Summary of fuel-carrying propulsion systems
I've looked around on the internet and found some statistics on existing
and proposed propulsion systems for spaceships. At the bottom of the page
is a discussion of the specific impulse we need for TEP.
I make no claim as to the accuracy of any of this stuff, beyond that it
mostly came from NASA sites. Some of the electric-propulsion ideas came from
"Plasma Propulsion in Space", Eric J. Lerner, The Industrial Physicist Oct 2000
pp. 16-19. Here are some definitions used in the table:
- has been flown or will be flown soon
- we could build one without any major technological breakthroughs
- we could build one with major technological breakthroughs but no new
- Specific impulse, which is for a mass m of fuel thrust * time maintained
divided by acceleration of gravity at earth's surface. This is measure of
the efficiency of a drive. It relates directly to the exhaust velocity:
specific impulse is exhuast velocity / g (9.8 m/s^2).
- Plume Temperature
- This and the mass flow rate determine how far away you can see one of
these drives firing. For many of these systems, the answer is "all the way
across the solar system."
||4500 F - 6500 F
||at least plasma
||Must be plasma
||Must be plasma
||Existing, not flown
||Must be plasma
||> 10000 F
||Probably at least plasma
||1E+8 - 1E+9 K
||I'd guess much more than 1E+9 K, the fusion plasma temperature
Here's more explanation of what the systems are:
- Chemical rocket
- These statisics are for the Space Shuttle Main Engine (SSME), "the most
efficient liquid rocket ever flown" (says NASA). The plume temperatures
given are for general big chemical rockets, not specifically for the SSMEs.
- Ion engine
- Ion engines, also called electrostatic engines, use an electric potential to
accelerate the ions in a plasma, which are then shot out the back at high speeds.
They characteristically produce very low thrust. The first row is for the
Deep Space 1 (DS1) vehicle, which I think has either flown or is about to be flown.
The second row is a claim I saw on a NASA site as an upper limit.
- Hall-effect engine
- Rather than using a static electric field to accelerate ions, they can be
accelerated through the interactions of magnetic fields and currents. This has
the advantage that it spits out a neutral plasma rather than a stream of same-
charge ions, so the spacecraft doesn't accumulated charge.
- Magnetoplasmadynamic engine
- Uses the Lorentz force from a radially outward directed electric current
and its self-induced magnetic field to accelerate a (net) neutral plasma.
These are pretty high-thrust devices, and one is serious use would presumably
be powered by nuclear fission. None have flown yet, but they are under
consideration for some upcoming NASA missions.
- Fission rocket
- The first row is a gas core fission reactor, in which the fission reaction
heats LH2 which is the reaction mass. This is also called "nuclear
thermal propulsion." The second row is for the NERVA system, a less
efficient version which was built in the 1970s. Other fission rockets
have been built with I_sp as high as 1100 s.
- Electric Fission
- NASA calls this "nuclear electric propulsion"; it is simply using a
fission reactor to power an ion engine. The numbers given are for the
"Leaky Magnetic Mirror" engine concept, which NASA has been playing with for
a while but currently can't be built.
- External Fission
- Throw small fission bombs out the back of your spaceship and let the shocks
push you. NASA's Orion project back in the '60s was based on this, and it
was very efficient at the time. This one has the emissions of a bunch of
small fission bomb blasts. NASA built a prototype using dynamite bombs.
- Fusion rocket
- I rate this as very theoretical because we can't make fusion reactors that
break even yet. It's just like a fission rocket but with fusion.
- Electric fusion
- There are many suggestions for how to do this; the numbers given are
theoretical numbers for a design called a "plasma focus", which uses a capacitor
bank to generate a pulsed sheet of current which runs down a barrel between
coaxial electrodes, and is pinched to a narrow focus at the end. The focus
produces enough compression to start fusing, and as they decay form focused
beams of ions that shoot away, propelling the ship.
- External Fusion
- This works just like external fission, but uses fusion bombs. Actually
most proposals use fuel pellets that are ignited with lasers, or something.
I rate this as theoretical because we could use our existing nuclear
arsenals to build one. NASA had one of these on paper only, called the
What specific impulse do we need?
Say ships are around 100t = 1E+5 kg. And, we want to be able to fly from
earth to jupiter ten times on a load of fuel in such a ship. The
distance is about 1E+12 m. At 1g acceleration we could make the trip in
2 * sqrt( d / a ) = 2 * sqrt( 1E+11 s^2 ) = 6E+5 s or a week, which isn't
so great. 10g would be more reasonable, cutting it down to 2 days. The
fuel would need to provide 1E+5 kg * 1E+2 m/s^2 = 1E+7 N of thrust for
2E+5 s. Asking for 10 trips, we need to maintain 10g for 2E+6 seconds.
I_sp (specific impulse) is thrust * time / ( mass * 10m/s^2 ), so
fuel mass = thrust * time / ( I_sp * 10 m/s^2 ) = 2E+12 / I_sp kg.
Say we can devote 50% of our ship mass to fuel, 50t = 5E+4 kg, so
I_sp = 5E+7 s. As we see below, an antimatter rocket may be able to
achieve 2E+6 s, so we could manage a single such trip on that. I'd say
any of the other systems should only be used briefly in situations where
the player has been stranded on a primitive world or something.