> On Fri, 27 Nov 1998, Tom Anderson wrote:
> oooh, *that* kind of fusion drive. right. there are two types of fusion
[snipped for brevity]
Right. The problem is that the internal-style combustion and NERVA
rockets son't produce the thrust needed to get anything anywhere. Their
Isps are down around 10-20 thousand km/sec. At that low an efficiency,
like 80% of the rocket has to be fuel just to get from Earth to Mars-
not exactly the ship design described in the FT rules. Ion engines put
out (note- I'm still going by the sources mentioned last time) around
160 km/sec, while an He(3)-D fusion pulse engine puts out on the order
of 8,000 km/sec, an efficiency good enough to push around those NAC
carriers and not worry about pushing around a bunch of liquid hydrogen. These
engines have already been designed (in theory at least) and work off
principles that don't change as your technology improves since they're based
on basic chemistry and the energy levels contained in the
atoms used (at least until you start getting into the grav-drive realm,
but then you discard the inefficient old reaction drives in favor of something
a little faster).
I guess what got me about the whole photon-drive thing is that if I
could focus enough photonic energy to drive a battleship around, I think I'd
rather sit in one place and spend my time pointing my drive at
soon-to-be-extinct enemy ships in my general neighborhood....
Cheers
in a really garbled message earlier, I said:
> The problem is that the internal-style combustion and NERVA
> out on the order of 8,000 km/sec...
Those Isps are actually:
- NERVA and fusion (steam exhaust): 10-20 km/sec.
- ion engines: 160 km/sec.
- fusion: 8000 km/sec.
For the curious, the Isp (specific impulse) of a rocket engine is a measure of
how fast the propellent is expelled from a rocket. This turns out to be a good
measure of rocket efficiency, since in space the faster you throw something
away from you the faster you're pushing yourself forward. The less efficient
your engine is, the more fuel you have to carry around (can anyone say Space
Shuttle?). With many of these advanced engines (especially the ion engine) the
mass flow rate, or the amount of propellent you can dump overboard to increase
your thrust, is also very important. The space shuttle couldn't get you to
Mars, but in a drag race between the shuttle and a high-tech ion engine
rocket, the shuttle's going to win every time because it can blow a lot of LOX
all at once.
A photon engine would be 100% efficient (no mass, all acceleration, Isp
approaching infinity) but I still think if I had a laser source that powerful
I'd just sit in the backyard here in Houston and burn those pesky Martians
from here without bothering to build a spacefleet.
standing corrected
> On Fri, 27 Nov 1998, Sid Jones wrote:
> >Ion engines put out (note- I'm still going by the sources mentioned
erm, i'm not quite sure that's right. i think specific impulse is the amount
of velocity change you can get from 1 kg of drive (or 1 kg of fuel, i'm not
sure) with a 1 kg ship. so, with a nerva which took up one fifth of your
ship's mass, you could burn at 1 gravity for 200 seconds (3 minutes 20
seconds). this ain't much good for a combat starship, so, yes i think exernal
fusion is the order of the day. i may have this all wrong,
though.
anyway, i can well believe that the daedalus/orion type drives are far
more powerful than thermal-expansion drives. however, the prospect of
setting off a nuke half a mile behind me every few seconds (a la orion) is
not too attractive. on the other hand, the daedalus - which uses many
small explosions, and in a very sophisticated form might almost use a
continuous stream - is quite neat.
> A photon engine would be 100% efficient (no mass, all acceleration,
well, photoons do have mass, that's the point. energy is mass and vice versa
(if you heat something up, it gets heaver by a tiny amount). a lot of the
energy from a photon drive is going into making photons; this potential energy
stays locked up until the photon hits something. i'm not sure a photon drive
is 100% efficient.
> but I still think if I had a laser source that
good point. the photon drive is well established, needs no fuel, and is still
a really daft idea.
Tom
> Thomas Anderson wrote:
You're correct, though Mr.Jones isn't all wrong.
If:
Isp = Specific Impulse (Seconds)
Ve = Exhaust Velocity (meters / second)
g = acceleration due to gravity ( 9.81 meters / second^2 )
Then:
Isp = Ve / g
Ve = g * Isp
The factor that every one is interested in is "delta-V"
or "change in velocity". In EFSB, this corresponds to how far one can move the
endpoint counter when the ship does a burn.
If:
Mbs = mass of ship at start of burn (kilograms)
Mbe = mass of ship at end of burn (i.e., Mbs - mass of propellant
used) 1n(x) = take the natural logarithm of x
Then:
deltaV = Ve * 1n( Mbs/Mbe)
Isps I have on hand:
CHEMICAL ROCKET Propellant Isp
===========================================
Hydrogen-Fluorine (F2/H2) ideal 528
Hydrogen-Oxygen (O2/H2) space shuttle 460
Hydrogen-Oxygen (O2/H2) ideal 528
(O3/H2) ideal 607
(F2/Li-H2) 703
(O2/Be-H2) 705
Free Radicals (H+H)->H2 2,130
Metastable Atoms (e.g. Helium) 3,150
SATURN V FIRST STAGE Isp = 430 seconds SPACE SHUTTLE Isp = 455 seconds
NERVA/DUMBO "Atomic Rocket" Isp = 850 to 2950 seconds
ORION "old Bang-bang" Isp = 1000 to 5000 seconds
ARCJET (Electrothermal) Isp = 800 to 1200 seconds MPD (Electromagnetic) Isp =
2000 to 5000 seconds ION (Electrostatic) Isp = 5e3 to 4e5 seconds LIQUID CORE
NUCLEAR REACTOR Isp = 1300 to 1600 seconds GASEOUS CORE NUCLEAR REACTOR Isp =
3570 to 7000 seconds GASEOUS CORE COAXIAL FLOW REACTORIsp = 1800 seconds
DAEDALUS (Fusion Microexplosions)Isp = 1e6 BORON FUSION Isp = 1e6 HYPOTHETICAL
FUSION TORCH Isp = 5e4 to 1e6 seconds ANTIMATTER Isp about 3.06e7 seconds
PHOTON DRIVE Isp about 3.06e7 seconds
> good point. the photon drive is well established, needs no fuel, and
Well, the photon drive needs no *reaction mass*, but it needs *power*, which
will come from fuel. In chemical rockets, the reaction mass and the power fuel
is the same thing. In an ion rocket, they are different. For instance, the
cesium reaction mass will be accelerated by power from a nuclear reactor
fueled with plutonium.
> On Tue, 1 Dec 1998, Nyrath the nearly wise wrote:
fair enough.
> Isp = Specific Impulse (Seconds)
aha! Isp is just a slightly more convenient way of stating exhaust velocity. i
get it.
force is rate of change of momentum, which is exhaust velocity times rate of
flow of mass. acceleration of a ship is given by the exhaust velocity times
the flow rate divided by the ship's mass. thus, if you multiply specific
impulse by flow rate and divide by mass, you get the ship's acceleration in
multiples of g. clever.
sid and my two definitions of specific impulse were in fact equivalent
(although they were both wrong, as we both left out the g factor).
> CHEMICAL ROCKET
eh? (a) how is helium metastable? (b) how does a helium-fuelled rocket
work?
> NERVA/DUMBO "Atomic Rocket" Isp = 850 to 2950 seconds
these are the options we have run over, i think. i have read of GCNRs. i
thought they were of similar efficiency to scnr (nerva type), but apparently
they are much, much better. better, in fact, than orion. still,
the daedalus-type drive is best by far. the orion type is used my the
human ship Michael in footfall; it has been said that the alien ship used a
daedalus type, but i don't recall that.
> ARCJET (Electrothermal) Isp = 800 to 1200 seconds
so the ion drive has a higher Isp than Orion; i suppose it would have to.
the thing is that it is very hard to have a high mass-flow rate (the
other factor in determining thrust) with an ion drive, so an orion will give
you a higher, less efficient thrust.
> BORON FUSION Isp = 1e6
hmm, now this is getting a bit blue-sky. ultraviolet-sky? i've never
heard of boron fusion before; sounds like a bit of a daft idea to me.
excellent stuff! added to my collection of reference works cribbed from
Nyrath ... shall i webbify this (and star-list?) or are you on the case?
> > good point. the photon drive is well established, needs no fuel, and
sorry, sorry. still, if it is fusion-powered it will not need much fuel.
feeling much wiser, Tom
> Thomas Anderson wrote:
Yep. You got it.
If:
F = Thrust (Newtons or kg mt/sec)
Ve = exhaust velocity ( mt/sec )
Isp = specific impulse (sec)
Mdot = propellant mass flow ( kg/sec )
A = acceleration ( m/sec^2 )
Mc = ship's "current" mass ( kg )
Then:
F = Mdot * Ve F = Mdot * g * Isp
A = F / Mc
> > Metastable Atoms (e.g. Helium) 3,150
Atoms can be "excited" by thrusting their electrons into higher orbitals. Such
electrons quickly fall back into
their base state, re-emitting the energy.
Metastable atoms have their electrons excited in such a way that they stay in
the higher orbitals. For a while, at least.
> > NERVA/DUMBO "Atomic Rocket" Isp = 850 to 2950 seconds
i
> thought they were of similar efficiency to scnr (nerva type), but
Yes, but Nerva rockets at least have the decency to keep their fuel elements
of glowing radioactive death sealed in the reactor. GCNR are much like slow
reacting Orions, spraying vaporized plutonium like a radioactive crop duster.
This wouldn't matter much in space, but it would be unwise to use one to lift
off into orbit. At least not from a spaceport based on a continent you like.
> > ARCJET (Electrothermal) Isp = 800 to 1200 seconds
Correctomundo. Once you figure in the ion drive's mass
flow, the thrust can be measured in humming-bird power.
> > BORON FUSION Isp = 1e6
No, actually boron fusion would be marvelous, if we could only figure out how
to ignite the reaction. Here's the relevant section from rec.arts.sf.science:
nyrath@clark.net (Winchell Chung) BORON FUSION 11 4
B5 + p -> 3( He ) + 16Mev
5 2
that is, bombard Boron-11 with protons. A complicated reaction
ends with helium and no pesky nuclear particles. 16 million electron volts
gives
an exhaust velocity of better than 10,000 km/sec, which translates into
a theoretical specific impulse of something over a million seconds. What's the
catch?
schillin@spock.usc.edu (John Schilling) The catch is, you have to arrange for
the protons to impact with
300 keV of energy, and even then the reaction cross section is fairly small.
Shoot a 300 keV proton beam through a cloud of boron plasma, and
most of the protons will just shoot right through. 300 keV proton beam against
solid boron, and most will be stopped by successive collisions without
reacting. Either way, you won't likely get enough energy from the few which
fuse to pay for accelerating all the ones which didn't.
Now, a dense p-B plasma at a temperature of 300 keV is another
matter. With everything bouncing around at about the right energy, sooner or
later everything will fuse. But containing such a dense, hot plasma for any
reasonable length of time, is well beyond the current state of
the art. We're still working on 25 keV plasmas for D-T fusion.
If you could make it work with reasonable efficiency, you'd get
on the order of ten gigawatt-hours of usable power per kilogram of fuel.
Paul Dietz <paul@interaccess.com> Unfortunately, this discussion ignores side
reactions:
11 12
p + B -> C + gamma
5
4 11 14
He + B -> N + n
5
Since we're on the subject...
I generally play a very "near-future" version of FT: no FTL drives at
all and the STL drives have realistic capabilities. This isn't as as different
from a "standard" version of FT as you might think; instead of requiring weeks
or months to travel between star systems, it takes weeks or months to travel
between planetary systems. At any rate, one of the things I've had to do was
figure out what the capabilities of real engines are and how they could be put
into the existing FT system. This has equal relevance to
star-spanning games as well. I assume that the engine percentages given
in FTFB are fixed (we don't, after all, want to change the design system) but
that the time and distance scales can be varied. As it turns out, varying the
time scale from the "standard" 15 minutes doesn't change things much, so
given that this works well with other parts of the FT/DS/SG system, I
decided to hold this fixed as well.
While I realize that 1 MU=1000 km is the most prevalent choice for people on
the list, a real ship simply can't maintain that acceleration for very long. I
assume that a ship needs to carry at least 10 combat turns worth of
fuel/reaction mass and this must be included in the "engine mass" of the
design system (this of course eliminates some promising engines, see the
discussion below). While I'm sure it's probably happened, I can't think of one
of my FT games where the battle has gone on longer than 10 turns. I'll put the
spreadsheet up my website fairly soon, but until then here are the results for
several different engines. I am using the mass and Isp's from "Rocket Flight"
just because the few I've been able to check independently from my sources
have matched up and it gives us a common base for comparison. If anyone knows
where the sources for "Rocket Flight" came from, I'd love to find them...
So without further ado...
Turn Length (sec):900 Combat Turns in Drive: 10 Ship Mass: 600
Ship Thrust: 8
Column headings:
Drive Type-Quantity-Ve (km/s)-Thrust (kN)-Engine Mass (tons)-Burn Rate
(Tons/Turn)-Total Mass (tons)-Max Accel(m/s^2)-Accel/Thrust Pt-km/MU
COMPARATIVE EFFICIENCY Liquid Chemical 0.072 0.33 120.10 0.144 24.02 240.34
0.2002 0.0250 10.13 Solid Core Fission 3.999 35.19 195.95 39.99 20.04 240.39
0.3266 0.0408 16.53 Mass Driver 1.54 46.20 30.80 231 0.92 240.24 0.0513 0.0064
2.60 Ion Electric 0.6 94.20 6 240 0.03 240.34 0.0100 0.0013 0.51 JxB Electric
2.17 160.58 10.85 238.7 0.13 240.02 0.0181 0.0023 0.92 Orion Fission Pulse
0.942 40.51 247.75 188.4 5.19 240.25 0.4129 0.0516 20.90 Gas Core Fission 6.51
195.30 149.73 195.3 4.49 240.22 0.2496 0.0312 12.63 MPD 0.156 48.99 3.12
240.24 0.01 240.33 0.0052 0.0007 0.26 Orion Fusion Pulse 1.018 74.31 297.26
203.6 3.66 240.25 0.4954 0.0619 25.08
D-T Fusion 4.43 97.46 478.44 44.3 19.57
240.03 0.7974 0.0997 40.37
H-B Fusion 0.799 783.02 48.74 239.7 0.04
240.15 0.0812 0.0102 4.11 Pion Annihilation 0.48 3763.2 23.52 240 0.00 240.03
0.0392 0.0049 1.98 New Drive 2 5000.0 11852 200 4.27 242.67 19.7533 2.4692
1000.01
I have assumed that each engine is scalable. For example, you wouldn't
normally talk about "0.6 of an ion engine". This is not too terrible since in
most cases you can get the same mass by not running the engine at full steam.
I have done these calculations using integer numbers of engines, the final
scales changed somewhat, but not the choice of engine. The last column is the
desired info. This is the number of km per MU assuming the engine limitations
specified by the FTFB design system. I have used the smallest,
fastest ship which should be the most restrictive case. AS you can
see, to
get 1000 km/MU you need a truly phenomenal drive if you want to keep
your engine and fuel within mass limits (there are other combinations possible
other than the "New Drive" shown, of course, but they're all pretty gross).
The best we can manage *today* would seem to be the Orion Fusion Pulse
(detonate hydrogen bombs behind the ship). This allows 1 MU=25 km. The
best near-future drive, not surprisingly, seems to be the D-T Fusion
torch, which allows 1 MU=40 km. Incidentally, this sets 1 thrust point equal
to
almost exactly 0.01g which is a convenient round number. Anti-matter
(pion
annihilation) drives, incidentally are -incredibly- fuel efficient, but
don't have the thrust to match what is seen in the game.
And that brings up an important point. I'm not saying that the D-T
Fusion
drive, for example, is the -best- engine for space flight. I'm just
saying that it is the best one that fits the performance parameters and rules
restrictions we see in the game. You could change the rules of FT (for
example, by not including fuel in the engine system mass), but then you don't
have the "basic" FT rules anymore. Remember, my goal is to see what
scales are necessary to make -FT- realistic, not to design a new game.
I think it's interesting that with 1 MU=25 or 40 km, sensors would have no
trouble identifying everything on the game board. This fits nicely with the
basic rules, as no bogey markers are ever needed. Sensors and stealth
would be critically important at the strategic level, however, and I am
working on a board game (for which FT can be used as the combat system if
desired) which uses this fact.
One other note, not relevant to star-spanning games, I use the mass
assigned to FTL drives in the design system to be extra fuel which is used for
strategic movement, i.e., travel between planets. Generally this fuel is
completely burned over the course of a few dozen combat turns and the ship
then coasts on an intercept orbit to its destination - exactly as it
would in the real world. Ships that have more fuel to burn would be able to
make intercept faster than would that had less. There's even the possibility
of using a different engine (such as an ion drive) for strategic movement, but
FTL drives in FT are so tiny that it turns out you're better off just using
the mass for extra fuel for your main drive.
Anyway, just something I thought I'd throw out for discussion.
> On Wed, 2 Dec 1998, Nyrath the nearly wise wrote:
sounds sort of feasible, even if i can't see any metastable energy levels in
helium. however, i did recently discover (ie in a quantum mechanics lecture
this morning) another alternative. we all know that helium does not form
chemical bonds: the He2 molecule is not stable. however, it turns
out that He2+ (two heliums, with one electron missing) is stable; a
mixture of He2+ and F- (fluorine is good at holding onto a -ve charge)
might be metastable, and could be 'burned' to helium and fluorine gas. just a
thought.
> > these are the options we have run over, i think. i have read of
they are designed to minimise fuel leakage, however; various schemes have been
developed whereby the coolant and fuel streams do not mix. i don't understand
it myself, but it is probably similar to the problems faced by fusion
engineers in getting the flow of plasma in a tokamak right.
> This wouldn't matter much in space, but
there is a semi-alternate history story called 'moon six' by stephen
baxter, in which the british have a nuclear moon rocket. they launch it from
australia...
> > > BORON FUSION Isp = 1e6
eh? how does going from boron to helium make energy? the binding energy per
nucleon goes up! i suppose that free proton must be the key. still, it
produces less energy per unit mass than proton-proton fusion.
> > excellent stuff! added to my collection of reference works cribbed
it will soon be done. see
users.ox.ac.uk/~univ0938/nyrath/
for an index to the Words of Nyrath archive!
Tom
Hark! I found it.
> Turn Length (sec):900
I don't recall why, but I decided at one point that the Natural Turn Length is
7.5 minutes, which still fits into s DS2 turn nicely.
D-T Fusion 4.43 97.46 478.44 44.3 19.57 240.03 0.7974 0.0997
40.37
> H-B Fusion 0.799 783.02 48.74 239.7 0.04 240.15 0.0812 0.0102
> The best we can manage *today* would seem to be the Orion Fusion Pulse
The
> best near-future drive, not surprisingly, seems to be the D-T Fusion
I'd assume an antimatter drive works by using the a/m to heat normal
matter to high temperatures and use that as the propellant, rather than use a
1:1
ratio of a/m to matter. I don't know a lot about advanced propulsion
types--does this proposal make sense?
In a message dated 99-02-13 16:35:08 EST, you write:
<< I'd assume an antimatter drive works by using the a/m to heat normal
matter to high temperatures and use that as the propellant, rather than use a
1:1
ratio of a/m to matter. I don't know a lot about advanced propulsion
types--does this proposal make sense? >>
It does, but I would say that an anti-matter/matter power source would
provid fast moving particles that would induce an electric field in a magnetic
coil, that would give you your generator, then you could use the electicity to
power your proposhion system.
-Stephen