The VASMIR/"The Martian" tactic to getting to Duna or other planets

[Gkirmathal]

Which really is a shame Because developing such power sources, would generate the early 'steppingstones' which could open up the near future of human long term exploration of our solar system and would enable VASIMR type systems.

Speaking of propulsion advances. What do you guys think is more achievable on today's tech level for human exploration: closed cycle NTR systems, or VASIMR/plasma based propulsion systems?

Personally I hinge towards the first, but that is because we don't have well enough power sources/heat dispensation means jet.

[peadar1987]

The Hermes as depicted in The Martian seems to have been conceived entirely as a plot device.

It takes six months to send crewed spacecraft to Mars using existing chemical propulsion. Cargo can be sent on lower energy trajectories such as an eight-and-a-half month Hohmann transfer. Using VASIMR thrusters and huge nuclear reactors drastically increases the mass, development time, flight time, and cost of the mission, while minimising the time spent actually exploring Mars.

Current long duration mission plans call for a six month type one fast ballistic transfer to Mars, with a year and a half spent exploring the surface, and a six month transfer back to Earth. Martian resources are leveraged to provide propellants for the return flight, fuels for surface vehicles, and life support consumables. Assets are concentrated on the Martian surface to establish a permanent base.

By contrast, the mission depicted in The Martian spends a few weeks (at best) on Mars and provides the astronauts with negligible mobility and power, reducing their mission to an effectively worthless "flags and footprints" exercise. The base they establish during their short stay is soon abandoned, never to be used again. By employing "MacGyver" techniques, Mark Watney is able to use the technology and resources available to him to survive for a year and a half. In other words, he is able to survive a standard length Mars mission by adapting inadequate hardware and supplies to perform a predictably more capable long-duration mission that should have been designed that way all along...!

In the book, NASA admit that Mark Watney has provided a far better mission and far better science return than could possibly have been provided with their existing programs.

Hermes provides several benefits (basically all arising from the greater ISp of the VASIMR propulsion):

-Larger habitat for the Mars transfer leg.

-Larger delta-V margins. Greater ISp means you can build in additional delta-V without increasing the mass of the spacecraft to stupid levels. A chemical rocket would struggle to make a Sol 6 abort if it was needed. This capability can be built into Hermes.

You've got a point about the duration of the surface mission, but that's not really related to Hermes. There's no reason a long-term Mars mission couldn't just park Hermes in Mars orbit for a longer period of time.

[mikegarrison]

Which really is a shame Because developing such power sources, would generate the early 'steppingstones' which could open up the near future of human long term exploration of our solar system and would enable VASIMR type systems.

Speaking of propulsion advances. What do you guys think is more achievable on today's tech level for human exploration: closed cycle NTR systems, or VASIMR/plasma based propulsion systems?

Personally I hinge towards the first, but that is because we don't have well enough power sources/heat dispensation means jet.

Orion nuclear pulse jets are probably the most achievable with today's tech.

[Streetwind]

Speaking of propulsion advances. What do you guys think is more achievable on today's tech level for human exploration: closed cycle NTR systems, or VASIMR/plasma based propulsion systems?

Personally I hinge towards the first, but that is because we don't have well enough power sources/heat dispensation means jet.

You'd be surprised... after an issue that cropped up recently with Near Future, I did some math, and it turns out nuclear engines are far less exciting than I thought they would be. Zero-boil-off cryogenic tanks kill them to the point where hypergolics have better total dV unless you really improve the specific impulse of NTRs significantly. =/

[ainamogel]

Orion nuclear pulse jets are probably the most achievable with today's tech.

And... there is a KSP mod adding Orion pulse propulsion. It looks like it hasn't been updated in a long time, though.

Take care,

Ainamogel

[More Boosters]

You'd be surprised... after an issue that cropped up recently with Near Future, I did some math, and it turns out nuclear engines are far less exciting than I thought they would be. Zero-boil-off cryogenic tanks kill them to the point where hypergolics have better total dV unless you really improve the specific impulse of NTRs significantly. =/

I don't get it, isn't boiloff a bad thing? Or do you mean that KSP modders balanced hydrogen tanks in weight to make up for how they don't boil off?

[davidy12]

You can do it in KSP just fine, thanks to the absurdly overpowered ion drives and tiny solar system with tiny dV costs. Flying from low Kerbin orbit to Duna in 39 Earth days (156 Kerbin days) requires only ~1130 m/s (about 50 more than a Hohmann transfer). Doing it in 39 Kerbin days requires 5,150 m/s, assuming you aerobrake, and that's possible (if inefficient) even with a single-stage chemical rocket. With nukes or ion drives, it's effortless. Of course, that kind of aerocapture will probably incinerate your craft, so let's do it without: 11,900 m/s. That barely still works with nukes, and would be quite comfortable with a two-stage rocket (one chemical, one nuclear). Ion drives could still do it effortlessly though. Add in mods like Near Future, offering advanced electric drives and power sources to go with them, and that kind of dV would be considered "low-ish"

It's IRL that it's completely impossible, VASIMR yes or no. There is no power source existing today that's even within an order of magnitude of the power/weight ratio required to pull off such a feat - and nothing in that direction is being actively developed either.

1130 M/S? How, I always use MJ and advanced planetary transfer.

PS: This ranks as my most viewed thread ever!

[Streetwind]

I don't get it, isn't boiloff a bad thing? Or do you mean that KSP modders balanced hydrogen tanks in weight to make up for how they don't boil off?

It's not even a modding thing, it's a RL thing.

Boiloff is a bad thing, yes. In launch vehicles, we can largely ignore it. But if you want to go to Mars, you need to store cryogenic liquid hydrogen for 4-6 months, not counting preparation time in LEO. If you want to go to Mars and come back again, you need to store it for 2-3 years. Considering hydrogen is one of the most difficult to store substances known to man, requiring absurdly low temperatures to liquefy and being eager to just casually drift through solid metal walls while also corroding them, this is a bit of a problem.

NASA drafted up a proposal for its Mars Design Reference Architecture detailing a NTR driven vehicle with 925s Isp. The proposal featured "zero boil-off" actively cryocooled hydrogen tanks specced for the full return mission duration. The fuel-mass ratio of these tanks is 3.15:1 (that is, 3.15 tons of fuel are stored in tankage with 1 tons of dry mass). Compare this to typical fuel tanks in KSP, which are 8:1, and typical RL launch vehicle grade tanks, which can get as high as 20:1. You can see just how extremely poor the fuel-mass ratio of the LH2 tanks really is, simply on account of hydrogen's low density and the need to make the tank walls extra thick and sturdy and carry active crycooling systems with pumps and radiators and coolant lines and cooling fluid. If you used something like kerolox or methalox in a chemical engine, the Isp would be a lot worse, but your tanks would get away with much less dry mass. And then there's hypergolics... some of those mixtures do not need any special treatment to stay usable pretty much indefinitely. That's why the Apollo CSM and Lunar Module used the stuff. Storing cryogenic propellants for multiple days simply took too much mass and effort.

To give you an idea of how the dry mass influences the total dV available, there's an easy calculation you can make. If you just plug the fuel-mass ratio of the tanks into the rocket equation, you can get the mathematical maximum dV that a stage can achieve with those tanks, assuming everything else besides the tanks has zero weight. It's a completely unrealistic figure you'll never reach, but it's useful as a comparison tool between engines and/or tank types.

Let's assume that there is a hypergolic engine we can use with 320s Isp. That's what the Apollo CSM engine did. We can probably build something better today, but for the sake of simplicity, let's use that number. Let's urther assume that even the hypergolic tanks are built a little bit more sturdy than the 20:1 launch vehicle grade. Now, let's pit that engine against the 925s Isp NTR at various tank fuel-mass ratios, and see what happens...

NTR, zero-boiloff grade:

dV = 9.81 * 925 * ln(3.15) = 10,411.82

Hypergolic, KSP grade:

dV = 9.81 * 320 * ln(8) = 6,527.78

Hypergolic, half launch grade:

dV = 9.81 * 320 * ln(10) = 7,228.28

Hypergolic, three quarters launch grade:

dV = 9.81 * 320 * ln(15) = 8,501.11

Hypergolic, full launch grade:

dV = 9.81 * 320 * ln(20) = 9,404.20

See how slim the advantage is? Considering the Mars reference mission requires a total dV of about 6,5 to 7 km/s, you can see that even the puny 320s Isp hypergolic engine has a chance to pull it off if the tank dry masses are low enough to keep the full vessel's fuel-mass fraction suitably high. Yes, the NTR has of course still more headroom, and is perhaps easier to assemble in LEO with fewer launches; but the NTR hasn't even developed yet and would cost at least 3-4 SLS launches worth of money to get operational, if not more.

And if you managed to build a storable propellant engine with ~350s Isp, it would pretty much equal the NTR in the launch grade tanks case. Is that realistic? Well, ESA is currently developing an upgraded upper stage engine for its Ariane series, which would do 340s with a gas generator cycle. We likely could get even more with staged combustion if we tried. And let's not forget the sheer dry mass of a nuclear reactor either; that NTR would drag down the full ship's fuel mass fraction quite a bit, which removes any apparent advantage the math above still leaves it.

And a further thing: staging. The NTR concept is drafted up to be single stage (albeit with a third of its fuel in a drop tank that gets ditched after the first periapsis kick), since you really don't want to stage away your super-expensive NTR. Hypergolics? Pff, toss em. With a three-stage vehicle, you might get a total vehicle weight that is lower than that of the NTR concept.

And all of this not because the NTR is bad, but because the tanks are. NTRs are great, but driving them with hydrogen for interplanetary trips feels like a total dead end if you cannot push the Isp significantly above 1000s. I'd be really curious how a NTR would do if fed with something like ammonia for ~600s Isp, but I have no idea to what fuel-mass ratio you can build ammonia tanks that can keep their contents for multiple years.

Edited September 18, 2015 by Streetwind

[davidy12]

Is there a faster way to do it than an advanced transfer to planet burn with mechjeb?

[Streetwind]

Is there a faster way to do it than an advanced transfer to planet burn with mechjeb?

I have never used MechJeb, but to plan high energy transfers I use this transfer planner or its ingame mod version by TriggerAU.

[davidy12]

I have never used MechJeb, but to plan high energy transfers I use this transfer planner or its ingame mod version by TriggerAU.

What's the shortest time you've ever gotten to duna

[Streetwind]

Uh. Frankly, no clue. I've never pushed quite as aggressively as 39 Kerbin days, though. Usually it's just not necessary, since Duna is already so close.

Fast transfers get really interesting with KScale2 and Outer Planets Mod though...

[The Yellow Dart]

It's not even a modding thing, it's a RL thing....

Thanks for this post, it was a fun read and good info (even of we are well into off topic territory).

[davidy12]

No more like 90-190 days

[FleshJeb]

Is there a faster way to do it than an advanced transfer to planet burn with mechjeb?

Don't use the ideal dV window it gives you. One axis is transfer window, the other is transit time. I usually pick the cheapest earliest window, and don't worry about transit time.

For a brachistochrone transfer, you're going to have to fiddle with the nodes. Start with your tangent pointed directly at your target and see where that gets you for the dV you're packing.

[Raptor9]

I don't worry about transit time either. The bad thing (IMO) about high-speed transfers to Duna is it results in a steep approach angle into the Duna orbital path. If you approach at a more shallow angle into the Duna orbital path, it requires much less delta-V to slow down to insert into Duna orbit. Otherwise you'll shoot right through the SOI without enough time to slow down. Not too critical with normal LF+O rockets, but using electric engines like the Ion's, it becomes more of an issue due to the longer "burn" times.

If you're a more advanced player and can time an aerobrake or gravity capture just right, this might not be a problem of course.

[Archgeek]

It's not even a modding thing, it's a RL thing.

[LENGHTY INFORMATIVE TREATISE ON LH2 TANKAGE SUCKING]

That is terribly unfortunate, but due in part to recent exposure to the game High Frontier, I think I smell a work-around.

What of a single, very small (say a quarter or fifth the expected size, or just big enough for reasonable burn times) zero-boiloff active cooled cryotank, that is refueled by an internal process from a much larger tank of something much denser and far easier to store?

High Frontier abstracts fuel and money both to great stonking orbital tanks of liquid water -- if a ship had water as its main fuel store, it could use solar power to electrolyze what's needed to produce the hydrogen needed for a burn, pumping the gas through a cooling system to liquify en route to the cryotank. This would perforce generate plenty of waste heat, but that could in part be run around the fuel store to prevent the water freezing, then just dumped with a radiator. The resultant oxygen could simply be breathed by the crew.

Heck, I'm not sure how the efficiencies would work out -- since you get four times the remass for the same volume but around half? the Isp -- but one could even forego the nukage altogether (save perhaps as a secondary power source), and have a second cryotank to use the oxygen for good ol' LHOx propulsion, effectively flying on really indirect solar power -- ripping water molecules apart with it, then slapping them back together in the engine's combustion chamber.

Yipe, do not try to diagram that sentence. It is not a run-on, but it is very complex with too many clauses.

Edited September 18, 2015 by Archgeek
Missing bracket, bad spacing.

[selfish_meme]

If you have lots of nuclear power and water why bother even liquifying it.

[mikegarrison]

Basically, if you have enough power you can do pretty much anything.

[Streetwind]

That is terribly unfortunate, but due in part to recent exposure to the game High Frontier, I think I smell a work-around.

What of a single, very small (say a quarter or fifth the expected size, or just big enough for reasonable burn times) zero-boiloff active cooled cryotank, that is refueled by an internal process from a much larger tank of something much denser and far easier to store?

High Frontier abstracts fuel and money both to great stonking orbital tanks of liquid water -- if a ship had water as its main fuel store, it could use solar power to electrolyze what's needed to produce the hydrogen needed for a burn, pumping the gas through a cooling system to liquify en route to the cryotank. This would perforce generate plenty of waste heat, but that could in part be run around the fuel store to prevent the water freezing, then just dumped with a radiator. The resultant oxygen could simply be breathed by the crew.

Heck, I'm not sure how the efficiencies would work out -- since you get four times the remass for the same volume but around half? the Isp -- but one could even forego the nukage altogether (save perhaps as a secondary power source), and have a second cryotank to use the oxygen for good ol' LHOx propulsion, effectively flying on really indirect solar power -- ripping water molecules apart with it, then slapping them back together in the engine's combustion chamber.

Some interesting ideas there. Another variant would be to use the oxygen for a LANTR, as a kind of afterburner.

The problem I see with this is that you might not get enough hydrogen out of that. Water is 80% oxygen by mass, so only 1/5th of the mass you carry would be actual NTR fuel. A tank with a 20:1 mass ratio goes right back down to 4:1 with that, and there is serious extra dry mass involved with the intermediate cryogenic storage tank. Additionally, the NTR could not run very long or on very high throttle settings because of the lack of constant hydrogen throughput to cool it. In this case I believe a nice 460s Isp hydrolox rocket would beat the NTR handily, so you can make use of all the fuel. Though most engines run either fuel rich or oxidizer rich for their own reasons. Not sure how viable an engine is that runs a perfect stoichiometric ratio. AFAIK that can get absurdly hot, especially with hydrogen/oxygen combustion...

Still a fun thought exercise though. Makes me wish I had the necessary background to investigate it in more detail.

[Gkirmathal]

It's not even a modding thing, it's a RL thing.

<snip-very-informative-post-snip>

Good and interesting read Streetwind!

This indeed again shows 'we' still have a long way to go in the area of more effective cryogenics, effective vacuum heat dispensation systems and lighter composite materials to construct with.

Btw, wasn't the Jet Prop. Lab. investigating/designing&testing a new cryogenic LH2 tank, based of composite materials instead of Ti, that would, if I remember the article of a few year ago correctly, limit the boil-off substantially compared to more conventional cryo tank designs and used materials? (I'll see if I can find the article somewhere)

[Pixel of Life]

Done what, exactly? Long burns with a low TWR spacecraft? If so, I've done that quite often. My standard interplanetary transfer maneuver for such things is as follows:

1) Timewarp until a few days before the actual transfer window

2) Initial burn at the correct spot to increase eccentricity so I'll have a reference point for subsequent burns, and to make sure I'll eject at roughly the correct angle.

3) 4, 6 or 8 minute burns at PE to raise AP. Burn duration depends on altitude of PE (since long burns can take you on a trajectory through the atmosphere if you're not careful).

From here on it's just like any other transfer.

[davidy12]

Okay, the reason why I care about time is:

1.) I'm kind of impatient

2.) More importantly, even though I don't use LS, I like to think that they're (kerbals) are being bombarded by radiation so they need to get there fast. Okay I will admit, 39-50 days is a stretch, I mean like 100-180 kerbal days and successfully enter orbit around Duna.

[Streetwind]

If you have lots of time to plan ahead, try the so-called Mangalyaan Maneuver. It's one of the best ways to ensure you can do an ejection burn with even the weakest of propulsion systems, it requires almost no math to go through, and you get away without having to estimate or measure your ejection angle. It works by repeatedly periapsis-kicking your orbit at a specific date ahead of the transfer window where your periapsis is aligned with the sun, so you cannot possibly miss the right angle. Then you just wait until the actual transfer window arrives to burn the final few dozen m/s towards your destination. Details in the linked thread.

(Just keep your growing orbital period in mind while raising your orbit.)

[Archgeek]

Some interesting ideas there. Another variant would be to use the oxygen for a LANTR, as a kind of afterburner.

The problem I see with this is that you might not get enough hydrogen out of that. Water is 80% oxygen by mass, so only 1/5th of the mass you carry would be actual NTR fuel. A tank with a 20:1 mass ratio goes right back down to 4:1 with that, and there is serious extra dry mass involved with the intermediate cryogenic storage tank. Additionally, the NTR could not run very long or on very high throttle settings because of the lack of constant hydrogen throughput to cool it. In this case I believe a nice 460s Isp hydrolox rocket would beat the NTR handily, so you can make use of all the fuel. Though most engines run either fuel rich or oxidizer rich for their own reasons. Not sure how viable an engine is that runs a perfect stoichiometric ratio. AFAIK that can get absurdly hot, especially with hydrogen/oxygen combustion...

Still a fun thought exercise though. Makes me wish I had the necessary background to investigate it in more detail.

Ah, but water is over 14.1 times as dense as liquid hydrogen, so not only can you use lighter tankage, you can use a lot less of it, cutting dry mass something fierce. I'm not sure why NTR heat would be a problem, though -- the intent is to prepare enough LH2 for the desired burn duration, including reactor spin-down. Or in the case of the possibly more efficient LHOx (which could be run fuel-rich, leaving some O2 for the crew), you could just use regenerative cooling as normal.

EDIT: I see with insane expansion ratios, it's possible to beat 480s: http://www.alternatewars.com/BBOW/Space/Propellants.htm

That requires the engine run a bit oxygen-heavy, though, but it looks like we can still beat 470 with our split water. Score another for good ol' LHOx.

Dang it, this is making me want to run the numbers on this tomfoolery.

Edited September 19, 2015 by Archgeek
interesting new ISp data