39 days to Mars possible now with nuclear-powered VASIMR.

[Exoscientist]

That's not how it works. All heat engines are limited by the Carnot efficiency of the cycle, given by ETA=1-(TSink/TSource). This is a hard law of thermodynamics and there is no way around it, save by avoiding using a heat engine altogether (fuel cells, for example, can achieve higher efficiencies).

You might be thinking of the propulsive efficiency of a rocket engine, which is not the same as the thermal efficiency. Propulsive efficiency can, in theory, approach 100%.

The "efficiency" you see quoted for turbopumps isn't for their ability to convert heat in the exhaust into mechanical work. It is what is known as the "isentropic efficiency", which is how they perform their pressurisation process compared to a thermodynamically ideal model.

For a nuclear reactor operating at, say, 1500K (which is extremely high, and will require a very high flow rate of coolant, which costs mass), you can easily calculate the Carnot efficiency of the power conversion cycle for a range of heat sink temperatures. For a heat sink of 500K, the Carnot efficiency is 67%. For a heat sink of 300K, the Carnot efficiency is 80%.

Now, the Carnot efficiency isn't a limit you can ever actually reach, only approach. And the closer to the Carnot efficiency you get, the more cumbersome and heavy your equipment becomes. The best that is usually achieved in practice is with a combined cycle gas turbine, which uses a gas turbine running on the Brayton cycle, with a bottoming Rankine cycle powered by the Brayton cycle's exhaust. These achieve thermal efficiencies of about 50% from a 1400K heat source, or in other words, about 60% of the Carnot efficiency.

...

The "turbine efficiency" I mentioned about the SSME turbopumps is indeed the isentropic efficiency. However, by exhausting the turbopumps to near vacuum you can get the ideal Carnot efficiency above 90%. Therefore an isentropic efficiency of 80% still gives an overall efficiency above 70%.

Also, the question of the short lifetime of the space nuclear reactors when run at full power has a couple of solutions. One way is to replace the nuclear fuel canisters when they burn out. Another is to just run the reactors at low power, which is known to extend their lifetime.

After running the numbers for plasma propulsion at short flight times to Mars, I came to the conclusion you can make the total missions size smaller than with the usual slow travel times. The reason for this unexpected conclusion is the short travel time allows a much smaller habitat for the transit, and therefore smaller propellant load:

Nuclear powered VASIMR and plasma propulsion doable now, Page 2.

http://exoscientist.blogspot.com/2015/10/nuclear-powered-vasimr-and-plasma.html

Bob Clark

[peadar1987]

The "turbine efficiency" I mentioned about the SSME turbopumps is indeed the isentropic efficiency. However, by exhausting the turbopumps to near vacuum you can get the ideal Carnot efficiency above 90%. Therefore an isentropic efficiency of 80% still gives an overall efficiency above 70%.

Also, the question of the short lifetime of the space nuclear reactors when run at full power has a couple of solutions. One way is to replace the nuclear fuel canisters when they burn out. Another is to just run the reactors at low power, which is known to extend their lifetime.

After running the numbers for plasma propulsion at short flight times to Mars, I came to the conclusion you can make the total missions size smaller than with the usual slow travel times. The reason for this unexpected conclusion is the short travel time allows a much smaller habitat for the transit, and therefore smaller propellant load:

Nuclear powered VASIMR and plasma propulsion doable now, Page 2.

http://exoscientist.blogspot.com/2015/10/nuclear-powered-vasimr-and-plasma.html

Bob Clark

Hi Bob,

I think you're starting to get it. You can indeed get far higher thermal efficiencies by exhausting through a high-expansion nozzle into vacuum (this cools the exhaust down big-time, and increases the Carnot efficiency). However, this still isn't practical for power generation, as in a power generation cycle, you don't throw away your working fluid, you recirculate it. If you expand it to extremely low pressures and temperatures, you have no way of getting rid of the heat and making the working fluid ready to be recirculated.

The Rankine Cycle is the most commonly used thermal cycles. The others are broadly similar. It consists of:

-Pumping liquid to high pressure -> Boiling the high-pressure liquid using a heat source -> Expanding the resulting high pressure vapour to produce power -> Condensing the resulting cool, low pressure vapour to a liquid, ready for the pump again.

If you want a closed-loop system (and you do, because otherwise you might as well just use a rocket engine and skip the extra step of generating electricity), you need that condensation/heat rejection stage.

If you expand your working fluid to near a vacuum, it's going to be cold. Very cold. The radiators needed to shift the necessary megawatts of heat from such a cold working fluid stream are going to be ridiculously large and heavy.

My concerns over the lifetime of the reactor aren't really fuel-based, they're more around the survivability of the components. It's pretty trivial to make nuclear fuel that will undergo fission for an extended period, and even have high burnup. What is harder is making heat exchangers, pressure vessels, expanders, etc. that will operate at high temperature for extended periods of time without problem.

The reason the burn time on the NTR you linked is so low is because, after that time, the nozzle will be soft, the moderator cracking, the turbopumps nearing the end of their fatigue life, the heat exchanger tubes beginning to warp. It's not because the nuclear fuel is no longer able to undergo fission. Making things that perform at high temperature for extended periods of time is very difficult, which is why earthbound power plants operate at only a few hundred degrees. It would be easy to get over 2,000 degrees out of a fossil fuel plant, but harnessing that temperature is very difficult.

[Kibble]

The title, "39 days to Mars possible now with nuclear-powered VASIMR." is highly misleading as there's no clear cut demonstration that it actually is possible. For all we know the power requirements are going to remain unrealistic for at least another decade.

Agreed! This should be renamed "39 days to Mars might be possible, some distant day, and it might be using something like VASIMR"

[RuBisCO]

A Rankine cycle simply would not do in space. You would need a Closed Brayton cycle, that means most likely very compressed Helium is the working fluid, maybe Helium-Xenon mix if directly helium cooled reactor, or better yet super critical CO2 if using a molten salt reactor. The reactor would need to operate at above 800°C to achieved efficiency above 50%, but would lead to much smaller turbines, as well as smaller radiators because the outlet temperature would be much higher and easier to radiate heat away.

here is a paper on the idea of supercritical Brayon cycle in space.

http://web.mit.edu/rsi/www/pdfs/papers/2005/2005-ianr.pdf

Edited October 4, 2015 by RuBisCO

[peadar1987]

A Rankine cycle simply would not do in space. You would need a Closed Brayton cycle, that means most likely very compressed Helium is the working fluid, maybe Helium-Xenon mix if directly helium cooled reactor, or better yet super critical CO2 if using a molten salt reactor. The reactor would need to operate at above 800°C to achieved efficiency above 50%, but would lead to much smaller turbines, as well as smaller radiators because the outlet temperature would be much higher and easier to radiate heat away.

http://beforeitsnews.com/mediadrop/uploads/2013/39/c5fe676ffd6ff32cb2390fd82b33258e6da57bb5.png

here is a paper on the idea of supercritical Brayon cycle in space.

http://web.mit.edu/rsi/www/pdfs/papers/2005/2005-ianr.pdf

[RuBisCO]

peadar1987,

Well how much? It is not like this has not be studied before:

http://ntrs.nasa.gov/archive/nasa/casi.ntrs.nasa.gov/19930005293.pdf

Now that is a 1993 study, just brayton, not supercritical brayton, and its coming up with masses between 56-43 tons for a complete 15 MWe powerplant, radiators and all.

[NuclearNut]

peadar1987,

Well how much? It is not like this has not be studied before:

http://ntrs.nasa.gov/archive/nasa/casi.ntrs.nasa.gov/19930005293.pdf

Now that is a 1993 study, just brayton, not supercritical brayton, and its coming up with masses between 56-43 tons for a complete 15 MWe powerplant, radiators and all.

Seems to be a paper reactor, not an actual reactor in existence today. I would predict, with the little knowledge on the subject matter that I have, that the reactor will weigh quite a bit more and have several design issues that are experienced after operation begins. Current reactor designs or designs for anything for that matter tend to absolutely refuse to go gracefully from drawing board to design and often appear to be a compromise with the design goals and reality. Or at least that is how I see it.

[RuBisCO]

Seems to be a paper reactor, not an actual reactor in existence today.

No... you don't say, I'm so surprised!

I would predict, with the little knowledge on the subject matter that I have, that the reactor will weigh quite a bit more and have several design issues that are experienced after operation begins.

or it is an already outmodded design and a smaller higher w/kg ratio reactor could be built, the super-critical braytons for example.

Current reactor designs or designs for anything for that matter tend to absolutely refuse to go gracefully from drawing board to design and often appear to be a compromise with the design goals and reality. Or at least that is how I see it.

yeah sure, please make a study on why it can't be done.

[peadar1987]

peadar1987,

Well how much? It is not like this has not be studied before:

http://ntrs.nasa.gov/archive/nasa/casi.ntrs.nasa.gov/19930005293.pdf

Now that is a 1993 study, just brayton, not supercritical brayton, and its coming up with masses between 56-43 tons for a complete 15 MWe powerplant, radiators and all.

Supercritical operation isn't necessarily going to increase the efficiency (it tends to increase the pump work significantly), its main benefit is operating at an increased pressure ratio, not because of some intrinsic property of supercriticality. Edit: You're right in that it will tend to give you smaller components though

The cycle you linked rejects heat at 433K, for a heat source temperature of 2000K, giving it a theoretical Carnot efficiency of 75%. I'd put actual thermal efficiency closer to the 45-50% mark. And it only achieves ~0.3 kg/kW.

[RuBisCO]

Supercritical operation isn't necessarily going to increase the efficiency (it tends to increase the pump work significantly), its main benefit is operating at an increased pressure ratio, not because of some intrinsic property of supercriticality. Edit: You're right in that it will tend to give you smaller components though.

Didn't say it would.

The cycle you linked rejects heat at 433K, for a heat source temperature of 2000K, giving it a theoretical Carnot efficiency of 75%. I'd put actual thermal efficiency closer to the 45-50% mark. And it only achieves ~0.3 kg/kW.

Which link?

[peadar1987]

Didn't say it would.

Which link?

There was only one link in your post!

[RuBisCO]

There was only one link in your post!

There was a previous post about the supercritical brayton.

The NASA link on the other hand I don't see where you getting those numbers. For example the chart on page 7 states heat source temp of 2000 K, but efficiencies of 45%-27% depending on design option, and mass to power ratios of 3.74 - 2.9 kg/kw, in fact a ratio of ~0.3 kg/kW would be phenomenal, a 15 MWe power plant would weigh 4.5 tons! These reactor designs are doing 57-44 tons for 15 MWe.

Edited October 5, 2015 by RuBisCO

[peadar1987]

There was a previous post about the supercritical brayton.

The NASA link on the other hand I don't see where you getting those numbers. For example the chart on page 7 states heat source temp of 2000 K, but efficiencies of 45%-27% depending on design option, and mass to power ratios of 3.74 - 2.9 kg/kw, in fact a ratio of ~0.3 kg/kW would be phenomenal, a 15 MWe power plant would weigh 4.5 tons! These reactor designs are doing 57-44 tons for 15 MWe.

:facepalm:

That should of course be 0.3kW/kg.

[RuBisCO]

Ok lets say we have a 250 ton space craft with 15 MW, 50 tons power plant. If we are using Dual-Stage 4-Grid at 100 kW/N that is 150 N of thrust or an acceleration of 6 mm/s^2, 2.16 m/s^2 per hour, and 51.84 m?s^2 per day, so it would take 19 days to do 1 km/s of deltaV. So certainly this option would not achieve 39 days to mars.

Although with 30 tons of propellant it could do 24.2 Km/s of DeltaV, but of course would take 467 days of continuous thrusting. Now it would need a minimum of 11.42 km/s to go from LEO to LMO and back to LEO again, twice that because not Hoffman transfer. It would take 123 days just to crawl out of earth orbit, 40 days to speed up to intercept mars and slow down again (not including cruise time), 56 days to crawl into LMO. Even if crews fly up to meet the ship in HEO or L2 that certainly at least 3 months to mars and having to go directly from mars intercept to landing, leaving the mother ship to slow down unmanned into LMO.

Edited October 6, 2015 by RuBisCO

[peadar1987]

Ok lets say we have a 250 ton space craft with 15 MW, 50 tons power plant. If we are using Dual-Stage 4-Grid at 100 kW/N that is 150 N of thrust or an acceleration of 6 mm/s^2, 2.16 m/s^2 per hour, and 51.84 m?s^2 per day, so it would take 19 days to do 1 km/s of deltaV. So certainly this option would not achieve 39 days to mars.

Although with 30 tons of propellant it could do 24.2 Km/s of DeltaV, but of coursed would take 467 days of continuous thrusting. Now it would need a minimum of 11.42 km/s to go from LEO to LMO and back to LEO again, twice that because not Hoffman transfer. It would take 123 days just to crawl out of earth orbit, 40 days to speed up to intercept mars and slow down again (not including cruise time), 56 days to crawl into LMO. Even if crews fly up to meet the ship in HEO or L2 that certainly at least 3 months to mars and having to go directly from mars intercept to landing, leaving the mother ship to slow down unmanned into LMO.

Yep, you need at least 1kW/kg for VASIMR to be an option, according to even its strongest proponents. That's at the very edge of our short to medium term capabilities with nuclear fission. It's definitely not possible now, or even with existing technology.

[NuclearNut]

No... you don't say, I'm so surprised!

To form a distinction, and a clear distinction between the stuff that has already been used or even prototyped is really important.

or it is an already outmodded design and a smaller higher w/kg ratio reactor could be built, the super-critical braytons for example.

You mean a Supercritical Water Reactor? Those have several issues that I know of. First and foremost is fuel rod durability in an truly extreme environment, this problem cannot be understated as almost all experience in power reactors is with zircaloy fuel rods or ceramic pellets held in zircaloy fuel cladding. From what I understand, and has been made abundantly clear at the Gen IV international forum, that a whole new fuel element design would be needed, down to the alloy involved. This means massive experimental reactors, test rigs, and other such things that will take years (and must take years to simulate a fuel cycle) to complete before a reactor is ready for a manned mission.

On earth if you have to replace the fuel rods or make repairs on your new reactor it is considered a minor design flaw, in space you have the serious possibility of not having the crew live through the power loss induced by such a problem.

yeah sure, please make a study on why it can't be done.

I am not saying that it cannot be done, rather saying that most studies are quite optimistic about what can be done, and in that lies a problem, primarily that of making things like designing a whole new reactor look easy, simple to build, and running exactly as planned. Looking back at the design of nuclear reactors for submarines one tends to find that neither would describe it.

Anyway, as peadar1987 said, the reactor has a crap (for VASMIR) power to weight ratio, so it is kind of a dead end in that direction too. From what I understand what would be best is a NTR, yes, not as efficient as VASMIR as far as fuel goes, however you could use the reactor for heat, power, and an engine, thus allowing the possibility of the thermochemical production of hydrogen from water on mars after you "cook out" the water from the regolith using reactor heat. And that would be far easier primarily because we already know how to build a NTR that would function well enough with the technology that we have now, and we already have built test rigs for it.

[RuBisCO]

To form a distinction, and a clear distinction between the stuff that has already been used or even prototyped is really important.

A distinction for whom?

You mean a Supercritical Water Reactor?

No, read my first link.

- - - Updated - - -

Yep, you need at least 1kW/kg for VASIMR to be an option, according to even its strongest proponents. That's at the very edge of our short to medium term capabilities with nuclear fission. It's definitely not possible now, or even with existing technology.

Well we could do it, just not in 39 days.

[Exoscientist]

Agreed! This should be renamed "39 days to Mars might be possible, some distant day, and it might be using something like VASIMR"

The title arises from the doubt raised by some such as Robert Zubrin that the required power systems would ever be available at the needed lightweight to make the 39 day to Mars transit time using VASIMR plasma propulsion viable:

The VASIMR Hoax by Robert Zubrin  July 13, 2011.

http://spacenews.com/vasimr-hoax/

The VASIMR would need 1,000 watts per kilo for its power source. But, argues Zubrin, this is a hundred times better in power-to-weight ratio than has been achieved so far for space nuclear power. He therefore argues this is a far off technology development.

However, I argue the thermal power per weight put out by space nuclear reactors already is well above this. So what is really needed is more efficient and lightweight conversion of the thermal to electric power. I'm suggesting such lightweight conversion also currently exists. Then we already have the technical means to build such a plasma powered Mars spacecraft for the proposed short travel time. It is not some far off technology.

Bob Clark

Edited October 9, 2015 by Exoscientist

[Exoscientist]

A Rankine cycle simply would not do in space. You would need a Closed Brayton cycle, that means most likely very compressed Helium is the working fluid, maybe Helium-Xenon mix if directly helium cooled reactor, or better yet super critical CO2 if using a molten salt reactor. The reactor would need to operate at above 800°C to achieved efficiency above 50%, but would lead to much smaller turbines, as well as smaller radiators because the outlet temperature would be much higher and easier to radiate heat away.

http://beforeitsnews.com/mediadrop/uploads/2013/39/c5fe676ffd6ff32cb2390fd82b33258e6da57bb5.png

here is a paper on the idea of supercritical Brayon cycle in space.

http://web.mit.edu/rsi/www/pdfs/papers/2005/2005-ianr.pdf

Thanks for the link, and the graphic. I noticed the supercritical turbine also converted a decent amount of the thermal to electrical, important for minimizing radiator size.

Bob Clark