Particle Puff Propulsion - 25%c in Our Lifetime

Isaac Kuo · Member

Particle Puff Propulsion - 25%c in Our Lifetime

Isaac Kuo · Member

Hmm...my post is cut short. Here's a link to my proposal: Google Groups Article

-- Edited by Isaac Kuo at 05:50, 2005-11-17

10kBq Jaro · Senior Member

Your concept does not seem entirely unfamiliar -- I believe that some variation of it will eventually be made to work....

10kBq Jaro · Senior Member

Isaac Kuo wrote:

These disposable emitters are laid out in a long line along the acceleration path of the starship.

Incidentally, you cannot simply "lay them out along the acceleration path." Especially not the ones near the end of the runway, where the ship will be going close to 25%c. Remember relativity -- the situation would be just like a stationary ship getting hit by an object going 25%c. VERY deadly !

The packets would have to be sent out ahead of the ship's departure, at various velocities, to closely match the speed of the ship when its going to pick them up.

Of course, accelerating anything as big as a nuke bomb (or "disposable emitter") to 25%c would take a HUGE amount of energy. That's why more common versions of this "fuelled runway" concept envision using small fission/fusion fuel pellets instead of entire bombs, to be set off, following pick-up by the ship, by an on-board inertial confinement gizmo using either laser or particle beams, possibly with a few antiprotons tossed in as an accelerant. The total energy required to lay out such a runway might turn out to be close to that using complete bombs, but being divided into a great many small pellets, its no doubt much more doable practically....

Isaac Kuo · Member

The ship doesn't pick up the pulse units nor does it collide with the pulse units. The pulse units detonate after the starship passes by. Due to the design of the pulse unit, it sends out a very fast pulse of hydrogen gas out its "nozzle" while the rest of the remnants explode outward in all directions at around 3%c or less.

In fact, the most dangerous time for the starship is near the start of the bombtrack, where the ship is most exposed to neutron radiation from the exploding pulse units. Once it gets up to around 5%c, the majority of neutrons can't even catch up with the ship.

When I say the pulse units are laid out along the starship's acceleration path, I don't actually mean they are DIRECTLY in the path. Instead, they are laid out off to the side of the path, maintaining proper position with their own stationkeeping thrusters. The starship needs to stay centered along the planned acceleration path also, of course. Rather than use stationkeeping thrusters, the starship can be maneuvered by introducing slight extra delays to the appropriate pulse unit detonations. A small extra delay means less of the particle puff hits the starship sail. Since the pulse units are off to the side of the starship's path, they are pointed slightly inward providing an inherent sideways thrust. By delaying the pulse units on one side, there's a net sideways push toward that side.

10kBq Jaro · Senior Member

OK -- got it now !

Hmmm -- certainly different

Any idea just how precisely the blasts would have to be timed when the ship blows by at ~25%c ? ....seems kind of challenging.

G R L Cowan · Veteran Member

How do they take photos of racehorses just as their noses reach the finish line? Similarly, I would guess, can be detonated these proton throwers just as the vessel is, like, 100 km past them.

But if it's a quarter of 'c' within our lifetime, ... how far does that line of exploders extend? How many tonnes of bomb-grade nuclear fuel is in them?

--- Graham Cowan, former hydrogen fan
boron as energy carrier: real-car range, nuclear cachet

GoogleNaut · Guru

...the short answer: a whole bunch!

Yes, offset blasts could be done. I'm not sure how much coupling efficiency you'll get. Project Orion used various figures for the net specific impulse, some as low as 4000 sec, and some as high as 300,000 sec for the ultra huge Orion with high yield thermonuclear devices for pulse units.

Position and pointing accuracy are a double must--first, the pusher plate of the ship must intersect the momentum stream from a pulse unit. Second, pointing errors will result in reduced momentum transfer, no momentum transfer, or possibly damage or destruction of the vessel.

100 km was used as the offset distance. At 1/4 c the ship would be moving at nearly 75,000 km/s. If the pulse unit could in fact achieve that much speed in its momentum stream (dubious at best, as even the most violent radiation implosion in Teller-Ulam configuration H-bomb implodes at no more than about 2000 km/s, and the fusion fireball expands at a similar pace initially, it seems impossible to achieve 1/4 c this way.) Anyways, even if they could achieve 1/4c, then they would have to fire at about a 45 degree intercept angle in order to intersect the ship. In otherwords, the pulse units would have to "lead" the ship like a skeet-shooter leading a clay pidgeon. The momentum would also be offset by 45 debrees at the pusher plate, resulting in not only a reduction of thrust by a factor of the cosine of 45 degrees, but would also push the vessel off course due to lateral momentum transfer. The only way to nullify this, is to have two tracks, side by side, popping off their acceleration bombs simulatneously. This would in effect negate the cosine 45 degree becuase we would regain one additional factor....

OK, I did a little figuring on some paper and came up with some numbers. I won't post all of the details, but on request I can email a fairly detailed description of what I've done. I used some analytical geometry and some basic physics concepts to obtain the following results. The window for a successful momentum transfer with a lateral offset of 100 km, for an object moving at 75,000 km/s, with a 'target length' of 100 m (length of the ship, lets say) with a target diameter of 100 m (width of pusher plate) with an interception angle of 45 degrees will require approximately:

a pointing accuracy of order phi=20.3 microradians, and a gate time of order 670 nanoseconds. That is, the error and the dispersion of the impulse jet cannot exceed 20.3 microradians for a hit to occur, is would probably have to be atleast 1/4 of this for good reliability. And the gate timing, the actual timing of the detonatiuon must be synchronized to less than 600 nanoseconds in order for the pulse to arrive at the moving pusher plate--a very, very, very restrictive window to hit reliably!

Realistically, relativisitc effects start to become important in this speed regime as well, which adds complications in both timing and position prediction, but also creates a momentum 'drop off' because the momentum coupled to the pusher plate will be slightly less because the ship is receding away from it. Eventually, diminishing returns will set in, and even if every momentum pulse scores a hit, less and less momentum will be added as the ship accelerates. Eventually, no momentum would be transfered at all.

It is much easier, in my opinion to carry a supply of 'fuel' (pulse units) onboard, and use them in a way such that every pulse imparts momentum. As the remaining pulse units are accelerated with the ship, their relativistic momentum will actually increase and the rest energy of each pulse unit is increased by the kinetic energy stored within its motion along the same path as the ship--what this means is that there is no momentum 'tail off effect' and each subsequent pulse unit imparts the same delta-v as previious units from the point of view of the ship. From a stationary observer, the momentum transfer and energy release of each pulse unit should appear to increase, however the mass of the ship is also increasing (due to relativistic effects) so the net result is, I think, a draw...however, relativity is pretty clear that the closer to the speed of light an object moves, then the acceleration will appear to drop of anyway...

...an interesting problem...!

-- Edited by GoogleNaut at 09:33, 2005-11-18

-- Edited by GoogleNaut at 09:36, 2005-11-18

Isaac Kuo · Member

I think a sideways offset distance of 100km is too much, unless we have developed M2P2 sail technology so that the actual size of the magsail is larger than 100km. With an advanced M2P2 sail, the sail is essentially a large field of very thin plasma. The pulse unit punches through that thin plasma causing a negligible loss in speed, and then it detonates behind the plasma sending its fast puff of hydrogen at the sail.

The reason very high speeds theoretically up to 40%c is possible in the proton puff is because of thermal rocket theory. The heavy fission fragments in the "reactor chamber" have an energy of around 84MeV per particle. If a small enough amount of hydrogen is heated by the fission fragments to not cool it down significantly, then that means heating the protons to 84MeV per particle--that means on the order of 40%c (actually lower since I'm not using the relativistic equation).

An alternate way of looking at it is that the imploding walls of the reaction chamber compression heat the hydrogen. After thinking about the undesirable effects of cooling of the fission fragments from neutrons, and concerns about neutron radiation hazard to the starship, I think this sort of indirect heating of the hydrogen may be desirable. The walls implode in the shape of a narrow tapered cone--squeezing out the hydrogen with a "zipper" effect. In this case, the fission stage acts as a very fast ablative fission fragment rocket--the outer layers get saturated by neutrons and as the reaction works its way inward the outer layers ablate away at ~3%c producing thrust inward. This forces a neutron reflecting/absorbing tamper inward which compression heats and squeezes out the hydrogen.

Because the imploding tamper will also be compression heated into a dense plasma, it needs to have a high atomic weight in order to be an effective proton reflector. The practical taper slope is determined by the square root of the atomic weight ratio.

A key question is how tight of a puff of hydrogen plasma is acheivable. My gut feeling is that the angular spread of the exhaust will be rather wide, so you don't want to detonate very far from the magsail. It may even be necessary to design the starship has a large magsail ring shape such that the bomb-bots actually pass through the center of the ring before detonating. In this case, the bomb-bots are indeed strung directly along the acceleration path, and the nozzles would be pointed in somewhat off-center directions in order to produce sideways maneuvers.

Oh--the timing of the detonations can be done in a number of ways. I like the idea of the starship emitting laser light in a circular fan all around pointing sideways to the path. Each bomb-bot has light sensors which can detect when this laser light hits it. The starship sends radio commands forward to "prime" the bomb-bots ahead of it, and tell them what time delays to use (to account for current velocity and the desire to move sideways). The primed bomb-bot simply detects when the laser light hits it and detonates after the commanded time delay.

Isaac Kuo · Member

Oh--another thing. It does NOT make sense to carry the pulse units on the starship, popping them out just before detonation. That's essentially the idea behind Orion, and this limits performance to the rocket equation.

The problem is that even our best H-bombs only really acheive energy densities about as good as fission. That means that if you use them in a rocket, you're limiting yourself to a theoretical exhaust velocity of at best around 3%c. In practice, losses will plausibly limit you to an effective exhaust velocity of 2%c or less. It would take an obscene mass ratio to get to high relativistic speeds.

The basic problem with rockets is that you have to carry the mass of your propellant with you. That implies a mass ratio which is exponentially sensitive to overhead in the pulse unit. With a bombtrack, the mass ratio is linearly sensitive to overhead (quadraticly sensitive if your starship carries a deceleration bombtrack).

With an Orion style rocket, you are limited by an average fragment velocity of around 3%c. With Particle Puff Propulsion, the design of the pulse unit takes its energy and disproportionately dumps a good fraction of it into hydrogen propellant 2-3 orders of magnitude less massive than the total device mass.

Isaac Kuo · Member

Jesus Christ! I just realized that using compression heating of hydrogen and an imploding tamper, near-C puff speeds are possible! That means that you can acheive ANY speed with Particle Puff Propulsion, so long as you're willing to put up with the energy cost.

Previously, I was assuming the fission stage would react more or less all at once, and that this energy is then thermally transfered to the hydrogen propellant. That's not actually what happens, though. What really happens is that the reaction starts off in the outer layers, and it works its way inward as the neutrons penetrate. The outer layers ablate away, acting as an ablative fission fragment rocket, reaching inward implosion speeds of perhaps up to 17%c (at which point it's moving as fast as the inward neutrons from the fusion stage). This provides the remaining uranium tamper with an inward velocity of up to 17%c.

Now, the uranium tamper is a dense plasma which acts as a very effective proton reflector. Because of its heavy atomic weight, it can squeeze out protons from the tapered imploding tube at up to 15 times its own velocity. Naturally, relativistic effects kick in long before 255% of the speed of light.

10kBq Jaro · Senior Member

GoogleNaut wrote:

the gate timing, the actual timing of the detonatiuon must be synchronized to less than 600 nanoseconds in order for the pulse to arrive at the moving pusher plate--a very, very, very restrictive window to hit reliably!

isn't that 0.6 micro-seconds ? ....that's the time between each link in a fission chain reaction, in a bomb. As I recall, something like 70 of those are required to produce a high-yield explosion. Trouble is that the first few are a crap shoot -- not very predictable. The chain reaction could get going a bit earlier or a bit later. So it seems that even if your trigger timing was perfect, you couldn't guarantee when a blast would produce the "particle puff" and therefore what its aim direction should be.

10kBq Jaro · Senior Member

It appears I was off by a factor of ten: Glasstone & Dolan write in their seminal book The Effects of Nuclear Weapons (US DoD, 3rd ed., 1977), on page 16, paragraph 1.54. that,

"the generation time is essentially equal to the average time elapsing between the release of a neutron and its subsequent capture by a fissionable nucleus. This time depends, among other things, on the energy (or speed) of the neutron, and if most of the neutrons are of fairly high energy, usually referred to as 'fast neutrons,' the generation time is about a one-hundred-millionth part (10^-8 ) of a second, i.e., 0.01 microsecond. [note 4: A microsecond is a one-millionth part of a second, i.e., 10^-6 second; a hundredth of a microsecond, i.e., 10^-8 second, is often called a "shake." The generation time in fission by fast neutrons is thus called roughly 1 shake.

And on page 17, paragraph 1.57,

"to produce 0.1 kiloton equivalent of energy....if the chain reaction is initiated by one neutron....would take approximately 51 generations [and] to release 100 kilotons of energy would require....about 58 generations. It is seen, therefore, that 99.9 percent of the energy of a 100-kiloton fission explosion is released during the last 7 generations, that is, in a period of roughly 0.07 microsecond."

This comes from the formula N = No * exp(xn), where x is the neutron surplus per fission (ie. about 1 for uranium), n is the number of generations, and No is the initial number of neutrons - also one ( N = 1.45 E23 fissions is equivalent to 1 ktTNT). This has practical implications in ACMF, where antiprotons are used to initiate the fission chain reaction with up to several dozen neutrons, making it possible to achieve high yields in very small imploded critical masses.

In paragraph 1.58 they have this interesting statement:

"In 50 generations or so, i.e., roughly half microsecond, after the initiation of the fission chain, so much energy will have been released -- about 10^11 calories -- that extremely high temperatures will be attained. Consequently, in spite of the restraining effect of the tamper and the weapon casing, the mass of fissionable material will begin to expand rapidly. The time at which this expansion commences is called the "explosion time." Since the expansion permits neutrons to escape more readily, the mass becomes subcritical and the self-sustaining chain reaction soon ends. An appreciable proportion of the fissionable material remains unchanged and some fissions will continue as a result of neutron capture, but the amount of energy released at this stage is relatively small."

Isaac Kuo · Member

In any case, I'm simply not thinking in terms of a narrow beam shot at the starship from a hundred kilometers away. It's no use thinking about pointing accuracy in terms of 20 microradians when the puff itself is maybe 20 degrees wide.

I'm thinking in terms of a probe maybe 1km in size, dominated by its 1km diameter aluminum coil electromagnet sail. Perhaps an M2P2 effect could expand the target size to much larger than 1km in diameter, but until M2P2 flies in space we won't know for sure. But let's say the target size is about 1km in diameter.

There are two reasonable possibilities for pulse unit layout. One is directly along the path, which requires the magsail to be perpendicular to the path and for the starship to be strung around it like beads on a necklace. Another possibility is for the pulse units to be just off the path, maybe by 100m to 2km off center. The magsail could laid out with its axis perpendicular to the path--this would be particularly beneficial if an M2P2 effect can be utilized.

In any case, the pointing angle should be mostly in the direction of travel. Firing at an angle steeper than 30 degrees would cut out too much of the forward component.

GoogleNaut · Guru

O.K., I see better now. I wasn't sure what kind of standoff distance we were talking about--so I used the 100 Km thing...

I see this scheme as a nuclear conceptual equivalent of a rail road. That in order to get someplace, you have to build the railroad there first...

The trouble is not so much in the accuracy of placing pulse units, gating the detonations, or even the technical realities of pulse unit design--the big problem is one of navigation and infrastructure. In order to build the 'railroad' you have to go a long-way out anyways. In order for you to achieve even 1/4c, assuming you sustain something like 1 g of acceleration, or possibly even 2 g's--if you want your passengers to be alive you can't go much higher than that. Using a little college physics we can estimate the length of this string:

D=1/2*a*t^2; where a=acceleration in meters/sec^2; t=acceleration (burn) time;

Now if we assume a=1 g (9.80665 m/s^2;) then the time needed to accelerate to close to 1/4 c, with c=299,792,458 m/s becomes:

t=(0.25*299792458 m/s)/(9.80665m/s^2)

t=7.64*10^6 seconds (or about 88.5 days)

At 2 g's., this time is halved: t=3.8*10^6 seconds.

Plugging these into the distance equation for uniform acceleration (good for a first approximation,) we get:

For 1 g acceleration: D=2.862*10^14 m or about 286 billion kilometers! This is 11 light-days long! Or to put it another way: 1700 astronomical units, the distance from the Earth to the Sun is one astronomical unit.

For 2 g acceleration: D=1.43*10^14 m, or again roughly half the length of the first track.

The logistics of transport suggests that building this track demands the use of a transport mechanism with an appreciable fraction of the speed of light anyway. So supply depots, construction fascilities, and placement tugs, service centers, must all be 'strung out' to something like half-way into the Oort Cloud. Even assuming Von Neuman machines were used for all phases of construction and service negating the need for human crews to be everywhere, we're still talking an absolutely enormous undertaking. I suspect that it would actually be easier to create a scheme first proposed by Dr. Robert L. Forward illustrated in his science fiction novel "Flight of the Dragonfly."

Dr. Forward came up with a scheme for harvesting solar energy by gigantic collectors in a sun synchronous orbit over Mercury and then converting the intense sunlight into thousands of laser beams. The lasers were beamed to a focus behind Mercury where a station was suspended by the pressure of the light above the dark side of Mercury. This station combined the beams into a single beam by using them as a pump for a really big dye laser and then directed that beam at a photon sail spacecraft that was larger than the state of Texas. It's a good book, I highly recommend reading it. And the science behind the book was about as sound as a doctor in physics can imagine...

Such a scheme is capable of fairly quickly accelerating spacecraft to 40% or so of the speed of light. No additional infrastructure needed...the logistics for such a mission, though huge, is still fairly manageable because the main resource base is where the resources are (metals, energy, etc.) and not out in deep space requiring enormous transportation efforts to get there.

-- Edited by GoogleNaut at 20:47, 2005-11-18

Isaac Kuo · Member

I'm assuming a relatively lightweight flyby probe--quite a few orders of magnitude less massive than a manned mission, and you don't have to worry about stopping at the destination or somehow getting back.

An unmanned probe can handle sustained acceleration of 10gees, which can get up to .25c in an acceleration run of about 1AU. That's quite long, but it doesn't require relativistic speeds to put in place. Note that you only need to launch the bomb-bots out to 1/2 AU--half in one direction, half in the opposite direction. Orbital mechanics can help you out also.

10kBq Jaro · Senior Member

In the more conventional fuelled runway concept, the nuclear fuel pellets are fired from some sort of a mass-driver-type gizmo, from the vicinity of earth. I would expect Isaac Kuo's concept to include a similar type of arrangement - not production plants stationed somewhere out in the Oort Cloud.

Presumably, at the far end of the runway, the bomb-bots would spaced much farther apart than at the near end, due to the much higher ship speed. They could all be launched at the same speed, only at varying delay intervals. But to get them out to 1700 AU in a reasonable amount of time, that launch speed would have to be pretty high -- which means that the ship would somehow have to be brought up to pretty high speed initially too, or else it will never catch up to the fast bomb-bots, in order to accelerate up to c/4.....

As for detonations gating not being of much concern, I would tend to disagree. Particularly since the scheme likely involves several thousand detonations, any one of which could potentially kill the ship's crew, if slightly mis-timed. Its interesting to compare this to the concept of in-tube scram-accelerator propelled LEO shuttles. This too involves high-speed gating, in this case to isolate sections of launch tube filled with mixtures of fuel-oxidiser gases (i.e. another kind of fuelled runway). The gate valve at the end of the launch tunnel must be opened a split second before the passage of the shuttle craft - traveling at hypersonic speed at that point. Needless to say, not too many people would be willing to take the risk - however remote - of slamming into a mulfunctioning gate ! .....how many do you think would be willing to take a similar risk - on a single trip - thousands of times over ?

<><><><><><><><><><><><><><><><><><><><><><><><>

Oooops ! ....missed your last post about lightweight flyby probes.

Yes, that makes more sense !

-- Edited by 10kBq Jaro at 00:11, 2005-11-19

Isaac Kuo · Member

Oh, for a ballpark idea of the cost of this interstellar mission, Dana Andrews wrote about the costs of particle beam magsail propulsion for an interstellar mission in his 2003 paper, "Interstellar Transportation using Today's Physics". Assuming a small 2.6 ton probe accelerated up to .3c, he determines that 3.5E19 joules of beam energy is required. Assuming 35% fission power efficiency, this adds up to 1E20 joules of fission energy--1000 tons of enriched uranium. He finds the cost of that uranium in 2003 dollar is around 3 billion dollars.

Of course, 3 billion dollars is a lot of money, but pretty cheap compared to a lot of things (like, say, a war).

Based on the relatively affordable cost of the fuel, Andrews notes that the real challenge is minimizing the costs of the particle beam emitters (including their power reactors). My proposal minimizes the cost of the particle beam emitters, although of course they can only be used once.

The big costs come not from the fuel, but rather from the manufacture of all the bomb-bot spacecraft and booster rockets (which, as noted, can launch a cluster of bomb-bots). But economies of scale will have a hugely beneficial effect.

GoogleNaut · Guru

Yeah, I was going to say that fuel (propellant) is typically only 5% or less of the launch cost of a typical rocket. So, even if fuel costs were 3 billion a truer picture of the cost would be 20 times as much, and probably more than that (unfortunately.)

O.K., a smaller automated probe makes sense. 2.6 tons put it in the Cassini class, possibly a Cassini on steroids as it were. The problem there, is that if power generation onboard the probe were too small, then communication with Earth would be a problem. The "Big Ears" of the deep space network have been upgraded many times so that they can continue to listen to Pioneer 10 and 11 and Voyager 1 and 2's signals which are getting weaker and weaker...

To "hear" a probe over light-years will almost certainly require larger, more sensitive radio telescopes to monitor probe transmissions--possibly space based or lunar based to cut out domestic radio interference, or if using a LASER transmitter--optical interference from the atmosphere. Still, I would imagine that probe transmit power would have to be bumped up from the current 10 W or so for the Pioneers, and 120-150W for Cassini to something like 100 to 200 KW for an interestellar probe. This almost requires a probe on the size scale of the British Interplanetary Society's Daedalus spacecraft.

I don't know--I think it is possible to launch small or even really tiny microprobes to the stars with near term technologies, but will it be possible for us to hear from them...? If they're too small, then they won't generate enough onboard power to shout their data stream back across the cosmos. And to travel all that way to fail---that's a trajedy!

-- Edited by GoogleNaut at 08:45, 2005-11-19

Isaac Kuo · Member

One idea for tiny "microprobes" is an interstellar precursor mission. The "microprobes" are simply a swarm of tiny spherical solar sail "bubbles", launched and released together to travel in a loose group into interstellar space at maybe 30-50km/s. Sensitive telescopes on/orbiting Earth stare at their path, looking for the signs of any collision explosions.

The goal is simply to get an idea of how much macroscopic debris is out there. Such debris could obviously be a serious threat to an interstellar mission traveling through 4 light years of interstellar space. The volume of space covered by millions of tiny "bubbles" at 30km/s can be far greater than that covered by the Pioneers and Voyagers.

Lord Flasheart · Member

Speculation:

It'd be interesting to have all those tiny sails linked together in the first stage of the trip. You could use them in conjunction with a laser for a little oomph, and afterward release them to check for interstellar debris.

Just a thought...

Isaac Kuo · Member

I've put together a graphical web page describing the new particle puff device mechanism:

http://members.cox.net/mechdan/ppd/index.html

This mechanism relies on more of a hybrid shaped charge/light gas gun effect than simple thermal mixing. As such, puff velocities of .5c or higher are acheivable with reasonable efficiency.

GoogleNaut · Guru

Hmmm. Yes, you seem to have hit upon a nuclear version of a Monroe shaped charge. I don't know what kind of columnation you could get, and I'm not quite sure what kind of velocity you could get with it either. 0.5c sounds a tad bit high. The hydrodynamic focussing is going to really be a function of the implosion geometry--true--but the speed of implosion isn't much higher than, oh, a couple of hundred kilometers per second. I don't see how, even in the incredible violence of a conical implosion, that a jet of material could be squirted out at a speed something like almost a thousand times the implosion speed. Hydrodyanamically speaking, your talking about increasing the kinetic energy of the projectile mass by one million times or so, over an above what it could get from the x-ray thermal bath from the fissioning primary. Even the incredible violence of a full on nuclear detonation could not achieve such a high concentration of kinetic energy in such a small volume. Atleast, by no mechanisms I can think of (which isn't to say that there could be mechanisms more energetic, but I'm not aware of any...)

10kBq Jaro · Senior Member

I see another problem -- in the llustrations:

You say that "X-rays ablate away the outer surface of the fusion stage tamper, imploding the lithium-deuteride fusion stage."

Fine, except that you've got a hollow cone shape. This will get collapsed long before the fusion reactions start. You can't do this ablation-implosion thing and expect to create an effect like a chemical shaped charge. The two are simply not the same. IMO, at best, the result will be something like a big chemical shaped charge (because you've got a little fission bomb driving it), possibly with a nice symmetrical fusion blast shortly afterwards.

Isaac Kuo · Member

Don't confuse the weak X-ray pumped implosion of the outer tamper with the violent implosion of the inner liner. The latter is driven by what's essentially a fission fragment rocket with an exhaust velocity of around 3.7%c, plus inward pressure from fast neutrons and protons moving at 17%c. An inward velocity of 5%c is an underestimate, if anything.

The Monroe effect isn't being used. The angle of the cone is far too steep, and the liner ends up just squishing itself into a slow moving rod-like lump. Instead, it's more of a light gas gun effect, where heavy U238 nuclei moving at 5%c compression-heat protons to 50%c.

As for the "problem" of the fusion stage being hollow--as I understand it, real life nuclear devices already feature hollow cores because only the outer layers of the fusion stage get a chance to be compressed anyway. The real problem is that the fusion stage gets started too early, not too late.

10kBq Jaro · Senior Member

From what I've read, the hollow cyllinder contains a plutonium pipe that gets imploded along with the rest of the cyllinder, and helps light the LiD fusion from within.... With a solid cyllinder (following the implosion), you get a nice strong shockwave rebound from the center. With the hollow cone, you get no such thing (or rather the rebound progresses from the near end to the far end, along with the implosion).

GoogleNaut · Guru

In Project Orion, the pulse units used the thermal x-rays within the radiation "bottle" to ablate a fair amount of a flat plate of high-z material, such as tunsgten or some other dense material. The abltation causes a strong shockwave to form--the "Rocket Effect"--which causes nearly instant acceleration of the remainder of the plate. A cool, dense plasma strikes the pusher plate with fairly uniform pressure. Good collimation is achieved by using dense materials--tungsten is good although osmium would be better for this (but far, far more expensive.) The collimation is maintained for the hundred or so meters from the pulse unit to the pusher plate.

Collimation, or maintenance of a jet over even a few kilometers may still be a challenge.

As for the nuclear shaped charge effect, I'd be more than a little concerned that columnation would be maintained to the degree that the target vessel, instead of pushed, would simply be holed and destroyed...

Isaac Kuo · Member

As I understand it, the inner "spark plug" used to be a solid rod, but it was found that only the outer layers played any role in the mechanism so it was replaced with a hollow tube. Note that the goal here isn't to maximize yield, but to propel the hydrogen pellet. The fusion stage can almost be considered mainly just a cheap neutron generator for the U238 fission "rocket". As long enough neutrons are generated to efficiently fission the final stage, that's enough.

The collimation of the puff would unfortunately be pretty poor. The hydrogen "projectile" leaves the "muzzle" in a terrificly compressed state. It's going to want to expand very very badly! I'd be surprised if the puff could be made tighter than, say, 60 degrees wide.

GoogleNaut · Guru

This is one reason why I think that it makes more sense to carry the pulse units onboard the spacecraft as 'propellant.' True, the vehicle is limited by the rocket equation, however, the efficiency by which momentum is actually transfered to the vehicle through the reactive forces generated be expelling propellant will more than make up for losses of a external pulse 'track' system.

I think one of the more promissing technologies is a kind of melding of Daedalus and traditional Orion concepts: the MiniMag Orion vehicle uses a z-pinch method to implode spheres of alternative nuclear materials: several isotopes of curium, cerium, and americium are possible fissile and fissionable materials that could be used with the MiniMag. This concept was put forth by Andrews Space at:
http://www.andrews-space.com/content-main.php?subsection=MTA2

The AIAA has an excellent white paper on the concept published in July 2003, entitled:

"Mini-MagOrion: A Pulsed Nuclear Rocket for Crewed Solar System Exploration."

Such a vehicle should have roughly the specific impulse of advanced VASIMR while delivering as much or more thrust that a SSME. The individual pulses yield about the equivalent of 3-5 tons of TNT. Not much, but with a repetition rate of 1 per second, the jet power developed is hundreds of GigaWatts. I think that MiniMag Orion, developed to the next logical step would include fission/fusion devices, and then we can seriously contemplate an interstellar mission.

Isaac Kuo · Member

Mini-mag and fission-fusion Orion style rocket propulsion might be good enough for fast interplanetary travel in the outer solar system, but they just can't acheive reasonable interstellar travel velocities without assuming nigh perfect efficiency with virtually no overhead (i.e. no mass of fuel tanks despite 1000:1 mass ratios, very high yields, little/no wasted products, etc). Alternatively, it requires new technology for efficient fusion yields, which current technology H-bombs can't acheive.

Personally, I'm not a big fan of nuclear pulse propulsion for interplanetary travel. Nuclear/solar electric propulsion is already demonstrated technology, and the potential advantages of nuclear pulse propulsion don't seem that great.

In the medium term, I see laser thermal propulsion to be far more promising than nuclear pulse units within the Solar System. Unlike nuclear rockets, laser powered rockets can be used SSTO as well as single stage to escape velocity.

For interstellar travel in the near term, though, there just isn't any other alternative to a bombtrack. It's either the nuclear bomb way or no way at all.