Throttleable NERVAs

[Whirligig Girl]

Suppose you wanted to set up a high-efficiency Lunar Fuel Depot, so you want to use a Nuclear Thermal Rocket. Is it practical to throttle a NERVA/NTR? If so would it be practical to use a NERVA in a LEO-Lunar Surface Rocket to ferry fuel and supplies mined from the Moon on a regular basis?

[K^2]

There is really no reason to use NERVA for that. It's absurdly expensive for the task. If you are going to be sending a lot of stuff from Moon to Earth or LEO, you should build a magrail on the surface. Launches to Earth can be done directly from surface, requiring only minor corrections along the way. Launches to LEO would be better done by using a VASMIR tug. In this scenario, cargo is launched by mag rail, uses chemical rockets to circularize and meet the tug, and the tug takes it to LEO gradually in the following weeks.

[Ralathon]

There is really no reason to use NERVA for that. It's absurdly expensive for the task. If you are going to be sending a lot of stuff from Moon to Earth or LEO, you should build a magrail on the surface. Launches to Earth can be done directly from surface, requiring only minor corrections along the way. Launches to LEO would be better done by using a VASMIR tug. In this scenario, cargo is launched by mag rail, uses chemical rockets to circularize and meet the tug, and the tug takes it to LEO gradually in the following weeks.

Why would you use a VASMIR tug to get to LEO? I can understand LEO-Moon but not the reverse. If time isn't an issue you can use atmospheric drag to lower your SMA, no need for anything but small course corrections.

[K^2]

Yeah, I forgot about aerobraking. Luna to LEO should be doable in under 100m/s, then. No need for VASMIR, you're right.

[Idobox]

I am a big fan of rotovators and momentum exchange tethers.

Magrails are nice, but can only be used to launch stuff from the Moon, when a rotovator can be used to land crafts without using any fuel, but sending stuff instead. Of course, they work best near the equator, which is not great for fuel depots on the Moon.

Anyway, NERVA is a nuclear reactor, it is relatively slow to throttle, but you can vent more or less hydrogen to change thrust.

[KASASpace]

Well, a magrail in a decently sized (a little higher than Hubble's) orbit, and one on the Moon.

Now, using orbital Solar Arrays (I like to call them power farms) that use some form of microwaves to transmit power to both stations, as well as Earth and the Mining Post, the power problem is better overcome.

[Rakaydos]

KASA, putting a magrail in orbit is pointless- every action has an opposite reaction, even magnets. :/

[shynung]

Throttling a NERVA doesn't necessarily mean throttling the reactor. Reducing propellant flow will do the trick, with a gain in I_SP. However, I agree that a NERVA is probably overkill for a Luna-LEO shuttle for cargo. If not a mass driver, an aluminium hybrid rocket will do the job just as well.

[Streetwind]

Somebody did the math and decided that while space elevators on earth are distant future visions, a space elevator on the moon could probably be realized with today's technology in a single launch of a rocket the size of a Falcon Heavy or SLS Block 1.

Basically you launch your vessel towards Earth/Moon L1 (the Lagrange point in between the two bodies), carrying the spent rocket stage with you instead of decoupling it. Once there, you fire a kind of harpoon drill at the lunar surface, which burrows and anchors itself into the rock. A thin fiber (something like dental floss, just a more sturdy material) connects the harpoon anchor to the L1 point station. Then, you attach the spent rocket stage to another such tether and carefully extend it back towards earth, a whole long way. More than half the way between L1 and Earth, in fact. Once there it becomes the counterweight, and Earth's gravity keeps the tethers straight. Because the Moon is tidally locked to the Earth, there is no rotation, and simple tether tension keeps the L1 station perfectly in place without a need for stationkeeping. From then on, you can deliver further construction materials to the L1 station. Tiny robots would crawl along the tether and draw additional fibers, one after another, to reinforce the tether. Once this has been done enough between L1 and lunar surface, you can start transporting cargo up and down, reinforcing the ground anchor and construct a base around it. The L1 station itself could be expanded into a whole spaceport with fuel depot and maybe even a permanent crewing. Virtually the only downside to this plan is the fact that the ground station can only be placed directly below Earth/Moon L1, not anywhere else on the lunar surface.

The guy started a company to make this concept a reality, but unfortunately it's been bumbling around for a while without any real progress being made. Kind of a shame, it would be cool to have such a thing. Low maintenance, fuel free lunar soft landing and launch capability without the need for a power-hungry magrail.

[Kryten]

Throttling a NERVA doesn't necessarily mean throttling the reactor. Reducing propellant flow will do the trick...

The propellant flow is also the coolant flow. Reducing coolant flow without reducing reactor output is not a good idea.

[shynung]

The propellant flow is also the coolant flow. Reducing coolant flow without reducing reactor output is not a good idea.

If the reactor does not have auxiliary cooling systems or used as a power generator, yes. Though, having a nuclear reactor solely for an NTR thruster doesn't seem like a good idea in the first place; why use all that power for something that is done occasionally?

[Kryten]

Though, having a nuclear reactor solely for an NTR thruster doesn't seem like a good idea in the first place; why use all that power for something that is done occasionally?

Why use all that weight for a power generator that'd likely be massively overbuilt for most ships? Most NERVA concepts are for tugs, their non-propulsion power requirements are going to be far below the actual reactor output. It'd be much more efficient for them to just use a standard solar+batteries system.

[Hasegawa]

Suppose you wanted to set up a high-efficiency Lunar Fuel Depot, so you want to use a Nuclear Thermal Rocket. [...] If so would it be practical to use a NERVA in a LEO-Lunar Surface Rocket to ferry fuel and supplies mined from the Moon on a regular basis?

Given how little shielding the NERVA style NTRs have, there may be issues with the radiation environment around the rocket. Both for the ground crew and support equipment from the neutron and gamma radiation from the reactor and for any crew or electronics on the ferry from back-scatter from the ground.

(Source: Atomic Rockets, Radiation page)

[shynung]

Why use all that weight for a power generator that'd likely be massively overbuilt for most ships? Most NERVA concepts are for tugs, their non-propulsion power requirements are going to be far below the actual reactor output. It'd be much more efficient for them to just use a standard solar+batteries system.

In the case of tugs powered by bimodal NTRs, the reactor could be throttled down just enough to run the internal systems. While one still needs to throttle the reactor in this case, it's not completely powered down, but more like being left idle. When needed burns are to be performed, powering up an idling reactor is much faster and easier than powering up an inactive one.

[Kryten]

In the case of tugs powered by bimodal NTRs, the reactor could be throttled down just enough to run the internal systems. While one still needs to throttle the reactor in this case, it's not completely powered down, but more like being left idle. When needed burns are to be performed, powering up an idling reactor is much faster and easier than powering up an inactive one.

If being used for the proposed reactor throttling, the radiator on the power generator would still have to be large enough to dump a large fraction of the full output, much higher than required during coast. That's still a lot of weight that otherwise wouldn't be needed, just so the engine can operate in regimes that would likely be suboptimal anyway.

[shynung]

If being used for the proposed reactor throttling, the radiator on the power generator would still have to be large enough to dump a large fraction of the full output, much higher than required during coast. That's still a lot of weight that otherwise wouldn't be needed, just so the engine can operate in regimes that would likely be suboptimal anyway.

It does, unfortunately. That's inherent in any NTR tug design.

Which is why using NTRs as tugs are quite an overkill. Aluminium hybrid rockets are somewhat more feasible for Luna-LEO shuttles, I think.

Though if we were talking about NTR for larger manned spacecrafts, this method would probably be more reasonable.

[Kryten]

It does, unfortunately. That's inherent in any NTR tug design.

No it isn't. All that's required is a separate power system and a bit of excess fuel for reactor cooling during the shutdown process.

[shynung]

No it isn't. All that's required is a separate power system and a bit of excess fuel for reactor cooling during the shutdown process.

I think you meant propellant, since it doubles as coolant in this case; the fuel would be the U₂₃₅ rods in the reactor. This would mean propellant would be flushed through the system as the reactor cools down (and any subsequent unwanted thrust be corrected via RCS). This design favors reactors that does not possess high thermal inertia, and can therefore be shut down and fired up quickly, to reduce propellant loss for cooling purposes. That means using reactors that have lower core temperatures, which results in lower I_SP in comparison to reactors with higher core temperatures.

I'm convinced, this design could work. Still needs some engineering tweaks, though.

Edited October 18, 2015 by shynung

[KASASpace]

KASA, putting a magrail in orbit is pointless- every action has an opposite reaction, even magnets. :/

That is exactly why you fire off two vehicles in opposite directions.

The vehicles are shot outward, and the kickback produced is countered by the other vehicles kickback.

The second vehicle is basically just a hunk of iron.

[Ralathon]

That is exactly why you fire off two vehicles in opposite directions.

The vehicles are shot outward, and the kickback produced is countered by the other vehicles kickback.

The second vehicle is basically just a hunk of iron.

So, you shoot a few tons of iron in LEO in a retrograde direction with about 4km/s. I'm fairly sure that's going to make a bunch of politicians VERY nervous. this is the basic idea behind a kinetic bombardment weapon.

[KASASpace]

So, you shoot a few tons of iron in LEO in a retrograde direction with about 4km/s. I'm fairly sure that's going to make a bunch of politicians VERY nervous. this is the basic idea behind a kinetic bombardment weapon.

Oh I know very much about that, but it wouldn't be in LEO any more, it would be either on a large orbit with Apoapsis near the moon, or on an escape trajectory.

Plus, it would be an international station, where it would have safeties and fail-safes to prevent kinetic bombardment attempts.

[Mecha Pants]

Still, the retrograde shot would have to come from somewhere, either Earth via rocket or luna via magnetic shot. I feel like it might be better served with some sort of hardened LEO -> Luna tug that could be refueled and refurbished on the moon and then shot back into LEO for its next cargo.

[KASASpace]

Still, the retrograde shot would have to come from somewhere, either Earth via rocket or luna via magnetic shot. I feel like it might be better served with some sort of hardened LEO -> Luna tug that could be refueled and refurbished on the moon and then shot back into LEO for its next cargo.

Why not from a captured asteroid?

In orbit over the Earth, at an Earth/Moon Lagrange point.

OH, and I think I have an idea about how to simulate Lagrange points in KSP.

Basically, a "ghost" sphere of influence, that moves at the same speed as the planet it is sprung from.

A tug would be more reliable, yes. Once used enough times.

[Javster]

Could you throttle it like a nuclear reactor, with control rods?

[Idobox]

Could you throttle it like a nuclear reactor, with control rods?

That's the problem, this type of throttle is slow. It takes days to stop a nuclear reactor.