Nuclear spaceship

[strangequark]

OK, if it doesn't simplify things it's probably not worth it. Here another wild concept I'd like to run by you:

What if instead of going all the way to NTR we only use a nuclear reactor to drive a turbopump and to preheat the propellant in an otherwise conventional chemical rocket? Would that be a practical stepping stone towards a more advanced LANTR system, leaving out the reverse ramjet complications? More like a nuclear assisted chemical rocket than the other way round.

More complex on every front, and you gain maybe 50s of Isp over straight hydrolox, and crash your T/W.

 An LH2 NTR is actually a pretty simple machine, if done the way it should be. If we're going to go nuclear, then design it for a reusable, in-space vehicle. It should be designed, from square one, to be simple, reliable, and long-lived. The ideal NTR in my mind is a 100kN engine, expander cycle, running at 900sec (to keep temps reasonable), with a CERMET reactor core. It needs to operate like a naval reactor, and not like a chemical rocket engine. It should not require replacement until the reactor core is decommissioned after 50 years of service. As for stepping stones, if we're ever going to get serious about being a space-faring society, we need long-term storage of cryogens. I don't think that is too much to ask.

[mmeijeri]

More complex on every front, and you gain maybe 50s of Isp over straight hydrolox, and crash your T/W.

Only 50s? Some playing around with Propep suggested that even with preheated temperatures below 1000K storable bipropellant could be competitive with purely chemical LOX/LH2 and preheated LOX/LH2 could be competitive with storable propellant NTR. Then again, I've gotten strange numbers out of Propep before, so that might be the real explanation.

And is it really more complex? You would have lower temperatures and you would need a smaller reactor. Or if it is, could it be a stepping stone towards LANTR? I won't be crushed if the answer is no, but I am curious.

An LH2 NTR is actually a pretty simple machine, if done the way it should be.

But it still needs storage and transfer of LH2 in orbit, and it has very bad mass fractions and T/W.

If we're going to go nuclear, then design it for a reusable, in-space vehicle. It should be designed, from square one, to be simple, reliable, and long-lived.

I think that should be the case regardless of whether we go nuclear. And we should start non-nuclear simply because it is simpler and more reliable.

As for stepping stones, if we're ever going to get serious about being a space-faring society, we need long-term storage of cryogens. I don't think that is too much to ask.

Agreed, but I'm wondering if the two can be usefully decoupled. Ammonia NTR suggests it could be. Delays to one technology should ideally not cause ripple effect on others.

[strangequark]

More complex on every front, and you gain maybe 50s of Isp over straight hydrolox, and crash your T/W.

Only 50s? Some playing around with Propep suggested that even with preheated temperatures below 1000K storable bipropellant could be competitive with purely chemical LOX/LH2 and preheated LOX/LH2 could be competitive with storable propellant NTR. Then again, I've gotten strange numbers out of Propep before, so that might be the real explanation.

That what I got. I wouldn't want to preheat much more than about 1200K, and the chemistry gives you a hard upper limit (this is why turbojets do not work hypersonically).

And is it really more complex? You would have lower temperatures and you would need a smaller reactor. Or if it is, could it be a stepping stone towards LANTR? I won't be crushed if the answer is no, but I am curious.

A smaller reactor isn't necessarily that much simpler. If you are using it in the manner you suggest, then the operating pressure will be much higher, and using oxygen as the working fluid brings lots of additional challenges you would not have to solve in any other system. LANTR requires very different challenges to be overcome. It's less a stepping stone than a complete detour.

An LH2 NTR is actually a pretty simple machine, if done the way it should be.

But it still needs storage and transfer of LH2 in orbit, and it has very bad mass fractions and T/W.

Which we need to develop, period. Mass fractions and T/W are less important than the raw Isp for most of the missions you'd want an NTR.

If we're going to go nuclear, then design it for a reusable, in-space vehicle. It should be designed, from square one, to be simple, reliable, and long-lived.

I think that should be the case regardless of whether we go nuclear. And we should start non-nuclear simply because it is simpler and more reliable.

You know, I don't know if I'll give you simpler or more reliable. The most instructive example for spaceflight is that of naval reactors. The safety and reliability record of US naval reactors is absolutely sterling. As for simplicity, you have half the feed system hardware, non-critical propellant injection, a gentle expander cycle for your turbomachinery, and a simple supercritical heating process. A rocket combustion chamber is much more chaotic and complicated.

As for stepping stones, if we're ever going to get serious about being a space-faring society, we need long-term storage of cryogens. I don't think that is too much to ask.

Agreed, but I'm wondering if the two can be usefully decoupled. Ammonia NTR suggests it could be. Delays to one technology should ideally not cause ripple effect on others.

Ammonia NTR may be useful to pursue, but an Ammonia NTR is going to look much different. For instance, concerns about nitriding may drive you toward a very different core.

[mmeijeri]

That what I got.

I'll try again when I have more time, maybe we can figure out what I'm doing wrong. And of course, that's not counting the gain from powering your turbopump with the reactor, is it?

A smaller reactor isn't necessarily that much simpler. If you are using it in the manner you suggest, then the operating pressure will be much higher, and using oxygen as the working fluid brings lots of additional challenges you would not have to solve in any other system.

Smaller may not be much simpler, but isn't it still easier or at least cheaper? And it would have much better T/W than an NTR. Much lower Isp too of course, and also much lower T/W than a purely chemical engine.

I didn't understand your point about needing higher pressure or the complications of oxygen. I wasn't suggesting running oxygen through the reactor, but using a nuclear preburner that discharges into a mostly conventional combustion chamber.

LANTR requires very different challenges to be overcome. It's less a stepping stone than a complete detour.

I didn't mean LANTR as a stepping stone towards NTR, but "nuclear augmented chemical rockets" as a stepping stone towards LANTR. Or maybe I misunderstood your point?

[93143]

strangequark, how much do you know about Dumbo?  (Or did I ask you this at some point already?)

[rusty]

... it remains my informed opinion that ultimately NTR/NEP will be the in-space propulsion system of choice.

I've been considering opening a thread on ways to reduce the mass of radiators for NEP, including the feasibility of existing studies and new ideas.

Here, there or of no interest?

[strangequark]

strangequark, how much do you know about Dumbo?  (Or did I ask you this at some point already?)

Just saw this. I don't know much about it at all. Do you have any interesting reading material?

[DarkenedOne]

Ultimately, all our in-space spacecraft will be nuclear powered imo. But before that happens I do believe that we will see a large number of attempts to sway the architecture to be all SEP. In the end however, I believe that SEP will continue to have its place but will fail as the standard-bearer of in-space propulsion simply because of the gargantuan size of the arrays necessary to produce enough power to move anything really heavy and the time it takes to transport things. Will SEP work? Sure, but it will be awkward. SEP's will likely remain the propulsion choice for things small and of non-time critical importance, but NTR/NEP will become the majority of in-space propulsion.

Properly constructed they are no more dangerous than any other powerful source. Proper attention to detail and professional conduct of duties will mitigate the dangers and NTR/NEP will become commonplace, especially where personnel transport and/or time-critical cargo are concerned. As far as radiation from the engine is concerned, it is not a problem as the exhaust is clean and not radioactive at all.

We should be looking to nuclear for the vast majority of our in-space transportation needs, not more and more wasteful chemical propulsion, nor larger and larger solar arrays. As for nuclear fuel, we can mine it from the moon's surface. There is enough thorium on the lunar near side to provide nuclear fuel for both spacecraft *and* lunar surface operations for literally thousands of years. See the thorium map below provided by NASA's Prospector spacecraft. As for the working fluid, once we have mining on the moon we will get all the hydrogen we need as well. Establish proper mining on the moon and we will never again have to lift propellant from the earth's surface for any of our spacecraft missions.

I know there are many out there in NSF land who will disagree with me because they believe that the answers all lie with SEP. To those hardy souls all I will say is that they are entitled to their opinions, but I have looked very carefully at the whole question and while I do see a place for SEP, it remains my informed opinion that ultimately NTR/NEP will be the in-space propulsion system of choice.

Guys you are missing a huge reason why nuclear propulsion was thought of in the first place. 

Many of the early ideas for NTR accepted ANY GAS OR LIQUID as a working fluid, so long as it does not damage the reactor in any way of course.  Unlike a chemical rocket, in which, the reaction mass and the energy source are the same thing, the nuclear reactor provides the energy in a nuclear propulsion system.  The nuclear reactor will heat the working fluid, no matter what it is, and since the ideal gas law states that all gases expand when heated, the NTR will work with any gas. 

This fact is very important and has huge ramifications for space exploration and development.  It makes in-site propellant usage incredibly simple and easy. 

Take Mars Direct for example.  They propose sending a unmanned, nuclear-powered refinery with hydrogen to produce methane and oxygen from Mar's atmosphere, then storing it for use by a chemical rocket on the return trip.  I have seen proposals for going to Mar's with a NTR where the plan was to simply pump in the gas from Mar's atmosphere, and stick it right into the NTR for the return journey.  It does not matter to the reactor that it is using CO2 rather than the ideal, which is H2. 

With nuclear propulsion you guys have to free yourselves from the mindset that you need specific, highly-refined propellant.  With nuclear power you need neither.

[strangequark]

Ultimately, all our in-space spacecraft will be nuclear powered imo. But before that happens I do believe that we will see a large number of attempts to sway the architecture to be all SEP. In the end however, I believe that SEP will continue to have its place but will fail as the standard-bearer of in-space propulsion simply because of the gargantuan size of the arrays necessary to produce enough power to move anything really heavy and the time it takes to transport things. Will SEP work? Sure, but it will be awkward. SEP's will likely remain the propulsion choice for things small and of non-time critical importance, but NTR/NEP will become the majority of in-space propulsion.

Properly constructed they are no more dangerous than any other powerful source. Proper attention to detail and professional conduct of duties will mitigate the dangers and NTR/NEP will become commonplace, especially where personnel transport and/or time-critical cargo are concerned. As far as radiation from the engine is concerned, it is not a problem as the exhaust is clean and not radioactive at all.

We should be looking to nuclear for the vast majority of our in-space transportation needs, not more and more wasteful chemical propulsion, nor larger and larger solar arrays. As for nuclear fuel, we can mine it from the moon's surface. There is enough thorium on the lunar near side to provide nuclear fuel for both spacecraft *and* lunar surface operations for literally thousands of years. See the thorium map below provided by NASA's Prospector spacecraft. As for the working fluid, once we have mining on the moon we will get all the hydrogen we need as well. Establish proper mining on the moon and we will never again have to lift propellant from the earth's surface for any of our spacecraft missions.

I know there are many out there in NSF land who will disagree with me because they believe that the answers all lie with SEP. To those hardy souls all I will say is that they are entitled to their opinions, but I have looked very carefully at the whole question and while I do see a place for SEP, it remains my informed opinion that ultimately NTR/NEP will be the in-space propulsion system of choice.

Guys you are missing a huge reason why nuclear propulsion was thought of in the first place. 

Many of the early ideas for NTR accepted ANY GAS OR LIQUID as a working fluid, so long as it does not damage the reactor in any way of course.  Unlike a chemical rocket, in which, the reaction mass and the energy source are the same thing, the nuclear reactor provides the energy in a nuclear propulsion system.  The nuclear reactor will heat the working fluid, no matter what it is, and since the ideal gas law states that all gases expand when heated, the NTR will work with any gas. 

This fact is very important and has huge ramifications for space exploration and development.  It makes in-site propellant usage incredibly simple and easy. 

Take Mars Direct for example.  They propose sending a unmanned, nuclear-powered refinery with hydrogen to produce methane and oxygen from Mar's atmosphere, then storing it for use by a chemical rocket on the return trip.  I have seen proposals for going to Mar's with a NTR where the plan was to simply pump in the gas from Mar's atmosphere, and stick it right into the NTR for the return journey.  It does not matter to the reactor that it is using CO2 rather than the ideal, which is H2. 

With nuclear propulsion you guys have to free yourselves from the mindset that you need specific, highly-refined propellant.  With nuclear power you need neither.

Use anything other than H2 or NH3 and your Isp goes down the toilet, to where it is better to use the chemical systems. If you want to make use of the Mars atmosphere, it would be a better idea to use a very small reactor and crack the CO2. The Isp is better.

[ChileVerde]

I've been considering opening a thread on ways to reduce the mass of radiators for NEP, including the feasibility of existing studies and new ideas.

Here, there or of no interest?

I'll vote aye. Radiators may well be the longest pole in that tent.

(During SDI, radiators were of interest for other reasons, but the funding was there to develop some interesting ideas.)

[DarkenedOne]

Ultimately, all our in-space spacecraft will be nuclear powered imo. But before that happens I do believe that we will see a large number of attempts to sway the architecture to be all SEP. In the end however, I believe that SEP will continue to have its place but will fail as the standard-bearer of in-space propulsion simply because of the gargantuan size of the arrays necessary to produce enough power to move anything really heavy and the time it takes to transport things. Will SEP work? Sure, but it will be awkward. SEP's will likely remain the propulsion choice for things small and of non-time critical importance, but NTR/NEP will become the majority of in-space propulsion.

Properly constructed they are no more dangerous than any other powerful source. Proper attention to detail and professional conduct of duties will mitigate the dangers and NTR/NEP will become commonplace, especially where personnel transport and/or time-critical cargo are concerned. As far as radiation from the engine is concerned, it is not a problem as the exhaust is clean and not radioactive at all.

We should be looking to nuclear for the vast majority of our in-space transportation needs, not more and more wasteful chemical propulsion, nor larger and larger solar arrays. As for nuclear fuel, we can mine it from the moon's surface. There is enough thorium on the lunar near side to provide nuclear fuel for both spacecraft *and* lunar surface operations for literally thousands of years. See the thorium map below provided by NASA's Prospector spacecraft. As for the working fluid, once we have mining on the moon we will get all the hydrogen we need as well. Establish proper mining on the moon and we will never again have to lift propellant from the earth's surface for any of our spacecraft missions.

I know there are many out there in NSF land who will disagree with me because they believe that the answers all lie with SEP. To those hardy souls all I will say is that they are entitled to their opinions, but I have looked very carefully at the whole question and while I do see a place for SEP, it remains my informed opinion that ultimately NTR/NEP will be the in-space propulsion system of choice.

Guys you are missing a huge reason why nuclear propulsion was thought of in the first place. 

Many of the early ideas for NTR accepted ANY GAS OR LIQUID as a working fluid, so long as it does not damage the reactor in any way of course.  Unlike a chemical rocket, in which, the reaction mass and the energy source are the same thing, the nuclear reactor provides the energy in a nuclear propulsion system.  The nuclear reactor will heat the working fluid, no matter what it is, and since the ideal gas law states that all gases expand when heated, the NTR will work with any gas. 

This fact is very important and has huge ramifications for space exploration and development.  It makes in-site propellant usage incredibly simple and easy. 

Take Mars Direct for example.  They propose sending a unmanned, nuclear-powered refinery with hydrogen to produce methane and oxygen from Mar's atmosphere, then storing it for use by a chemical rocket on the return trip.  I have seen proposals for going to Mar's with a NTR where the plan was to simply pump in the gas from Mar's atmosphere, and stick it right into the NTR for the return journey.  It does not matter to the reactor that it is using CO2 rather than the ideal, which is H2. 

With nuclear propulsion you guys have to free yourselves from the mindset that you need specific, highly-refined propellant.  With nuclear power you need neither.

Use anything other than H2 or NH3 and your Isp goes down the toilet, to where it is better to use the chemical systems. If you want to make use of the Mars atmosphere, it would be a better idea to use a very small reactor and crack the CO2. The Isp is better.

ISP will suffer, but I doubt it will suffer enough to make chemical systems better have a better ISP.  In any case I believe the ISP will still be quite reasonable. 

Adding in chemical reactions and refinement makes the whole operation 100x more complicated, risky, and expensive.  It necessitates the need for supporting infrastructure for reacting, refining, and storage.  These mechanisms must work flawlessly over significant periods of time, 26 months in the case of Mars Direct, to prepare enough propellant for use.  Of course, this means more development and more mass that has to be launched.  It also means more  risk. 

With NTR you just land, and pump and CO2 into the propellant tanks, then your able to take off again.  No supporting infrastructure other than the pump are required. 

[strangequark]

ISP will suffer, but I doubt it will suffer enough to make chemical systems better have a better ISP.  In any case I believe the ISP will still be quite reasonable. 

Adding in chemical reactions and refinement makes the whole operation 100x more complicated, risky, and expensive.  It necessitates the need for supporting infrastructure for reacting, refining, and storage.  These mechanisms must work flawlessly over significant periods of time, 26 months in the case of Mars Direct, to prepare enough propellant for use.  Of course, this means more development and more mass that has to be launched.  It also means more  risk. 

With NTR you just land, and pump and CO2 into the propellant tanks, then your able to take off again.  No supporting infrastructure other than the pump are required. 

The chemical reactors involved are solid-state, and pretty robust as a result. They're also not that heavy, especially when you can drop a heavy NTR from your spacecraft. You can believe what you want on Isp, I have done the math.

[93143]

strangequark, how much do you know about Dumbo?  (Or did I ask you this at some point already?)

Just saw this. I don't know much about it at all. Do you have any interesting reading material?

Project report:

http://www.dunnspace.com/00339489.pdf

Later talk on the subject:

http://ntrs.nasa.gov/archive/nasa/casi.ntrs.nasa.gov/19920001882_1992001882.pdf

More recently:

http://ntrs.nasa.gov/archive/nasa/casi.ntrs.nasa.gov/20090008640_2009008062.pdf

What I like about Dumbo is mostly that it seems to offer the potential for very high T/W (at least compared with NERVA) even with tungsten/cermet core designs, thus ameliorating one of the major disadvantages of NTR as compared with chemical.

It has been alleged that Dumbo was downselected for "non-technical" reasons - one person said that the NERVA nozzle was mandated and didn't match Dumbo's needs, and that was that - but not everyone agrees on this subject...

Use anything other than H2 or NH3 and your Isp goes down the toilet, to where it is better to use the chemical systems. If you want to make use of the Mars atmosphere, it would be a better idea to use a very small reactor and crack the CO2. The Isp is better.

ISP will suffer, but I doubt it will suffer enough to make chemical systems better have a better ISP.  In any case I believe the ISP will still be quite reasonable.

I've explained this to you before.  The chemical system does indeed have a better Isp, mostly because the chamber temperature is higher.  Combustion generally produces temperatures well in excess of what an NTR core can reasonably be expected to withstand, and while regenerative cooling can protect a combustion chamber, an NTR core has to be hotter than the gas it's heating.  Hydrolox rockets gain an additional advantage over steam NTRs from the ability to optimize the mixture ratio, since the maximum Isp is obtained far from stoichiometric in this case.

Chemical also has a higher T/W, and is easier to protect against corrosion from high-temperature dissociation of steam or carbon dioxide.

The major potential advantage of your scheme is ease of propellant processing and storage.  Water is way easier to store than LO2/LH2 (though the latter is within the state of the art).  Unfortunately CO2 is not so easy, because you need to pressurize it to at least 75 psi or it doesn't have a liquid phase.

As for the claim that NTRs can run anything, well, chemical rocket engines can run anything too.  A specific NTR cannot run just anything; it has to be designed for the propellant it will be using.  CO2 and hydrogen, for instance, should never be used sequentially in the same engine because they require mutually exclusive design solutions, notably in regards to chemical compatibility, but also in regards to such parameters as nozzle throat size; it turns out that running water or CO2 through an NTR designed for hydrogen would not significantly increase the thrust, which means the core would have to be throttled way back.

[Using CEA, I get between 12% and 19% thrust increase for CO2 vs. H2 in a 3100 K core at 30 bar with a 250:1 nozzle in vacuum.  For reference, the Isp for CO2 was in the range of 259-285 s, probably below 270 s, compared with 999-1037 s for H2, or 270-303 s for a LOX/CO rocket with the same chamber pressure and expansion ratio - keep in mind that 3100 K is actually pretty optimistic for an NTR...]

Besides, how much flexibility do you really need?  There are only a few plausible ISRU propellant sources - water ice on the moon or Mars or Ceres, or Mercury, or Callisto, carbon dioxide on Mars or Venus, methane and water ice on Titan...  you could generate methane on Mars...  an RL-10 does hydrolox out of the box, and could easily be modified to burn methane/LOX - probably more easily than a pump-fed NTR could be converted from either H2 or H2O to CH4.  LOX/CO can't be that tough; even if you had to design an engine from scratch, it's still easier than designing a CO2 NTR from scratch.  What's left?

Adding in chemical reactions and refinement makes the whole operation 100x more complicated, risky, and expensive.

You're comparing the inclusion of a CO2 cracking operation with the inclusion of a nuclear rocket core capable of withstanding the corrosive action of the hot propellant, and completely ignoring the knock-on effects of using such a heavy, poorly-performing rocket as your main propulsion system.

Doesn't sound obvious to me at all.  I'm going to have to call in Akin's First Law...

[Patchouli]

The best argument I saw for running a NTR on CO2 was for a ballistic NTR Mars hopper Zubrin proposed which would have allowed exploring sites much farther and faster then possible with a rover alone.

CO2 has a poor ISP in an NTR engine but the advantage is the propellant can be acquired at low cost in energy about 85 KW per ton.
Source http://salotti.pagesperso-orange.fr/Zubrinvehicles.doc

This means in theory the hopper can replenish it's propellant in a few days to a little more then a week.
Assuming the NTR engine has specs similar to the TRW Triton which has a power generation mode of 50 KW x 24 hours =1200kwh
1200kwh/85kwh per ton =14.11 tons of propellant per day.

[93143]

I notice he handwaves the required engine T/W.  I'm dubious about particle bed rockets; Timberwind had problems with thermal instability.  Dumbo might help, but no one seems to know about it for some reason, and those who do don't seem to care...

Also, his Isp seems optimistic.  Using CEA, I get 236-250 s vacuum for 2800 K and 100:1 expansion.  (The lower number is for frozen composition, and the higher number is for equilibrium composition; the true value is generally somewhere in between.)  That's at 30 bar.  Dropping to 10 bar boosts it to 237-256 s.

But if you make the chamber pressure too low, the engine gets large and possibly too heavy (if it isn't already).  At 10 bar you're getting ~42,000 lb/ft² of throat area and an exit pressure similar to that of Mars' surface atmospheric pressure, which sounds like it might be reasonable, but at 1 bar you're only getting about a tenth of those numbers, which is not reasonable, and Ivac at 1 bar is still only 241-271 s, which doesn't seem likely to get up to 264 seconds in a real system.

Still, he makes a good point about the effective Isp bonus from not having to bring enough propellant for the return trip.  I don't know if it's worth developing a high-T/W CO2 NTR just for this application, but it does look at least mildly interesting...

Also, it occurs to me that it might be possible to make a flex-fuel NTR that can take either CO2 or water (I believe the chemical compatibility challenges are similar).  Water in an NTR doesn't perform as well as hydrolox, but it's a heck of a lot better than LOX/CO, and if the engine isn't unmanageably heavy...

[Sith]

Is this better than VASIMR?

http://fti.neep.wisc.edu/iecworkshop/PDF/TECHNICAL_TALKS/webber.pdf

[aero]

Is this better than VASIMR?

http://fti.neep.wisc.edu/iecworkshop/PDF/TECHNICAL_TALKS/webber.pdf

"Better?" IMO "better" should be qualified, maybe future potentially better or some similar caveat. VASIMR is advanced to the stage of flight or nearly flight qualified and the reference is a paper study. There is no reasonable way to compare the two, in terms of "better." A more specific question is in order.

[DarkenedOne]

Use anything other than H2 or NH3 and your Isp goes down the toilet, to where it is better to use the chemical systems. If you want to make use of the Mars atmosphere, it would be a better idea to use a very small reactor and crack the CO2. The Isp is better.

ISP will suffer, but I doubt it will suffer enough to make chemical systems better have a better ISP.  In any case I believe the ISP will still be quite reasonable.

I've explained this to you before.  The chemical system does indeed have a better Isp, mostly because the chamber temperature is higher.  Combustion generally produces temperatures well in excess of what an NTR core can reasonably be expected to withstand, and while regenerative cooling can protect a combustion chamber, an NTR core has to be hotter than the gas it's heating.  Hydrolox rockets gain an additional advantage over steam NTRs from the ability to optimize the mixture ratio, since the maximum Isp is obtained far from stoichiometric in this case.[/quote]

Wait a minute are you trying to say that chemical propulsion systems are better than NTR systems.  Last time I checked with hydrogen ISPs in the 900s range were achievable with NTR systems.  As far as chemical is concerned they cannot seem to break out of the 400s-600s range. 

Chemical also has a higher T/W, and is easier to protect against corrosion from high-temperature dissociation of steam or carbon dioxide.

No one doubts that chemical propulsion has the highest T/W ratio of any propulsion system, however that is most valuable when you are fighting gravity at lift off.

The major potential advantage of your scheme is ease of propellant processing and storage.  Water is way easier to store than LO2/LH2 (though the latter is within the state of the art).  Unfortunately CO2 is not so easy, because you need to pressurize it to at least 75 psi or it doesn't have a liquid phase.

You have seen dry ice makers right.  CO2 can easily to stored in the form of dry ice.  The cold conditions of mars make storing CO2 as dry ice quite feasible.

As for the claim that NTRs can run anything, well, chemical rocket engines can run anything too.  A specific NTR cannot run just anything; it has to be designed for the propellant it will be using.  CO2 and hydrogen, for instance, should never be used sequentially in the same engine because they require mutually exclusive design solutions, notably in regards to chemical compatibility, but also in regards to such parameters as nozzle throat size; it turns out that running water or CO2 through an NTR designed for hydrogen would not significantly increase the thrust, which means the core would have to be throttled way back.

First of all, there is a difference between not being able to run at all, and not being able to run optimally.  Most car engines can accept a variety of fuels including vegetable oil fuel, even though they are not optimized for that fuel.  M1 Abrams turbine engines are supposedly able to burn anything from jet fuel to diseal.  That does not mean it can burn all those fuels at optimal efficiency. 

Second of all, chemical rockets rely on their fuel as an energy source, whereas NTR does not.  With chemical rockets you have to store multiple fuels and react them in the right proportions. 

[Using CEA, I get between 12% and 19% thrust increase for CO2 vs. H2 in a 3100 K core at 30 bar with a 250:1 nozzle in vacuum.  For reference, the Isp for CO2 was in the range of 259-285 s, probably below 270 s, compared with 999-1037 s for H2, or 270-303 s for a LOX/CO rocket with the same chamber pressure and expansion ratio - keep in mind that 3100 K is actually pretty optimistic for an NTR...]

Well that would put the specific impulse of CO2 close to that of the lunar lander engines. 

Besides, how much flexibility do you really need?  There are only a few plausible ISRU propellant sources - water ice on the moon or Mars or Ceres, or Mercury, or Callisto, carbon dioxide on Mars or Venus, methane and water ice on Titan...  you could generate methane on Mars...  an RL-10 does hydrolox out of the box, and could easily be modified to burn methane/LOX - probably more easily than a pump-fed NTR could be converted from either H2 or H2O to CH4.  LOX/CO can't be that tough; even if you had to design an engine from scratch, it's still easier than designing a CO2 NTR from scratch.  What's left?

Its not the rocket that is the expensive part.  It is the infrastructure for reactor, refinement, and storage of those materials on the surface of Mars.

Adding in chemical reactions and refinement makes the whole operation 100x more complicated, risky, and expensive.

You're comparing the inclusion of a CO2 cracking operation with the inclusion of a nuclear rocket core capable of withstanding the corrosive action of the hot propellant, and completely ignoring the knock-on effects of using such a heavy, poorly-performing rocket as your main propulsion system.

Doesn't sound obvious to me at all.  I'm going to have to call in Akin's First Law...

It comes down to the values of explorers.  The prospect of a rocket engine that is able to take a vast variety of fuels from a large selection of ISRU sources is more valuable than a better performing rocket engine that requires highly refined fuel that needs to come from specific sources.  Versatility is more valuable than outright performance.

[93143]

Read my post again, carefully this time.  Read my subsequent one too.  Then see if you want to make any changes to yours.

[RanulfC]

I'm dubious about particle bed rockets; Timberwind had problems with thermal instability.  Dumbo might help, but no one seems to know about it for some reason, and those who do don't seem to care...

First let me state I'm a "DUMBO" fan so I "might" be a bit biased

I'll also note that the LANL library seems to be "available" again so here are the original DUMBO files:
http://lib-www.lanl.gov/la-pubs/00339489.pdf

http://lib-www.lanl.gov/la-pubs/00350007.pdf

Timberwind, and from what I understand most particle-bed NTR concepts ended up with thermal issues as well as thrust-to-weight/isp far lower than predicted most probably due to turbulent flow within the particle-bed (which I note in the second reference Bill Kirk dismiss as unimportant) causing high-drag, and uneven flow within and through the bed.

IIRC the final concepts for the PDR-NTR involved spinning the reactor to force the propellant flow to flow evenly through the bed which helped ISP but killed any hoped for T/W advantage.

"I" am not a nuclear engineer, but I'm rather "disturbed" by Mr. Kirk's rather off-hand dismissal of the effect of turbulent flow in a reactor, especially in regards to one being used for rocket propulsion.

Turbulent flow is known to cause drag in high-speed liquid/surface interactions, especially in something like an NTR. I know there where questions on the actual mechanical and neutronic stability of the DUMBO "washer" design but it has been my understanding that both mechanical function, including design, fabrication and testing, as well as criticallity of the design were tested and found to be achievable.

NERVA had turbulent flow due to the core-chamber size required for self-supporting graphite-elements. This dictated the minimum length of the elements since turbulent flow does not allow even heating of the propellant in a predictable manner. Laminar flow on the other hand allows a predicatable and consistant heating curve to be used allowing much shorter passages and a higher heating flux within the passages.

Questions have been made regarding the fact that the propellant flow "folded" through two 90-degree turns in the DUMBO design. Such "turns" of course are avoided as much as possible in flow-design because each turn robs the flow of speed. However in the DUMBO design this helps stabilize the pressure on the heat-exchanger and assists in the Dynamic Cooling allowing higher core temperatures to be achieved. Nice "side-effect" of this was a higher chamber pressure without having to have a larger turbo-pump system.

Donald Kingsbury, in a December 1978 Analog article gave an overview of the entire DUMBO program, and as I understand it actually talked to several of the original engineers involved with the project. That's where it's mentioned that a fundamental reason the DUMBO got cancled was the incompatability of the NERVA nozzle and the Dynamic Cooling needs of the DUMBO engine.

The Grooved Ring reactor concept has some more information in here:
http://my.fit.edu/~dkirk/

In this presentation:
http://my.fit.edu/~dkirk/NTP%20Research%20Summary.ppt

Randy