Stats of a Nuclear-Powered VASIMR

[gutza1]

I've heard about the first VASMIR engine having a specific impulse of approximately 6000 seconds and a thrust of 4 Newtons. However, I read that, for a manned Mars mission, the VASMIR engine would have to be powered by a nuclear reactor. Does anybody know what the thrust and specific impulse for a nuclear-powered VASMIR would be. I would be very grateful if someone could help me.

Edited February 27, 2014 by gutza1

[Guest]

I've heard about the first VASMIR engine having a specific impulse of approximately 6000 seconds and a thrust of 4 Newtons. However, I read that, for a manned Mars mission, the VASMIR engine would have to be powered by a nuclear reactor. Does anybody know what the thrust and specific impulse for a nuclear-powered VASMIR would be. I would be very grateful if someone could help me.

They would be the same as what you quoted since the engine isn't changing, the powerplant is.

[Horn Brain]

To get power requirement (roughly) multiply the thrust (in Newtons) by the Isp by 9.81 m/s/s. That puts you in the ballpark. Then you divide by the efficiency (about 60% for VASIMR, I think) and that gives you the proper power. The thrust and Isp numbers are often fudged, though, since VASIMR means VARIABLE SPECIFIC IMPULSE Magnetoplasma Rocket. So you have a peak Isp (with low thrust) and a peak thrust (with low Isp). Usually they will quote both at the peak values to make the thing sound better. The efficiency is also a function of the mode that the engine is operating in, so there is a lot of variation.

Usually, the power requirement is the best place to start. Take the power requirement, multiply by the efficiency and divide by 9.81 m/s/s, then divide by peak thrust to get Isp at peak thrust, or divide by peak Isp to get thrust at peak Isp. These will give you your two modes of operation (assuming equal efficiency).

There is another problem with VASIMR, though that they haven't completely nailed down: the plasma may not detach from the magnetic field of the rocket, causing it to loop around and hit the spacecraft. It's not really dangerous, I don't think, but it does cost you thrust. That means that the efficiency that is quoted is likely an overestimate. This effect is not easy to pin down, though.

So in the OP, we have an example of a 6000 s Isp, 4 N thrust engine. If we do the math with 60% efficiency, we see that the power requirement for that would be 400 kW. That's twice as much as the VASIMR they are planning to fly on the space station in 2015. Solar panel power production mass is quoted (by the VASIMR people) as between 2 and 7 kg/kW. That means we're looking at ~2 tons of (high tech, never before flown) solar arrays, plus heat rejection systems (radiators), plus the VASIMR itself, plus tankage. If we fly that vehicle with no payload, we are looking at a burnout acceleration (acceleration on the last drop of fuel) of about 3.6 m/s per HOUR (give or take a factor of two). or about 0.001 g. That means if we had a vehicle accelerating at that rate from LEO starting today, it would take it until between July and October to escape the Earth's gravity and build up speed for an interplanetary transfer (delta-v requirement for these low thrust, spiral escapes is much higher, like 10-20 km/s, than for impulsive burns in LEO).

We really need some better power options.

[Streetwind]

We really need some better power options.

I suppose that's precisely why gutza1 asked about the prospect of using nuclear power.

The question is then - can you build a 400 kW fission reactor weighing less than the ~2 tons the solars would weigh? And would that require less or more heat radiators? And if there's beneficial scaling in reactors, would maybe 10 of these VASIMRs on a 4 MW reactor provide a better thrust-weight ratio?

[MaverickSawyer]

God, if there was ever a time for the Arc Reactor from Iron Man...

Honestly, we do need to develop some new space-rated reactors. Sadly, no one wants to be the one that starts the project, for fear of the public backlash...

"You want to put a nuclear reactor WHERE?????"

[falofonos]

From http://www.world-nuclear.org

Using a "SAFE-400" reactor, you can get 100Kw electric for 512 kg of power source. The "only" problem then is getting rid of the 400kW of waste heat and that all references to it talk about an additional 72kg of heat exchanger without saying how many would be required.

Edited February 27, 2014 by falofonos

[MaverickSawyer]

Yes, but VASIMRs require kilowatts to run. Nukes are about the only option that's compact.

[KvickFlygarn87]

That's VASIMR, not VASMIR. It's one of my many pet peeves.

Bothers me when people write "VASMR", "VASMIR", "VASIMIR", "VAMSIR" and the likes.

[Idobox]

If you're going to put a nuclear reactor on board, why would you go through the complex process of turning the heat into electricity, then the electricity into thrust, when you can simply use thermal nuclear rockets?

[Scotius]

Higher Isp i guess. VASIMR with a tank of xenon will take you further than equally heavy thermal engine using hydrogen as accelerant. And VASIMR can be called 'electric engine' - thus becoming easier to swallow by nukephobic media and public. Call it 'nuclear engine' and watch fireworks explode.

[Streetwind]

Because plasma thrusters have a much higher specific impulse and use more convenient fuel.

You have to keep in mind that the nuclear thermal rocket doesn't magically shoot some sort of "nuclear energy" out the nozzle. All it does is simply pass hydrogen gas over the hot reactor core, which picks up some heat and therefore expands as it is guided through the nozzle. That's far from a 100% efficient process either, and it requires you to carry hydrogen as fuel. Hydrogen is the lightest and least dense material you could possibly carry. The rocket equation cares only about the mass of the stuff you throw out back, not its volume or other measurement. Plus, hydrogen generally doesn't like being in containers and requires specially insulated tanks to store it for more than a few hours. Which means you need expensive, gigantic tanks to get any real delta-V, which in turn increases construction costs and dead weight. Plasma thrusters however can use the other noble gases, all of which are much heavier and denser and effortlessly stored long-term. So even if the Isp was the same, you could cram the same amount of reaction mass into much smaller, cheaper tanks.

But yeah, the main draw is the big Isp. Remember, in the rocket equation the effective exhaust velocity is an exponent over the mass change. Even small increases in Isp create comparatively large differences in dV. And here you have a huge increase in Isp. The best possible nuclear thermal rocket that has been theorized (the nuclear lightbulb) would roughly sit in the lower 2000's; the ones we actually built were more around 1000-1200. The VASIMR described above has 6000.

[Horn Brain]

So it looks like SAFE is the lightest nuclear power plant (it was designed for space, after all) at ~500 kg / 100 kWe. That means it's about 2 tons for 400 kWe just like the solar farm. It would be more compact, though. This energy problem is hard!

[Idobox]

Because plasma thrusters have a much higher specific impulse and use more convenient fuel.

You have to keep in mind that the nuclear thermal rocket doesn't magically shoot some sort of "nuclear energy" out the nozzle. All it does is simply pass hydrogen gas over the hot reactor core, which picks up some heat and therefore expands as it is guided through the nozzle. That's far from a 100% efficient process either, and it requires you to carry hydrogen as fuel. Hydrogen is the lightest and least dense material you could possibly carry. The rocket equation cares only about the mass of the stuff you throw out back, not its volume or other measurement. Plus, hydrogen generally doesn't like being in containers and requires specially insulated tanks to store it for more than a few hours. Which means you need expensive, gigantic tanks to get any real delta-V, which in turn increases construction costs and dead weight. Plasma thrusters however can use the other noble gases, all of which are much heavier and denser and effortlessly stored long-term. So even if the Isp was the same, you could cram the same amount of reaction mass into much smaller, cheaper tanks.

But yeah, the main draw is the big Isp. Remember, in the rocket equation the effective exhaust velocity is an exponent over the mass change. Even small increases in Isp create comparatively large differences in dV. And here you have a huge increase in Isp. The best possible nuclear thermal rocket that has been theorized (the nuclear lightbulb) would roughly sit in the lower 2000's; the ones we actually built were more around 1000-1200. The VASIMR described above has 6000.

The nuclear lightbulb could have up to 3000s of ISP, when the VASIMR is measured as 5000s, without counting the plasma that curls back.

On the energy point, efficiency for thermal to electric power is about 30% on Earth, 20% for SAFE, combined to 60% efficiency of the VASIMR, your get 12 to 18% overall efficiency. That's terrible. On the other hand, converting thermal energy in the core to thermal energy in the reaction mass can be done with very high efficiencies (just wrap the reactor in thermal insulation so that the heat can escape only through the working mass), and you don't have to worry about heat dissipation anymore.

Hydrogen storage is an issue, but because the thrust is so much higher, you don't have to store it for months (I'm assuming a probe that will need only one pig burn). If you need longer storage, there are options other than LH2, like compressed H2, metal hydrures or using a different working mass, although they have negative impact on either weight or ISP.

Most importantly, you save on mass. A nuclear thermal rocket is much lighter than a nuclear power plant, even without the VASIMR. If you have a big ISP, but a crappy mass ratio, you won't go very far.

A better option for electric propulsion, in my opinion, would be beamed microwaves. A parabolic reflector made of metalized Mylar would weigh a fraction of an equivalent solar panel, and could receive more power.

[Nuke]

saying that vasimr has an isp of 5000 is somewhat misleading. the whole point of vasimr is you can trade thrust for isp as mission needs dictate. its isp actually can be set i believe in the low 1000s to around 10000. with higher thrust being available at the lower end of that spectrum. you might use the low setting when you have a short window for a maneuver, or you need to power through a hazzard (such as the van allen radiation belts), but the rest of the time you will be using it at a high isp setting where its thrust will be very miniscule. of course the mission design becomes more complex because isp can change at the turn of a knob. its the perfect engine for unmanned operations, but i doubt it will be of much use a manned platform.

if you are going through the trouble of having a nuclear reactor on board, you might as well just use an mpd thruster. a 1MW thruster might be able to push 100n @ 10000s. nobody has ever flown a reactor that powerful in space, but its not like we have been working on that problem since the 50s and 60s.

http://www.nasa.gov/centers/glenn/about/fs22grc.html

one of the reasons i hope polywells make it is because it would give you a small (3 meters, mostly hollow) fusion reactor that could push 100MW. imagine what you can do with that and a cluster of those mpd thrusters.

tldr:

wikipedia is a horrible place to get engine stats.

Edited February 27, 2014 by Nuke

[Streetwind]

if you are going through the trouble of having a nuclear reactor on board, you might as well just use an mpd thruster. a 1MW thruster might be able to push 100n @ 10000s. nobody has ever flown a reactor that powerful in space, but its not like we have been working on that problem since the 50s and 60s.

http://www.nasa.gov/centers/glenn/about/fs22grc.html

The problem with those regular MPDs is that they suffer from cathode erosion, which not only pollutes the exhaust with tungsten and lowers effective Isp below the theoretical numbers, but also causes gradual performance loss and eventually renders the thruster inoperable after a certain firing duration. It will easily last the duration of a mission, true - maybe even three or four missions. But the VASIMR is envisioned to be placed on robotic space tugs that ferry cargo between Earth, Moon and EML1/2. Those would remain autonomous for a very long time, so a drive system that doesn't erode is needed. The VASIMR plasma thruster doesn't have any electrodes that can erode. It's pretty much attributed to that fact that it isn't quite as good as a static MPD. In addition, the variable power of the VASIMR is well suited for a cargo tug which flies heavily loaded in one direction but may return light and empty. I actually made plans to build exactly such a craft for exactly that reason using the Near Future Propulsion addon (but I got distracted by RSS).

As for fusion - we've been 20 years away from it for the past 60 years, as the saying goes I'm not expecting anything like a polywell reactor before 2030 at the very very earliest, and then it'll be another 5 years minimum before they launch one into space and another 5 years after that for productive adoption. The VASIMR on the other hand exists now. They'll ship one to the ISS in 2015. Thus I think in the short term it makes sense to look into fission power instead.

[SargeRho]

VASIMR exists right now in the same sense as Fusion Reactors do. So far, there is no power source that could power VASIMR on a manned mission, and it has to be operated using capacitors when mounted on the ISS. Similarly, Fusion Reactors aren't even a rarity these days, but they don't produce power yet.

In the inner solar system, nuclear thermal rockets or inertial-magnetic fusion engines are probably better.

[Nuke]

if we had this technology now we would be on mars already. that said a ready to test in space vasmir engine is sitting in a hanger waiting for a launch so it can be demoed in space.

thats a lot closer than any fusion tech. polywells are a reactor of interest for space craft, since its small enough to be integrated into a space craft. you might use the iter timeline as an indication of when fusion will be ready for use in space, but that is a totally erroneous way of thinking. a tokamak will just be too damn heavy to use in space, and its not a machine you can miniaturize. the cross section of the toroid needs to be several meters across, and this is a constraint placed upon the design by the laws of physics. you cant just make it smaller (in fusion, it actually pays off to have a larger machine).

since a polywell is spherical, the reactor only needs to be about 3 meters across, plus a little bit extra for the cages you need for direct conversion and your ion and electron guns. you can dispense with the heavy vacuum chamber for space applications. so its definitely something that would fit on a rocket stack. direct conversion also gets around the need to have a bunch of heavy thermodynamic equipment, such as turbines, radiators, pumps, and coolant (same stuff you have to have with a nuclear reactor). a tokamak also wont be compatible with direct conversion.

i should also point out that the time to develop a demo polywell is less than 10 years, and would only cost a couple hundred million. i for one cant wait to see 100MW fusion-electric space craft.