Expansion-Condensation Closed-Cycle Nuclear Turbine Generator

[MatterBeam]

Hi. I've been thinking about this configuration of a nuclear reactor, a set of nozzles and a turbine generator.

The Expansion-Condensation Closed-Cycle Nuclear Turbine Generator.

A nuclear reactor heats a coolant such as nitrogen to high temperatures and pressures, like a nuclear thermal rocket. Efficiency 90%. The coolant is expanded through a nozzle to moderate temperatures, low pressures and high velocity, like a propulsive nozzle. A multi-stage turbine converts the gas's energy into mechanical energy. This slows down the gas. Efficiency over 80%. Electric generator converts turbine motion into electricity, efficiency 90%. Second expansion nozzle condenses and liquefies the gas. This creates near-vacuum pressures, reducing back-pressure at the first nozzle. The liquid coolant is collected and pumped back into the reactor. Overall efficiency is 65% to 80%. Nuclear reactor can have GW/ton specific power. Turbine can have several dozen MW per ton specific power, coupled with electric generator.

Can this work? Will it compare favourably to Stirling engines and thermocouples?

Is the concept sound? Is there something I am missing as to why direct-cycle nuclear turbines have not been used for power generation?

In space, specific power is very important, especially when several MW of power is needed for electric rockets, such as VASIMR. So far, concepts have described thermocouples or Stirling engines for electric power. Can this concept compete with current designs?

After discussion:

-Expanding the gas to hypersonic velocities is not desireable. Using a very-high-pressure, sub-Mach flow is much more efficient.
-A radiator will be needed at the end of the Brayton cycle to remove the residual waste heat. At low temperatures, this radiator will be huge. Therefore a much lower temperature difference is needed between the reactor and the second nozzle, to be compensated for with a large pressure difference... much like actual nuclear reactor.
-A similar concept has been tested.
-Overall, this concept will be more effective due to the direct-cycle involving the nuclear reactor shooting directly into the turbine, rather than because of the possible temperature difference.

Edited March 9, 2017 by MatterBeam

[Tex_NL]

A system like that will work. Because that is pretty much exactly how a common nuclear power plant works. Except yours is a single loop and most modern reactors use a double loop for safety.

Edited March 9, 2017 by Tex_NL

[Elthy]

You have a carnot cycle here. In theory you can get the efficency that high if you can dump the waste heat fast enough, but thats a big issue in space, radiators are heavy...

[magnemoe]

Yes, benefit of double loop is that your turbine and don't get radioactive, you will have to do service on it from time to time.
You are also more free to select how your condenser work. 
And you can keep all the radioactive water inside the containment structure with short pipes making an leak less likely and less of an problem. 

[MatterBeam]

6 minutes ago, Tex_NL said:

A system like that will work. Because that is pretty much exactly have a common nuclear power plant works. Except yours is a single loop and most modern reactors use a double loop for safety.

I thought this would be slightly different. Modern reactors create a loop of high temperature isobaric coolant that loses heat through an exchanger. It is limited by the temperature limits of its various components.

The Expansion part allows for a lot more energy to be put into the gas, while only subjecting the blades to moderate temperatures. The condensation at the end allows for us to 'cheat' the brayton cycle by allowing an arbitrarily high pressure drop across the turbine.

2 minutes ago, Elthy said:

You have a carnot cycle here. In theory you can get the efficency that high if you can dump the waste heat fast enough, but thats a big issue in space, radiators are heavy...

I think it is a brayton cycle. At 90% or so efficiency at each step, the waste heat to be removed is quite low.

Just now, magnemoe said:

Yes, benefit of double loop is that your turbine and don't get radioactive, you will have to do service on it from time to time.
You are also more free to select how your condenser work. 
And you can keep all the radioactive water inside the containment structure with short pipes making an leak less likely and less of an problem. 

In space, mass is at a premium. I think a single-loop reactor that operates at high reactor temperatures and yet has low temperatures at the turbine end might end up having very high specific power (kW/kg).

[magnemoe]

4 minutes ago, MatterBeam said:

I think it is a brayton cycle. At 90% or so efficiency at each step, the waste heat to be removed is quite low.

In space, mass is at a premium. I think a single-loop reactor that operates at high reactor temperatures and yet has low temperatures at the turbine end might end up having very high specific power (kW/kg).

In space you can go single loop I agree, its lighter, you don't have environmental issues and will not do much service. For an large installation I would still run the cooling loop separate. 

[TheSaint]

It's a closed-cycle gas turbine, they've actually already run one with a nuclear reactor, ML-1. It's a sound concept, although I think your projected efficiencies are a little high. I also think you may want to look at reliability as well. A gas-turbine will be, by its very nature, less reliable than, say, a Stirling engine, and if it were to fail it would fail far more spectacularly.

[MatterBeam]

3 minutes ago, TheSaint said:

It's a closed-cycle gas turbine, they've actually already run one with a nuclear reactor, ML-1. It's a sound concept, although I think your projected efficiencies are a little high. I also think you may want to look at reliability as well. A gas-turbine will be, by its very nature, less reliable than, say, a Stirling engine, and if it were to fail it would fail far more spectacularly.

Ha! They even chose the same coolant (nitrogen) as I did! Thanks for the find. 

Well, regarding reliability, I have started discussing this design elsewhere, and found that a hypersonic gas flow at the first nozzle was too optimistic and could cause huge drag losses. Mach 0.85 is the optimal gas flow velocity, because then we can use ultra-reliable, ultra-efficient turbofan designs from modern aviation.

Turbofans run for thousands of hours without a hitch.

[shynung]

44 minutes ago, MatterBeam said:

we can use ultra-reliable, ultra-efficient turbofan designs from modern aviation

Kind of a nitpick here. Aviation turbofans derive most of their thrust from the fan, not the engine.

On most common current-gen aviation turbojets, about 1/6 of the air entering the engine ever encounters the combustion chamber, while the rest goes straight to the exhaust nozzle; a bypass ratio of 5:1. Some newer engines such as some models of the PW1000G has bypass ratios that goes up to 12.5:1. That's how they get their efficiency: only a small amount of air entering the engine actually gets combusted, the rest is simply pushed backwards by a big fan.

Power generating turbines differ from aviation ones, mostly because they don't have to bother with generating aerodynamic thrust and size/mass limitations, except maybe for ships.

This is a steam turbine from power plant. High-pressure steam would enter from the middle, where the red-shirt person is standing, and exit through both axial ends. This reduces thrust loads on the bearings.

Edited March 9, 2017 by shynung

[MatterBeam]

After discussion:

-Expanding the gas to hypersonic velocities is not desireable. Using a very-high-pressure, sub-Mach flow is much more efficient.
-A radiator will be needed at the end of the Brayton cycle to remove the residual waste heat. At low temperatures, this radiator will be huge. Therefore a much lower temperature difference is needed between the reactor and the second nozzle, to be compensated for with a large pressure difference... much like actual nuclear reactor.
-A similar concept has been tested.
-Overall, this concept will be more effective due to the direct-cycle involving the nuclear reactor shooting directly into the turbine, rather than because of the possible temperature difference.