The World Ends in approximately 24-28 months.

[jeffb]

Back at the main event, I'd have to go with Orion's. As in Larry Niven and Jerry Pournelle's Footfall.

Sadly, nope. The basic rule of this scenario is that you have 2 years before DOOOOM. Massively insufficient time to develop Orion. Remember, you'd be basically developing Orion from scratch. There is virtually no tribal knowledge of the technology left. ten years, I could see it. Six... maybe. Two? better to devote your efforts to building rockets, and using the nukes to rain destruction upon your enemies.

One of the big selling points of the Orion concept was its simplicity. Drive bombs would be delivered to the gun that launches them beyond the pusher plate by rail systems similar to those used all over manufacturing currently. The US certainly has the simulation capability to design an appropriately sized drive bomb very quickly. The rest is all heavy engineering which is carried out daily in ship yards all over the world. Building them out of steel wouldn't be ideal of course, but the delta-v performance means a reduced period of coast between here and mars anyway and the massive cargo capacity means larger protected zones within the ships.

If you go with rockets though, your problems soon multiply. Sea dragon was designed to put up to 550 tonnes into LEO. But that payload has to be a ship capable of crossing to and landing on mars. So you gained a whole interplanetary space craft project that you've got to stand up in 2 years.

Just my 2 cents.

 

[Orionblamblam]

One of the big selling points of the Orion concept was its simplicity. Drive bombs...

Ah, yes. The incredibly simple drive bombs. Bombs capable of fairly precise sub-kiloton yields. Do we have those? Nope. Do we have any *designers* of sub-kiloton nukes? I'd be surprised. How long would it take to generate that capability, since we lost it *generations* ago?

The US certainly has the simulation capability to design an appropriately sized drive bomb very quickly.

Sez who? And how reliable would these simulations be? How do you know? The last time the US set off a nuclear detonation was in 1992. Do we rely on nuke simulations from the early 90's and before, or *modern* simulations that have no testing under their belts?

The rest is all heavy engineering which is carried out daily in ship yards all over the world.

What shipyard has years of experience dealing with woven pneumatic "gas bag" shock absorbers that can handle thousand-ton steel lates oscillating at plus and minus 50,000 g's?

 

[jeffb]

One of the big selling points of the Orion concept was its simplicity. Drive bombs...

The US certainly has the simulation capability to design an appropriately sized drive bomb very quickly.

Ah, yes. The incredibly simple drive bombs. Bombs capable of fairly precise sub-kiloton yields. Do we have those? Nope. Do we have any *designers* of sub-kiloton nukes? I'd be surprised. How long would it take to generate that capability, since we lost it *generations* ago?

Sez who? And how reliable would these simulations be? How do you know? The last time the US set off a nuclear detonation was in 1992. Do we rely on nuke simulations from the early 90's and before, or *modern* simulations that have no testing under their belts?

Well the test ban goes straight out the window obviously, you create a new design based on simulation and test it. Straight fission designs are supposed to be quite straight forward and there are a large number of pre existing, calibrated designs so, given the time constraints, yep.
http://www.purdue.edu/newsroom/research/2012/120605BagchiWeapons.html

But you don't really need a new design, a number of the existing artillery nuclear rounds and warheads are in the right ball park. The W-72 (0.6 kT), the W-84 (variable 0.2kT - 150kT) and the W-79 (Variable 100T - 1.1kT) all seem like reasonable candidates. Doubtless the blueprints are lying around somewhere.
http://nuclearweaponarchive.org/Usa/Weapons/Allbombs.html

The rest is all heavy engineering which is carried out daily in ship yards all over the world.

What shipyard has years of experience dealing with woven pneumatic "gas bag" shock absorbers that can handle thousand-ton steel plates oscillating at plus and minus 50,000 g's?

None of course, but they do have loads of experience building the large steel structures required by the rest of the ship. And the pneumatic shock absorber is hardly a show stopper, Dyson and crew thought this was achievable back in '58! And who says it has to be a pneumatic shock absorber?

Interesting read:
http://www.projectrho.com/public_html/rocket/realdesigns2.php#id--Project_Orion

 

[Orionblamblam]

But you don't really need a new design, a number of the existing artillery nuclear rounds and warheads are in the right ball park. The W-72 (0.6 kT), the W-84 (variable 0.2kT - 150kT) and the W-79 (Variable 100T - 1.1kT) all seem like reasonable candidates.

Sadly, no. Nukes in the sub-kiloton range are pretty vague in their yield. This is because low-yield nukes are by design *bad* nukes. This results in their yield having some pretty wide error bars, and Orion has to have pretty tight control over the yield. If you want to use a SADM to blow a dam to smithereens, the difference between 0.1 kilotons and 0.12 kilotons is nothing, but if you have a finely calibrated mechanical spring system, whacking it with variable yields is a fast way to break the thing.

Doubtless the blueprints are lying around somewhere.

Uh-huh. Even if there were... look up "Fogbank."

Also: we don't have too many nukes these days. We don't have a whole lot of plutonium processing capability these days. We could yank the pits out of other nukes, but processing them into the *thousands* of new bombs needed for even one launch would require not just work but new facilities.

None of course, but they do have loads of experience building the large steel structures required by the rest of the ship. And the pneumatic shock absorber is hardly a show stopper, Dyson and crew thought this was achievable back in '58!

Yes, indeed, and it would only take a few years of hard work to work out the issues. All that work is effectively lost now, so we'd be starting from scratch. They were also working during the Golden Age of atomics, when there were a whole lot of people working in a whole lot of atomic fields... fields that barely exist anymore. Who is the modern day Ted Taylor with years of successful atomic bomb design work under his belt?

And who says it has to be a pneumatic shock absorber?

It could be electromagnetic. And would that be faster to develop and less prone to fail than pneumatic?

Interesting read:

Try this:
http://www.aerospaceprojectsreview.com/ev1n4.htm
and
http://www.aerospaceprojectsreview.com/ev1n5.htm
and
http://www.aerospaceprojectsreview.com/ev2n2.htm

Orion is awesome. If we *had* Orion, it would be a slam-dunk for the Colonial Marines for emergency transport. But it would take more than two years to develop, while rockets exist *now.*