive doesn't rely completely on enormous power generation, but it might manage to power a photon drive. But I'm still leanign towards ion. Its perfectly reasonable for the ship, which would be pretty big, to employ a very large ion drive.
Well, a photon drive big enough to provide decent thrust is pretty much the largest application of power i can think of in terms of spaceships. Plus, its usability as a weapon is so much higher than for an ion drive.
It is important to note what kind of ion drive. An ion drive that uses a static charge to accelerate the ions has a very low upper limit for thrust, so you'd have to use a really large number of those (relatively small) drives. Otoh if you use one that accelerates the ions with magnetic coils, you can have one big one, basically a kind of big particle accelerator (though the ions won't get nearly as fast as particles in a cyclotron). I think on Project Rho they're found under the heading of plasma drives not ion drives.
Another question this brings up is what type of drive you would use for something like, say, attitude adjustments. Would small ion drives be effective, or would you be better off with chemical rockets?
The probably best choice are arcjets. I.e. the reaction mass is simply heated using electricity from the main generator, and the resulting expansion provides thrust. Drives of that type have lower efficiency than ion drives (need more reaction mass) but require less power and have a higher thrust to power ratio. They're also really small, so it's no problem to plaster them all over a ship as directional thrusters.
You could even use an arcjet as a main engine if you like the concept, the efficiency is not too bad, definitly better than chemical rockets, though that depends on the amount of power you can dump into the reaction mass. A possible modification would be to heat up the fuel using lasers instead of an electric arc, which makes it possible (or easier) to reach extreme temperatures, giving it very good mass to thrust ratio. Though there are some construction problems -you don't want to vaporize your heating chamber walls- which can be explained away by futuristic materials. ;)
 
Joined
Jun 29, 2008
Messages
11,042
Reaction score
841
Location
Second star on the right and on 'til morning.
Website
atsiko.wordpress.com
Polywell might scale better, but I don't think that a net-power gain has yet been proven on a larger model. Correct me if I am wrong.

The arcjets look like a keeper, though. I was originally unsure about power-consumption, but I don't think it would be too much of a problem if I'm running an enormous ion drive anyway. Which I'm planning to have be electromagnetic, not electrostatic, which means it could be classified as a plasma thruster.
 
Last edited:

Lhun

New kid, be gentle!
Super Member
Registered
Joined
Jan 30, 2007
Messages
1,956
Reaction score
137
Polywell might scale better, but I don't think that a net-power gain has yet been proven on a larger model. Correct me if I am wrong.
You're not wrong but that's not something that actually needs to be tested. Since volume scales as cube of the radius and surface area as square, the actual question is how small you can build fusion reactors and still get a net power output, not how big. It's the same for tokamak designs, that's what makes ITER so interesting, it's the first tokamak that actually got enough funding to be built in a reasonable size. And it looks like tokamaks need to be really big to get a net power output. In theory, polywell might still provide power at pretty small sizes, a meter or so, but there's no question you'd get a net power gain in a big one.
The arcjets look like a keeper, though. I was originally unsure about power-consumption, but I don't think it would be too much of a problem.
If you can power an ion drive, you can power arcjets. After all, all that inreased efficiency of an ion drives comes from simply pumping more power into the reaction mass. That's basically the tradeoff for all drives. The more reaction mass you burn, the less power you need, and vice versa.
 
Joined
Jun 29, 2008
Messages
11,042
Reaction score
841
Location
Second star on the right and on 'til morning.
Website
atsiko.wordpress.com
Okay. I'll tentatively assign polywell fusion reactor(s) as the power system. Though some secondary system power schemes might run better on other generators.


The ion drive would do best with a dedicated generator, right? Then you wouldn't have to run through all that power re-routing crap that's so popular in scifi, like on star trek. Not that I'm holding up Star Trek as a useful example for a hard sci-fi system.

BTW, what are the major dangers of this sort of fusion reactor?
 
Last edited:

Lhun

New kid, be gentle!
Super Member
Registered
Joined
Jan 30, 2007
Messages
1,956
Reaction score
137
Okay. I'll tentatively assign polywell fusion reactor(s) as the power system. Though some secondary system power schemes might run better on other generators.
As far as i know, inertial containment fusion reactors are more interesting for small scale, while continuous reactor designs are better for big ones. Though polywell is not the only possibility there, tokamak is the most common example and should scale just as well. Heck, even some technobabble like weber uses (gravity containment something) would be possible. The interesting point here is that starting fusion needs quite a lot of energy, so when you want a large power output over a long period of time, it's the better choice to leave the fusion process running continuously, as opposed to fusing small pellets of fuel repeatedly.
The ion drive would do best with a dedicated generator, right? Then you wouldn't have to run through all that power re-routing crap that's so popular in scifi, like on star trek. Not that I'm holding up Star Trek as a useful example for a hard sci-fi system.
That's probably best asked of an electrical engineer with some experience in high power systems, but as far as i'm aware there's no problem with having a single reactor powering everything. Though a very useful feature of such a reactor would be to have variable power output, that shouldn't be a problem for a sufficiently advanced fusion reactor. It should be able to function with various levels of fuel input, and thus various levels of energy output (and heat generation).
As far as i know, "power rerouting" is just so much bullshit anyway. You can't expect of any normal type of machinery to work better if you just pump more energy into it. Most would burn out instead. Or at least blow the fuses. ;)
And if you don't have enough electricity to run everything that's hooked up to your generator/battery/power outlet, you usually deal by simply switching one off. ;)
Though some (smaller) backup power systems can be sensible for other practical purposes. For example starting a fusion reactor would require quite a bit of energy. So you'd want at least some good capacitors to store enough for that. And you'd probably also want to be able to run life support independently, in case there's some problem with the main reactor that is still repairable but takes time.
The best choices i can see here are either capacitors, which would need to be a lot more efficient than the ones we have today, though even now there's research into that so i'd call it plausible speculation to have really good capacitors. Or, second choice, fission reactor. The advantage of those is that they're pretty foolproof, require refueling only every few dozen years, and are safe. Toshiba currently has some designs in testing that could provide 200kw for around 40 years and is about 24m³ in size. Would be perfect for running lifesupport on a ship whenever the fusion reactor is powered down. Though that assumes pretty big ships already, for a small sized ship, a simple fossil-fuel powered generator might be the best choice, though miniaturization might be the most reliably predictable of all technological advances.
BTW, what are the major dangers of this sort of fusion reactor?
Not a lot of dangers really. Fusion reactors of all kinds do not explode. Since the big problem in getting them to run is to actually contain the fuel so that fusion can happen, any major damage will cause the fusion to fizzle out resulting in no explosion whatsoever. If whatever type of containment happens to be switched off abruptly (magnetic in tokamak, electrostatic in polywell) you still have a big ball of hot plasma though. Which will immediatly expand, burn anything in it's path, while slowly getting colder. But if you have decently thick walls around the room where the reactor is placed, it should only damage what's inside. Heck, you probably have enough radiation shielding around such a reactor anyway.
The reactor "overheating" or some such thing, as often seen on StarTrek, is also not a big problem. If it's decently designed, it should be possible to switch off the fuel to the reactor with the press of a single button, making the plasma ball go *fzzt* almost immediately. Obviously, you'd design such a system with many redundancies. Heck, you could even have a simple manual valve on the fuel lines somewhere, in case absolutely everything else fails and you need to switch it off.
Depending on what fuel is used for fusion that might be volatile. The easiest type of fusion is with hydrogen isotopes, which means hydrogen tanks on board. The cleanest type of fusion would be boron-proton, still requiring some hydrogen tanks, and lot better containment, since the plasma needs to be about 12 times as energetic for the fusion to start. "Cleanest" because it releases a lot less harmful radiaton than deuterium fusion. Which is the biggest hazard of fusion reactors, most types of fusion release neutrons and gamma rays, which will turn most materials they hit into radioactive isotopes with varying half-lifes. So after some time, pretty much every part of the machinery exposed to the fusion plasma will be highly radioactive. Not a problem in space, just chuck it in the nearest sun and replace the parts, but people definitly shouldn't walk around inside the (switched off) reactor without protection.
 
Joined
Jun 29, 2008
Messages
11,042
Reaction score
841
Location
Second star on the right and on 'til morning.
Website
atsiko.wordpress.com
Okay. I found an earlier thread where dommo spouts off about polwell, so hopefully that will help.

On the subject of type, I've heard that tokomak is much more expensive than polywell is projected to be. Since the ship itself is built by a trade organization for survery-exploratory missions, I'd imagine that lower costs would be preferable.

I still have a lot more research to do, now that I have some good choices laid out in front of me, though. There's just so much more than the propulsion to consider in terms of what combination of systems would work best.
 
Last edited:

Vincent

Cheers
Poetry Book Collaborator
Super Member
Registered
Joined
Feb 12, 2005
Messages
1,934
Reaction score
468
Easy. Nuclear Pulse Propulsion.

Allows for ships of up to into the multiple million ton range to be ground launched from earth(assuming you can live with the fallout, or perfectly doable if you had clean nukes). Doable with current technology, and allows for a top speed of somewhere between 5-10% light speed depending on the type of nukes used.

If I were ever going to build a colony on mars, this is how I'd do it. I could take a ship that's larger than an aircraft carrier, and with that much space and cargo ability, it would allow for a viable founding population to start on a colony(say a thousand people), with all of the needed equipment and tools to build a community around the ship and start mining the planet for resources.

http://en.wikipedia.org/wiki/Antimatter_catalyzed_nuclear_pulse_propulsion

http://en.wikipedia.org/wiki/Project_Orion_(nuclear_propulsion)

Yeah, this method is a favorite of mine. How else can you a 10,000 ton lunker constructed like a battleship into orbit?
 

Vincent

Cheers
Poetry Book Collaborator
Super Member
Registered
Joined
Feb 12, 2005
Messages
1,934
Reaction score
468
By building it there?

Not that building it like a battleship is necessarily a good idea in the first place.

Sure, but this way you get to do it with 1950s tech. Which is all sorts of retro-cool.
 

Albedo

Alex
Super Member
Registered
Joined
Dec 17, 2007
Messages
7,376
Reaction score
2,958
Location
A dimension of pure BEES
Though that assumes pretty big ships already, for a small sized ship, a simple fossil-fuel powered generator might be the best choice, though miniaturization might be the most reliably predictable of all technological advances.

Wouldn't you need to carry massive amounts of oxidiser as well as fuel to run a fossil fuel-burning engine though? I don't see how it would offer any advantage over something like a souped-up version of the nuclear batteries we use on satellites now.
 

White-Tean

Illustration Student
Super Member
Registered
Joined
Apr 25, 2009
Messages
145
Reaction score
16
Location
Australia
Website
white-tean.livejournal.com
Oh gosh, I love this thread, thanks for starting it.
It's actually been something I've been needing to research for my own SF - there may be a tiny bit set in actual space, most of it is on the planet but I'd like my main character to be able to have a better than layman understanding of space travel technology and other forms of advanced technology. I mean, I'm working with the assumption that the average person (in this futuristic novel) won't have any deeper understanding of how starship propulsion systems work - because microwaves have been around for ages and are used everyday and most people only have a vague understanding of how they work.

Anyway, much props to everyone and their vastly greater knowledge of science than myself.
:D
 

Lhun

New kid, be gentle!
Super Member
Registered
Joined
Jan 30, 2007
Messages
1,956
Reaction score
137
Wouldn't you need to carry massive amounts of oxidiser as well as fuel to run a fossil fuel-burning engine though?
No different than carrying fuel for a chemical rocket. ;) If you have one of these as engine/directional thrusters you could even use the same fuel tank.
I don't see how it would offer any advantage over something like a souped-up version of the nuclear batteries we use on satellites now.
Compared to nuclear batteries, chemical fuel delivers a whole lot more power in a shorter time. The only advantage nuclear batteries have is a really long lifetime. If you want a decent mass/energy ratio from nuclear fuel, you need to start a chain reaction.
 

dgiharris

Disgruntled Scientist
Super Member
Registered
Joined
Aug 24, 2006
Messages
6,735
Reaction score
1,833
Location
Limbo
Looks like you got some decent advice on this.

Sub-light drives aren't really a problem, we have that technology already. It is merely a matter of fuel and propulsion.

The biggest problem will be acceleration and decceleration. You will need some type of inertia dampner or you will be forced to accelerate and deccelerate at a slower rate that can be tolerated by your passengers, probably 1G.

just use the simple physics equation of velocity equaling acceleration to figure out how long it woudl take you to accelerate and deccelerate to whatever % of c you are shooting for.

Another problem would be fuel. You could 'refuel' at your local neighborhood gas giant or you could develop some theoritical fuel that has a kick ass weight to thrust ratio.

And as has been mentioned above, nuclear propolusion or something equivalent would be realistic.

Mel...
 
Joined
Jun 29, 2008
Messages
11,042
Reaction score
841
Location
Second star on the right and on 'til morning.
Website
atsiko.wordpress.com
Looks like you got some decent advice on this.

Sub-light drives aren't really a problem, we have that technology already. It is merely a matter of fuel and propulsion.

The biggest problem will be acceleration and decceleration. You will need some type of inertia dampner or you will be forced to accelerate and deccelerate at a slower rate that can be tolerated by your passengers, probably 1G.

just use the simple physics equation of velocity equaling acceleration to figure out how long it woudl take you to accelerate and deccelerate to whatever % of c you are shooting for.

Another problem would be fuel. You could 'refuel' at your local neighborhood gas giant or you could develop some theoritical fuel that has a kick ass weight to thrust ratio.

And as has been mentioned above, nuclear propolusion or something equivalent would be realistic.

Mel...

*sigh* I didn't even bother to bring up acceleration. Choosing from realistic sub-light engines is bad enough. I haven't decided whether to use an inertial dampener or not. I probably will. As much as it feels like cheating, realistic acceleration just complicates things so much, at least for the story I'm planning. This is especially true since the main two methods I'm considering for FTL require acceleration.

As an example (and everyone please feel free to correct my calculations):

In order to make a trip to Alpha Centauri in the timescales I'm planning on using, it has to be done in about four hours. If I did the math right, that's 8760*c. Ignore for the moment that that's impossible by physical laws, and you understand the issues of acceleration involved. No way that's happening without some kind of excuse, like a massive inertial dampener. I'm not interested in using an actual hyperspace system. My FTL exists at least partially in normal space.

Though, for sub-light travel, I don't imagine significant percents of c are practical anyway. I'm honestly not bothered by differences in intrasystem travel time. For my puposes, 8 minutes an AU is not all that much better than 32, or 64, or even 128. Also inter-system would have to be FTL already, to meet time constraints.
 
Last edited:

Vincent

Cheers
Poetry Book Collaborator
Super Member
Registered
Joined
Feb 12, 2005
Messages
1,934
Reaction score
468
*sigh* I didn't even bother to bring up acceleration. Choosing from realistic sub-light engines is bad enough. I haven't decided whether to use an inertial dampener or not. I probably will. As much as it feels like cheating, realistic acceleration just complicates things so much, at least for the story I'm planning. This is especially true since the main two methods I'm considering for FTL require acceleration.

As an example (and everyone please feel free to correct my calculations):

In order to make a trip to Alpha Centauri in the timescales I'm planning on using, it has to be done in about four hours. If I did the math right, that's 8760*c. Ignore for the moment that that's impossible by physical laws, and you understand the issues of acceleration involved. No way that's happening without some kind of excuse, like a massive inertial dampener. I'm not interested in using an actual hyperspace system. My FTL exists at least partially in normal space.

Though, for sub-light travel, I don't imagine significant percents of c are practical anyway. I'm honestly not bothered by differences in intrasystem travel time. For my puposes, 8 minutes an AU is not all that much better than 32, or 64, or even 128. Also inter-system would have to be FTL already, to meet time constraints.




At speeds of a light year a hour, you're out of the realm of real physics anyway, and I can't imagine that mundane matters like acceleration and deceleration come into the equation. I don't think you really can ignore that. An inertial dampener larger than the known universe wouldn't help you. You're cruising into the realm of the exotic, on engines powered by weapons grade balonium, where you get to make the rules.

If that's your thing.
 
Joined
Jun 29, 2008
Messages
11,042
Reaction score
841
Location
Second star on the right and on 'til morning.
Website
atsiko.wordpress.com
At speeds of a light year a hour, you're out of the realm of real physics anyway, and I can't imagine that mundane matters like acceleration and deceleration come into the equation. I don't think you really can ignore that. An inertial dampener larger than the known universe wouldn't help you. You're cruising into the realm of the exotic, on engines powered by weapons grade balonium, where you get to make the rules.

If that's your thing.


It was one option. The FTL was already bolognium anyway. It's the sub-light that has a chance to be somewhat realistic.
 

Lhun

New kid, be gentle!
Super Member
Registered
Joined
Jan 30, 2007
Messages
1,956
Reaction score
137
It was one option. The FTL was already bolognium anyway. It's the sub-light that has a chance to be somewhat realistic.
Imo it's important not to ignore STL physics, even if the ships are going FTL. One example why it's important for consistency are weapons. If some kind of STL drive can easily, quickly and most important of all cheaply get a vehicle to significant portions of c, every weapon used in that universe will be some kind of kinetic kill missile. And if they're big portions of c, they're doomsday weapons.