Slashdot Mirror
Nuclear Rockets Moving Along
AKAImBatman writes "Bruce Behrhorst of NuclearSpace.com recently stumbled across a new engine from everyone's favorite Jet Engine maker, Pratt & Whitney. Unlike P&W's previous engines, however, this engine is not a jet, and is powered by Nuclear Fission.
It seems that P&W has responded to the need for Mars transportation by inventing the first commercially viable nuclear thermal rocket. They have heavily improved upon the NERVA NRX design from the 60's, and have even solved the graphite ablation problem! With this new engine, it seems that an inexpensive trip to Mars is now firmly within our grasp. Will we rise to the challenge?"
This design is significantly different from the NRX. For one, they didn't attempt to build the most powerful reactor in the universe. For another, they took advantage of LHOx afterburners. With both of those design choices in mind, they were then able to use a titanium shell to act as the heat sink for the reactor. Not only does it not ablate, but the titanium will melt and scram the reactor long before the reactor itself experiences meltdown.
:-)
In other words, this is an extremely safe reactor design.
Javascript + Nintendo DSi = DSiCade
Well, you could always RTFA. Here:
BB: Is there a 'fail safe' operation in the event the reactor core must
be shut down exiting a planetary 'gravity well' or on approach to a
'gravity well' ?
RJ: There are several features that we have adapted and evolved into the
current 'TRITON' design to handle risk mitigation for the Uranium
Dioxide (UO_2 ) fuel element core in a Nuclear Thermal Rocket (NTR).
We have approached this by providing an integrated, robust design the
uses dual turbopumps (turbopumps provide coolant flow to the reactor in
propulsion mode).
In thrust mode where you have high power operation, is where this
concern has been typically addressed.
The safety features that have been taken into account for risk reduction
entail constant supply of reactor coolant by using dual turbopumps. This
means turbopumps with their moving parts like bearings, shafts, turbines
etc. may cavitate and over speed, if for some reason one of the
turbopumps showed signs of malfunction or not operating within
appropriate parameters, you could effectively shutdown or bypass the
offending turbopump and still have coolant flow going to the reactor.
This is one of the key features for propulsion mode operation to make
sure coolant is available to ?flush; the reactor if it needs to be shut
down when it has gotten to the full thermal power level. In power mode
it's [core] sitting at an idling power-level so the amount of time for
the reactor to over-heat if starved of coolant (i.e. He/Xe gas) is
extremely negligible because you are running the reactor core at nearly
half the maximum temperatures the core is design for. So, if in the
event of something like let's say, a minor leak in the radiator during
power-mode operation, you can do a shut-down of the reactor from a very
moderate control state without over-heating the reactor core. Other
failure mode mitigation would be to have a segmented radiator design, or
have a coolant purge circuit in the design, or actually split the
coolant circuit to provide redundancy. We also have several valve
arrangements so that in the event of leakage in idle power mode you
could shut a section of the radiator down; the temperature of the
reactor is so low it would cool down on its own. This works to our favor
in the ?TRITON? design because the CERMET core materials have high
maximum operating temperatures since it's designed for exit temperatures
near 2,700-K in the propulsion mode.
Another feature is the nature of going to a fast spectrum reactor. It
allows issues such as criticality and impact immersion (e.g. wet sand or
salt water) to immediately be mitigated because of the reactor neutron
flux levels and the use of only a reflector and no moderator to
thermalize a bulk of the neutrons. Essentially it helps to 'poison' the
internal nature of the reactor so in the worst case event at launch, if
the reactor were to end up in sand or saltwater it will keep it from
resorting to a super-critical state. If it shuts down after a brief
period of operation, like for some reason and I had to shut it down
during an early phase of a human Mars mission, the 'burn-up' (fission
product build-up) is so low. Even if I run it for only 5 minutes or, 10
minutes I'd have built up only a minuscule amount that could barely be
measured with regards to build-up of fission products in the core. So if
it did for some reason re-enter the earth?s atmosphere, the radiation
levels are only slightly higher than typical naturally occurring levels.
Now, you would have to methodically go through a full risk analysis, or
a whole mission point-to-point to define the 'What if scenarios' along
the mission's plan to properly build in aborts for all the most probable
failure modes.
For example, one 'What if scenarios' would look at the failure modes for
an orbit capture high-thrust burn at a planet Mars or for Lunar
transport. In essence, an inve
Javascript + Nintendo DSi = DSiCade
Did you read the article? It has 15000 pounds of thrust, at nominal output. Totally useless for ground-orbit missions. It is designed to fly from orbit here to orbit somewhere interesting.
"I do not agree with what you say, but I will defend to the death your right to say it"
Not to sound paranoid, but when the reactor overheats and falls off where does it go?
Launch profiles are designed so that everything falls into the ocean. NASA has aborted quite a few launches, and has never dropped anything on people's heads. China on the other hand...
What happens if the reactor falls off over a populated area?
Well, since it's not supposed to be activated until the craft is already outside of the atmosphere, I suppose someone gets a bump on the head. Even if we assume that the reactor overheated, the titanium shell will melt down and scram the reactor before the reactor itself melts down. It should be nice and cool (and still wrapped in titanium shielding) by the time it hits the water.
Say the reactor falls off on the way to mars. Unless there is a shift in the momentum of the ship or the reactor it'll just melt down beside the ship. Then imagine the case where the ship can separate itself from the reactor. Now how do they get back?
The mission profile suggests three engines. Unless there's a critical failure in all three, a modified flight path could be developed.
While this is probably an improvement, I'd hardly consider it safe.
Consider a chemical rocket on the way to Mars. What happens if the tanks explode? That's right, you've got no way back. Even the failure of one engine could spell doom for the mission. This engine is more powerful, and FAR safer than any chemical engine. Even if the tanks leaked on the way, fuel could still be scooped from Mar's atmosphere. No chemical rocket can make that claim.
Javascript + Nintendo DSi = DSiCade
Very simply put, the NERVA engines ran so hot that the Graphite used to transfer the heat from the reactor into the exhaust would flake off and end up in the exhaust. The problem is that while the hydrogen exhaust cannot be made radioactive, the graphite can. So you'd get little specs of radioactive graphite raining down behind you. It wasn't so much graphite as to be a major concern, but many of us would rather not exhaust anything radioactive if we can help it.
Javascript + Nintendo DSi = DSiCade
Did you bother checking the track record of nuclear material that has already rained down? Seems the US has done a fairly good job containing such materials. (That is, right after they figured out that it might be a good idea to do so. :-))
Javascript + Nintendo DSi = DSiCade
Well, yeah...
You're more right than you (may) know. I served on a nuclear sub, as a reactor operator. In the two years of schooling we get, there's much emphasis on rote memorization as well as understanding. One list we had to memorize is "negative public consequences if there were an accident", one of them being "negative public reaction to the naval nuclear program". We were operating in secret. We were taught that a major part of the reason that the naval nuclear program even still exists is because it's never (ever) had an actual accident. ("Accident" being a strict government policy-defined term.) The only reason we can get away with six nuclear reactors bobbing up and down in San Diego's bay right at this moment is because people really honestly don't know they're there. They're not in the news, they have a low physical profile. "Well yeah", nuclear subs have been operating in secret.
If aspiration is a virtue, achievement cannot be a vice.
Ya need to know that sub-critical nuclear fuel is never going to produce a mushroom cloud. Producing a runaway nuclear reaction is extremely difficult. You'd require the right isotope of uranium, first. Then you'd need two sub-critial lumps separated enough so the radiation engendered by their proximity wouldn't simply vaporize the engine before a chain reaction could take off. The two-sub-critical masses have to be brought into close proximity quickly, usually by firing the masses into each other with two high-explosive devices; picture a tube with HE on each end, with a uranium "shell" on each charge. You'd fire both shotgun shells down the tube to meet each other. The temperature and the radiation caused by their increasing proximity tries to vaporize the assemblage, but the sheer speed at which they collide enables the neutron levels to increase to a the point where a runaway chain reaction released enough energy to raise the temp to a few million degrees. Boom.
If a nuke Challenger went down, the LH2 used as propellant would ignite with the O2 from the air, and you'd get a big boom. Not as much as the Challenger with it's perfect blend of LOX and LH2, but it'd be pretty big, as booms go. But the reactor would simply fall like a radioactive Geo Metro. No boom. Wrong isotopes, no way to go critical.
It's pretty safe to say that the likelihood of a nuclear reactor crushing into a critical configuration despite the normal measures taken to keep it "off" (neutron-absorbing control rods inserted, etc) is vanishingly small. In that you are correct.
In a gun design you only need to move one mass. This only appears to be feasible with U-235. Faulty thinking; the temperature and radiation (which turns the bomb core into high-pressure gas and pushes it apart again) are caused by the reaction; they are not separate from it.One point you appear to be missing is that the nuclear reaction takes a certain amount of time; neutrons are not infinitely fast, nuclei do not fission instantaneously, the exponential change rate of the reaction (whether growth or decay) is controlled by the composition of the material and its geometry. The geometry controls whether a splitting atom has a > 1 or < 1 probability of causing another fission. If the probability is >>1, you've got an explosion in progress; if it is < .5, you've got a lump.
The goal of the bomb designer is to turn the sub-critical mass into a prompt-supercritical mass before a chain reaction can begin and take the mass apart again; to this end they design implosion mechanisms and neutron generators to make everything happen when desired and not a microsecond before. The goal of the reactor designer is to make certain that the chain reaction is always under control. We can see that this isn't overly difficult; even Three Mile Island had a nicely-controlled reaction (its problem was lack of coolant), and only the Russians appear to have been careless enough to have a major incident (and without any containment building either, tsk tsk).
Sustainability and energy independence essay