How Come Nuclear Saltwater Rocket Engines Do Not Melt?

[Spacescifi]

NSWR are the only theoretical realistic torchdrive rockets that have ever been designed.

It is my understanding that the energy release is equal to a constantly exploding nuke.... only directed as a rocket plume.

This reaction somehow does not occur until what?

Once it's already leaving the nozzle?

[RCgothic]

"Constantly Exploding Nuke" is a very poor descriptive term.

It's a nuclear reactor. We have plenty of nuclear reactors that deal with the heat produced by energy released.

Nuclear saltwater rocket engines are a type of reactor where the fuel is suspended in the propellant and made to undergo a quick reaction in a chamber before the nozzle throat. The heat of the nuclear energy released is carried away in the propellant exhaust.

[Shpaget]

When dealing with fission, neutron management is the name of the game. This is achieved by usage of different materials to absorb, reflect, moderate, propagate etc. the neutron to achieve the desired effect. Also, geometry is a big factor so by properly designing the reactor, and supplying the correct amount of fission fuel, in the correct geometry with correct combination of support materials you can achieve the desired result.

In case of NSWR, this specific combination is achieved only for a relatively small amount of fuel/propellant in a specific part of the engine that is designed to withstand these energies.

[AckSed]

This is true. However, I pessimistically imagine how a hard start, pump cavitation or fluid hammer might go, and I shudder.

[mikegarrison]

TRL2, at best.

[farmerben]

What do you call electrothermal propulsion?  You could run a kilopower reactor as primarily an electric generator then run tungsten heating elements with hydrogen propellant.  It would still be combined cycle, as the waste heat from the reactor would warm the cryogenic propellant into a warm gas before running it though the primary heating element in the nozzle.  Would some sort of ion drive be better?

[darthgently]

16 minutes ago, farmerben said:

What do you call electrothermal propulsion?  You could run a kilopower reactor as primarily an electric generator then run tungsten heating elements with hydrogen propellant.  It would still be combined cycle, as the waste heat from the reactor would warm the cryogenic propellant into a warm gas before running it though the primary heating element in the nozzle.  Would some sort of ion drive be better?

Depending on temperatures, that could be (nearly?) an ion engine.  Sort of, right?  Not sure, but would plasma be involved?

[sevenperforce]

6 minutes ago, farmerben said:

What do you call electrothermal propulsion?  You could run a kilopower reactor as primarily an electric generator then run tungsten heating elements with hydrogen propellant.  It would still be combined cycle, as the waste heat from the reactor would warm the cryogenic propellant into a warm gas before running it though the primary heating element in the nozzle.  Would some sort of ion drive be better?

Any nuclear reactor needs two things: a working fluid and a heat sink. Otherwise, it can't generate power. To spin a turbine, you've got to move the heat out of the reactor and into a fluid, and then you've got to allow that fluid to expand through the turbine, and then you've got to do something with that fluid. Usually that means condensing the fluid back down by cooling it with some outside heat sink so you can run it back through the loop.

Space doesn't have any readily-available heat sinks, though; any heat rejection has to be done via radiators, and that is limited by surface area (not to mention mass). If you're in space and you're using liquid hydrogen as a propellant already, then you should just be running the liquid hydrogen through the reactor and using it as your working fluid, then venting it through the engine as your engine exhaust. If you're doing that, though, you should just skip the turbine altogether and boom, you've got an ordinary nuclear thermal rocket.

Is there any use for adding something further? Maybe. The limiting factor on the efficiency of a nuclear thermal rocket is typically the temperature of the reactor. You don't want it to melt, after all. The temperature inside the RS-25 is actually higher than the temperature inside a nuclear thermal rocket for this reason; the NTR is just more efficient because it doesn't need to use any of that heavy oxygen in the reaction. In theory, I could imagine a design in which the waste heat from an NTR is pulled off by a propellant preheating loop which turns a turbine before sending the preheated propellant into the engine, and that turbine somehow sends electricity into the propellant downrange of the reactor, heating it up beyond the melting point of the reactor before it is exhausted. But that's a LOT of complexity for not a lot of utility. 

Just now, darthgently said:

Depending on temperatures, that could be (nearly?) an ion engine.  Sort of, right?  Not sure, but would plasma be involved?

If you can get up to around 5000 K, most of the hydrogen molecules will be disassociated into neutral hydrogen atoms, cutting the exhaust molecular weight in half and thus increasing exhaust speed and specific impulse. This is already used in other situations; the RS-25 actually runs extra fuel-rich by adding more hydrogen than is required to decrease the average molecular weight and thus increase specific impulse even though this drives down temperature a little.

Up over 7000 K you get ionization of the individual hydrogen atoms. You can get near-complete ionization north of 10,000 K.

A true ion engine, however, operates by electromagnetically accelerating charged particles. Because the acceleration is electromagnetic rather than thermal, an ion engine can "skip" the thermodynamic limitations of the Carnot cycle and achieve much higher particle energies. There's no real gain from having thermally ionized exhaust in a thermal rocket.

[farmerben]

https://en.wikipedia.org/wiki/Kilopower

The kilo power reactor uses sodium as a working fluid, and radiates heat into space.

[kerbiloid]

33 minutes ago, farmerben said:

and radiates heat into space.

and has radiator panels which radiate heat in space, and uses LNa as a coolant to transfer the heat from the reactor to the panels.

The reactor itself is too small and dense to be an effective radiator.

[sevenperforce]

1 hour ago, farmerben said:

https://en.wikipedia.org/wiki/Kilopower

The kilo power reactor uses sodium as a working fluid, and radiates heat into space.

Certainly, but again that's to get around the whole "lack of open-cycle working fluid" problem. If your goal is to use hydrogen as propellant, then there's no reason to use the reactor to heat the sodium to convert to electricity to run through resistors to heat tungsten to heat the hydrogen; just heat the hydrogen with the reactor directly.

The conversion of heat to electricity and back to heat is extremely inefficient.

[farmerben]

If you're running a combined cycle power plant into a mostly thermal engine, the inefficiency doesn't mean much.   The key is whether a pure tungsten nozzle can operate at higher temperatures.  Or if you can get some ion thrust by running cathode ray tubes out parrallel to your plasma nozzle.

[StrandedonEarth]

2 hours ago, farmerben said:

If you're running a combined cycle power plant into a mostly thermal engine, the inefficiency doesn't mean much.   The key is whether a pure tungsten nozzle can operate at higher temperatures.  Or if you can get some ion thrust by running cathode ray tubes out parrallel to your plasma nozzle.

Yeah, I was thining it would make more sense as combined cycle, since spacecraft need power. But would it not be more efficient to just keep heating the plasma with microwaves, like VASIMR?

[Terwin]

15 hours ago, StrandedonEarth said:

Yeah, I was thining it would make more sense as combined cycle, since spacecraft need power. But would it not be more efficient to just keep heating the plasma with microwaves, like VASIMR?

While it may make sense to supplement the heating with microwaves if your nozzle can handle higher temps than your reactor, converting from heat to electricity to microwaves to heat cannot be as efficient as just conducting the heat directly.

So heating up your reaction mass to the reactor temp while simultaneously using it to cool the reactor is much more efficient.

[magnemoe]

2 hours ago, Terwin said:

While it may make sense to supplement the heating with microwaves if your nozzle can handle higher temps than your reactor, converting from heat to electricity to microwaves to heat cannot be as efficient as just conducting the heat directly.

So heating up your reaction mass to the reactor temp while simultaneously using it to cool the reactor is much more efficient.

I agree, now this could let you increase reaction mass velocity I suspect adding turbines and radiators to generate meaningful power to boost this is not more efficient than simply carry more reaction mass. You could focus the microwaves to only heat the center of the escaping reaction mass flow making it hotter than the nozzle could handle.  Using the reactor on low power to generate power to run the ship is much easier as its just say 20 kw who is nothing compared to the engine. 

As for nuclear saltwater engines. They an nuclear meltdown as an rocket engine, or you want orion pulse nuclear to look very safe and down to earth. One issue is settling of the fuel at any point and it might go critical who will result in an loss of ship. 

Orion pulse nuclear only issue is that fails to pusher plate or shock absorbers will result of an loss of engine who is also an loss of ship in practical terms but more in the way an Apollo mission would fail if 3rd stage RUD during lunar injection.
Granted this would be worse if you went on a 1 year trip to Saturn as you would be on an solar escape trajectory but you could send an rescue ship. 

Tangent: An Issue I have in KSP 2 is that orbital paths break down if your trajectory is an solar escape one. Done Jool missions in KSP 1 with an solar escape trajectory  
Ran into this trying to return from Duna with 8 km/s dV and no reason to keep any, ship class was obsolete. 

[farmerben]

If the reactor core operates at 700 degrees, and the nozzle operates at 4000 degrees then it makes sense.

Here is your core for kilopower

 

  

[sevenperforce]

17 hours ago, farmerben said:

If the reactor core operates at 700 degrees, and the nozzle operates at 4000 degrees then it makes sense.

I'm not sure if you can get enough power.

Since you were talking about tungsten before and tungsten melts at 3,400°C, I'm assuming you mean 4000°F. This paper has a lot of analysis on an electric resistance-based rocket engine. Figure 2 gives something on the order of 690 seconds specific impulse for hydrogen heated to that temperature. We know the biggest kilopower reactor would have generated ten kilowatts; ten kilowatts divided by ~6800 m/s is a thrust of 1.47 Newtons. That's not much.