How Long to Make an Orion Rocket?

[Rhadamantus]

I was reading Seveneves, and I was wondering, in a situation where resources are no object, how long would it take to make an Orion rocket (for example, the Battleship Micheal)?

[Simon_Jester]

Decades, honestly. There are a LOT of unsolved engineering problems in the design, shock absorbers being one of the most obvious ones. Sadly, it's not something you could improvise, especially since no detail design work has ever been done on one.

[Jub]

To put it simply, something like the F-22, orders of magnitude less complex than an Orion rocket, took thirty years from design competition to entry into US service. The F-35 program has been running for around twenty years now and still isn't done. Sure, these projects likely could have been rushed if design goals remained static, but even then that's saving a few years at best. Designing cutting edge stuff is expensive, time-consuming, and exceedingly difficult. This is even truer for something like Orion which has never even been fully planned out, let alone attempted.

[Simon_Jester]

I'm not actually sure an F-22 is that much easier to design properly than an Orion drive ship; the bleeding edge avionics and aerodynamics of the F-22 present a lot of challenges.

On the other hand, an Orion drive ship will NEVER work without considerable amounts of prototyping; we don't really know enough about what happens when you nuke big steel plates to be able to simulate them very accurately.

[Jub]

I'm not actually sure an F-22 is that much easier to design properly than an Orion drive ship; the bleeding edge avionics and aerodynamics of the F-22 present a lot of challenges.

On the other hand, an Orion drive ship will NEVER work without considerable amounts of prototyping; we don't really know enough about what happens when you nuke big steel plates to be able to simulate them very accurately.

True, but at the same time we've built jets and stealth capable aircraft before. The F-22 is a massive leap forward over existing designs, but we've at least built more primitive versions of them.

With an Orion driven craft, it's completely uncharted ground that could be fraught with challenges we can't currently anticipate. That said, it might also end up being easier than we may currently imagine. It's completely beyond me to even speculate.

[Simon_Jester]

Well, it might be doable with, say, 'only' ten times the engineering man-hours that went into the F-22. I don't know. Still going to be immensely difficult and time-consuming.

[Adam Reynolds]

Also, if the objective were simply a quick response to an odd phenomenon, building something using more conventional propulsion with rockets or an ion engine would make more sense. Scaling those up in a manner like Hermes from The Martian would be a better idea in the short term.

What is the need for an Orion engine specifically?

[Eternal_Freedom]

Presumably, the need is for something that only an Orion rocket could do, like provide frankly absurd amounts of delta v as well as high thrust. Plenty of hypothetical designs, and even some real ones, can provide enough delta v for interplanetary flights, Orion does that but with thrust that gets you there in useful timescales (as in, an Orion rocket could maintain 1g accelerations for the trip, something most designs can't do. In fact, the design studies done that I can recall anticipate this).

I'm curious though; everything I've read on Orion said that it requires no fundamentally new technology and could have been built in the '50's. IIRC they even built a small chemical-explosive-driven testbed that flew well enough.

[Rhadamantus]

In the book seveneves, the moon exploded (via ASB), and in two years all life on Earth will die. In the book, Humanity needs to get as much stuff and people off Earth as quickly as possible. Orion would be a good solution.

[Iroscato]

In the book seveneves, the moon exploded (via ASB), and in two years all life on Earth will die. In the book, Humanity needs to get as much stuff and people off Earth as quickly as possible. Orion would be a good solution.

Would it? As mentioned, it would take decades to get a working prototype, let alone rolling it out for mass-production. By that time, humanity is little more than a carbon stain on the Earth.

[Sea Skimmer]

With only two years you'd have to build the ship with no large scale testing of anything. Which is not good when you need giant shock absorbers, a giant nuke proof plate, a mechanism to shoot nukes through that plate proof against hundreds of nukes, and some kind of enormous control system for stability. Even then your going to have trouble launching a useful payload in the 'life is doomed' context, so why bother.

Honestly 2 years would just mean life probably is doomed. We could convert ICBMs to get small chunks of extra payload into orbit quickly, but only a few manned launches would be possible with something close to existing craft. Trying to build spacecraft would probably be a waste of time compared to trying to build some underground shelters that might actually withstand the moon debris and have some kind of way to dig back out. If you had underground water and geothermal power + gardens and the impact's didn't disrupt you might have some kind of a long term chance at this.

[Simon_Jester]

I'm curious though; everything I've read on Orion said that it requires no fundamentally new technology and could have been built in the '50's. IIRC they even built a small chemical-explosive-driven testbed that flew well enough.

There's a difference between a one ton thing that flies when you blow up hunks of dynamite, and a hundred thousand ton thing that flies on exploding nuclear bombs. A huge difference. The chemical testbed can act as proof of some parts of the concept, but it doesn't solve all the engineering issues.

And "no new technology" is very, very iffy. There isn't any category of thing we haven't invented in an Orion drive. But there are many specific things which would have to be built, that never existed before. Like gigantic shock absorbers, and steel plates designed to get nuked five hundred times without being seriously damaged, and nuclear bomb launchers that will still operate reliably after being nuked five hundred times, and so on.

If we tried to build one without decades to perfect the design, it would almost certainly crash, fall apart, or otherwise fail badly.

In the book seveneves, the moon exploded (via ASB), and in two years all life on Earth will die. In the book, Humanity needs to get as much stuff and people off Earth as quickly as possible. Orion would be a good solution.

We're screwed. Two years is not enough time to build new spacecraft of any type that doesn't already exist. An Orion-drive ship built in less than two years would simply not fly, period. We probably couldn't even write the simulation code to model how the ship would behave in two years.

[Starglider]

It's not so much a question of 'resources are no object', it's also 'what level of risk, to human life and the environment, is acceptable?'. If you were going with 1960s Soviet standards, where test vehicles blowing up, killing personnel and seriously polluting bits of the countryside are all considered minor setbacks, you can take more risks and progress faster.

[Simon_Jester]

Yes, but two years is still not enough time. Because physically constructing a vehicle that has even a theoretical chance of flying will probably take that much time- and your first attempt will predictably fail and crash and burn.

There won't be time to build a second one.

[Starglider]

Yes, but two years is still not enough time. Because physically constructing a vehicle that has even a theoretical chance of flying will probably take that much time- and your first attempt will predictably fail and crash and burn.

There won't be time to build a second one.

Yes that was a counterargument to Jub saying it would take 'longer than the 30 year F-22 program', I was not saying that the ~2 year timeline from 'Footfall' was reasonable.

[Jub]

Yes, but two years is still not enough time. Because physically constructing a vehicle that has even a theoretical chance of flying will probably take that much time- and your first attempt will predictably fail and crash and burn.

There won't be time to build a second one.

Yes that was a counterargument to Jub saying it would take 'longer than the 30 year F-22 program', I was not saying that the ~2 year timeline from 'Footfall' was reasonable.

Yeah, if we're willing to take any risk to get things done that would cut the time down significantly. However, the OP never made it clear that this was an at all costs project. I was picturing something more like the Apollo program where the team is given a tight deadline and large amounts of funding, but where the project still needs to be conducted in a sane fashion.

[Flagg]

Nukes in space don't work like in an atmosphere. Are there any real studies that show an Orion type ship would even be more efficient than ion drives combined with solar sails? I mean you have to get all of those tens of thousands of nuclear warheads into space, there just plain WILL be rockets exploding at launch (as in fuel, not nukes) so you have to construct more launch pads which takes awhile.

Even ignoring the stupid moon exploding 2 year deadline, I don't think an Earth constructed (barring a game changing development like a space elevator or antigravity tech) Orion ship is cost or resource effective.

[Imperial528]

And "no new technology" is very, very iffy. There isn't any category of thing we haven't invented in an Orion drive. But there are many specific things which would have to be built, that never existed before. Like gigantic shock absorbers, and steel plates designed to get nuked five hundred times without being seriously damaged, and nuclear bomb launchers that will still operate reliably after being nuked five hundred times, and so on.

Now I'm not sure about the shock absorbers, but the latter two were solved back in the 50s-60s. Coating the plate with a layer of oil will prevent ablation of the plate (discovered by accident when they noticed that on test plates fingerprints had become embossed on the plates; the steel around them had been ablated via vaporization, but where oils had rubbed off from people's fingers the steel was undamaged) and from my memory, the bomb launcher (and oil sprayer unless you decide on the ablative plate) is positioned either in a blind spot where the pulse won't reach, or behind a deflection shield, depending on the pusher plate's design.

As for making an Orion today, personally I think the biggest factor would be getting the plutonium to make enough charges. While the sufficient material surely exists, it's not exactly being mass produced and much of it is sequestered in nuclear weaponry.

Engineering wise I think the biggest challenge would be making sure the plumbing stays where it should be during launch. And keeping Greenpeace away from the facility long enough to actually build it. /s

The thing about Orion though is that you can afford to over-engineer by a margin far greater than chemical rockets allow. You have the thrust and delta-v potential to just put in ten tons of structural reinforcement when the math says you only need two or three tons worth. It will cut into your payload and d-v, but you have so much of it in comparison to conventional rocketry that it's a small price to pay for a little extra safety.

Oh, and here's the chemical Orion testbed: https://media.giphy.com/media/dheTP1wdTDAT6/giphy.gif

Nukes in space don't work like in an atmosphere. Are there any real studies that show an Orion type ship would even be more efficient than ion drives combined with solar sails? I mean you have to get all of those tens of thousands of nuclear warheads into space, there just plain WILL be rockets exploding at launch (as in fuel, not nukes) so you have to construct more launch pads which takes awhile.

Orion may be affectionately referred to as a scaled-up firecracker under a tin can, but it's considerably more sophisticated than that. Orion's propellant charges are designed to focus the majority of the energy release from the nuclear charge into a propellant mass, usually powdered tungsten, which is propelled forward in a narrow cone. The propellant mass striking the pusher plate is what delivers thrust, rather than any direct product of the nuclear blast. The effect is similar to a HEAT warhead in basic function (though due to the propellant being powdered instead of solid the end result is much different), and if scaled up can be used to make the only practical plasma weapon -the casaba howitzer- ever conceived of.

Another note, you don't actually have to get the nukes into space as a dedicated project ahead of time. A full-blown Orion, unlike the NASA mini-mag Orion, can be launched from the surface, though the launching charges tend to have a different yield and some design tweaks compared to the propellant charges.

Orion's biggest strength ironically is high-thrust rather than exhaust velocity (as far as near-torchships go it's fairly poor in that regard), making it an excellent heavy lift vehicle. Just so long as you don't mind detonating kiloton-yield charges to get your payloads into orbit.
That said, it is most definitely better than solar sail or ion drives for any application that is time sensitive, requires a high payload, or both. Orion specific impulse can vary between 3,355s to 12,237s depending on the design you look at, and delta-v is generally measured in the double-digit km/s, but you can get triple-digit km/s d-v if you sacrifice payload for more propellant. And this is while possessing very good mass ratios; for a 10km/s d-v fuel load, you have a spacecraft with a ratio of about 11.5 mass units of payload per unit of propellant.

In comparison, ion drives can manage much higher specific impulse (Isp) and usually delta-v as a result, but have abysmal thrust, with massive power requirements. Whereas with Orion, all the power you need comes with the propellant charges themselves.

Here's a design of a propellant charge, though I believe it's for one of the NASA designs, not the USAF one:

[Flagg]

Ahh, ok. That makes a lot more sense than what I was under the impression the charges were (just warhead blasts which in space just create heat and other forms of radiation). I was under the impression Ion drives were slow to speed, but relatively good for long distance while requiring low power. But that's based on the one NASA used. I don't have anything against Orion drives per se, just not a fan of detonating nukes in the atmosphere. More like "get craft past moon, then go nukey nukey".

Thanks a bunch for that info.

[Imperial528]

Well, low power ion drives do work, and work well. They just work best on satellites that can afford multiple years of staggered engine burns (burn for part of the orbit, go around, burn again when lined up, repeat) or on very low mass craft.

Great for orbital science applications. But terrible if you're trying to get somewhere outside of Earth orbit before your crew starts getting old.

[Simon_Jester]

Now I'm not sure about the shock absorbers, but the latter two were solved back in the 50s-60s. Coating the plate with a layer of oil will prevent ablation of the plate (discovered by accident when they noticed that on test plates fingerprints had become embossed on the plates; the steel around them had been ablated via vaporization, but where oils had rubbed off from people's fingers the steel was undamaged)...

That's "solved" in the sense that we know how to protect a steel plate from a nuclear blast once. Thing is, the oil will prevent ablation of the plate once, but will it prevent ablation over and over? Sure, you can in theory rig a sprayer, but the sprayer has to withstand hundreds of nukings in a short amount of time.

I'm not saying it's impossible, but it's only "solved" in the vaguest of senses; no one ever did detail design, nobody ever actually tested it (or anything) under realistic conditions of getting nuked every few seconds for minutes or hours at a time.

...and from my memory, the bomb launcher (and oil sprayer unless you decide on the ablative plate) is positioned either in a blind spot where the pulse won't reach, or behind a deflection shield, depending on the pusher plate's design.

In addition to this, it also needs to be highly, massively, serially shock-resistant, and also resistant to any reflected flash from the repeated nuclear events.

Engineering wise I think the biggest challenge would be making sure the plumbing stays where it should be during launch. And keeping Greenpeace away from the facility long enough to actually build it. /s

This is actually a huge problem, and one that overbuilding the structure won't help with.

Oh, and here's the chemical Orion testbed: https://media.giphy.com/media/dheTP1wdTDAT6/giphy.gif

It's lovely, but it is far, far smaller than a real one and won't experience even remotely similar stresses and material forces.

Again, there is a massive gap between the level of 'engineering solved problem' where you've got at least an outline of how your vehicle will work, and actually making the vehicle work as planned. This is why real life airplanes and ships take years or decades to design.

[Starglider]

Space-based Orion propulsion can use a big sail (on a long shock-absorbing tether) for considerably better ISP than the pusher plate design. The latter is more for ultra-heavy-lift and planetary orbit maneuvering.

[Simon_Jester]

So... you use a big bag to capture the nuclear explosion, instead of a pusher plate?

Wow. Never thought of that.

[Sea Skimmer]

That's "solved" in the sense that we know how to protect a steel plate from a nuclear blast once. Thing is, the oil will prevent ablation of the plate once, but will it prevent ablation over and over? Sure, you can in theory rig a sprayer, but the sprayer has to withstand hundreds of nukings in a short amount of time.

I'm not saying it's impossible, but it's only "solved" in the vaguest of senses; no one ever did detail design, nobody ever actually tested it (or anything) under realistic conditions of getting nuked every few seconds for minutes or hours at a time.

For that still classified Orion drive battleship project they might have done some non trivial detail design of that sort of thing, but it'd be with 1960s technology and manufacturing, not of the greatest use on short notice.

For testing the main concern would be neutron KE shock, and the embrittlement from said neutrons. The idea behind the oil spray is the oil has lots of hydrogen in it, so that will completely convert some neutrons into gamma rays before they hit the actual metal. It's also a thermal insulation against the worst of the direct thermal effects and gamma rays from the nuke itself. The oil vaporizing then creates thrust, but also lots of vibration. That's where the shock absorbers become a nightmare.

I don't think its in question we could build the blast shield without testing though if it was just allowed to be very heavy, we aren't really talking about structural requirements higher then face hardened battleship armor crossed with a forged nuclear pressure vessel lid. If you can forge it all into one piece it's going to hold up. But it's like trying to forge a whole battleship turret at the same time.

If you used nuclear shaped charges though you need a super reliable way to ensure the nuke always detonates in the correct direction! It can't just tumble out, so functionally I think the only likely solution is a nuclear autocannon. Very workable in theory, in practice it would take 10 years if you were allowed to live fire nuke test for developmental work.

...and from my memory, the bomb launcher (and oil sprayer unless you decide on the ablative plate) is positioned either in a blind spot where the pulse won't reach, or behind a deflection shield, depending on the pusher plate's design.

In addition to this, it also needs to be highly, massively, serially shock-resistant, and also resistant to any reflected flash from the repeated nuclear events.[/quote]

Yes and this is where doing things like just forging a big cup become a real pain in the butt, you need to put in some kind of hole blank to machine out in the work for the oil spray system and the nuke chute. That's a ripe chance for cracking, and any crack screws you. For an emergency small cracks could be welded, but a big one is just fatal fail.

[Imperial528]

So... you use a big bag to capture the nuclear explosion, instead of a pusher plate?

Wow. Never thought of that.

Yeah, it's known as Medusa. Actually much more efficient than Orion, it's kind of the nuclear method applied to solar/laser sails.

Though less useful to warships due to the sail being a very good target, while an Orion pusher plate makes excellent armor (and radiation shield to boot).

Also, don't get me wrong, I don't think we could build one *NOW*, I just think that it'd be possible to have a working prototype in a decade and production model within the next, presuming the requisite heavy industry is up to task.

With the seveneves scenario of two years, well, honestly I think you'd take two years just to get the logistics in order. I mean the classic Orion design is basically a spaceship with a tonnage similar to the displacement of large cargo ships. Two years might get you a working mini-mag Orion, but that's a very tight timescale to design and test one, let alone build it in orbit.

I think bare minimum for any new heavy lift spacecraft of any design is five years, most of which would be spent on getting the industry in place for it.

If you used nuclear shaped charges though you need a super reliable way to ensure the nuke always detonates in the correct direction! It can't just tumble out, so functionally I think the only likely solution is a nuclear autocannon. Very workable in theory, in practice it would take 10 years if you were allowed to live fire nuke test for developmental work.

The details are probably still classified, but from what has been declassified they were designing just that. The charges even have drive bands to engage with a rifled or grooved barrel that would presumably launch them.