Talk:Variable specific impulse magnetoplasma rocket

From Wikipedia, the free encyclopedia

	This article is within the scope of WikiProject Physics, which collaborates on articles related to physics.	Physics Portal
???	This article has not yet received a rating on the assessment scale. [FAQ]
???	This article has not yet received an importance rating within physics.
Help with this template Please rate this article, and then leave comments here to explain the ratings and/or to identify the strengths and weaknesses of the article.

From the article: "Current VASIMR designs should be capable of producing specific impulses ranging from 10,000-300,000 m/s". Why is specific impulse expressed in m/s? Shouldn't it be in seconds? Maybe those numbers actually refer to the effective exhaust velocity, as listed here? --Grnch 23:23, 7 Jun 2005 (UTC)

It's good to see NASA using SI once in a while, as their own Inspector General told them to do, consistently. Seconds are SI units of time; seconds are not SI units of specific impulse. Specific impulse and effective exhaust velocity are not different things, even though they may be expressed with different numbers and different units in Fred Flintstone units.

Yes, when it is expressed in meters per second, it often is called effective exhaust velocity. When it is called specific impulse, the units are often stated in SI as the units of impulse (newton-seconds) divided by the units of mass (kilograms), or N·s/kg. But in the case of the SI units, the numbers for either are the same number: 1 N·s/kg = 1 m/s, so it doesn't matter if you reduce the N·s/kg to m/s because that is what they are in terms of base units. In other words, 300,000 m/s is 300,000 N·s/kg. But while 30,000 "seconds" is the same as 30,000 lbf·s/lb or 30,000 kgf·s/kg, you get different numbers if you express it as an exhaust velocity in terms of speed: it is 294,000 m/s or 965,000 ft/s. Two different numbers, but really measuring the same thing, calculated from the same values for the input variables.

Those "seconds" are really lbf·s/lb, or the equivalent kgf·s/kg. But there are no kilograms force in the modern metric system, so these are not the proper SI units. Of course, some people have made a post-hoc amendment to the formulas, gratuitously thowing in a metrological conversion factor which doesn't belong there because of the physics, to make the dimensional analysis work out for using "seconds" for specific impulse. Ends up being the same result as what actually happened originally, when the rocket scientists just used "pounds" to cancel out "pounds", never mind that in one case it is pounds-force and in the other case it is pounds-mass. ____

[edit] When will it be practical?

How long until such a craft will be practical?

It's semipractical now. Unfortunately, it's severely limited by the power/weight ratio of energy sources... even nuclear power is inadequate, due to weight of the reactor and shielding. So, in practice it probably doesn't work better than conventional ion drives. What's really needed is lower ISP than ion drives, but VASIMR gives same/higher ISP. Wolfkeeper 00:34, 2005 Apr 22 (UTC)

According to the article: The radio waves and magnetic fields would be produced by electricity, which would almost certainly be produced by nuclear fission. Any conceivable chemical fuel used in a fuel cell or to activate a generator would be more efficiently used in a conventional rocket. Solar energy could be more efficiently used in a solar thermal rocket. However, wouldn't a nuclear thermal rocket also use nuclear energy more efficiently than VASIMIR? In any case, VASIMIR has a higher maximum specific impulse than solar thermal, so solar power should not be ruled out.

I believe a nuclear thermal rocket's efficiency would be limited by the temperature its core elements could withstand; exotic ideas like gaseous cores aside, that's only a few thousand Kelvin compared to VASIMIR's 10 million K. I'm no expert, though, so if someone more authoritative comes along pay me no mind. Bryan 01:18, 31 May 2004 (UTC)

I'm no expert either, but the whole point of a VASIMR is that you can have outrageously hot exhaust gases (and let them cool when you're in a hurry). Of course, this guzzles power, so you need lots. It certainly is easier to build a high-power nuclear thermal rocket than a nuclear power plant producing the same power output that can be carried aboard a spacecraft. But on a long trip, you worry much more about reaction mass than consumption of nuclear fuel. So a fission-powered VASIMR seems reasonable.

No, probably not IMHO. There's a relationship between ISP and optimal mission delta-v- they should be comparable. VASIMR in high gear has an ISP of 100,000 m/s. I'm not aware of *any* mission that needs that. In low gear, it's more like 30,000 m/s, which is a better match for a Mars mission. Trouble is, lots of other drives have similar performance as low gear, at similar power consumptions.

If I understand correctly, what you mean by "optimal" here is "minimum total energy".

Correct, but the energy generator equipment comes out of the payload (as does the fuel of course). And energy efficiency is inversely proportional to exhaust velocity. The big problem with any really high ISP drive is its gross inefficiency- almost all of the power ends up in the high speed exhaust and very little ends up as kinetic energy of the vehicle (except where the vehicle ends up going similar speed to the exhaust, then the exhaust stops and the vehicle gets all the energy, actually the maths is even more bizarre than that...)

Well, gross energy inefficiency, which is balanced against high efficiency in reaction mass. If you have plenty of power, but limited mass, then this is a good tradeoff. --Andrew 18:54, Apr 22, 2005 (UTC)

This is not necessarily optimal for all mission plans - in particular, an energy-rich design (perhaps built with a large fission reactor - Kiwi-2A produced 4 GW) might use more energy than necessary in order to save on reaction mass.

You seem to be effectively assuming that the generator is lighter than the propellent. VASIMR's power requirements in high gear are so high that that assumption isn't valid- the power/weight ratio of current nuclear power isn't much better than solar.

Yes, I'm assuming that the generator is lighter than the propellant. For a long enough voyage, of course, this is almost certainly true, since the fuel requirements increase exponentially with delta-v, but I think your point is that maybe there aren't any long enough voyages to make this trade-off worth it.

Nevertheless, current nuclear systems are RTGs, which produce far less power than a critical nuclear reactor. --Andrew 18:54, Apr 22, 2005 (UTC)

If you're mass-limited and you have to carry your energy store with you, your optimal exhaust velocity (that is, for obtaining maximum delta-v from a fixed mass) is that where you eject a piece of depleted energy store at a veolcity where it carries off all the energy it contained as kinetic energy (in your frame).

Sure. But that's not VASIMR you're describing.

Think of it this way: suppose you get your energy from matter-antimatter annihilation, so there's plenty. You should then run your engines at huge Isps so you don't waste propellant (since energy is plentiful). On the other hand, if you're really energy-starved, lower Isps are a good idea.

Yes. Right now we don't have useful antimatter drive, and even if we did, it's unclear that VASIMR would be the best system to use it.

But maybe I don't understand what you mean by an optimal Isp. --Andrew 02:04, Apr 22, 2005 (UTC)

On the other hand, a VASIMR needs a certain minimum energy input to maintain its magnetic bottle and to heat its plasma. It is quite likely that the solar cells required would be enormous and unsuited to acceleration, defeating the purpose of the variable specific impulse.

In principle, the magnetic bottle doesn't need any energy to maintain it; superconducting coils should work.

Well, sort of. There's cryogenics, support and control circuitry, and so on. I mean, solar cells are really low-power when you're talking about megakelvin plasmas. --Andrew 02:04, Apr 22, 2005 (UTC)

As for chemical power, if you're using up the chemicals anyway, why not use them as reaction mass? And if you're doing that, then you have enough reaction mass to afford a low specific impulse.

That's correct. If you generate power by a chemical reaction, then sending the chemicals out the exhaust by using the energy liberated is a win. Wolfkeeper 00:34, 2005 Apr 22 (UTC)

All this suggests that that part of the article could use some clarification. --Andrew 05:54, 2 Jun 2004 (UTC)

If your propellant is hydrogen then surely the best source of power is a fuel cell. Of course you'll have to carry oxygen which will increase the weight of the rocket but there's two things that can mediate that. At launch time you can use atmospheric oxygen. For manned flights you'll want to carry oxygen to breath. The output of a fuel cell is water, which can be seperated back into hydrogen and oxygen at high efficiency using membranes and solar heating. Not only is the production of hydrogen and oxygen in this manner likely to be more efficient than powering the rocket using solar panels, it is also a lot less complicated (and therefore less prone to failure) and can be used like a battery to store solar energy when the rocket is not in use.

Keep in mind that for such an engine there are two things needed: energy and reaction mass. The difficult question is where to get the energy. If you get it chemically (from hydrogen and oxygen, say) what do you do with the reaction products? If you discard them, they might as well serve as reaction mass; this is most efficiently done in a straight chemical rocket - the high specific impulse requires more energy per unit reaction mass than can be obtained from chemical reactions. Perhaps more simply, you will need so much of the unreacted chemicals that you will lose the advantage of the low reaction mass requirements. In order to re-use the chemicals, you must obtain the energy to return them to their unreacted state from somewhere, so you're back to needing an energy source.

It may be appropriate to use a hydrogen-oxygen fuel-cell/electrolysis closed cycle as an energy storage system, or as part of a solar (or nuclear) energy collection system, but ultimately the energy needs to come from some other source. Moreover, the primary usage of such an engine would be for continuous low-level use, which requires a long-term energy input; a storage system won't help with this. Of course, a VASIMR has another mode, a high-thrust mode, and in this case it might be appropriate to use some sort of stored energy, but it will be difficult to store enough to be useful in a reasonable amount of chemicals.

High specific impulse engines are designed to take advantage of plentiful energy and scarce reaction mass. Almost any system based on chemical energy will require far too much mass to carry enough energy - chemical energy is not really dense enough to carry around. Nuclear energy can be obtained from much less material, and solar energy can (sometimes) be obtained from the environment. --Andrew 05:19, 11 Jul 2004 (UTC)

As for when a VASIMR-propelled craft will be practical, the technology exists today, more or less; NASA has run prototypes. But they're big, complicated, and fragile. I believe that they're also poorly suited to launch applications, at least as used now; the power consumption at launch is huge, and if you're not getting it from the reaction mass, you have to have some huge power plant aboard. So if they're no use for launch, then they're mainly suited to long space-to-space voyages. And since they're big, they're suited to big vehicles. The market for large, complex space-to-space vehicles suited for long voyages is, so far, pretty small. --Andrew 05:54, 2 Jun 2004 (UTC)

The bottom line to me is that only the very lowest gears for VASIMR is that much use, the high gear might get you a small speed boost when you're more or less coasting, but the 1000 seconds end determines whether it's useful or not. In fact, I've a horrible feeling that VASIMR is inefficient for some reason at 1000 seconds, which is why they're not talking about it. If so, then a system using Hall effect thrusters might in fact be better- and there's no reason at all why Hall effect thrusters can't use VASIMR's power supply.

But none of the papers are emphasising this point; they talk about the high end. Rockets are like drag racers- they need to accelerate. VASIMR is like advertising a drag racer's fuel efficiency at high speed. Who cares?

Still, VASIMR is an ongoing research project, and it's been ongoing for 30 years. That's worrying in and of itself. Wolfkeeper 14:50, 2005 Apr 22 (UTC)

---

One idiots questions. Bear in mind this is massively speculative. But answers would be greatly appreciated. This is definately discussion.

VASIMR seems to have two ionization systems. First, hydrogen is ionized through radiowaves and magnetically threaded into the core. Second, magnetic containment and further radio bombardment serve to raise core temperature and presumably ionization state to far-exceeding-astronomical levels. (see Saha ionization equation). Essentially fancy ways of dumping lots of energy into a exceedingly ionized plasma core.

How would VASIMR handle earth launch characteristics?

It wouldn't. You can't strike the plasma at earth atmosphere pressures. At very high altitude, it's theoretically possible though. But, the thrust/weight ratio of VASIMR, with or without the power source is *very* much less than 1, so it wouldn't do much good.Wolfkeeper 00:34, 2005 Apr 22 (UTC)

Some comments to this effect are now in the article. --Andrew 02:04, Apr 22, 2005 (UTC)

How does earth's atmosphere come into play for starting the plasma? the plasma is extremely dense, I cant see earth as being any way signficantly different from the relative zero pressure of space.

The main reason I think VASIMR is interesting for launch is because it partly escapes the conventional aeronautical limitations of being bounded by reaction mass. There's two (correlated) adjustable limitations; the amount of energy you can dump into your system at any given time and how many ions you are generating. VASIMR isnt designed to generate enough nearly enough ions for high thrust and it doesnt pour nearly enough energy into the system in the first place to support all those ions, but these are just current limitations. VASIMR is really three systems in one; a plasma generator, a plasma exciter, and a magnetic nozzle system. With a scaled ion/plasma generation systems and much higher energy output (when the exciter becomes a full fledged fusion reactor), it doesnt seem entirely implausible for a launch system; one that can be scaled very well. Particularly when you can ignite the (extremely costly fusion reaction planetside.

--User:Myren

Would there still be a "completely" ionized plasma core, or would the core start shifting towards more merely-astronomical temperatures & ionizations (or even, gasp, cold plasma states)? In launch conditions, the core trades off ion speed & presumably/consequentially temperature for sheer ion quantity, but then how do ions get made? This is probably cyclical, but isnt a very hot plasma core instrumental to rapidly exciting more plasma?

Bear with me; you want to be able to (more or less) dump equal amounts of energy into the primary plasma core whether you're lifting off or interplanetary. Earth-bound launch is mainly special because it requires a huge number of already-excited ions coming into the system. Do you have to generate these ions as you're ejecting them, or does it make any sense to have a secondary ion/plasma containment-or-generation system? Containment makes sense in terms of building up plasma pre-launch, but we're not aiming for high-velocity, just lots of ions. What is the efficiency of radio-wave ion generation compare to, say Pulsed inductive thruster or other ion generation systems? Pulsed inductive thruster is interesting because it seems like you could scale it three dimensionally + with relatively little weight. How different would the plasma flow be for earth launch? Lower energy makes me think the containment would be easier, but how adaptable are the magnetic containment systems? Again, massive speculation, but wouldnt velocities be far different?

I'd like to know a lot more about the radio-excitiation. jt60's rf excitation is the only source of information i've found so far, and is somewhat disturbing. the ion generation / general heating system runs just shy of 110mhz at only 10 MW... and it looks massive. I've found little discussion on the topic at large, even with Plasma source. I'm presuming Electron cyclotron resonance heating is the general topic, found from Talk:Fusion_power.

Can we get details on the magnetic nozzle? Surely such a system must consume just-as-many gobs of power.

If you did promote fusion inside a plasma core, how would you keep random elements from forming and ionizing into deadly ions? The neutron issues would be horrendous. Is there any way to prevent fusion from occuring in ultra-high-temperature cores? Would there be any way to share plasma or to pre-excite ions by leaching off the fusion core (again without launching ions of every sort into space?) Someone please shoot me down on this next one; but what about hybridizing fusion+vasimr into a single system?

Um. VASIMR started out as that, and was watered down because nobody knows how to do fusion, never mind being able to turn it into a vehicle thruster.Wolfkeeper 00:34, 2005 Apr 22 (UTC)

It seems beneficial because you could avoid many of the difficult power-recovery steps associated with fusion. Instead of running some nuclear power source to generate RF to bombard ions, you "just" maintain fusion while dumping in more plasma and ejecting some out the back. The fusion sustains the excitement process rather than having to RF excite your ions. A couple ionic cyclotrons (VASIMR sans nozzles) feeding a plasma reactor with a magnetic nozzle out the back of the plasma reactor. But yes, way beyond current science.

A ³he fusion/vasimr hybrid would be the system to end all systems, but i have no idea if its in any way feasible.

Now you're dreaming in technicolour. The ignition temperature/pressure is *way* higher for ³he fusion than conventional fusion.Wolfkeeper 00:34, 2005 Apr 22 (UTC)

Above and beyond this, there are concerns with bremsstrahlung losses that may make any of the aneutronic reactions infeasible without truly gargantuan (kilometers of plasma) reactors. See Fundamental limitations on plasma fusion systems not in thermodynamic equilibrium thesis of Dr. Todd Rider at MIT. --Andrew 02:04, Apr 22, 2005 (UTC)

Wow, fantastic paper. I think i'm going to go home and cry now. I always assumed fusion was just a ignition/containment problem. I'm amazed the impossibility of useful aneutronic fusions remains unknown.

Between varying containment needs and reactant maintenence, i have no idea if fusion + VASIMR is in any way possible. VASIMR seems so similar to fusion its hard to image the two reactors couldnt be muxed somehow.

How is the core thermally contained? I'm presuming magnetic containment is the first line of thermal containment, I cant imagine how else you'd hold 10m K. the goal is to keep as much heat in one place, how do is this accomplished? this is especially important in space where you really have no choice but to carry your heat with you or toss it out the back.

Yes, cooling is known to be a problem :-) Wolfkeeper 00:34, 2005 Apr 22 (UTC)

--User:Myren 4:30, 21 April 2005 (ECT)

[edit] Extra detail

The article statse "The ions spiral around the magnetic field lines with a certain natural frequency; by bombarding them with radio waves of the same frequency, the system heats the ions to 10 megakelvins."

The article should either go into detail on how the ions are heated to such a high temperature, or should link to an article that explains in more detail.