Nuclear Shotgun
Last night Blake and I were talking about staged explosive propulsion after reading about the Operation Plumbbob test launching a manhole cover past solar escape velocity (estimated at ~56 km/s vs. solar escape of ~43km/s). Anyway, we got to thinking and decided that this is now the most awesome (if maybe not the most practical) asteroid defence.
The idea is: Dig a shaft (or a cluster of shafts) somewhere. Case the shaft in reinforced concrete and steel pipe. Place a nuclear warhead at the bottom. The yield probably doesn't need to be that great - Plumbbob was only a 300 ton yield - so a 50 kT warhead would probably get you a pretty good punch. Put a several ton projectile in the shaft. It'll probably need to be something with a lot of tungsten and carbon shielding on the bottom. Once you're lined up on the asteroid's flight path, fire the nuke. You'll get a multi-ton projectile up to (hopefully) several hundred km/s or better and destroy the target with kinetic energy. You could probably scale this thing up a lot, use a multi-megaton warhead and get that projectile really moving. You've not done your job if the projectile doesn't mostly vaporize on impact.
So we were wondering why this idea never gets any play in NASA and other studies on plans for taking out an Earth-bound asteroid. There are two main drawbacks that I can see with this plan: The first is that you'd have to construct your shafts specifically for each target. Most proposed plans could work on any asteroid in any orbit. The second (and in my mind, the big killer) is the orbital calculations required to get a projectile on target. Presumably this rock is not going to hit us for a while, so it's pretty far out. At the distances we're talking, our projectile being off by 1/1000th of a degree on initial launch would cause it to miss. Since we don't have terminal maneuvering capability, that would pretty much suck for Earth. So Blake suggested just building like 100 of these things (or more) and firing all of them in hopes that one hits. This would still probably be cheaper than mounting a Deep Impact style mission. Launching a robot booster ship would probably still be cheaper. The other problem is that we might just shatter the rock, leaving a lot of fragments that might still be dangerous.
The idea is: Dig a shaft (or a cluster of shafts) somewhere. Case the shaft in reinforced concrete and steel pipe. Place a nuclear warhead at the bottom. The yield probably doesn't need to be that great - Plumbbob was only a 300 ton yield - so a 50 kT warhead would probably get you a pretty good punch. Put a several ton projectile in the shaft. It'll probably need to be something with a lot of tungsten and carbon shielding on the bottom. Once you're lined up on the asteroid's flight path, fire the nuke. You'll get a multi-ton projectile up to (hopefully) several hundred km/s or better and destroy the target with kinetic energy. You could probably scale this thing up a lot, use a multi-megaton warhead and get that projectile really moving. You've not done your job if the projectile doesn't mostly vaporize on impact.
So we were wondering why this idea never gets any play in NASA and other studies on plans for taking out an Earth-bound asteroid. There are two main drawbacks that I can see with this plan: The first is that you'd have to construct your shafts specifically for each target. Most proposed plans could work on any asteroid in any orbit. The second (and in my mind, the big killer) is the orbital calculations required to get a projectile on target. Presumably this rock is not going to hit us for a while, so it's pretty far out. At the distances we're talking, our projectile being off by 1/1000th of a degree on initial launch would cause it to miss. Since we don't have terminal maneuvering capability, that would pretty much suck for Earth. So Blake suggested just building like 100 of these things (or more) and firing all of them in hopes that one hits. This would still probably be cheaper than mounting a Deep Impact style mission. Launching a robot booster ship would probably still be cheaper. The other problem is that we might just shatter the rock, leaving a lot of fragments that might still be dangerous.