Transterrestrial Musings

Comments

VASIMR isn't a practical space drive at the moment; and may never be.

The exhaust velocity is enormous; this means that VASIMR needs a massive powerplant. Current powerplants or solar panels manage under IRC 50 watts/kg, VASIMR needs many orders of magnitude more power than that from the powerplant; it needs megawatts- that weighs many tonnes- the weight of the powersupply means the rocket accelerates very slowly because it is so heavy.

Even if you could get a lightweight powerplant to work, you would in fact get far more acceleration from ganged up ion drives with the same powerplant- ion thrusters are not especially heavy, and more energy efficient than VASIMR- they would use more fuel; but then again, Ion drives don't use terribly much fuel.

Ultimately, with thrusters, you want the thruster exhaust velocity to have a few times the mission delta-v at most, otherwise you are wasting power, and not saving much fuel. Ion drives have exhaust velocities of 30 km/s- a round trip to Mars orbit is maybe 8km/s. VASIMR has an exhaust velocity of up to 300km/s- way, way, way too fast for any practical use around the solar system. Maybe if you wanted to go to Pluto.

Posted by Ian Woollard at August 30, 2004 05:09 PM

Quote from Ian Woolard: "VASIMR has an exhaust velocity of up to 300km/s- way, way, way too fast for any practical use around the solar system. Maybe if you wanted to go to Pluto."

From sometime in the past:

"No one will pay good money to get from Berlin to Potsdam in one hour when he can ride his horse there in one day for free."

-- King William I of Prussia, on hearing of the invention of trains,

Or,

"What can be more palpably absurd than the prospect held out of locomotives traveling twice as fast as stagecoaches?"

-- The Quarterly Review, March edition, 1825

Posted by Hefty at August 31, 2004 05:46 AM

What if your powerplant isn't solar-cells-aimed-at-the-sun, but solar-cells-powered-by-laser-battery?

What sort of power/mass ratio do you end up with in that case?

Posted by Al at August 31, 2004 11:49 AM

Hefty- the point isn't the exhaust velocity- it's the acceleration of the vehicle. For a fixed power supply, you get 10x *lower* acceleration of the vehicle at 300,000 m/s, than at 30000 m/s. It's exactly the same problem as trying to use a rocket instead of a jet engine to go from New York to London. You can't economically it, because the exhaust velocity of a rocket is about 10x higher- so you run out of energy (and hence fuel), 10x quicker. But try to go to orbit- which requires going much faster- then a rocket comes into its own.

Al- solar panels tend to overheat and give really low efficiency if you point a laser at them, but it's possible in principle. But its pretty difficult to focus the laser beam on the vehicle from large distances, particularly from the earths surface due to atmospheric effects and diffraction limits; and it's even harder when the object keeps orbiting round and round the earth.

Posted by Ian Woollard at August 31, 2004 12:56 PM

A laser system is technically possible, and far easier if not from the earth's surface. For such a high energy density, a receiver heat engine/generator scheme might be more effective than PV. An alternative idea is to use a microwave antenna transmitter array with a very large effective diameter, and a receiving rectenna. Either scheme would require very good feedback, of course. In ANY system dealing with large amounts of power, handling waste heat will be a very big deal.

We will eventually want much higher acceleration + delta-v for passenger travel. My bet would be on some type of nuclear system, but that would have to wait for technical advances and an attitude change.

For cargo that isn't time sensitive, the most economical choice will be to keep the delta-v to a near minimum. I expect something like the cargo ship/passenger aircraft division will eventually develop, but the distinction will be far larger.

Posted by VR at August 31, 2004 01:34 PM

Actually, the Isp of VASIMR may potentially go UP to the really high values, but the point is that the Isp can be tuned. The initial test version was going to have 10,000 sec Isp (purposely fixed to save mass on the test engine). That's the beauty of it, you use lower Isp/higher thrust deeper in a gravity well, where you need it and you can tune to be more efficient in deep space.

The problem that I recall being raised about VASIMR is the penalty one pays for the power generation and conditioning system mass. I don't know that there's a general warm fuzzy that you'll get to the equivalent overall mass efficiency of more conventional electric propulsion. The case for VASIMR gets worse when you look at recent analysis showing that you can get most of the variable Isp benefit by using a conventional EP engine that has a high and low Isp operating point and switching between the operating points (and coast segments) as needed.

The one and only really good argument I've seen in favor of VASIMR is that it pioneers some of the technology needed for eventual fusion rockets. One can also make relatively convincing arguments that VASIMR-type thrusters scale up well to large (presumably crewed) ships. But then, Lorentz force accelerators and possibly Hall thrusters might scale just as well.

So, does that cover things??

- Eric.

Posted by Eric S. at August 31, 2004 05:27 PM

I just think that VASIMR is a short term fix to a short-term problem (for large values of short term).

It's predicated on the idea that fuel is expensive, but energy is cheap.

In reality, by the time we can reach Mars in a frequent way, mining exploration is going to be well underway, and refuelling throughout the solar system is going to be extremely common.

That means that we don't need high thrust-high delta-v drives- you just tank up till you reach the next fuel stop, rather than try to carry all your fuel for the entire trip.

Posted by Ian Woollard at August 31, 2004 05:29 PM

OK, sorry to follow up on myself, but I forgot to correct a misconception in the original comment. Although VASIMR may not yet be as mass efficient as conventional EP (although, to be fair, it's not at the same TRL either), that does NOT mean that it requires some massive multi-megawatt powerplant just to work. The proposed space test engine was really quite modest in size and power. Your kitchen microwave stands at about a kilowatt, right? Picture, what, maybe 10 microwave ovens? That's what the test was going to be.

- Eric.

Posted by Eric S. at August 31, 2004 05:40 PM

Ian, what is your definition of "high thrust, high delta-v"? Refueling or not, if it takes years to reach your destination, I doubt there will be many passenger flights.

Posted by VR at August 31, 2004 09:26 PM

I read an article which suggested a combination of high thrust NTR and low thrust EP was the fastest way to reach Mars. Faster than either Nuclear Thermal Rocket or Electric Propulsion by itself. I think a trip time of 90 days was possible with the combination drive.

Posted by Brad at September 1, 2004 08:16 AM

I also believe that aerobraking and refueling are key to superior interplanetary flight. Even the Orion drive mission to Enceladus was planned using those methods. Orion aside, some sort of multi-propellent long-lived NTR could be a key technology. Of course NTR of some kind are THE key to future manned deep space exploration.

Posted by Brad at September 1, 2004 08:31 AM

found the link

http://www.spacedaily.com/news/spacetravel-04w.html

Correction - It's only 60 days to Mars using Nuclear plus Electric propulsion, not 90 days.

Nuclear by itself is 130 days trip time and chemical is 240 days trip time.

Posted by Brad at September 1, 2004 08:44 AM

Yes, for Mars you might get things down to four months or so with solid core nuclear thermal, and somewhat less than that with fairly conventional nuclear/ion, assuming you work within the launch window.

However, if you want to go to the outer solar system, flight times increase dramatically and you can't always wait for the best launch window.

Posted by VR at September 1, 2004 01:52 PM

"Actually, the Isp of VASIMR may potentially go UP to the really high values, but the point is that the Isp can be tuned. The initial test version was going to have 10,000 sec Isp"

Um. That's precisely missing the point. Normal ion drives have an ISP of 3000 seconds, and even that is too high. That's a lot to do with why they give such low thrust. The optimum mission delta v for an ion drive is about 48 km/s from an energy point of view. I don't think getting to Mars needs even that much. Multiplying the exhaust velocity by 3 is just... insane.

I'm not saying the engine doesn't work, or can't be tested or isn't clever. I'm just saying that ordinary electric propulsion devices or conventional chemical or NTR are better matches to anything we are likely to want to do in the near future; like our lifespan.

And once you get away from things we currently do- technologies like Orion start to look interesting.

If they could rework VASIMR to give an ISP nearer 1800 seconds with high energetic efficiency that *would* be interesting.

It's not just me that says this either- The Space Show recently had Zubrin on it; he has multiple degrees and has invented a couple of space drives and planned missions to Mars- and he *hates* VASIMR too, for basically the same reason I do.

Posted by Ian Woollard at September 1, 2004 03:18 PM

Ian,

It's not so much that fuel is expensive. Even if the fuel is there free for the taking, it is energetically expensive to push it around.

BTW, I never said that the *lower* Isp limit for VASIMR was the 10000 sec of the test engine. That was just the fixed operating point that was chosen for demonstration purposes. If you just dump more hydrogen into your hot plasma, you decrease the net Isp and increase the mass flow, yielding increased thrust.

Finally, things in this thread are getting a bit confused on the whole Isp vs. thrust thing, perhaps because we're not comparing all candidate propulsion systems at the same power levels. EP is a process for converting input electrical power to output exhaust jet power (kinetic energy rate) -- with mass flow and exhaust velocity in the expected relationship. Much of the reason that, for example, the European spacecraft is taking so long to even reach the moon, is that to reduce the fuel mass they've chosen an engine/operating point that provides a very low thrust to local gravity force ratio, necessitating a prolonged outward spiral. This does NOT mean that the Isp is "too high", it's simply the outcome of a design trade.

One more thing... I've met and spoken with Zubrin several times at various meetings. He's a bright guy, with a lot of really good ideas, but I'd hardly consider him anything like an expert in EP and I certainly don't recall him inventing any "space drives". Like many bright, forceful people with strongly held views, he sometimes blurs where his expertise ends and his opinions begin. His opinion on VASIMR, without hearing the details (if any) of his arguments, don't really carry a lot of weight.

- Eric.

Posted by Eric S. at September 1, 2004 08:23 PM

Even if the fuel is there free for the taking, it is energetically expensive to push it around.

It is only energetically expensive only if you try to push it away really, really fast like VASIMR does. Kinetic energy is a SQUARE LAW; whereas thrust is simply proportional to exhaust speed.

In fact, in free space, if you continuously adjust the exhaust velocity of the fuel to be equal and opposite to the velocity of the vehicle relative to the initial launch at each moment, then the exhaust carries *no* energy away because it stops dead in space- *all* of the energy ends up in the dry mass of the vehicle (neglecting thermal losses due to finite nozzle length).

In practice this strategy is never followed exactly, as this strategy is highly efficient of energy, but it uses *lots* of fuel- but the further you get away from it, the more energy you need; and the bigger your powerplant needs to be. And powerplants are *heavy*.

In the case where the exhaust is much faster at all points than the vehicle is going; then most of the energy ends up being carried backwards and away in the exhaust- and that's the problem with VASIMR- it literally pisses away energy at a tremendous rate.

In fact, an excessive ISP reduces your thrust, because of power limits in your generator. That's the *real*, underlying reason that thrust is low in EP.

By contrast, if you fix the exhaust velocity at 2/3 the mission delta-v then you roughly minimise the overall energy for a fixed exhaust speed. (Go to http://www.neofuel.com/optimum/ for a derivation, and discussion.)

In fact, there's an inverse relationship between exhaust speed and thrust; that's why VASIMR needs such a huge powerplant to get good thrust. If you stick the same VASIMR powerplant on ganged conventional Ion drives- you get 3-10x more thrust than VASIMR.

Oh yeah, about Zubrin, you might like to look into magnetic solar sails and airbreathing methane launch vehicle, both are concepts that Zubrin has been intimately involved in. Zubrin is a clever guy, he's way too clever not to see through VASIMR. Actually, if you read some of the webpages you find there are two schools of thought on VASIMR; the nuclear engineers who don't know the maths- they really love it; the space engineers usually pan it. Zubrin is in the latter camp.

Sorry, the emperor has no clothes.

Posted by Ian Woollard at September 2, 2004 11:26 AM

In the case where the exhaust is much faster at all points than the vehicle is going; then most of the energy ends up being carried backwards and away in the exhaust

Um, maybe I'm missing something, but these statements appear to contradict conservation of momentum.

If a rocket begins at rest, it is not possible for it to accelerate away and leave its exhaust behind at rest, right? That would imply an acceleration of the center of mass of the entire rocket-plus-exhaust system, which, of course, must remain at rest since it began at rest. . .

Posted by Astro Boy at September 2, 2004 12:15 PM

Good. Sounds like you're starting to think about and understand this stuff. Adjusting the ISP in that way is not really a practical technique at all- the fuel use is way too high- it's a thought experiment to illustrate the physics involved. And even in theory you need to start with a push start, otherwise the exhaust has zero velocity and you dont even start moving. After the push start the fuel then stops and the, much smaller payload ends up with *all* that momentum you got from the push start.

But this statement:

"In the case where the exhaust is much faster at all points than the vehicle is going; then most of the energy ends up being carried backwards and away in the exhaust"

Talks about energy, not momentum.

Rockets are momentum machines; but understanding where the energy goes is critical to understanding EP, and all other rockets that separate the fuel and energy source.

Posted by Ian Woollard at September 2, 2004 12:37 PM

Finally, things in this thread are getting a bit confused on the whole Isp vs. thrust thing

Yup. Actually, more important questions are: What is the destination? What is an acceptable flight time for the spacecraft? From there you can decide the best options to achieve the mission requirements.

Zubrin is a clever guy, he's way too clever not to see through VASIMR.

Zubrin is a clever guy. He also has said some incredibly silly things, such as his argument against O'Neill type space colonies. I enjoy reading his books, I have a soft spot for Mars, but it becomes obvious very quickly that he has an agenda, and is often not objective.

Posted by VR at September 2, 2004 01:56 PM

Ian,

Sigh... My bad on how I stated it. The bit about "pushing it around" was meant regarding *carrying the fuel mass*, not exhausting it at high velocity. That is, free fuel has nothing to do with the problem -- how you use that fuel is, in fact, the point of the whole propulsion system design problem.

Your mention of K.E. going like v-squared but thrust going like exhaust velocity struck me as a bit of a non sequitur. Remember that this is a design point you choose by selecting power, mass flow, and Isp (in the case of something like VASIMR). You *choose* the thrust - Isp operating point, subject to any mass & power & mission time constraints you may face. BTW, it's interesting that you bring up the whole "leave the exhausted mass at rest" thing... That's essentially what VASIMR is trying to do a better job of. Don't focus strictly on the high end Isp's that VASIMR is potentially capable of. The point of VASIMR isn't the extreme Isp, but the *variable* Isp. In essence, all your arguments against VASIMR primarily apply if you were to choose to operate it at the high end of the Isp range.

As far as Zubrin's propulsion credentials, I'm quite familiar with the MagSail, and I seem to recall that the actual idea was something Dana Andrews came up with after talking w/ the late Bob Forward. (Admittedly, it's been a few years since the conference session where that was discussed -- I think it was the JPC in Seattle, which would be '97 or so.) Bob Zubrin certainly contributed to fleshing out the idea, though, but I never had the impression he invented it. As for the other, I'm not sure what you're talking about. There's the whole "Black Colt" thing that he worked on, but that's really Mitchell Burnside Clapp's thing. Now, if we were talking nuclear thermal propulsion, I most definitely *would* take Zubrin's word. But of all the things of his that I'm familiar with, I'd not count him as a premier expert on EP.

Just to put this to bed, though, I worked for several years for someone who I *do* consider to be a premier expert on EP and he was skeptical regarding VASIMR. And in an earlier post I mentioned the basic argument -- the mass penalty VASIMR faces w.r.t. the magnets and klystrons, etc. But, as I also said, you really can't make a fair comparison in that regard until VASIMR has been allowed to climb a rung or two up the TRL ladder. And certainly, trying to argue against VASIMR based on the erroneous notion that "its Isp is too high", just doesn't cut it.

- Eric.

Posted by Eric S. at September 2, 2004 08:25 PM

"You *choose* the thrust - Isp operating point, subject to any mass & power & mission time constraints you may face."

Yeah, sure, in principle. In practice there are limits to how high or low the ISP and thrust go.

"In essence, all your arguments against VASIMR primarily apply if you were to choose to operate it at the high end of the Isp range."

That's because it is. In all the literature on VASIMR I've read the ISP goes from 5000 seconds (*much* too high for good thrust and economical/low mass powerplants) to 30,000 seconds (much, much, much, much too high).

"BTW, it's interesting that you bring up the whole "leave the exhausted mass at rest" thing... That's essentially what VASIMR is trying to do a better job of."

No. At no point in any of their proposed missions does the vehicle velocity go as *high* as an order of magnitude below the exhaust velocity. So they are throwing away >99% of the powerplant energy in the exhaust. And then they have the problem that they can't get anything like a powerful enough powerplant... it's just totally insane. They're going totally the wrong way.

"the mass penalty VASIMR faces w.r.t. the magnets and klystrons, etc."

They've mostly solved that problem. But they still don't have a solution for the powerplant problem. And even if they solve that problem; you actually get higher thrust from ganged ion drives or Hall effect thrusters with the same powerplant than from VASIMR.

I'd be impressed if they came out with a high thrust at 1500 seconds ISP; but they keep bragging about ISPs at 5000-30000 seconds which is of no use to man or beast, given the current power plant technology.

"I worked for several years for someone who I *do* consider to be a premier expert on EP and he was skeptical regarding VASIMR."

There are multiple problems with VASIMR. You should listen to him. All I know is that the engine isn't a magic box; if the inputs and outputs of the box aren't enough to do what they say, then the box contains snake oil; no matter how many PhDs designed the magnetic fields or the coils or drew pretty herringbone patterns of transfer orbits.

Posted by at September 5, 2004 10:38 AM

"At no point in any of their proposed missions does the vehicle velocity go as *high* as an order of magnitude below the exhaust velocity."

By that logic, DS-1 was a dismal failure, because it pissed away energy in the high speed exhaust.

"They've mostly solved that problem."

If so, great! Because then they're on an even footing with other EP. For the last time, THERE IS NO POWERPLANT PROBLEM! If I want 25 kw of power into my propulsion system, regardless of the propulsion type, I build a 25 kw powerplant. The powerplant doesn't somehow 'know' a VASIMR is attached and suddenly get heavy.

Now, let's try and get you over your hangup on "high thrust" vs. "low thrust". These are relative terms determined by the 'thrust/local-force-of-gravity' ratio (i.e. the local thrust/weight ratio). On one end there's the high values -- burn & go sorts of things where the delta-V is applied very quickly and you're on an escape trajectory. On the other end is the low end, for example the European SMART-1 mission that's taking a year or so to reach lunar orbit via a long spiral out, a few lunar resonances, then a long spiral in. There are, however, many possible *intermediate* thrust trajectories.

The point is: inability to do 'high thrust' doesn't rule out a propulsion technology.

Oh, and one thing I missed in your original post... You mention the well-known rule of thumb that mission delta-V should be something around 2x the exhaust velocity. But then you quote a quite small value for earth orbit to mars orbit delta-V -- one typical of a 200+ day one-way transfer. Try running the numbers on, say, a 90-100 day one-way transfer. Then add in the return trip. Now you're beginning to see numbers that justify something like a VASIMR. And if you consider relatively short duration trips to the asteroids or Jupiter, the numbers are even more convincing.

BTW, I *did* have numerous long talks w/ my now-former boss regarding VASIMR. That's why I know that there are not "multiple problems" with VASIMR. There are certainly some open questions that require further research. The main and only real objection has always been in the area of wall plug power/exhaust jet power and mass of engine/exhaust jet power. The notion of VASIMR somehow only working with an enormous powerplant arises from the quite reasonable feeling that VASIMR scales better to large payloads than other EP. Chang-Diaz does himself and VASIMR no favors sometimes in the way he 'markets' it, but I think most arguments on this technology by knowledgable individuals boils down to where the crossover point is.

And just to restate, again, I wholeheartedly agree that something like Hall thrusters or Lorentz-Force Accelerators are more likely technologies -- these look scalable to megawatt exhaust power and are of much higher TRL than VASIMR. Which is not to say that VASIMR shouldn't be researched at some level, just to see where it goes.

Oh, and if you want to see a mismatch between mission delta-V and exhaust velocity, look at JIMO... For this mission, the engines are chosen because they're essentially the only game in town at the moment.

- Eric.

Posted by Eric S. at September 5, 2004 09:24 PM

"If I want 25 kw of power into my propulsion system, regardless of the propulsion type, I build a 25 kw powerplant. The powerplant doesn't somehow 'know' a VASIMR is attached and suddenly get heavy."

It just doesn't accelerate as fast as if you had attach a Hall Effect thruster. You get to the destination slower with VASIMR than a HET. So what's the point?

Posted by Ian Woollard at September 8, 2004 01:11 PM

Again, that all depends on the assumptions you're making in the overall propulsion system design and also on the mission requirements. I'm, in fact, inclined to agree with you for many foreseeable missions. At this stage, though, nobody can really do more than make an assertion or speak from their intuition or gut feelings. VASIMR can't really be fairly compared to HETs or LFAs because the latter are at much higher TRLs. Having started out in particle physics, I've seen the progress that's been made in the last 20 years on magnets and intense RF production. As a result, I firmly believe that you really can't rule out an eventual VASIMR that's as 'light' as the other devices.

- Eric.

Posted by Eric S. at September 8, 2004 07:42 PM

- Eric.

Posted by Eric S. at September 8, 2004 07:45 PM

Sorry for the double post... phone rang, I jumped, mouse got clicked twice. You know, that same sad story...

Posted by Eric S. at September 8, 2004 07:47 PM

Hi. I would like to learn some of the stuff you have been talking about. What books/websites would you recommend?

Posted by Yauco at October 26, 2004 06:15 PM

VASIMR should be used for now as the best choice for a propulsion system for a mission to mars. electrostatic engines are way too slow at accelerating, and chemical reaction propulsion systems are definitely a no. however, a NERVA is a possible system while we experiment with VASIMR.

Posted by at January 4, 2005 02:12 PM

Eric, it seems you are hanging up on power plants and ISP. The important factors to consider in any mission are how much delta V do you get for how much fuel do you use (Hall Effect will use more) and power you consume for a given payload.

The high thrust 'gear' of the VASIMR is essentially a helicon thruster on steroids, while the high Isp mode uses the microwaves to heat the fuel up massively.

The way I read Diaz' work, it seems he's selling the technology for the Mars mission to keep the research and development alive for a Jupiter mission, when VASIMR will really shine, and missions beyond (as someone said, going to Pluto). Higher Isp propulsion shows a profit the further you have to go.

The ship he's proposing for Mars would actually do quite well on a Jupiter mission in which the ship rendezvoused with comet Tempel 1, refuelled using comet ice, and bypassed the comet trajectory out to Jupiter, which the comet has a 2:1 resonance with. The refuelling would give this ship the delta v it needed to get around the Jupiter system and back out again, then refuel again on the way back to make the return leg to Earth.

Posted by MikeL at July 25, 2005 10:59 PM

I am confused. Isn't the higher ISP the better? I mean, its a measurement how effective a rocket is, how can it be "too high"?

Posted by A stranger at March 30, 2007 12:25 PM

Post a comment