You are not logged in.
Robertdyck-- I think we're approaching the safety issue in different ways. I don't disagree that, in the normal course of operations, an NTR is "safe" insofar as normal operation does not spew radioactivity everywhere like open cycle does, and it doesn't spew out toxic chemicals. While the exhaust may contain an elevated amount of deuterium, one could hardly call that a significant issue. What it does do, however, is produce a lot of radioactivity at a very high rate while it's firing. I am aware that aside from its tendency to burn vigorously in the elemental form, Uranium is not particularly poisonous. However, I would like to add that the Uranium used in CANDU reactors is mostly natural Uranium, meaning that it contains just .7% of the good stuff, U-235. Even U-235, before it's been fissioned, is not particularly bad from a radioactivity sense (I believe its half life is many millions of years, perhaps a billion?)
Coming out of a reactor of any kind is a different story. The materials out of which the engine are built will become highly radioactive; And I don't care what ceramic the Uranium is dispersed in, I bet you anything that it won't survive impact with ground or water at 5 km/s. And forget about landing in an inhabited area-- that engine is going to be far too hot, from a radioactivity standpoint. For that matter, forget about reusability entirely, because you'd have to go through the understandably expensive and time consuming procedure of replacing the intensely radioactive nuclear fuel at each launch. A 1% failure rate is pretty good for disposable rockets, I'd be surprised if adding the reentry step decreased this for any reason. The idea of having one of these crash anywhere near an inhabited area is rather terrifying.
It might be acceptable to use in the upper atmosphere of Venus however for returning astronauts to Orbit from Venusian Aerostat stations floating 50 km up in the atmosphere, otherwise it would be very difficult to get astronauts back into space after dropping them in the atmosphere, a nuclear jet engine might be just the thing for cruising in the atmosphere while a Nerva rocket with hydrogen propellent can be used to get into space. Radioactivity being a nonissue on Venus, radioactive fallout will drop to the surface where nobody lives.
Offline
Like button can go here
There have been some investigators who assert that the engines of nuclear rockets are so heavy they would not be suited to SSTO, despite the high Isp.
My opinion, if you do want SSTO a faster route to it is developing altitude compensation methods for standard chemical propulsion such as the aerospike.
Bob Clark
Old Space rule of acquisition (with a nod to Star Trek - the Next Generation):
“Anything worth doing is worth doing for a billion dollars.”
Offline
Like button can go here
I tend to agree with the latter, but not so much with the former. It depends just how efficient your engine is, of course, but for example the Timberwind proposal (Isp 1000 s, T/W 30) seems like it would be well-suited to an SSTO, although it's very questionable if you would get much payload, safety, or reusability out of the vehicle that resulted.
-Josh
Offline
Like button can go here
I think we'll be stuck with some kind of chemical system for surface-to-orbit for a long while. Fission is out for obvious reasons (yes, in normal operation it shouldn't pose any problems, but then neither does hydrazine), lightcraft have their own inherent problems, and fusion is not going to achieve the sorts of T/W needed unless we have a breakthrough in something like muon catalysed fusion.
Use what is abundant and build to last
Offline
Like button can go here
Well, the idea I've been mulling over lately (It led, indirectly, to my post about Beamed Propulsion) was to do beamed with Sodium Vapor lamps instead of vapor. It's a much lower technology level but probably a higher power density, while still producing relatively easily focused monochromatic light. It's more of a long term solution but it's an idea nevertheless.
Then there is also the possibility of some hybrid system, perhaps. One never knows. Or Airship to Orbit. I still haven't gathered the time for the calculations, but it's a fascinating idea even if it's not a practical one.
-Josh
Offline
Like button can go here
I tend to agree with the latter, but not so much with the former. It depends just how efficient your engine is, of course, but for example the Timberwind proposal (Isp 1000 s, T/W 30) seems like it would be well-suited to an SSTO, although it's very questionable if you would get much payload, safety, or reusability out of the vehicle that resulted.
I know very few about nuclear thermal rocket: NERVA or Timberwind based spaceships have to be orbit to orbit spaceships, with habitat and rocket separated by a long truss with propellant tanks, to reduce crew radiation exposure, or may be compact and able to perform atmospheric entry and landing?
Last edited by Quaoar (2013-12-23 11:29:49)
Offline
Like button can go here
Well, in an engineering sense it's perfectly feasible to use a Timberwind style engine for earth liftoff. For safety and political reasons, not so much. The radiation that the crew would be exposed to is pretty small because the hydrogen fuel between the reactor and the crew increases both distance and is a great shield against most kinds of radiation.
-Josh
Offline
Like button can go here
By the time NERVA finished, they had a plume very clean of radioactivity, and a nice Isp, just a crummy T/W. Projections for the next design, based on what they did achieve, were Isp 900-1000 s at T/W near 5.
Timberwind was a particle-bed variant of the fuel loading design. It never got tested as an engine, so there are no performance figures for it, only design analysis estimates. It appeared to have about the same Isp potential as NERVA (in the vicinity of 1000 s), but at a much better T/W, because the reactor and chamber design was so much lighter. Whether that would have actually translated into a lighter engine ready-to-fly is what the testing that was never done would have answered.
There was another fuel-incorporation design called Dumbo, that used some sort of metallic ribbons instead of ceramic rods. There was also some sort of change in the nozzle design for Dumbo, relative to NERVA. I've heard tell that a better nozzle could have been incorporated into NERVA as well, but I honestly don't know the truth or falsehood of that. Like Timberwind, Dumbo was never tested, so all we have are design projections, not trustable performance. It, Timberwind, and Nerva all had about the same Isp. Both Dumbo and Timberwind were supposed to be lighter than NERVA for better T/W.
As for plume radioactivity, again only NERVA ever got tested. Initially, it was really bad due to core cracking and erosion. But, they fixed that by the time the program was stopped. Stopped just short of first flight, I might add. Whether the other two designs would have had clean plumes is something that only testing would verify, and that was never done.
All three had the same abort/crash problems, especially after first use. You are faced with handling a very radioactive core in an emergency situation, all the way from the pad to orbital speeds. That, rather than plume radioactivity, is the real problem with using nuclear rockets for Earth surface launch. It's not that it cannot be handled, because it was, with the demonstrated 500-mph crash integrity of the reactor vessel flown on the NB-36 airplane. But that's very heavy, leading to engine T/W under 1 as a nuclear rocket.
The abort/crash problem is "solved" by the open-cycle gas core concept, but not by the closed-cycle "nuclear light bulb" concept. However, even with open-cycle gas core, you trade crash/abort safety for a radioactive plume. Plus, gas core never got to a testable device, only separate academic lab demonstrations of a couple of the tinkertoys (a gas phase core was controlled, and a plasma device demonstrated sufficient relative confinement of heavy species to assure 100% burnup).
It's not like you will contaminate the entire planet if a nuke rocket has a problem. But the irrational fear of anything not "guaranteed safe" (even though there is no such thing) is a huge obstacle that has prevented the application of this technology. And the idiotic mistakes of the nuke power plant people (all over the globe) have done nothing to help assuage these fears.
GW
GW Johnson
McGregor, Texas
"There is nothing as expensive as a dead crew, especially one dead from a bad management decision"
Offline
Like button can go here
Irrational people believed Mars was invading in 1938. How seriously should we take irrational people? Those same irrational people were afraid of the astronauts bringing back "Moon germs", I think they saw to many horror films like Apollo 18.
Last edited by Tom Kalbfus (2013-12-24 12:23:08)
Offline
Like button can go here
Offline
Like button can go here
Which is why I like the idea of a privately funded Mars mission, because those funding it don't have to take into account the irrational opinions of stupid people who vote. I think the Apollo model of a Mars Mission is becoming obsolete. Space is turning from an activity conducted by superpowers to the playground of rich people and businesses. Eventually if we wait long enough, people will finance their own trips to Mars and other places. I think the Chinese may land a man on the Moon within a decade, its just a matter of developing the right vehicles, I think when that happens however, the Chinese won't be the only ones there. Getting to the Moon is not that much more difficult than launching a communications satellite, there is no reason why a lot of private firms couldn't establish a presence there, all one needs are the proper financial incentives. I think robotics is a great enabler. We got two legged robots in development, we could send those to the Moon and have them controlled by operators on Earth in near real time. Instead of sending an astronaut, how about a robot with a operator controlling its every movement from Earth. The robot of course will have to do somethings by itself such as balance and walk, but the operator would tell it where to go. I would like to see something like the Atlas Robot on the Moon, if we get enough of them on the Moon, we could build a base there out of Lunar materials, and when the base is complete and fully functional, we could then send live humans. We could do all kinds of stuff there by remote robotics, that would be much more difficult on Mars due to the time lag. Robots on Mars would have to be much more autonomous than ones on the Moon.
Offline
Like button can go here
Erm, it doesn't matter if you're private, people still won't be happy with you launching a nuclear rocket.
Use what is abundant and build to last
Offline
Like button can go here
GW Johnson: I really like the T/W ratio of Timberwind. It did have the problem that spherical fuel elements of it's pebble bed would agglomerate, making it non-restartable. Fine, reduce the temperature a bit. The upgraded NERVA of 1990 (design only) had calculated Isp of 925 seconds, while Timberwind had calculated Isp of 1,000 seconds. Reducing temperature to NERVA level doesn't loose much. But the big question is how did they do that? The mass of Timberwind is incredibly attractive! NERVA used axial flow with ceramic fuel elements, Timberwind used radial flow with pebbles. But what were those pebbles? Were they nothing but pure uranium metal? That would not contain fission fragments (nuclear waste). NERVA would contain fission fragments within its ceramic fuel elements. NERVA had several additional safety features: I could fall all the way out of orbit and the fuel elements would not break. They would get scattered over the debris field, but each element wouldn't break. And even if they did, the ceramic didn't just encapsulate uranium, rather the uranium was evenly mixed with ceramic, so braking one would just expose a new face. Like taking a loaf of bread and cutting it in two. Furthermore, NERVA was intended as an upper stage only; it would be launched cold. The reactor would only be started for Trans-Mars Injection. That is departure from LEO toward Mars. Yes, NERVA was intended as the engine for a new third stage for a Saturn V, specifically for a human mission to Mars. But Richard Nixon cancelled it. (grumble, grumble!) But Timberwind was designed as an upgraded Titan missile, an ICBM to carry a nuclear bomb to Russia. It was part of Ronald Regan's Star Wars. So that means a ground launched solid core Nuclear Thermal Rocket. Nuclear weapons guys would consider a radioactive exhaust plume insignificant compared to radiation from a thermo-nuclear bomb. Was Timberwind to use unshielded uranium fuel pebbles?
Also, you mentioned gas core NTR. That has extremely high Isp! It looks extremely good, until you realize all the fission fragments are expelled out the exhaust stream. Again, radioactive exhaust.
I would consider a ground launch NTR to be a good idea. However, I argue for solid core. The reason is to contain fission fragments, so there's no radioactive exhaust. And something like NERVA's ceramic fuel elements would be nice, to address the abort/crash problem. But if that's what increases NERVA's weight, then an alternative has to be found.
Offline
Like button can go here
There is something, after all, that we're forgetting about pretty much all of the technology used in space: Thus far it's been the explicit and sole domain of Space agencies and the Military. Remember that the Falcon 9 would make an immensely powerful ICBM (Even if filled with normal chemical explosives, or for that matter an Iron slug, it could deliver 13 tonnes of material moving at 8 km/s to a bunker anywhere on Earth. 13 tonnes of Iron moving at 8 km/s corresponds to 400 GJ of energy. That's 100 tonnes of TNT equivalent. Not quite a nuclear explosion, but more than enough to be terrifying. Falcon Heavy will be equivalent to 500 tonnes of TNT), and you see why it took so long to get any kind of private space going.
Now, what about nuclear technology? There are privately operated nuclear reactors in this country. However, their design hasn't changed in 50 years, and each reactor has to be individually certified and tested by the NRC (Nuclear Regulatory Commission) before it actually begins operation. Inspections are supposed to occur regularly because of both proliferation fears and safety concerns regarding a meltdown. Newer reactor designs would reduce the need for these significantly but there is really no impetus (and because of the uncertain regulatory hurdles, the impetus would have to be political in nature) to design or commission new kinds of reactors.
Now take this environment, and try to design a reactor ten times more powerful than any ever built in the US. Try to put it on top of a rocket moving at multiple kilometers per second. Try to get it passed when it is basically open to the air.
If it seems unlikely, that's because it is. Nuclear rockets are a wonderful technology, but the regulatory, design, and safety hurdles with implementing them for SSTO, while far from insurmountable, are significant and probably not worth the investment.
The goal of increasing Isp is a good one but I think it's best to go about it by chemical and physical means.
-Josh
Offline
Like button can go here
If you want a power plant for terrestrial use, why not simply buy a CanDU reactor from Canada? It has several advantages over the Three Mile Island design. It's a heavy water reactor, which means if a run-away reaction occurs leading to a melt-down, then loss of coolant simply causes the reaction to stop. Heavy water is both primary coolant and moderator; without moderator the nuclear reaction is not self-sustaining. That is a deliberate design feature; it simply cannot melt-down. Secondly, the Three Mile Island design is based on a reactor developed for a submarine. It must be dismantled to refuel, requiring about 3 months of shut-down. But the CanDU reactor is design so a big door is opened while the reactor is still operating; a robot arm grabs spent fuel elements and replaces them. All while the reactor simply doesn't stop. You don't want any human in that room while the door is open; in fact no electronics would survive. They have heavy electrical cables operate electric motors; all electronics are in another room behind a concrete wall several feet thick. The Pickering nuclear power plant had the worst accident possible in 1983. A tube that holds fuel rods ruptured, resulting in primary coolant leak. Coolant contaminated with tritium leaked. Do you remember that? No? Never heard of it? That's how benign it was. All they did was close all the beaches for Lake Ontario for 2 weeks. Zero fatalities. And the CanDU is designed to use uranium oxide, purified but not enriched at all. The ratio of isotopes is exactly as dug out of the ground. This is deliberate, primarily so a country can operate one of these reactors without the capability to enrich uranium. An unintended but beneficial side effect is the reactors cost less to operate, simply because you don't have to pay for enrichment. Canada has considered updating this design several times, but no one wants to buy any. Development cost cannot be justified if there are no customers. But as-is, it's far better than a Three-Mile-Island design.
Offline
Like button can go here
The CANDU reactors are better than the PBWRs used in the USA, but better than both designs is the Molten Salt Reactor. Compared to the cost of actually building a few reactors, the cost of development is not actually very high; We better get on with it. Every additional kilogram of CO2 we put into the atmosphere will cause more damage to the climate, and solar PV doesn't really cut it as a grid energy source.
-Josh
Offline
Like button can go here
Every time I read a news article about Iran and their nuclear enrichment, I keep thinking of CanDU reactors. Why not just replace the Russian reactors with Canadian ones? Then Iran can have nuclear power for industry without the ability to enrich. They many want weapons; if so the argument of Canadian reactors would reveal their ambitions.
Offline
Like button can go here
There's no reactor from which it is truly impossible to enrich, regardless of what you may have read. The CANDU reactors almost certainly produce large amounts of plutonium. This could be purified and separated from Uranium using only chemical means.
-Josh
Offline
Like button can go here
The thing with CanDU is uranium enrichment is not necessary or useful. There's no purpose for it. Enrichment is necessary to make a uranium bomb. The CanDU does produce plutonium, yes, that is a down side. Extracting plutonium from spent fuel is not easy. And you have to be careful how you do it. With normal reactor operation, plutonium produced will have multiple isotopes, so the isotope balance will be messed up so badly you can't use it for a bomb. You would have to expose a fuel rod for a short time, then take it out and extract that plutonium. That would give the right isotope. That's not operating the reactor the normal way. Iran could do it, but wouldn't inspectors notice?
Offline
Like button can go here
Not if it kicked the inspectors out. Alternat8ve answer: I have no idea and possibly not.
Having said that, it's certainly preferable to their program now. I wonder why nobody has proposed this yet. Perhaps we're not comfortable helping Iran with any of its reactor building.
In any case, we've gone right off topic and should probably go back to the non political uses of nuclear power.
-Josh
Offline
Like button can go here
Simply looking at the numbers:
1 orion launch = 1-10 hypothetical deaths from radiogenic cancer.
Gasolene/coal air pollution in the US = ~50,000 cancer deaths each year. This is to say nothing about the heart attacks that it causes. God only knows how many people die from air pollution in China. How many people in the US die each year from pollution from Chinese coal burning plants and visa versa? I bet its a lot more than 10. There is no pollitical storm. No mass protests. People are aware of it and live with it.
I would suggest that if public acceptance with ground launch nuclear propulsion is a problem, then the solution lies not in technological development, but in PR. At the end of the day, we all live with pollution, because the benefits we gain as individuals livining in a technological society justify taking the risks. I would humbly suggest that the same is true of nuclear propulsion.
Offline
Like button can go here
An exhaust velocity of 10 km/s implies that you're adding roughly 70 MJ/kg to the propellant including inefficiency. 10 km/s implies a mass ratio of roughly 2.7. If your T/W at liftoff is 20 N/kg and your powerplant consists of 10% of total liftoff mass, it needs to have a specific power of 1.4 MJ/kg.
The corresponding numbers for 6 km/s (assuming that the power plant constitutes the same fraction of the dry mass) are:
E=25 MJ/kg
R=5
T/W=20 N/kg
PP%=5.4%
P=1.5 MJ/kg
So pretty much the same. These calculations assume that the power source does not consume any fuel of any kind. If the power source does consume some kind of fuel, that fuel would need to have an energy content of at least 440 MJ (for 1000 s) or 370 MJ/kg (for 600 s). This assumes that the mass allocated to a power plant is actually used for fuel which reacts with the propellant in a hypergolic fashion.
Nuclear fission is the obvious answer, but I'd like to set that aside for a moment for the reasons of safety discussed above, and also because I'd prefer to speculate on other things.
The amounts of energy we're talking about are high, but not inconceivably higher than the chemical fuels we're used to. For reference, 440 MJ/kg is 4.6 eV/amu and 370 MJ/kg is 3.8 eV/amu. Nuclear fission gives off a recoverable energy around 675,000 eV/amu and fusion around 3,500,000 eV/amu. Antimatter gives the maximum possible value for energy content at 1,863,000,000 eV/amu (double the mass of the antimatter). But this energy is released as very-hard-to-capture gamma rays.
The energy we want is really just outside the range of chemical, as these things go. Hydrogen's ionization energy is 13.6 eV, which is more than enough if we could find a dense, light way to store plasma. Helium's first is 24.6 eV and its second is 54.4 eV, so in total that's 19.75 eV/amu (using He-4, as I'd recommend) or 26.3 eV/amu (using He-3, which seems like extremely expensive overkill).
The problem here is storage. Plasmas (especially if you're opting to go after doubly-ionized Helium) are hot and therefore diffuse. As an example, Helium at 10,000 K and 25 atm has a(n ideal) density of 0.12 kg/m^3. For comparison, LH2 is 70 kg/m^3 and water is canonically 1,000 kg/m^3.
Hypothetically, let's say you're using ammonia as propellant (Isp 600 s). You will need 4 kg of Ammonia per kg of dry mass, with a volume of 0.0059 m^3. To provide energy, you will need 0.013 kg of He per kg of NH3. If you can achieve a pressure of 1000 atm and the ideal gas law provides a reasonably accurate approximation (it doesn't, but I don't have a good way of calculating plasma pressure so I'm going to run with it as a very rough guess) your plasma density will be 4.8 kg/m^3 and take up 0.0027 m. In other words your plasma fuel will take up roughly as much volume (order of magnitude) as your propellant. It's not an issue, per se, but it's an interesting result.
1,000 atmospheres is a whole lot of pressure to contain with plasma, too. The temperatures we're looking at are low (10,000 K is roughly 1 eV, as compared to 10,000+eV in fusion plasmas) but considerable. As a sidenote, I pretty much pulled 10,000 K out of a hat and it may not be enough to ionize.
I've done a little more research based on the spectral emissions of stars. Based on this, the Hydrogen line shows up starting in G9 stars (5400 K). He I (i.e. Helium's first ionization) starts at B9 stars (10,900 K). He II (Helium's second ionization) starts in O9 stars (30,000 K).
These temperatures are not for total ionization, just when the ionization becomes detectable. Hydrogen lines are strongest at 11,500 K. He I lines appear to be strongest at 21,000 K. Wikipedia does not give a temperature at which He II lines are strongest, presumably because there are no known stars at that temperature. This means that He II lines peak at temperatures above 52,000 K.
Even at the peak spectral emittance, the species does not exist completely in an ionized state. A better estimate is that it comes roughly 50% ionized at that temperature.
-Josh
Offline
Like button can go here
By the way, I agree that CANDU type reactors would be very difficult to use as generators of plutonium or other bomb making materials. However, their heavy water filled jackets do generate quite a lot of Tritium, which is just the stuff for upgrading to Hydrogen weapons if you already have atomic bomb technology licked. They are also very expensive to fill in the first place.
Offline
Like button can go here
I think the best prospect for doing better than hydrogen is to take the power source off the vehicle. Beamed power seems like it's the way to go. If we restrict launches to the daytime, solar power is probably the cheapest source of energy to power the beam.
Particularly if it's put up in the stratosphere, where the sunlight is a lot stronger. I'm imagining a large aerostation that is covered in solar cells (or using parabolic mirrors to power turbines...), launching 1 tonne payloads into orbit whenever it's in sunlight.
Use what is abundant and build to last
Offline
Like button can go here