VASIMR - Solar Powered?

Sax Russell · 2012-11-11 17:06:19

Bob Zubrin is big critic of VASIMR, one of his main points has been that VASIMR doesn't have a power supply. I was looking at wiki the other day, among other places, and I was seeing total solar systems in the range of 300W/kg - 700W/kg. That's a wide range, but the largest VASIMR engines currently proposed, the ones they are going to try on the space station, are 200kW. That's only 670 kg of solar arrays (and associated junk), even at the unoptimistic end of that range. That's not much, so why is it I am always hearing that VASIMR needs nuclear to be effective. It seems like solar electric works fine for the inner solar system. My basic question here is, what am I missing? Why is solar electric VASIMR written off?

They are going for 5N on the 200kW engine, which is capable of accelerating 1 ton 3.5 km/s in 8.1 days. This is certainly sufficient for smaller payloads, and even for large payloads, a number of such engines could be used, or a scaled up engine. The scaling from 50kW to 200kW apparently went from 0.5N to 5.0N, so, this seems promising. I will agree with Zubrin we don't NEED a VASIMR-type engine, but it certainly could be useful, especially in making Earth-Mars travel cheaper.

Note: This was the closest I could find to what I was seeing the other day, which was a NASA study of existing panels c. around 2007, but this is a NASA study for outer solar system missions. Even here you can extrapolate >105W/kg from the figures, for a complete solar system, including all the unexpected side changes you have to make. This changes the game some, not much though.
http://www.lpi.usra.edu/opag/nov_2007_m … _power.pdf

JoshNH4H · 2012-11-17 21:02:52

VASIMR is a great thruster for low-thrust, high power applications. Stationkeeping at the ISS is pretty decent, though it's worth keeping in mind that the full solar power system will probably have a higher mass than just the panels, for fairly obvious reasons (batteries for when the array goes into shadow, plus machines to control the voltage and various support structures).

I believe that Zubrin's primary reason for opposing VASIMR (And I agree with him on this to some extent) is that their claims for its capacity are way overstated. For example, their claim that it makes 39-day transits to Mars possible are simply not borne out by the math. I do believe that any reasonable power system will outweight the engine itself. In addition to fuel and payload this means that VASIMR is only useful for stationkeeping or very long mission times, which is to say low-value unmanned cargo missions.

Sax Russell · 2012-11-18 14:52:39

I don't really care about speed - the only real reason to be concerned there is radiation and time, and if you are planning on going to Mars, you have time, and radiation is not a major concern (as Curiosity recently showed). I am far more concerned with price/kg. For cargo delivery, any electric system, if the engine and power system are not too heavy or too expensive to develop, is going to be better than chemical rockets for this.

As I mentioned though, the power systems I was seeing did include the complete system. In the case of the study I posted, they included batteries, and everything you need on a Juno or Cassini like probe. The only difference I can think of would be the high voltage, which would not weigh that much. So,

Power systems: I saw several studies showing systems (structures, batteries, wiring, panels, everything, reaching 300-700 W/kg). The study I posted was looking at outer solar system, cold optimized panels, and they were showing 105W/kg at Earth orbit and definitely accounting for all the incidental needs produced by a large solar system.
Thrust: Wikipedia shows VASIMR as having 5N of thrust at 200kW. If delta-v to Mars is 3.5 km/s, 5N is enough to push 1000 kg of cargo to Mars transfer orbit in 8.1 days. Of course, multiple engines could be used, and, though some inefficiencies arise with longer acceleration periods, a 40 or 50 period is certainly acceptable, and would allow for 5-6 times as much cargo to delivered. Also, as I mentioned, the engine scaled nicely from 50kW to 200kW (10x thrust/ 4x power).

Overall: 5 200kW engines could get 25 tons of cargo to Mars. From what I was seeing, the power systems required would mass <10,000kg (based on a 100W/kg figure, which, due to the outer planet missions study I saw (http://www.lpi.usra.edu/opag/nov_2007_m … _power.pdf) seems to be a hard lower limit) and possibly as little as 1400 kg (based on 700W/kg, a figure I saw referenced for Earth orbiting satellites in a NASA paper from what I think was 2007 as well). Bottom line, the power systems are probably not the problem, and it, like all electric engines, will be more efficient at delivering cargo, which will be important for Mars colonization. I think it should definitely be looked into for cycler. Bottom line, we may not need it, but for it seems like a good investment, which means, for the little we are spending on it, we should continue, if not do more. Why are so many people bemoaning the need for light nuclear power systems to power the thing?

JoshNH4H · 2012-11-19 00:57:19

I think it's worth noting that if you're thrusting to Mars transfer over a period of eight days, the transfer velocity is going to be significantly higher. Just as importantly, the fuel, and later tankage and engines for chemical rockets can be shipped from the Moon, which has between proven to have volatiles. Alternatively but equally feasible is to get fuel, ranks, etc from Phobos or Deimos. Further, your pessimistic estimate of mass involved ten thousand kilograms of solar power power one thousand kilograms of payload. That is a problem. Even the optimistic estimate of 1400 kilos (plus tank mass, fuel mass, engine mass, payload shroud, and actual payload). Even in the optimistic scenario you have thrusting times (accounting for the higher mass and delta-V caused by the absence of the Oberth effect) likely around 50 days. In the pessimistic scenario, closer to 150 or more. You also have to ask about the relative cost of this. Shipping up solar panels us going to be more expensive than buying fuel on the Moon.

Sax Russell · 2012-11-19 21:37:26

First, it wasn't 10000 kg of solar panels for 1000 kg of cargo, it was, at the most pessimistic, 10000 kg to 25000 kg of cargo, and this included the complete power system, not the panels alone. Also, I think I can narrow it down a bit more now. This NASA project http://sbir.gsfc.nasa.gov/SBIR/abstract … SBIR_11_P1, is building 200-500W/kg panels. Assuming this, 350W/kg, is a reasonable estimate, we get 2900kg/MW of power, which is 25 N of thrust.

Also, you are right about low thrust engines requiring higher delta-vs (as I mentioned). I did not take this into account. If you can find accurate figures for this, I will use them. This page has some, but they are heavily contradicted by other numbers from throughout the internet. http://www.lr.tudelft.nl/en/organisatio … a-v-value/ But I will use 8 km/s to be conservative.

The Numbers:
(This is a long section, but please at least read the last few sentences)
So, let's say we use five ISS class 200kW VASIMR engines. The arrays weigh 3,000 kg and overall it probably weighs from 10,000-15,000 kg dry. Add 25,000 kg of cargo. Assume VASIMR has an exhaust velocity of 50 km/s and a thrust of 5.7N/200kW, which is 28.5N/MW. (http://www.adastrarocket.com/aarc/VX200). Using the Tsiolkovsky equation and a delta-v of 8 km/s, we get Mo/Mf = 1.173, so 15% of the mass is fuel. So, we have a 15,000 kg rocket producing 28.5 N of thrust. This is easily capable of carrying 25000 kg to Mars orbit with little fuel expenditure. Even assuming the 53,000 kg payload of a Falcon 9 Heavy, we can show that it takes ((53000 kg + 15000 kg * 1.17 (mass fraction)) * 8000 m/s)/28.5N = 260 days. A long trek, but acceptable for cargo. If you use a round trip delta-v of 16 km/s, you get a mass fraction (Mo/Mf) of 1.377, or 27% fuel by mass and takes ((53000kg + 15000kg * 1.377 * 8000 m/s)/28.5N = 303 days. However, now the engine and the panels can be reused, so we have a cargo cycler. The cycler can be fueled with argon in LEO and later argon from Mars at the end of it's trip and reused several times at the very least, if not more. Ideally, you would scale it up to larger scale, however, it is may be much more efficient than using chemical rockets for Martian cargo, especially for large payloads, which would require massive rockets. It is debatable which would be cheapest, but it is worth investing in. And it is definitely not the power requirements that make it problematic.

Also, fuel, tankage, and volatiles from the Moon? How much will it cost to develop that kind of lunar industry? Also, NEAs are better, closer targets for such resources. They have just as much water, and more heavy/useful metals, and they are not as far from Earth delta-v wise.

JoshNH4H · 2012-11-22 15:06:01

Firstly, I apologize for misreading your post. As much as I love my phone, I suppose it is less than optimal for posting with

Now, to VASIMR: 3.5 km/s is the delta-V assuming "infinite" thrust, meaning that the impulse is provided over the course of a few minutes so that the position (and thus gravitational potential energy) of the rocket does not change much. Any reasonable VASIMR system is better approximated by "infinitesimal" thrust, that being more or less the opposite. This will result in a transfer orbit where the cargo spirals slowly out from LEO*. Once the ship is in a solar orbit, the impulse will (hopefully) be more similar to an "infinite" thrust model due to the length of time that interplanetary transfers take. I believe that's mostly irrelevant, though. For the record, I will use a figure of 3.8 km/s for LEO to Mars transfer as presented on Wikipedia and the Atomic Rocket website.

The Oberth effect can be represented with a pythagorean triangle. In this case:

Where:

A is the velocity of a LEO, for the purposes of this discussion we'll call it 8,000 m/s.
B is the ΔV required to get from that LEO to an earth escape trajectory. I believe that this will be the same regardless of whether you thrust quickly or thrust slowly. If I am incorrect in this, the ΔV will be larger for low thrust and therefore this assumption is either right or inappropriately benefits VASIMR. The value of this ΔV is 3,200 m/s
C is the additional ΔV for a Mars Transfer Orbit given a high thrust transfer. 3,200 m/s + C = 3,800 m/s; therefore C is 600 m/s.
D is the additional ΔV relative to achieving escape velocity for a low-thrust transfer.

Using the Pythagorean theorem, D is equal to 3,700 m/s, indicating a total transfer velocity of 6,900 m/s for slow VASIMR transfer.

Part of the reason why PV is generally underrated for space applications is that the progress in PV has mostly occurred recently with government incentives and a focus on "green" energy. Though the most recent systems have yet to find application in space, they are a significant improvement over older systems. At this point, I think that for mass-limited power generation systems in the inner solar system PV power is probably the system of choice in an environment with constant insolation. (Louis, take note).

Anyway, between fuel, structure, engines, and panels the largest share of the composition of the ship will be engines and panels, which are expensive items, and further are more likely to fail than tank and fuel.

Wrt Lunar fuel production, the Moon has at least one benefit that no other body besides Earth has: Frequent launch windows to Earth orbit. While it may not have the absolute lowest ΔV, constant production and transport to LEO (GEO, L1, or wherever else it makes sense to send fuel from the Moon, that generally being places within Earth's Hill sphere) means that it is more economical to produce fuel on the Moon. The ΔV from lunar surface to LEO is 5.9 km/s but much of that could potentially be provided by a launch assist device or by something similar to a cannon. I would not rule out solar thermal either, potentially in the form of beamed, reflected solar energy.

My main point: Chemical rockets are basically just a tank and a combustion chamber. If you use pressure-fed rockets it is quite simple to use a single-stage methlox (there is almost certainly Carbon in the craters at the lunar poles) to get the propellant to LEO. Lunar fuel production might not be the closest to Earth in terms of
delta-V, but the frequency of launch windows is simply unbeatable and the availability of numerous different kinds of resources on the same body makes it a good target for the establishment of a fuel production facility.

*I should point out that given the radiation belts this presents some technical difficulties; further, it will either require a lot of calculation and possibly unpredictability or a quick burst of chemical to navigate the fuzzy gravitational boundary at the end of the Earth's hill sphere. Neither of these are insoluble but they are an annoyance.

Terraformer · 2012-11-22 18:44:35

I didn't have you pinned for a Loony, Josh...

If you have cheap fuel available, using high Isp drives for relatively low mission delta-V's becomes rather pointless. Who cares if you need to use 300 tonnes of fuel to deliver 100 tonnes of payload to Mars, when that fuel will only cost you a million or so dollars; pennies compared to the value of the cargo you're going to be launching.

I love Luna. Especially since we can do fancy mission tricks that allow us to halve the travel time for a manned mission to Mars if we bring Lunar fuel mining into play, as well as opening up the possibility of launching manned missions to Ceres. Not that I'd discount solar thermal, of course.

JoshNH4H · 2012-11-25 20:05:06

I'm most certainly not a loony. But if there's one thing that makes even less sense, it's electric propulsion for volume interplanetary transportation.

The reasons? Time spent in the radiation belt is one; many components will be destroyed by radiation. People will be too, plus the extra time it takes to boost would represent an unacceptable amount of time spent in transit. Extra time spent in transit also increases the risk of a solar flare while en route, which is a cancer risk. Likewise, the slow acceleration means that people at Mars will have to wait even longer, sometimes an unacceptably long amount of time, to receive whatever is being sent. Meanwhile, the VASIMR unit will still need significant amounts of fuel delivered to it at both ends, and because of the very high mass of engines and power generation simply comparing the mass ratios does not elucidate this. While a plain chemical rocket might have 5% of the fuel mass in structure and engines, it's going to be much, much larger than that for a VASIMR. This works out to mean a whole lot more fuel than one would expect.

For example, given 25 tonnes of cargo to Mars desired, a chemical rocket (Methane; Isp 3.7 km/s) will probably look something like: 25 tonnes of cargo, 3.5 tonnes of structure and engines, and 70 tonnes of fuel, for a total initial mass of 98.5 tonnes. A VASIMR (Having done more research, I have come to the conclusion that low thrust escape trajectories also have a higher delta-V, so I will use a delta V of 8 km/s for the round trip) assuming a Vex of 50 km/s (Isp ~5000), and 200 W/kg (this site says that, as of 2004, even that is in the realm of the untested; seeing as thin film has not flown yet and has some pretty difficult issues dealing with radiation, I do not feel comfortable going higher), as well as an engine specific power (suggested by this paper) of 830 W/kg. I will additionally assume a tankage penalty equal to 2% of the fuel mass. I will assume 210 days of thrusting (7 months!) to attain the mission delta-V.

What does our interplanetary VASIMR look like now? Well, we have 25,000 kg of payload, 3.5 tonnes of structure and engines, and 6000 tonnes of fuel, for a total initial mass of 34.5 tonnes. Of course, one must add 210 days to the travel time; And the 6,000 kg of Argon will have to be taken up to orbit from Earth because there are not significant reserves of argon on the Moon.

So: If you posit lunar propellant production the mass lifted from Earth ought to be roughly comparable for a VASIMR and chemical propulsion. However, a VASIMR unit will be significantly more expensive to produce and will take much longer to get from one place to another. I would also like to add that tankage can be used as a heat shield for aerocapture because presumably you want to reuse your rocket engine once you get to Mars. I would urge you to at least consider that it would be possible to manufacture pressure-fed rockets on the Moon as most of their components are relatively simple. It would also be very possible to do some kind of beamed propulsion or whatever to get the rocket and the fuel to lunar orbit (delta-V=1.6 km/s) followed by solar thermal or, I suppose, VASIMR, to get it to LEO. I favor chemical propulsion until a more robust space infrastructure, at least capable of repairing a VASIMR system if not building a new one, exists. That is to say, the foreseeable future.

GW Johnson · 2012-11-26 12:38:52

Not being a nuclear rocket engineer, I do not know what the design of a "water NERVA" might have in the way of performance. I know there is a square root of molecular weight decrement of Isp for the heavier water. But, the far higher heat capacity of the steam, even as it ionizes, should allow a lot more reactor power and thus a higher effective "chamber" temperature. An LH2 NERVA, if resurrected from the old plans, would be around 900-1000 sec Isp at engine T/W 3.6. A wild guess for a "water NERVA" might be 600-700 sec Isp at similar engine T/W. Maybe the "atomic rocket" site has data on this, maybe not. I dunno.

The thing is, our probes and observations are telling us that there is water available almost everywhere. All you have to do is dig it up, melt it, and filter out the dirt (mostly just a settling tank). It probably does not matter very much if it is contaminated with admixtures of ammonia, methane, and/or carbon dioxide. You have readily available "fuel" everywhere you want to go, and in a form (water or ice) that is easily stored and transported, easier than any other imaginable propellant.

Try those numbers (600-700 sec) and engine T/W = 3.6 in your mission calculations, knowing that vehicle T/W's way above 0.1 are very easily possible. This is impulsive-burn stuff. There is no low-acceleration effect on velocity increments or on transit times.

My educated guess is that the original NERVA could be resurrected and made flight-ready in about 5 years. About 5 years or so after that, there could easily be a water NERVA made ready by the same team. That's about 1 decade or so to some very practical interplanetary propulsion, even for manned vehicles. Water for propellant is just about everywhere.

Then, dream about a gas core nuclear thermal engine running on water propellant. 1500-6000 sec Isp at engine T/W from 4-ish on down to maybe 0.01 at the high-Isp end. "Filling stations" nearly everywhere you go. And the water NERVA (or its gas-core offspring) can be your reusable landing boat propulsion, too.

Working on atomic rockets looks to me like it holds a lot more promise than an SLS made of recycled shuttle technology.

GW

Sax Russell · 2012-11-27 00:54:05

I am short on time, so this will probably be a short post.

I agree with you on the refueling in space. If it can be done, it could be much cheaper than a more expensive low thrust drive (unless you can get the fuel for the same in space as well, which might be worth looking into). You do have to consider the fuel source (which I will come back to). It's also a nice near term solution which develops some off world industry in the course of solving a problem, which is good.

On the Oberth effect - if you could explain the Oberth effect and the triangle more in the future, it would be helpful. I believe what it comes down to is the energy difference of a delta-v at a high velocity is greater than at a small velocity, because K = 0.5mV^2. But I may be wrong.

For radiation - the extra time spent in the belts is a problem, but, like you said, a small chemical burn can solve that. Even some large-retros should be enough to solve it for cargo. As for radiation concerns for humans, I thought Zubrin already killed that worry. If you're going to Mars, you're probably willing to take some risks. It is not optimal, but it's really not likely to reduce your lifespan a lot. I would be a bit more concerned about red lung disease among early colonists.

On escape velocity - I couldn't help but correct, but certainly the delta-v required should not vary if you depart from the same orbit. Escape velocity is, after all, simply an expression, in terms of kinetic energy, of the depth of the gravity well, which should not depend on using a high-thrust or low-thrust engine. I think this would bring the 8 km/s delta-v you were using down to 6.7 km/s (as you previously showed).

On your VASIMR calculations - how many engines and how much power were you using (a single 200kW). About argon, it is not present on the Moon, but it in the atmosphere on Mars, and on Earth (not sure how you could harvest the latter, but it could be doable).

Now, about the Moon. I don't have time to lay out everything I have to say on the moon, but let's examine the choice between NEOs and the Moon for fuel and general development. The asteroids possess most heavy metals, with rare-Earth metals in spades, even among the stony ones. Carbon compounds are also present on the asteroids. While some may be present on the Moon, we don't know how much or in what form. Though the LCROSS results did show CO and CO2 (http://lcross.arc.nasa.gov/observation.htm) which may be promising. Carbonaceous chondrites have pretty much all kinds of organics, and stony chondrites do have carbon, albeit in very low (0.3-2% by weight) concentrations (again, the Moon probably does little better, and accessibility is hard to compare). This is important because it is possible we may need to manufacture CH4 (example, Musk's Mars rocket). However, the asteroids have something to drive their development - rare Earth metals. The Moon, on the other hand, does not. Moreover, if one is to mine the moon robotically, it offers a slew of challenges. Asteroid mining, as planned by groups like planetary resources, on the other hand, is intended to be a robotic affair. The presence of organics on asteroids (and certainly on carbonaceous NEOs, which should be accessible by the same technology as the stony ones containing rare earth metals) as well as water makes them great fuel stations that can pay for their own development. The Moon will have a much harder time doing so. Also, the technological challenges are greater. On the Moon, you will have to not only develop mining equipment and processes, but also land them with precision, have them operate in -200 C cold, have them perform chemical operations to manufacture fuel, and have them export this fuel to Earth using either reusable rockets, perhaps based on something similar to Musk's Falcon 9 reusable, or have the industrial base to build rockets. Now, doing all this robotically is certainly doable, and it is aided greatly by the opportunity to use telerobotics, but it is much more difficult than mining NEOs. And doing either with humans is almost not an option. The costs would be astronomical and make the cost of fuel produced, while still advantageous, not the revolutionary change we need. The NEOs can pay for their development, the Moon cannot. That is a big deal.

Sorry GWJohnson, I will respond to your comment next chance I get.

Last edited by Sax Russell (2012-11-27 00:56:00)

JoshNH4H · 2012-11-28 22:26:05

GW- The Isp of a nuclear-water rocket would probably be somewhere around 400-450 s, give or take. However, water decomposition does have to be taken into account. Obviously, water decomposes into hydrogen and oxygen. At high temperatures, you will have significant amounts of both (as well as a certain amount of their monatomic forms) in the engine. While this does decrease the mean atomic weight (Assuming that the diatomic form is prevalent, the mean atomic weight of dissociation products will be 12 amu, compared to 18 amu for water; for the monatomic dissociation products the mean is 6 amu.) However, then you also have both hot hydrogen and hot oxygen in the engine. This problem is exacerbated when there is a higher proportion of the monatomics. This makes engine design harder because you have to fine some way to defend against attack both by chemical oxidation and reduction. However, neither of us were talking about NERVA. Refueling from the Moon is potentially an option for NERVA; I prefer Ammonia as a fuel for its higher Isp and because N2 molecules tend not to break up.

Sax, I will certainly address your post at a later time. I'm pretty busy right now but you have a lot of salient points which I would like to address.

JoshNH4H · 2012-11-28 22:26:51

Oh, also welcome to Newmars and I'm pleased to meet you! I hope to have many interesting and productive conversations in the future

GW Johnson · 2012-11-29 09:05:49

What got me started was the problem of extreme low vehicle thrust/weight (acceleration) that seems to be inherent with VASIMR and all the other electric propulsion devices. When you include the power supply for a very long "burn", the engine system T/W seems to be a very tiny number: N's of thrust vs tons of equipment. You never get to climb out of a gravity well rapidly that way, so mission delta-vees get increased, since you have what amounts to a huge gravity loss term.

For cargo-only, long transit times are no real problem, but they certainly are with men. Either you pack a whole lot more life support supplies, or you figure out how to provide (reliably) a closed recycling ecology. Both seem to be very heavy items, adding greatly to dead-head payload. This compounds the departure weight exponentially, high Isp notwithstanding. That's why I think the cargo can be sent with some-or-all electric propulsion, if advantageous, but probably not the men. For them, you need high-thrust / high Isp propulsion. We gotta fly much faster with men.

Thus, from a zero-order standpoint, what one should focus on for future manned fight developments should be devices with high-thrust / high Isp potential. I know of only two that exceed current chemical propulsion, and which could actually be built and tested quickly (less than a decade). Those are solid-core nuclear thermal, and nuclear pulse propulsion. Both have downsides: the nuclear thermal we knew as flight-ready 4 decades ago has less T/W and Isp than we would like, uses a propellant with availability and storability issues, plus has real safety issues with retained reactor cores; while nuclear pulse propulsion seems to work best at gigantic ship sizes for which we as yet have no use (we aren't ready to do major colony flights yet).

Yet, those two propulsion items seem to me to be the place to start. I would like to see NERVA put into service ASAP with a quick follow-on to a simple water NERVA. I'd bet a talented team could get both T/W up above 5 and Isp up above 500 sec in a very few years with a water NERVA. By the time they get that working, I'd bet they would have a path to a gas core version identified. There's the real potential for much higher T/W and Isp out of nuke thermal (think engine T/W way above 10 and Isp way above 2000 sec, although these are probably inversely related). The pulse propulsion needs refinement and update to something beyond 1950's fission technology. That was over half a century ago. We can do better, and quickly, too. We just need a good place to test stuff like this, and that's the moon.

Nuke propulsion tests, space/planetary science, and possible mining-for-profit; there's some good reasons to plant bases on the moon. No one, but all 3 together, make a pretty good case. And we can get there with rockets we already have, or will have soon.

We should have been doing this all during the last 4 decades after the Apollo landings, instead of poking our heads in the sand the way we did. Every decade lost not doing these things was another decade's delay getting men to Mars and other places. (20-20 hindsight is wonderful, ain't it?)

GW

Last edited by GW Johnson (2012-11-29 09:09:40)

Sax Russell · 2012-12-02 22:20:17

GW Johnson, I should have time to respond to your post(s) now.

First of all, your technical arguments seem sound, my main point of disagreement would be political and cost-wise.

Nuclear engines (and chemical engines for that matter, albeit to a lesser extent) do have the advantage that they can refuel anywhere with pretty limited infrastructure, and I would agree with you, high thrust, high-Isp is ideally the way to go, but at present one of the only ways to do this is nuclear (I am sure I am missing some very important propulsion systems here). For the near future, nuclear seems like a bad choice for me. Current politicians don't seem to support manned spaceflight to begin with, and the NASA bureaucracy and even larger aerospace companies (to a lesser extent) seem frightened of nuclear. I don't really want to convince the space community that nuclear is the way to go, because even if it is, space advocates are a small group, and I doubt our collective yearnings will push policymakers. So, if we are going to push nuclear, I think it is sort of a Dorsa Brevia moment - we need to decide on the desired method of propulsion (the goal) but we really can't even think about that without addressing the means of making it happen.

The fears over NERVA are not entirely poorly founded. Just as importantly, the fears over nuclear in general are irrational in both strength and breadth. It seems everyone has qualms about it - it is not politically correct to feel otherwise. How this came to happen, I am not sure, but it is a problem that must be dealt with. If we want nuclear, we will have to have a strong, widespread push asking Congress - why not? Why is it dangerous? Is it dangerous to the budget? Why not cut some defense budget to open space to economic development, colonization, and exploration? It will take a big push, and I think any discussion of it needs to expand to include that. It may be doable, but we should not put all of our eggs in the nuclear basket.

Cost-wise, we should avoid confusing mass and delta-v with cost - they are correlated, but not absolutely related - and the key number is definitely the cost. At the end of the day, the $/kg to destination is, imho, the measure of a propulsion system. Launch window frequency, travel time, etc. are also important factors, but I think we can all agree $/kg is what makes or breaks the big-time development of space in the near term.

For instance, you mentioned the low T/W ratios of VASIMR and electric engines increasing delta-v, which does happen - but for a sufficiently high isp engine, a doubling of delta-v doesn't matter much. Your fuel load increases, you have to carry more life support for manned trips, but if you can reuse any of the machinery associated with that, that is all chump change. The key is - can the engine be refueled and reused. Can the same be said for any machinery you bring along? I see chemical, electric, and nuclear all being capable of this. Chemical we have the most experience with, and it can be refueled anywhere in the solar system, with somewhat more difficulty than nuclear if you want to use, for instance, CH4 instead of H2. Electric we have little experience with, so they may be expensive, but it has great Isps. Unfortunately, the noble gases are on Earth and Mars, not in space, so refueling in orbit may be difficult. However, they use very little fuel so this may not be an issue, and it may be cheaper than refueling chemical rockets (that is a whole other topic, which it would be nice to have a thread on). Nuclear is the end all, be all, because of the capability for high isp and high thrust/weight, but the political challenges are difficult. That is a war to be fought - rationalists against irrationalists - and one that with sufficient forces is hard to loose. So, I concede that electric may not be the best method of propulsion. However, what advantages does nuclear possess that refueling chemical in orbit does not, in terms of cost. Nuclear will have to be redeveloped, at great cost, and the issues of reusability from is still there from Earth to LEO. If that problem can be solved for nuclear, it can be solved (VTOL, probably) for chemical rockets. Once in orbit, their advantage is primarily in terms of Isp. The refueling locations for chemical and nuclear are mostly the same, but we have less development cost and a nearer-term implementation. Overall, the cost is the cost of development + the cost of the booster + fuel / how many times you reuse it. If interplanetary boosters could be reusable chemical or nuclear rockets, the cost would plunge, just as for launchers. For the near term though, it seems to me that the cost of developing a reusable interplanetary chemical booster would be basically 0. Nuclear could cost a lot. It seems to me that the cheapest option is to develop cheap launches and/or asteroid mining to get cheap fuel to orbit and then use a reusable chemical booster, which would not have to be much different from current boosters, it could just circle around and return home. It seems like it would just require bigger, longer fuel tanks to store the extra fuel. That shouldn't matter if you are going to reuse it, and if the fuel is cheap, both of which could be true.

As for testing propulsion on the Moon. That would be extremely expensive. Given how little money NASA has, I don't think we could do that. As for mining-for-profit, what exactly would we mine on the Moon that we couldn't get from NEAs? What is there, other than He-3, that is valuable enough to bring back to Earth? Like I said in my last post, the Moon seems a very costly diversion from developing the technologies that will allow us to begin working towards those kinds of big goals. It seems like an economic backwater - launch windows aside, as JohnNH4H brought up.

Also, just want to check - are we all agreeing now that VASIMR would not require nuclear reactors? That was the original post, though I am pleased with how this discussion has diverged from that.

And thank you JohnNH4H, by the way.

GW Johnson · 2012-12-03 10:53:35

I think you could do a solar VASIMR, but I'd bet it ends up about as heavy as a nuclear source VASIMR, by the time one builds arrays to generate power in the dimmer sun out near Mars. It gets worse the further you go from the sun.

I don't see chemical VASIMR as a viable option. You burn the same propellants either way, why not get the 99% thermal efficiency of a high-thrust chemical rocket engine instead of the Carnot efficiency (in the 30-40% range) of an electric power plant, and for low thrust that increases the delta-vee required?

I was hoping VASIMR might be a breakthrough in terms of thrust for the power, but it is not. It is instead just a very practical arrangement to build an electric thruster. It will have about the same propulsive characteristics as are typically calculated for any of the electric items.

What VASIMR and the other electrics need is a high-power power supply that is lightweight. That's the needed breakthrough. Hasn't happened yet, to my knowledge.

GW

Terraformer · 2012-12-03 19:24:28

Beamed microwave power, perhaps? When not in use for powering spacecraft, we can sell that extra electricity that the solar array is putting out.

I still think chemical shouldn't be discounted any time soon. 3 months to Mars using a Terra flyby, departing from EML1, doesn't sound that bad, and that's with a reasonable mass ratio. Add in Arcjets using beamed microwave power...

Sax Russell · 2012-12-03 23:15:02

Why would you think it would end up being as heavy? For a 1MW VASIMR, we have been finding a few tons, at most, at Earth orbit, and, more importantly >200 W/kg in terms of power/weight, which is better than the small nuclear reactors we currently have (I don't remember the figures, but I remember them being very small). Even at Mars orbit, you can only divide that 2.25. It seems that shows that solar is simply better than small nuclear reactors we have, and is continuously developing already, without any additional expenditures. More importantly, it seems to me that the weight of the engines and fuel is as large as that of our power supply, so I don't think the needed breakthrough is in power system weight, I think the question of one of which is better - chemical rockets with fuel depots, or electric engines with solar - and though it seems to me that the former may be better, I don't think the power system is the deciding factor.

And I didn't mean chemical VASIMR - that's just a bad idea.

Terraformer, beaming seems like a good idea, although the infrastructure might be expensive. Although I think I remember reading that a very large (and heavy) antenna is required to beam much power, so that might defeat the purpose it is bad enough. Perhaps laser power beaming, though? Difficult, but possibly more efficient. Also, I doubt space based solar power can gain much of an advantage over Earth based power, even with launch costs as low as $100/kg, at least not with solar already at grid parity at many places on Earth.

Last edited by Sax Russell (2012-12-03 23:17:26)

Terraformer · 2012-12-04 03:57:24

It won't just be Terra that's in the energy market...

Having a big antennae doesn't matter if it's not on the spacecraft itself. I'm fairly certain the receiver doesn't have to mass much.

Sax Russell · 2012-12-04 14:09:41

Maybe in a few decades, but there won't be a massive off-world electricity market for quite some time, so it won't help much in the near term. And sorry, I meant rectenna (receiver). I remember reading that they had to be surprisingly large for a sizable amount of power.

Terraformer · 2012-12-04 18:08:13

Large doesn't necessarily mean heavy, though...

Anyway, I suspect it will still stack up well against solar electric, simply because both need a hefty solar array and the former allows you to leave it at home and use it for other missions.

Sax Russell · 2012-12-04 22:31:17

What I was originally thinking was apparently based on safe levels for humans for microwave transmission. Apparently, in 1978 design studies were looking at a 10km rectenna for 750 MW, so that's about 350m for a 1MW electric thruster or VASIMR, and that almost has to be heavier than 10,000 kg. Look under microwave transmission en.wikipedia.org/wiki/Wireless_power_transmission (I know, wikipedia).

Of course, in space we could definitely do more. Plus more recently, it looks like they've been trying a lot more concentrated beams. (see this paper http://www.sspi.gatech.edu/wptshinohara.pdf). So that could work.

Last edited by Sax Russell (2012-12-04 22:32:34)

Terraformer · 2012-12-05 19:08:31

Why would a 350m rectenna have to be over 10 tonnes in mass? I wouldn't expect it to mass greater than one tonne... ah yes, it's not length you're talking about, but diameter. Well, given that it's not going to be solid, due to the wavelength... plus, we can probably operate at higher power densities, since we're not worrying about frying people on the ground...

I do suspect at the moment that beamed power will be the best way to get high velocity spacecraft going. Once we've got a network throughout the solar system...

GW Johnson · 2012-12-06 10:53:44

If microwave is the preferred way to transmit electricity without wires, could this be done with a maser instead of a noncoherent beam antenna? That way, the collateral damage potential and the transmission losses would be a lot less, because the beam spreading is massively reduced, like with a laser. I don't know anything about maser technology, really, except that they built one right before they built the first laser, about 1960-ish.

GW

Rusakov · 2012-12-22 13:16:35

The main problem with powering VASIMR engines via solar is that you'd need a prohibitively large array to power it (assuming you put the array on the spacecraft and not use beamed energy). That implies current efficiencies. A company called Redwave is working on a nantenna based design for solar panels which could be far better than the best commercial cells today. If/when high efficiency nantenna based solar power becomes feasible, then you could reduce the array size to one-third of a traditional solar array. Still big, but much better than before, and potentially feasible.

Impaler · 2013-01-11 20:45:46

Sax (nice name btw ) I addresses a lot of this in my thread http://www.newmars.com/forums/viewtopic.php?id=6177 ware I detail why Solar-Electric Propulsion has already decisively won the contest to be the 'next' propulsion technology. All these people pining for Nuclear propulsion are living in a outdated sci-fi fantasy. Zubrin has utterly no patience for technology development and attacks everything that is not a big-dumb rocket that launched yesterday.

New Mars Forums

Announcement

#1 2012-11-11 17:06:19

VASIMR - Solar Powered?

#2 2012-11-17 21:02:52

Re: VASIMR - Solar Powered?

#3 2012-11-18 14:52:39

Re: VASIMR - Solar Powered?

#4 2012-11-19 00:57:19

Re: VASIMR - Solar Powered?

#5 2012-11-19 21:37:26

Re: VASIMR - Solar Powered?

#6 2012-11-22 15:06:01

Re: VASIMR - Solar Powered?

#7 2012-11-22 18:44:35

Re: VASIMR - Solar Powered?

#8 2012-11-25 20:05:06

Re: VASIMR - Solar Powered?

#9 2012-11-26 12:38:52

Re: VASIMR - Solar Powered?

#10 2012-11-27 00:54:05

Re: VASIMR - Solar Powered?

#11 2012-11-28 22:26:05

Re: VASIMR - Solar Powered?

#12 2012-11-28 22:26:51

Re: VASIMR - Solar Powered?

#13 2012-11-29 09:05:49

Re: VASIMR - Solar Powered?

#14 2012-12-02 22:20:17

Re: VASIMR - Solar Powered?

#15 2012-12-03 10:53:35

Re: VASIMR - Solar Powered?

#16 2012-12-03 19:24:28

Re: VASIMR - Solar Powered?

#17 2012-12-03 23:15:02

Re: VASIMR - Solar Powered?

#18 2012-12-04 03:57:24

Re: VASIMR - Solar Powered?

#19 2012-12-04 14:09:41

Re: VASIMR - Solar Powered?

#20 2012-12-04 18:08:13

Re: VASIMR - Solar Powered?

#21 2012-12-04 22:31:17

Re: VASIMR - Solar Powered?

#22 2012-12-05 19:08:31

Re: VASIMR - Solar Powered?

#23 2012-12-06 10:53:44

Re: VASIMR - Solar Powered?

#24 2012-12-22 13:16:35

Re: VASIMR - Solar Powered?

#25 2013-01-11 20:45:46

Re: VASIMR - Solar Powered?

Board footer