Fun with: Ion Engines

[Howedar]

Lemur, not to be an ass, but I am a rocket scientist. You are not. You really should stop posting in this thread, because you're making a bloody great fool of yourself. Just a friendly suggestion.

So, let me get this straight: You want me to stop posting in this thread because you say I'm wrong, without even bothering to explain why I'm wrong, even in a single instance? And you back up your assertion with a completely unsubstantiated claim of superior expertise? And I'm not even arguing with you; I'm arguing with other people who have presumably not osmotically absorbed your supposedly superior knowledge.

I don't want you to do anything. It's pretty amusing. I'm just letting you know you look like an idiot.

If you want to call me a liar, go ahead and ask folks around here if I'm in aerospace engineering. I didn't just bring this up off the cuff.

[Kuroneko]

You've just described a hydrogen-based ion engine.

I *know*; I was referring specifically to modern xenon-based ion engines. How many times do I have to repeat that?

None; I'd settle for saying it just once. Given that this is the first time you've made any explicit restrictions on the ion drive fuel, your indignation is a bit misplaced.

True, but then it has the advantage of getting power from an external source.

So what? What does that have to do with how good it is?

It's not always feasible. Over interstellar ranges, it's just plain infeasible except as an initial stage.

For very long-term missions, getting the suitable reliability out of the Orion design without immensely increasing the mass of the ship is even less believable than ion thrusters (which are bad enough by themselves).

What problem are you referring to specifically?

For interstellar flight, we essentially have centuries of nuclear bombardment. Ablation and spalling would be unavoidable, as well as extreme radiation shielding because it will build up.

[TheLemur]

Given that this is the first time you've made any explicit restrictions on the ion drive fuel, your indignation is a bit misplaced.

Eh, you're probably right. I was just tired of responding to every person who said it.

Over interstellar ranges, it's just plain infeasible except as an initial stage.

Why? Wouldn't you just need a larger lens? You can build ridiculously large, very thin lenses in space because there is so little force on the material.

For interstellar flight, we essentially have centuries of nuclear bombardment.

Centuries? How do you get that? Orion is a very high-acceleration drive; even at .1 G, over a century you would accelerate to highly relativistic velocities, which is far beyond Orion's capability.

Ablation and spalling would be unavoidable, as well as extreme radiation shielding because it will build up.

Radiation shielding is not cumulative; a block that protects you against 99% of radiation now will still protect you in ten years. But ablation and spalling are definitely problems.

[TheLemur]

I'm just letting you know you look like an idiot.

Well, if I am an idiot, could you at least direct me to a decent guide to rocketry and orbital mechanics that will cure my idiocy? Do you like idiots and want them to remain ignorant, or what?

[Sikon]

However, regarding ice rockets - Sikon, or someone else, could you describe this idea a bit more, or post a link where I may read about them?

Nuclear and solar heated water rockets are a concept with extraordinarily good potential economic performance for a lot of missions.

One application is retrieving materials including ice from near earth bodies. The result of being able to send out a "space water truck" rocket that over repeated trips brings back orders of magnitude more than its own mass is that the cost of material in earth orbit can be vastly reduced. Such is helpful for transport of commercial satellites between LEO and GEO, for starting space colonization, and for other applications.

The 1992 discovery of a water-ice, near-Earth object (NEO) in the space near Earth is evaluated as a source of rocket fuel and life support materials for Earth orbit use. Nuclear thermal rockets using steam propellant are evaluated and suggested. The space geological formation containing such water-rich NEO's is described. An architecture couples near-Earth object fuels (neo-fuel) extraction with use in Earth orbits. Preliminary mass payback analyses show that space tanker systems fueled from space can return in excess of 100 times their launched mass from the NEO, per trip. Preliminary cost estimates indicate neo-fuel costs at Earth orbit can be 3 orders of magnitude below today's cost.

[...]
Architectural simplicity provides a relative advantage for steam propulsion.
[...]
The steam architecture needs only to extract pure water from ice and store it in a tank.

[...]
Space electric generators are characterized by the system mass in kilograms to generate one kilowatt of power, expressed as the reciprocal of megawatts per ton. The most optimistic electric generator proposed to be practical was quoted at 7.8 kg/kw (0.12 Megawatts per ton) (reference 15). Typical space electric generators cannot do better than 15 to 200 kg/kw. (0.066 to 0.005 Megawatts/ton).

In comparison, a steam rocket does not require electricity, and it [... may generate up to] 200 Megawatts per ton [thermal].

[...]
Solar thermal rockets are similar to nuclear heated rockets. Inflatable mirrors focus the sun on a heat exchanger.

[...]
Issues of nuclear fuel element temperature and power density are discussed. Results suggest that power density of order 75 megawatts per ton of rocket at a specific impulse of 195 seconds (mixed mean outlet temperature of order 1100 K) are both achievable and will provide highly competitive payloads.

There are huge numbers of near-earth objects, including cometary bodies.

[... One example:]
Comet (4015) 1979 VA = Wilson Harrington, shown here with a tail in a 1949 plate, is about 5 km across and may have about 100 Billion (1E11) metric tons of water ice. It's gravity is very low and about 1/10,000 that of Earth, which is crucial for it to be useful to us. Its orbit perihelion is 1.003 AU (Earth is 1.00000) and has a 4.296 year period.

[... Comets] are covered by a layer between 10 cm and 10 m thick of dirt and/or extremely dark carbonaceous sooty material. The composition is approximately ~50% water ice, ~10% CO and CO2, and ~0.5% of a conglomerate of Carbon, Hydrogen, Oxygen and Nitrogen (CHON) materials. See Huebner (1990), Fanale (1991) and Lebofsky (1991). These are the raw materials to make rocket propellant, construction materials and plant food, and are crucial to sustaining life in space.

[...]
We can mine the entire NEO because it has negligible gravity. For example, the difficulty of lifting mass from a 1000 meter deep pit is 10,000 times easier than on Earth. Similarly, a small, 1 ton thrust rocket would launch 10,000 tons away from the comet.

Dr. Zuppero's website

[The above image] sketches the basic elements of a space tanker. The propellant or fuel, contained in the innermost bladder, is fed to the rocket engine attached to the structure. The payload, water, is frozen by space and engulfs the propellant bladder as an armoring shield.

[...]
The figure shows that specific impulses between 0.150 and 0.250 kilo-seconds can result in payback ratio's in excess of 100 to 1. The tank performance in the range of 500 to 4000 represents a tank similar to a garbage bag bladder. A 1/2 pound garbage bag can hold about 32 gallons of water, or about 500 times it mass. But only if the bag is in zero G, as one could simulate by filling the garbage bag with water in a swimming pool. Calculations indicate that water bladder tanks holding in excess of 4000 times their mass can be readily constructed for these applications.

The data for this figure is patterned after Zuppero (1992). A ship using 20 tons tanks, 20 tons engines and structure and developing 8 GWatts thermal for 1 day would deliver about 10,600 tons to HEEO.

Note that a water reactor operating between 500 C and 1200 C provides the required specific impulse. These reactors use well developed technology.

Dr. Zuppero's website

Lunar ice based on the Clementine and Lunar Prospector results has also been suggested, but it was always more uncertain, particularly so now with more recent investigation finding a lack of evidence for such existing in usable quantities. Near-earth objects, however, are quite suitable.

One need not worry about running out of space resources. Here's an illustration:

• Near earth objects:
  A random example is comet 1979 VA where its perihelion of 1.003 AU means that it is sometimes only a half million kilometers further from the sun than earth's orbit, and there are many bodies like that where the delta-v involved in reaching them and shipping material back is low. For some near-earth bodies, delta-v to earth orbit is sometimes even under 3 km/s as illustrated here, almost as easy as shipping material from the Moon, actually easier if one is using a transport with a low-thrust engine not suited to liftoff against any significant gravity. The delta-v to different objects varies, but, in total, there are trillions of tons of objects relatively close to earth's orbit, some having ~ 10% water as a hydrated mineral, some ice, some kerogen (hydrocarbons, good for plastics), and some having nitrogen materials (desired eventually for plant fertilizer).
• Asteroid Belt:
  Still greater resources are in the asteroid belt. Here there is a vast selection of asteroids and comets, some with relatively concentrated valuable heavy metals like platinum and iridium that mainly sank to earth's core during planetary formation but are more accessible in the asteroids. The mass of objects in the asteroid belt amounts to literally thousands of quadrillions of tons.
• Moons:
  Aside from varying locations, there isn't always much of a practical difference between large asteroids and small moons, the latter still having low enough gravity to have readily accessible resources. The total mass of various moons amounts to millions of quadrillions of tons, and a large portion of that mass is ice, such as in the case of some of Jupiter's large moons having ice more than a hundred kilometers thick over millions of square kilometers of surface area.
• Kuiper Belt and Oort Cloud:
  Giant Kuiper Belt objects are starting to be discovered. A random example is the object Eris discovered a couple years ago that is around twenty billion billion tons of mass. It is estimated that there are 35000+ Kuiper Belt objects more than 100-km in diameter. Statistical observations suggest the Oort cloud may contain on the order of a trillion comets with a combined mass on the order of a trillion trillion tons, though such are inactive so far from the sun and undetectable with current equipment. Of course, such objects wouldn't be used in the directly foreseeable future without motivation to go that far out, but they exist if their materials are someday desired.

In other words, a space civilization would likely start with near-earth objects, potentially later moving onto the asteroid belt. Mass driver transports or ice rockets can have practically unlimited quantities of free propellant.

[Kuroneko]

Why? Wouldn't you just need a larger lens? You can build ridiculously large, very thin lenses in space because there is so little force on the material.

It's a question of energy delivery across light-years. Try to calculate the requisite precision. As an initial stage, it's a good idea, but at interstellar distances, it really stretches credulity.

Centuries? How do you get that? Orion is a very high-acceleration drive; even at .1 G, over a century you would accelerate to highly relativistic velocities, which is far beyond Orion's capability.

Taking the extreme optimistic end of the Orion-like design on atomic rocket, {F = 8.00e6N, u= 9.80e6m/s}, a century of operating time gives us about 2.58e9kg of fuel (not counting inefficiency). If the total mass is about 2.59e9kg, the 1.0e7kg payload is the same as that in the calculation I've done on the previous page. It is very optimistic, given the bulk required by Orion designs, but it is probably fair for comparison purposes, since the previous page assumed multiple engines, which will likewise need more support structure. Accelerating for 93.17 years gets us 3.17 light-years with a cruise velocity of 0.0852c, comparable to that on the last page (about the same speed, less distance, but in half the time). The remaining 6.83 years will be used for deceleration back to rest, with braking distance of 0.405ly.

Radiation shielding is not cumulative; a block that protects you against 99% of radiation now will still protect you in ten years.

The block will absorb radiation and hence itself become radioactive. Over small time-frames, it's not an issue, but for centuries of close contact with continuous nuclear blasts, it may become a legitimate concern. Although, it may be mitigated by having the shielding made of atoms for which neutron capture still yields fairly stable isotopes (but I can't think of an example right now, if any).

[TheLemur]

The result of being able to send out a "space water truck" rocket that over repeated trips brings back orders of magnitude more than its own mass is that the cost of material in earth orbit can be vastly reduced.

Nice. NTRs can also be used to launch stuff from the surface, using LH2 as propellant.

[TheLemur]

It's a question of energy delivery across light-years. Try to calculate the requisite precision. As an initial stage, it's a good idea, but at interstellar distances, it really stretches credulity.

If we can get a very large power station to the star, you can launch a ship from Earth, use a magnetic or electric sail for deceleration, then launch it back from the star's own power station. But this is obviously for use only after a large infrastructure has already been developed.

Accelerating for 93.17 years gets us 3.17 light-years with a cruise velocity of 0.0852c,

That's only a force of 0.0009 G; why are there such long intervals between launching bombs? Didn't I already go through the rant about how faster acceleration gets you there quicker for the same amount of fuel?

The block will absorb radiation and hence itself become radioactive.

Only if it's neutron radiation, and then only if it's made out of a high-activation material, which obviously it won't be. Activated materials emit primarily beta particles anyway, which are easily shielded.

Although, it may be mitigated by having the shielding made of atoms for which neutron capture still yields fairly stable isotopes (but I can't think of an example right now, if any).

It also depends on the reaction cross-section. Boron, for instance, is very good at absorbing neutrons without becoming radioactive, while helium doesn't absorb neutrons at all.

[rhoenix]

Sikon, that was excellent, and deeply fascinating reading - I thank you. The ice rocket idea I like very much, because it's concise, and seemingly efficient and powerful enough to meet my needs.

Kuroneko, I also thank you for some of the concepts, and most of the math involved as you showed. It helped me understand a few aspects of what I was reading and thinking about that admittedly I didn't quite understand before.

However, with apologies to those that feel that require it, I'm afraid my next thread is going to involve handwavium.

[Kuroneko]

That's only a force of 0.0009 G; why are there such long intervals between launching bombs?

At the above parameters, average engine power is 3.92e13W, which is about a 9.4 kilotons per second, neglecting efficiency concerns. The proper acceleration at the beginning of the journey would be about 3.1e-4g and a very uncomfortable 2.7g just before deceleration. Let us hope it is an automated ship.

Didn't I already go through the rant about how faster acceleration gets you there quicker for the same amount of fuel?

You have, but it was very misguided. Since a higher acceleration is typically accompanied by a lower thrust, missions with time dominated by a period of no thrust will actually favor higher-impulse, lower-thrust designs. Just try a sample calculation for the same amount of fuel and engine power--with the above extreme Orion parameters, doubling the thrust and burning all the fuel would enable one to reach 0.0852c and cover 0.397ly in 25 years; keeping the same thrust will give us 0.169c and cover 3.17ly in 100 years. So, in a 100 years, the former will be at 6.47ly while the latter at only 3.17ly, yes, but this doubling of the distance came at the price of halving the speed for any further progress beyond that point.

So, if you're trying to go farther than the immediate stellar neighborhood, getting more acceleration for the same engine power and fuel will be a very inadvisable thing to do. A hundred years of fuel is obviously very excessive; for a more reasonable amount of fuel, even our closest neighbors would favor a lower-thrust, higher-impulse rocket if we keep engine power and amount of fuel constant.

Only if it's neutron radiation, ...

No. The gamma radiation from a nuclear blast will pose a similar danger as well, and it is energetic enough to potentially transmute the shielding material as well.

... and then only if it's made out of a high-activation material, which obviously it won't be. Activated materials emit primarily beta particles anyway, which are easily shielded.

Good point.

[Sikon]

If one takes the magnetic confinement fusion figures on atomic rocket as fairly precise (despite the fact that we don't even have a working prototype), {F = 5.00e4N, v_e = 8.00e6m/s, m_e = 6.00e2kg}, let's have 144 engines with enough fuel for 200 years of operating time: 5.68e9kg fuel + 8.64e4 kg engines + ? payload = 5.69e9kg. If ships accelerates for 192.04 years, it will cover 4.42 light-years and have a final velocity will of a respectable 0.0848c for coasting for an arbitrary amount of time; the deceleration phase will take 7.96 years, with braking distance of 0.492 light-years.

I was curious as to what could be possible with optimistic, but still somewhat realistic, technology.

[...] fair for comparison purposes [to Orion concepts] [...]

The Atomic Rockets web page giving those MC fusion engine figures lacks proper supporting engineering analysis, actually not giving so much as a reference for the figures. It states the imaginary magnetic-confinement fusion engine could obtain 200 GW exhaust power with a mass of 0.6 metric tons.

For perspective, that's implicitly implying the 600-kg device can internally confine and deal with a rate of internal energy release more than the energy of 50000 kilograms of TNT high-explosive exploding per second ... as that's the energy involved in 200 GJ / sec. (1 kilogram of TNT =~ 4 MJ). It's no surprise that there are no engineering calculations supporting such being structurally and thermally workable.

For perspective, the ITER magnetic-confinement concept is to hopefully manage to obtain a 0.5 GW burn in a tokamak inside a cyrostat of a size 24m high and 28m diameter, corresponding to ~ 1+E4 tons for that component of the overall facility, let alone the total mass. (That's with cyclotron and beam systems running off outside electricity delivering up to 0.11 GW of heating power to heat plasma and drive plasma current in it). Based on actual analysis (including suitable heat removal), that real-world concept has literally not a millionth of the power to mass ratio of the imaginary magnetic-confinement engine. Obviously, anything attempted to be used in a starship engine would undergo far more mass minimization than a terrestrial system, but there is no reason to treat as plausible a 200-GW 600-kg MC engine unsupported by any proper engineering calculations.

One could get into issues like the degree of break-even in the magnetic confinement system versus driver and electrical power requirements, and so on. Among hypothetical concepts for lightweight space electricity generation based on at least some engineering analysis, they tend to max at on the order of 3 MW / ton as described before, not good for this when its 0.6 tons of mass make that ~ <= 0.002 GW of electricity generation.

Anyway, a 200-GW 0.6-ton magnetic-confinement fusion engine being workable is unsupported.

In contrast, part of the reason I talk about Orion is because it has undergone real engineering calculations for heat transfer, ablation, induced radioactivity, etc. The $11+ million spent on it wasn't just about any guy making up performance numbers without such calculations like the average sci-fi engine, rather the opposite. See earlier discussion.

[Kuroneko]

... Anyway, a 200-GW 0.6-ton magnetic-confinement fusion engine being workable is unsupported. ...

A just criticism. I wasn't overly concerned about having a bit of unobtanium, as this is in the context of sci-fi, but my comment was indeed unfair to Orion.

[TheLemur]

and it is energetic enough to potentially transmute the shielding material as well.

What are you smoking? Gamma radiation of any kind can't physically change a nucleus except by splitting it apart, and splitting a nucleus typically requires an activation energy of ~40 MeV or so. The cross-sections for lower energy gamma rays are thus atrocious; for a 2.6 MeV gamma ray, even for a fairly favorable reaction (the splitting of Be-9 into a neutron and two alphas), the cross section is only thirty microbarns. In comparison, the neutron absorption cross section for a metal is typically a few barns, and for a few materials (boron, gadolinium) it can be in the thousands of barns.

At the above parameters, average engine power is 3.92e13W, which is about a 9.4 kilotons per second, neglecting efficiency concerns.

According to my calculations, that acceleration for that mass gives a force of around 8e6 N. This seems to be very low for the detonation of a 10 kT weapon; where are you getting your figures from?

Since a higher acceleration is typically accompanied by a lower thrust,

A higher acceleration is accompanied by a lower thrust?! Did you forget F=ma? Acceleration is directly proportional to thrust for any given mass.

Just try a sample calculation for the same amount of fuel and engine power--with the above extreme Orion parameters, doubling the thrust and burning all the fuel would enable one to reach 0.0852c and cover 0.397ly in 25 years; keeping the same thrust will give us 0.169c and cover 3.17ly in 100 years.

How do you figure that? You're obviously getting different delta-V figures for the two ships, even though they have the exact same mass ratio and the exact same propulsion system.

[Kuroneko]

According to my calculations, that acceleration for that mass gives a force of around 8e6 N. This seems to be very low for the detonation of a 10 kT weapon; where are you getting your figures from?

P = [1/2]Fu, where F is thrust and u is exhaust velocity. I've used the figures for thrust and exhaust at the atomic rockets page; given that it is already nearing ten-kilotons per second, it seems quite reasonable.

A higher acceleration is accompanied by a lower thrust?! Did you forget F=ma?

For "thrust", read "exhaust velocity". My apologies.

How do you figure that? You're obviously getting different delta-V figures for the two ships, even though they have the exact same mass ratio and the exact same propulsion system.

No, they're not the same. To keep the power constant (the only way this is going to be a fair comparison), the exhaust velocity needs to be halved. This also means the fuel is burned out in one-quarter of the time, since to be fair we should also keep the same amount of fuel. I've already covered how to do the velocity calculations on the previous page.
---
Edit:
Arg, somehow missed seeing this part...

What are you smoking? Gamma radiation of any kind can't physically change a nucleus except by splitting it apart, and splitting a nucleus typically requires an activation energy of ~40 MeV or so.

Why would you think that I was referring to nuclear fission in the radiation shield itself? Nuclear binding energy is only 7-8MeV/nucleon. I was under the impression that a nuclear bomb can produce gammas in that range, especially if it is thermonuclear, although perhaps my memory is failing me. I'll check later; right now I'm going to sleep.

[TheLemur]

I've used the figures for thrust and exhaust at the atomic rockets page; given that it is already nearing ten-kilotons per second, it seems quite reasonable.

Okay, the exhaust velocity I can buy, but it seems that an A-bomb would exert more thrust than that.

To keep the power constant (the only way this is going to be a fair comparison),

Why does the power need to be constant? If you're using up your A-bombs twice as fast, obviously the power is going to be twice as large!

Why would you think that I was referring to nuclear fission in the radiation shield itself?

How else would a gamma ray transmute elements?

Nuclear binding energy is only 7-8MeV/nucleon. I was under the impression that a nuclear bomb can produce gammas in that range,

It can. Binding energy and activation energy are not the same thing.

[Wyrm]

TheLemur, in regards to your post, the horse has already left the gate for a blow-by-blow, and I'm tired, so I'll summarize.

One, totally you're full of shit on your assertion that time is minimized by fixing total delta-V and letting acceleration approach infinity, because you pretend that total delta-V is not a function of thrust (and thus, acceleration). Kuroneko provides the key equation:

P = Fu/2

where P is the power (in terms of energy input into the engine per second, by way of fuel or what have you), F is the thrust and u is the exhaust velocity. It's clear that there is a fundamental trade-off between thrust and exhaust velocity, and therefore total delta-V. The only way to increase thrust (and therefore, acceleration) without decreasing exhaust velocity is to increase power... but of course, if the power gets too high, your engine will melt.

If you actually bother to set up a relation between the total time you spend on an acceleration/deceleration trip and the fraction of the time you spend accelerating, you find that the maximum occurs when you spend half the time accelerating, which implies that you spend half the time decelerating, and thus no time coasting. Therefore, if you can sustain your maximum acceleration throughout your trip, then you minimize your trip time.

This, by the way, jives with the observation that the more delta-V you have availible during a trip, the faster transfer trajectories are availible to you. My hypothetical 0.0003 G ion driven spacecraft accumulates 231 km/s delta-V on its flyby trip to Pluto. It therefore (realistically speaking) has faster trajectories availible to it than the ~50 km/s orbit (which would be availible to you, as you were actually bothering to slow down at the end).

Even if I take the SMART-1, with its 3100 seconds specific impulse at 20 mN on a 367 kg spacecraft (assuming this is fully loaded weight), and assume that I have a battery of sufficient energy to power this bugger as long as necessary, and can reduce the dry mass of the craft to 26 kg, that spacecraft would be able to reach Pluto in 16 years, 4 months (delta-V = 28.21 km/s, enough to escape the solar system). The calculation is rather simple, because I can deduce the flight time from the minimum acceleration, and from that and the propellant consumption rate the propellant mass.

Anyway, on to transfer orbits. Actually, for continuing accelerations, there is no orbit; the trajectory is spiral, not elliptical. In fact, under constant acceleration, it is a logarithmic spiral. You're constantly adding energy to them as you go through these sustained thrusts. Yes, they'll begin tight, but they will slowly straighten out as the spacecraft climbs out of the planet's gravity well. This straightening out happens faster than you think.

Speaking of trajectories, it's also clear you don't know what escape velocity means. If you achieve escape velocity for a body at a certain point, it doesn't matter what the craft's trajectory is; the object will escape the body's gravity, as long as you don't smack into it. (True, you'll get some velocity adding depending on your trajectory, but there is a lower limit to the amount of KE you need to add to achieve escape.) That you think that some special trajectory is required to escape Earth's gravity betrays a deep misunderstanding of the concept. That means that, if you add 8 km/s worth of kinetic energy to an object in LEO (which takes 9-10 km/s to reach from the surface), you will escape the solar system as long as you don't crash into anything. Robert Heinlein's statement that, once you're in Earth orbit, you're "halfway to anywhere," isn't just a clever maxim.

(In fact, interplanetary craft do indeed fire their engines at specific times when they escape Earth's orbit, but only because they want to escape along a path that will take them to their destination; if you just want to escape Earth's gravity, you can leave any time.)

I consider an engine type as having been "proven to work" when we have an example of a system flying and working. "Almost" only counts horeshoes and handgrenades. Ion drives have the highest specific impulse of those proven engines.

I also find it hallarious that you think that photon drives as being on your list of "better engines"; as a photon engine requires a Chernobyl in space (~300 MW) just to get one lousy Newton of thrust, it's even more pathetic than an ion drive.

Finally, rhoenix has not yet stated what he wants his spacefaring civilization to look like. Sure, you can't do some things with ion drives as your state of the art engines, but the sword cuts both ways: there will be "hard" sci-fi stories that rhoenix could not write if his engines are too powerful. For one thing, you can forget unregistered tramp freighters getting anywhere near spaceports — because an engine capable of pushing a spacecraft across the solar system in a week is a weapon of mass destruction. Only certified pilots would be able to fly them! Limiting the technology can be challenging and make the story more interesting. If you're using ion drives as the state-of-the-art propulsion, then giant rockets with poor acceleration with crew in hypersleep (read, sitting ducks) are prey to pirates lying in wait with high-thrust, low-efficiency chemical rockets. "Hard" sci-fi is, after all, a thinking man's fiction, and we can at best advise rhoenix, not command him.

Thank you.

[TheLemur]

Kuroneko provides the key equation:

P = Fu/2

where P is the power (in terms of energy input into the engine per second, by way of fuel or what have you),

Why do you assume the power is fixed? Obviously if you're burning fuel at twice the rate, you're going to get twice the power. This makes sense for an ion engine or other engine with a fixed, external power source, but for Orion the power is provided by the bombs themselves. Essentially, we made different assumptions; you assumed fixed power and fuel mass, I assumed fixed delta-V and exhaust velocity.

that spacecraft would be able to reach Pluto in 16 years, 4 months (delta-V = 28.21 km/s, enough to escape the solar system)

Could you please go into further detail on this calculation? What trajectory are you taking and what assumptions are you making?

Yes, they'll begin tight, but they will slowly straighten out as the spacecraft climbs out of the planet's gravity well.

This is a *bad* thing, because acceleration is most fuel-efficient at high velocities and big circular orbits have low velocities. Why do think SMART-1 used its thrusters only at perigee?

That means that, if you add 8 km/s worth of kinetic energy to an object in LEO (which takes 9-10 km/s to reach from the surface), you will escape the solar system as long as you don't crash into anything.

This is true. However, if you add 1 m/s to an object in LEO, and then 1 m/s to an object in LEO+1km, and then 1 m/s to an object in LEO+2km, etc., even if you still add the full 8 km/s, you're not going to escape the solar system, because at a certain point you will escape from the Earth's gravity with almost zero additional velocity and wind up in solar orbit, from which it takes a minimum of 12 km/s to escape onto a hyperbolic trajectory.

as a photon engine requires a Chernobyl in space (~300 MW) just to get one lousy Newton of thrust, it's even more pathetic than an ion drive.

Yes, but it also requires no fuel other than the power source. Did you forget about that part? An ion engine trying to accelerate to relativistic velocities will run out of fuel very quickly; a photon engine will not.

I consider an engine type as having been "proven to work" when we have an example of a system flying and working. "Almost" only counts horeshoes and handgrenades. Ion drives have the highest specific impulse of those proven engines.

Then your 260 km/s Earth-to-Pluto example is bunk, because we've only flown ion engines with a specific impulse of a few thousand seconds.

[Ender]

You think walking is more efficient then a car, bike, or airplane?

Maybe not a bicycle, but it's sure as hell more efficient than a car or airplane. Automobiles and airplanes use roughly 4 MJ per passenger mile. (source) Assuming you're taking your time, you can walk two miles in one hour. In one hour, at a very high power consumption of 500 W, you would consume 1.8 MJ, versus 8 MJ for automobiles and airplanes.

1) Wikipedia is not a valid source here
2) Look at the reference, the source for that information is stted as a year, not a publication, which means for all I know someone just made it up.
3) That is a comparison of total fuel efficiency. You compare their fuel efficiency to energy expended, which deliberately ignore the inefficiencies of biological processes. The stomach is only about 18% efficient at converting the energy in your food into energy you can use. So your dishonest 1.8 MJ should really be 10 MJ, which makes a car far more efficient.

4) you are shifting the goalposts by redefining efficient when it is convienent to you. Odd that for your arguements that non instant acceleration is inefficient you base it off speed, but when confronted here, you try to compare energy efficiency (though deliberatly exclude information to skew the answer)

[Ender]

Ablation and spalling would be unavoidable, as well as extreme radiation shielding because it will build up.

Radiation shielding is not cumulative; a block that protects you against 99% of radiation now will still protect you in ten years. But ablation and spalling are definitely problems.

You are full of shit. Neutron radiation will steadily break down your shielding by activating it. That is why such high precautions are taken during refueling operations - the shielding itself has become radioactive. Of particular concern is the cobalt 59 in the steel becoming cobalt 60, one of the most dangerous isotopes reactor operators deal with. Amazing how you keep tactically forgetting important pieces of information like this until someone calls you on it.

The block will absorb radiation and hence itself become radioactive.

Only if it's neutron radiation, and then only if it's made out of a high-activation material, which obviously it won't be. Activated materials emit primarily beta particles anyway, which are easily shielded.

That's why we don't use iron, cobalt,argon, flourine, xenon, cessium, or nitrogen in our shielding today, right? Oh wait, those are all used in nuclear reactors and are all nuclides of concern, particularly the iron and cobalt, which make up most of the highly radioactive "crud" in nuclear reactors.

[TheLemur]

1) Wikipedia is not a valid source here

If you actually spent thirty seconds figuring out where the information came from, you'd realize it was from the US Bureau of Transportation Statistics, complete with a .gov website.

(direct link)

2) Look at the reference, the source for that information is stted as a year, not a publication, which means for all I know someone just made it up.

See above. The BTS may distort things, but they don't just make them up.

You compare their fuel efficiency to energy expended, which deliberately ignore the inefficiencies of biological processes. The stomach is only about 18% efficient at converting the energy in your food into energy you can use. So your dishonest 1.8 MJ should really be 10 MJ, which makes a car far more efficient.

Firstly, even if it was 10 MJ a car wouldn't be "far more efficient", since a car would use 8 MJ. Second, I deliberately used a very generous estimate of 500 W total power (including waste heat), which is equivalent to around 430 kcal/hr; this link gives 265 kcal/hr burned for walking at two MPH even when the guy weighs 120 kg.

4) you are shifting the goalposts by redefining efficient when it is convienent to you. Odd that for your arguements that non instant acceleration is inefficient you base it off speed, but when confronted here, you try to compare energy efficiency (though deliberatly exclude information to skew the answer)

Acceleration for human walking is damn close to instant; it's zero to full speed in less than a second. Please provide an example of information I've supposedly excluded and any redefinitions I've supposedly made.

[TheLemur]

You are full of shit. Neutron radiation will steadily break down your shielding by activating it.

Okay, just the fact that you used the word "break down" demonstrates that you have no idea how radiation shielding works. It's not like a Star Trek shield that can be at 60% health or a video game shield with a certain number of hit points. Radiation shielding is deliberately manufactured to have a very low neutron-activation cross-section.

Of particular concern is the cobalt 59 in the steel becoming cobalt 60, one of the most dangerous isotopes reactor operators deal with.

Okay, firstly cobalt-60 emits beta radiation, which is very easily blocked (you can block it with only a few millimeters of aluminium, iron or plastic). Secondly, the vast majority of steel does not contain significant levels of cobalt (source).

That's why we don't use iron, cobalt,argon, flourine, xenon, cessium, or nitrogen in our shielding today, right?

We don't. You quite clearly have no clue whatsoever what you're talking about. Argon and xenon are noble gases that are totally useless for shielding of any kind because they're GASES. Nitrogen is also a gas, but it can be fixed into various chemical compounds. It is not used in radiation shielding because it's not good for much (low density, low neutron absorption). Fluorine is a ridiculously reactive gas; in fact, it's the closest thing we have to a universal solvent. It will eat through glass, most plastics, most metals, and organic compounds. It is not used in shielding for the same reason as nitrogen. Cesium is another ridiculously reactive element that melts at a very low temperature and is too expensive to use for bulk shielding. Iron is used in radiation shielding, but it is fairly immune from activation as only 0.2% can transmute and the isotope it forms is a quick-decaying beta producer. Cobalt is not used because it had a problem with neutron activation, as you pointed out, and it is also quite expensive.

[Wyrm]

Kuroneko provides the key equation:

P = Fu/2

where P is the power (in terms of energy input into the engine per second, by way of fuel or what have you),

Why do you assume the power is fixed?

Power is variable, up to a point. At this point, the engine is absorbing so much heat from the exhaust that, in order to maintain sufficient thermal conduction, the walls have to be at the melting point... which is not good.

Obviously if you're burning fuel at twice the rate, you're going to get twice the power. This makes sense for an ion engine or other engine with a fixed, external power source, but for Orion the power is provided by the bombs themselves.

If you increase bomb rate, you of course increase thermal and mechanical stress on the pusher plate. Eventually, given some level of technology and ship mass, the plate's gonna fail. Any given Orion engine will a maximum practical bomb rate, beyond which the pusher plate breaks down so fast that you will not be able to complete your mission. This, of course, limits the amount of power availible.

You seem to imagine that you can just increase power consumption of some kind of engine indefinitely. You cannot. With any engine, there will be a limit to how much power you can give the exhaust.

Essentially, we made different assumptions; you assumed fixed power and fuel mass, I assumed fixed delta-V and exhaust velocity.

And of course, delta-V is a function of how long you burn your engines, which is the point at issue — how long do you burn your engines to minimize time?

Given my assumptions are sensible and yours aren't, I believe my conclusion.

that spacecraft would be able to reach Pluto in 16 years, 4 months (delta-V = 28.21 km/s, enough to escape the solar system)

Could you please go into further detail on this calculation? What trajectory are you taking and what assumptions are you making?

That the dry weight of the mission can be brought down to 26 kg (the rest of 367 kg is propellant — hey, it's sci-fi), that the ion engine has a battery made by the MONDO Company, able to power the ion engine throughout the trip, and the engine has a thrust of 20 mN at a specific impulse of 3100 seconds. All of these I've stated. Also, I assumed no gravity and a trip distance of 7.3 billion km, and the mission is a flyby.

The acceleration of the craft is no less than 20 mN/367 kg = 5.45e-5 m/s. Take the familiar D = at²/2 and solve for t, which you get 5.176e8 seconds, which is about 16 years and a bit. Multiply acceleration by flight time to get the change in velocity: ∆v = at = 28.21 km/s. This is lowballing; the acceleration is an increasing function of time.

To get propellant consumption, find the exhaust velocity, which is specific impulse multiplied by 1 G, here u = 30.38 km/s. A chunk of the craft's mass (propellant), ∆M, is thrown at exhaust velocity u for a gain in momentum ∆p = u∆M. The increment of momentum per increment in time ∆t is ∆p/∆t = u ∆M/∆t, and if we let ∆p/∆t become infinitesmial, it becomes dp/dt = u dM/dt. dp/dt is of course, thrust (force). So 20 mN = u dM/dt -> dM/dt = 20 mN/(30.38 km/s) = 6.58e-7 kg/s. Multiply by the number of seconds and you get 340.75 kg, which is lower than the beginning mass of the mission by about 26 kg.

I could continue, making acceleration a function of mass, and therefore time using the fuel consumption to get a distance as a function of time with the functional acceleration, but that was a bit too much for so late at night. Please note that because of the increasing acceleration, time will be reducd.

Yes, they'll begin tight, but they will slowly straighten out as the spacecraft climbs out of the planet's gravity well.

This is a *bad* thing, because acceleration is most fuel-efficient at high velocities and big circular orbits have low velocities.

Did you flunk high school physics, or have you not just taken it yet? The amount of kinetic energy you have to expend to accelerate is independent of your initial speed.

Why do think SMART-1 used its thrusters only at perigee?

Because it wanted specifically to get into a lunar transfer orbit, which would minimize the delta-V needed? I did say that orbit considerations are important if you want to get to a specific destination, did I not?

This is true. However, if you add 1 m/s to an object in LEO, and then 1 m/s to an object in LEO+1km, and then 1 m/s to an object in LEO+2km, etc., even if you still add the full 8 km/s, you're not going to escape the solar system, because at a certain point you will escape from the Earth's gravity with almost zero additional velocity and wind up in solar orbit, from which it takes a minimum of 12 km/s to escape onto a hyperbolic trajectory.

Did you get your celestial mechanics degree out of a cracker jacks box? C'mon! Make more shit up. I could use a laugh.

Yes, but it also requires no fuel other than the power source. Did you forget about that part? An ion engine trying to accelerate to relativistic velocities will run out of fuel very quickly; a photon engine will not.

Neither will the power source, I see. Oh, wait; It will.

And stop moving the goalposts, fucktard. One of your requirements for "better engines" was "higher thrust than ion engines." Photon engines do not fit this critera.

I consider an engine type as having been "proven to work" when we have an example of a system flying and working. "Almost" only counts horeshoes and handgrenades. Ion drives have the highest specific impulse of those proven engines.

Then your 260 km/s Earth-to-Pluto example is bunk, because we've only flown ion engines with a specific impulse of a few thousand seconds.

If you want. (It was 231 km/s, BTW.) It was a gedanken experiment, anyway, assuming that we can spiffy up the ion engine to high but not completely outrageous figures. My point still stands: ion engines have the highest specific impulse of the proven engines.

PS, even I know that your spew is not what Ender meant by "breaking down".

[Kuroneko]

Okay, the exhaust velocity I can buy, but it seems that an A-bomb would exert more thrust than that.

Why? (Particularly since this is the time-averaged thrust, while the actual detonation lasts something on the order of microseconds.)

Why does the power need to be constant? If you're using up your A-bombs twice as fast, obviously the power is going to be twice as large!

Wait... so the entire point of your previous "rant" was that if one uses more power and more thrust, one gets there faster? I seem to have been under the mistaken impression that you were at least trying to say something non-trivial.

It can. Binding energy and activation energy are not the same thing.

The gamma energy is enough to make radioactive isotopes by knocking off neutrons. That's what I meant; apologies for using the wrong word; I was in a rather sleep-deprived state.

---[another edit]---
Addendum: since the question has attracted some interest, let's try to find the distance traveled as a function of time, assuming Newtonian mechanics . Let R = 1 - Ft/[Mu] be the mass fraction at t. Then v dt = -u log R dt = Mu²/F log R dR. Integrating, the distance traveled for zero initial velocity is D = [Mu²/F][Rlog(R)-R+1], where R is now the mass fraction at some particular t>0.

---[yet another edit]---

If you actually bother to set up a relation between the total time you spend on an acceleration/deceleration trip and the fraction of the time you spend accelerating, you find that the maximum occurs when you spend half the time accelerating, ... .

This is true for a constant acceleration, but not a constant thrust rocket. If the exhaust velocity and thrust are constant, then the only relevant parameter is the mass fraction. Thus, taking a line segment OM for total mass, with subsegments OT for turnaround fuel, OP for total fuel, and PM for payload, we must have the proportion TM:OM :: PM:TM. Therefore, thus the amount of fuel burned before turnaround is the geometric mean of the initial mass and payload. That's why the acceleration and deceleration times in the calculations above were so skewed (93 years accelerating and 7 years decelerating for a 100-year fuel burn).

[Howedar]

Okay, just the fact that you used the word "break down" demonstrates that you have no idea how radiation shielding works. It's not like a Star Trek shield that can be at 60% health or a video game shield with a certain number of hit points. Radiation shielding is deliberately manufactured to have a very low neutron-activation cross-section.

You're absolutely right. That's why the US NRC has Power Reactor Regulation Guide 1.99, "Radiation Embrittlement of Reactor Vessel Materials".

[Wyrm]

Addendum: since the question has attracted some interest, let's try to find the distance traveled as a function of time, assuming Newtonian mechanics . Let R = 1 - Ft/[Mu] be the mass fraction at t. Then v dt = -u log R dt = Mu²/F log R dR. Integrating, the distance traveled for zero initial velocity is D = [Mu²/F][Rlog(R)-R+1], where R is now the mass fraction at some particular t>0.

Cool! I didn't think of using the mass fraction when I was doing the same integration (the result looks similar). Damn thing looks a pain to invert, though. Ugh.

I assume that the mass fraction R is the ratio of initial mass at the beginning of the burn and final mass at the end. Am I correct?

This is true for a constant acceleration, but not a constant thrust rocket. If the exhaust velocity and thrust are constant, then the only relevant parameter is the mass fraction. Thus, taking a line segment OM for total mass, with subsegments OT for turnaround fuel, OP for total fuel, and PM for payload, we must have the proportion TM:OM :: PM:TM. Therefore, thus the amount of fuel burned before turnaround is the geometric mean of the initial mass and payload. That's why the acceleration and deceleration times in the calculations above were so skewed (93 years accelerating and 7 years decelerating for a 100-year fuel burn).

Yeah, well, I figured the constant thrust case would be a bit different. The point of my particular tirade was that flight time is minimized when you spend all your time with your engine on. No coasting.