[Mars]

If nuclear reactors failed as often as black box's, then we'd have had several minuture Chernobyl s.

[FreeSpaceFreak]

No. The one thing that counts in radiation shielding, is mass. AFAIK. The one thing that counts in airplane design, is also mass. The Soviets did it, yes, but they had so little shielding that after a couple of years, most of the crews that flew it were dead.

The Americans converted a B-36 to carry a nuclear reactor, and it had all the shielding required. Of course, that meant that it needed a crapload of engines (It had a combination of jets and props, IIRC). (My bad, all B-36 had jets and props.)

In May 1946, the Air Force began the Nuclear Energy for the Propulsion of Aircraft (NEPA) project which was followed in May 1951 by the Aircraft Nuclear Propulsion (ANP) program. The ANP program required that Convair modify two B-36s under the MX-1589 project. One of the modified B-36s studied shielding requirements for an airborne reactor to determine whether a nuclear aircraft was feasible. The Nuclear Test Aircraft (NTA) was a B-36H-20-CF (Serial Number 51-5712) that had been damaged in a tornado at Carswell AFB on 1 September 1952. This plane, designated the NB-36H, was modified to carry a 1 MW, air-cooled nuclear reactor in the aft bomb bay, with a four ton lead shield between the reactor and the cockpit. The cockpit was encased in lead and rubber, with a 6-inch (15 cm)–thick acrylic glass windshield. The reactor was operational but did not power the plane; its sole purpose was to investigate the effect of radiation on aircraft systems. Between 1955 and 1957, the NB-36H completed 47 test flights and 215 hours of flight time, during 89 of which the reactor was critical.

I know that. But there were about 5 people in the whole of the huge aircraft, and the power still came from normal fuel. Commercial airliners with a nuclear reactor will not work until 2367 As for fusion reactors: the bigger they are, the better they work. The JET in Oxford, including all its heating-up equipment, is too big to fit inside a plane, but too small to produce more energy than it needs to stay functioning.

If we could make a black box survive almost any wreck, surely we could with a reactor, add to that an automatic shutdown routine when the plane gets to a certain distance from the ground.

The bigger something is, the more vulnerable it is. A fission reactor is a bit bigger than a black box. And an automatic shutdown routine doesn't remove the radioactivity. As for reflecting the radiation: when, on a crash, the reflector would break, all of the accumulated radiation would be released, no?

[Mika]

True, but then again the reactor was also designed and built with late 1950's technology, surely we have progressed at least a little in that regard.

I don't think that the development has progressed in those areas. They have improved the reliability and security of the reactors, but not sizes or weights.

Radiation can also be reflected, right? So maybe it's not a question of piling on more lead but rather using some kind of crazy nanoscale engineering on what we've already got to reflect it.

In practise there is no such thing as mirror surface for x-ray photons, much less for gamma photons. While the x-rays can be operated with diffractive elements, gamma photons are much more difficult, them affecting the nucleus themselves and pretty much ignoring all the electrons on the atomic shell. If one would add reflectors around nuclear reactor and by some miracle got it working, the reflectors would heat up quickly due to x-rays, be structurally weakened by gamma rays, and would also turn radioactive also by themselves not to mention mess up the chain reaction itself by launching stuff back there. This time, radiation would not come from the reactor directly, but from the shielding itself. This is the reason there is need for huge thicknesses to really absorb those nasty thingies and to keep them protecting stuff for extended period of time.

[Off topic]
Some interesting non-related links:
http://www.youtube.com/watch?v=O_TuqaK3B0U
http://www.ejectionsite.com/stapp.htm (holy cow, human testing of the early ejection seats!)
http://www.youtube.com/watch?v=unz8Tf3ZMbA (nuclear rifle - yes, nuclear rifle)
[/Off topic]

Mika

[Retsof]

Carbon nanotubes/buckyballs seem to be good at everything else, would they be any good at sheilding?  What about ceramics?  Ah screw it, just have someone invent energy sheilds already.

[Mika]

Carbon nanotubes/buckyballs seem to be good at everything else, would they be any good at sheilding?  What about ceramics?  Ah screw it, just have someone invent energy sheilds already.

Unfortunately no, nanotubes and buckyballs don't work, neither do ceramics. This is caused by the fact that the higher the energy of a photon, the more deeper it will travel in the target atom. At gamma ray's case, translations happen in the very core of the atom, not in the electron shell. This effectively means that the area the photon has to struck in an atom in order to become absorbed is a lot smaller. So, you have to improve the odds that the photons will hit an atom in a blocking material. That is done by adding more material around the shield, another factor improving the odds is the density of the atoms in the material.
Carbon nanotubes are strong, but not dense (hence the weight saving). Same applies to ceramics.

This is then of course, what I think would happen, not a general answer from the community called Scientists.

Mika

[Kosh]

Scientists

Where is Herra Tohtori?

[NGTM-1R]

The bigger something is, the more vulnerable it is. A fission reactor is a bit bigger than a black box. And an automatic shutdown routine doesn't remove the radioactivity. As for reflecting the radiation: when, on a crash, the reflector would break, all of the accumulated radiation would be released, no?

Nuclear reactors are inherently sturdy objects. More than a few nuclear-powered submarines have been lost without a dangerous reactor failure ("crush depth" does not exist on this planet for reactor containment vessels manufactured in the US, and presumably in most other places). Studies of the problem indicate that even in the event of combat damage nuclear-powered ships would not pose a danger 99.9% of the time. This is an impressive feat when you consider that they were testing against a ton of Torpex or three thousand pounds of high explosive from a surface-to-surface missile. It can be done.

[Mika]

Studies of the problem indicate that even in the event of combat damage nuclear-powered ships would not pose a danger 99.9% of the time.

This begs for further questions: what is meant by not "a danger" in this case? How does this apply to airplanes?

Mika

[FreeSpaceFreak]

A nuclear submarine is not dangerous, unless it flies over NYC

[NGTM-1R]

This begs for further questions: what is meant by not "a danger" in this case? How does this apply to airplanes?

Even in the case of a direct torpedo hit against the enigneering spaces, the reactor will automatically scram and the containment vessel will not be breached. The point is, it's possible to make them safe against extremes of impact and pressure.

[Kosh]

Regarding the nuclear powered bomber, there are a couple of things to note: the aircraft where the reactor was fitted is large even by today's standard. But, it took away a lot of payload of the aircraft, and required quite a bit of shielding around the crew compartment.

It's also worth mentioning that they refitted an airplane designed for a completely different kind of propulsion system instead of building a new one from scratch, which is why its payload was taken away. Something along the lines of a flying wing designed from the beginning to use this instead of traditional jet engines and fuel tanks might work better.

Here is something that came out recently that might be somewhat better suited for this task:
http://www.physorg.com/news145561984.html

[IceFire]

Regarding the nuclear powered bomber, there are a couple of things to note: the aircraft where the reactor was fitted is large even by today's standard. But, it took away a lot of payload of the aircraft, and required quite a bit of shielding around the crew compartment.

It's also worth mentioning that they refitted an airplane designed for a completely different kind of propulsion system instead of building a new one from scratch, which is why its payload was taken away. Something along the lines of a flying wing designed from the beginning to use this instead of traditional jet engines and fuel tanks might work better.

Here is something that came out recently that might be somewhat better suited for this task:
http://www.physorg.com/news145561984.html

I remember reading about that.  Sounds pretty good if it works as advertised.  Sounds like they have the problems anticipated at least.

[Mika]

Even in the case of a direct torpedo hit against the enigneering spaces, the reactor will automatically scram and the containment vessel will not be breached. The point is, it's possible to make them safe against extremes of impact and pressure.

Depends on the actual location. I recall reading they tested a ferro-concrete wall with a fighter plane to determine if combat aircraft can breach the outer wall of nuclear power plant. Turned out that it can't (penetrated only a couple of centimeters).  Testing with high explosives, depends on the actual test. Doing things with penetrating shaped charges could yield results with low amount of explosives. Results of those tests can just as easily not be generalized to aircrafts. The accelerations and stresses are different on aircraft, say a momentarily acceleration of 300 gs is pretty high for anything constructed by man.

It's also worth mentioning that they refitted an airplane designed for a completely different kind of propulsion system instead of building a new one from scratch, which is why its payload was taken away. Something along the lines of a flying wing designed from the beginning to use this instead of traditional jet engines and fuel tanks might work better.

I stand by my reasoning that using nuclear reactor there would only add the weight of the aircraft as the whole passenger cabin should be protected from radiation, leading to usage of heavy materials.

Mika

[FreeSpaceFreak]

Apart from that, the reactor in submarines is probably heavily shielded, something you can't afford in airplanes.

[Kosh]

I stand by my reasoning that using nuclear reactor there would only add the weight of the aircraft as the whole passenger cabin should be protected from radiation, leading to usage of heavy materials.

Ok, so how much extra weight is there?

[NGTM-1R]

I stand by my reasoning that using nuclear reactor there would only add the weight of the aircraft as the whole passenger cabin should be protected from radiation, leading to usage of heavy materials.

The radition protection would be built into the reactor containment vessel itself, not the passenger cabin. This is a much more rational and lighter-weight design. Similarly applying the same design principles originally developed for quieting measures on a submarine would increase its safety by not making it part of the structure or truly connected to it at all. The weight is a significant factor, the typical pressurized water reactor weighs a good twenty tons, but there are aircraft flying a lot heavier than that.

[Kosh]

I also did some checking on how many pounds of fuel several kinds of airplanes are carrying around for comparison (keep in mind 2000 pounds per ton, so a 20 ton reactor weighs 40,000 pounds):

Based on wiki answers, 1 gallon of kerosene weighs 6.82 pounds

b-52: 47,975 U.S. gal * 6.82 = 327,190 pounds

747-400: 57,285 U.S. gal * 6.82 = 390,684 pounds

737-700: 6,875 US gal * 6.82 = 46,887.5 pounds

So something like that is actually practical for something the size of a 737 or bigger.

[FreeSpaceFreak]

What's the power output of your 20-ton reactor? And can it survive a plane crash without containment failure?

[Kosh]

What's the power output of your 20-ton reactor? And can it survive a plane crash without containment failure?

Nghtm1r already stated it could survive a crash intact. The power output depends on the size, I'm not sure what a 20 tonner could do (nghtm1r?), but the one they were originally thinking about using back during the Air Forces nuclear bomber program had a 1 MW output, but I don't know how heavy it was.

[Mika]

Nghtm1r already stated it could survive a crash intact. The power output depends on the size, I'm not sure what a 20 tonner could do (nghtm1r?), but the one they were originally thinking about using back during the Air Forces nuclear bomber program had a 1 MW output, but I don't know how heavy it was.

I'm pretty sure it cannot withstand a crash intact. Black box flight data recorders are required to withstand momentarily accelerations of 3200 gs. The data recorder inside the box weighs about 30 grams, so the force experienced by the support structure is 0,04 kg * 3200 * 9,81 m/s^2 ~ 1256 Newtons, so 30 gram chip weighs about 120 kgs (numbers are from Wikipedia). And sometimes the flight data recorder has not survived the crash intact!

So you'll have a hard time to convince me that it would be possible to build an reactor to withstand such accelerations and keep an airplane airworthy.