Sign in to follow this  
sevenperforce

Ideal manned interplanetary spaceship

Recommended Posts

If you're going to build an actual spaceship -- a craft you can take to another planet as easily as a pirate could sail across the Mediterranean -- there are a few things to keep in mind.

  • Gravity. You're going to need artificial gravity if you want to be able to manage long trips, so you'll need to either have a spinning hab, or you'll need to spin the whole ship.
  • Power. You need high-thrust engines to get on and off of planets, high-impulse engines to make your transfer burns propellant-efficient, and energy to run your ship in transit. Trimodal nuclear thermal engines are your only real choice (a trimodal NTR has three modes: high-impulse, where low-density propellant is heated and ejected by the nuclear core, high-thrust, where LOX is injected into the propellant stream to increase thrust at the expense of impulse, and thermal-electric, where the circulation of coolant generates electricity).
  • Shielding. Your hab needs to be shielded from both solar radiation and the nuclear radiation of your engine(s).
  • Volume. You need a large internal volume to carry an enormous amount of fuel if you're going to be able to make a transfer, land, take off again, and head back. Granted, you'd use in-orbit refueling wherever possible, but you need the flexibility to make a round-trip to unvisited worlds.
  • Surface area. Although you need a large volume, you also need a form factor with a blunt-body surface area, allowing re-entry heating to be as minimal as possible. Further, large surface area will help with radiating heat away in space. Finally, a lifting-body shape will make launch and re-entry a bit nicer.

What, then, is the optimal shape and configuration?

Here you go.

perspective.png

Yep, it's a flying saucer.

The hab is located in the center column and is a single floor, preventing any unpleasant gravity gradients. Because the center is open, it allows windows to be embedded in the ceiling, which will feel more natural.

hab_internal.png

The hab is far more "open" overall than most designs, while still being well-protected from radiation and micrometeoroid strikes due to its location in the center of the ship. It will feel very natural to have the sky "up" and the ground "down".

hab_external.png

The hab is shielded by wrapping the ship's tankage completely around it. The triangular cross-section maximizes internal volume while also having the optimal shielding profile. Obviously, the entire ship rotates.

The ship is powered by six small nuclear reactors, feeding three linear exhaust nozzles:

reactors.png

Each of the three exhaust nozzles is capable of running on a single reactor, so you still have maneuverability even if you need to scram one or two of your reactors. The coolant cyclers and generators are also in this area. Only minimal shielding is necessary, due to the placement of the large internal tanks.

For on-orbit burns, the three nozzles all fire together, providing moderate thrust even at the highest operating impulse:

on_orbit_burn.png

During any such burn, there will be a slight misalignment of the apparent gravitational field, but it will likely be no more disorienting than standing on a train while it starts to move. It is also likely that in most cases, the ship will only "spin up" after its transfer injection burns, relieving this issue entirely.

Takeoff and landing use the same orientation as on-orbit burns, but with the injection of liquid hydrogen or another oxidizer to dramatically increase thrust at the expense of specific impulse:

vertical_burn.png

During takeoff, however, maintaining this thrust orientation would make drag losses altogether unmanageable. For this reason, the nozzles are able to change orientation in order to thrust backward during in-atmosphere climbs (for reaching orbit from Terra, Mars, etc.):

gravity_turn.png

Because the "flying saucer" shape is one of the only shapes which is capable of achieving reasonable lift in subsonic, supersonic, and hypersonic flight, this allows for the higher-specific-impulse burn to be used (when applicable) to achieve orbital velocities. Obviously, on worlds without atmospheres, liftoff would be purely vertical without any intermediate horizontally-oriented burn stage.

Re-entry uses the large blunt surface area underneath to dissipate heat passively (although active cooling could be used, in principle):

re_entry.png

Passengers are seated in the upper ring during takeoff and landing, both for gravitational orientation and for safety reasons. Launch abort escape would be achieved either through individual ejection seats or through the ejection of the entire upper ring using built-in thrusters. It is possible that the upper ring could also be configured to serve as an ejectable lifeboat in the case of an on-orbit accident.

control_ring.png\

Attitude control could be achieved either by vectoring the thrusters, or by venting coolant.

Placing the heavy engines at the outside isn't ideal, but given that this is going to be designed with enough structural integrity for powered landing, it shouldn't prove too problematic.

The hab would have nearly 8500 square feet of floor space under artificial gravity, with more than 30 times the pressurized volume of the Space Shuttle crew cabin. I'm estimating a nominal dry weight of 1200 tonnes. The body encloses enough space for 9800 cubic meters of tankage; using a dense propellant like hydrazine, this corresponds to 9,900 metric tonnes of fuel. With pebble-bed reactors giving a specific impulse of around 520 seconds, you get a lovely 11 km/s of dV. Enough to reach orbit as SSTO (if you use LOX-injection augmentation). Also enough to fly to the moon, land, take off, and come back.

Now to build the damn thing...

Edited by sevenperforce

Share this post


Link to post
Share on other sites

I would totally fly in this. The only problems are cost and the fact that it requires use of nuclear rockets. It would most likely be ok, but people don't like nuclear rockets for some reason so I don't think it will happen anytime soon. This ship as you describe it could get to orbit, and to other planets if it's already full in space. The only problem is refueling it in space. It would have to be justifiably cheaper to launch a new rocket with enough fuel and instead of just bringing that rocket to [destination], transferring the fuel into this ship already in orbit, and having that one go. 

Share this post


Link to post
Share on other sites
19 minutes ago, cubinator said:

I would totally fly in this. The only problems are cost and the fact that it requires use of nuclear rockets. It would most likely be ok, but people don't like nuclear rockets for some reason so I don't think it will happen anytime soon. This ship as you describe it could get to orbit, and to other planets if it's already full in space. The only problem is refueling it in space. It would have to be justifiably cheaper to launch a new rocket with enough fuel and instead of just bringing that rocket to [destination], transferring the fuel into this ship already in orbit, and having that one go. 

Nuclear engines are mostly a problem for takeoff from Earth. Expelling radioactive propellant into the atmosphere, even if it is nearly harmless, isn't something people are happy about. For interplanetary transfers, or for landing on other worlds, people won't care as much about the use of nuclear engines. The average layperson wouldn't have the first clue about the difference between an NTR and an RTG. 

Also, launches from Earth wouldn't be very common. This is, after all, a long-term, extended-persistence interplanetary transfer vehicle capable of supporting a large crew for a long time. The initial launch would be followed by a series of missions to other worlds, returning to low Earth orbit to refuel and take on more supplies, but "dry dock" on earth wouldn't happen very often. Much easier to treat the ship like an orbiting space station and simply ferry up and down using capsules.

It would definitely be more economical to refuel, even using inefficient chemical rockets, because this contains its own engines, tankage, and hab. This ship would be able to support a large crew for a long time, drop them on another planet, and then come back. Hard to beat that. 

Another option (not pictured) would be to attach a cluster of ion engines with their own fuel tanks, either passing through the open center of the ship or in a toroidal ring mated underneath. The ion engines would run off the nuclear reactors, which would continuously expel small amounts of coolant in order to keep from overheating. The internal fuel tanks would still end up partially depleted by the end of the burn, since the reactors would be using propellant as their coolant, but you could develop an extremely high delta V.  This eliminates some of the major problems with using ion engines for interplanetary transfer (the high power requirements of ion engines typically result in either extremely low thrust, extremely large solar panels, or extremely large radiators and coolant loops with nuclear-thermal designs). Plus, you arrive with enough propellant to still use your engines the normal way.

In theory, this ship could mate to any number of additional cargo containers, pods, or fuel reserves. Big, overpowered engines are a large initial investment, but they pay great dividends. Especially because a moon shot ends up being as simple as flying to the moon, landing, taking off, and coming back.

Share this post


Link to post
Share on other sites
7 hours ago, sevenperforce said:

If you're going to build an actual spaceship -- a craft you can take to another planet as easily as a pirate could sail across the Mediterranean -- there are a few things to keep in mind.

  • Gravity. You're going to need artificial gravity if you want to be able to manage long trips, so you'll need to either have a spinning hab, or you'll need to spin the whole ship.
  • Power. You need high-thrust engines to get on and off of planets, high-impulse engines to make your transfer burns propellant-efficient, and energy to run your ship in transit. Trimodal nuclear thermal engines are your only real choice (a trimodal NTR has three modes: high-impulse, where low-density propellant is heated and ejected by the nuclear core, high-thrust, where LOX is injected into the propellant stream to increase thrust at the expense of impulse, and thermal-electric, where the circulation of coolant generates electricity).
  • Shielding. Your hab needs to be shielded from both solar radiation and the nuclear radiation of your engine(s).
  • Volume. You need a large internal volume to carry an enormous amount of fuel if you're going to be able to make a transfer, land, take off again, and head back. Granted, you'd use in-orbit refueling wherever possible, but you need the flexibility to make a round-trip to unvisited worlds.
  • Surface area. Although you need a large volume, you also need a form factor with a blunt-body surface area, allowing re-entry heating to be as minimal as possible. Further, large surface area will help with radiating heat away in space. Finally, a lifting-body shape will make launch and re-entry a bit nicer.

What, then, is the optimal shape and configuration?

Here you go.

perspective.png

Yep, it's a flying saucer.

The hab is located in the center column and is a single floor, preventing any unpleasant gravity gradients. Because the center is open, it allows windows to be embedded in the ceiling, which will feel more natural.

hab_internal.png

The hab is far more "open" overall than most designs, while still being well-protected from radiation and micrometeoroid strikes due to its location in the center of the ship. It will feel very natural to have the sky "up" and the ground "down".

hab_external.png

The hab is shielded by wrapping the ship's tankage completely around it. The triangular cross-section maximizes internal volume while also having the optimal shielding profile. Obviously, the entire ship rotates.

The ship is powered by six small nuclear reactors, feeding three linear exhaust nozzles:

reactors.png

Each of the three exhaust nozzles is capable of running on a single reactor, so you still have maneuverability even if you need to scram one or two of your reactors. The coolant cyclers and generators are also in this area. Only minimal shielding is necessary, due to the placement of the large internal tanks.

For on-orbit burns, the three nozzles all fire together, providing moderate thrust even at the highest operating impulse:

on_orbit_burn.png

During any such burn, there will be a slight misalignment of the apparent gravitational field, but it will likely be no more disorienting than standing on a train while it starts to move. It is also likely that in most cases, the ship will only "spin up" after its transfer injection burns, relieving this issue entirely.

Takeoff and landing use the same orientation as on-orbit burns, but with the injection of liquid hydrogen or another oxidizer to dramatically increase thrust at the expense of specific impulse:

vertical_burn.png

During takeoff, however, maintaining this thrust orientation would make drag losses altogether unmanageable. For this reason, the nozzles are able to change orientation in order to thrust backward during in-atmosphere climbs (for reaching orbit from Terra, Mars, etc.):

gravity_turn.png

Because the "flying saucer" shape is one of the only shapes which is capable of achieving reasonable lift in subsonic, supersonic, and hypersonic flight, this allows for the higher-specific-impulse burn to be used (when applicable) to achieve orbital velocities. Obviously, on worlds without atmospheres, liftoff would be purely vertical without any intermediate horizontally-oriented burn stage.

Re-entry uses the large blunt surface area underneath to dissipate heat passively (although active cooling could be used, in principle):

re_entry.png

Passengers are seated in the upper ring during takeoff and landing, both for gravitational orientation and for safety reasons. Launch abort escape would be achieved either through individual ejection seats or through the ejection of the entire upper ring using built-in thrusters. It is possible that the upper ring could also be configured to serve as an ejectable lifeboat in the case of an on-orbit accident.

control_ring.png\

Attitude control could be achieved either by vectoring the thrusters, or by venting coolant.

Placing the heavy engines at the outside isn't ideal, but given that this is going to be designed with enough structural integrity for powered landing, it shouldn't prove too problematic.

The hab would have nearly 8500 square feet of floor space under artificial gravity, with more than 30 times the pressurized volume of the Space Shuttle crew cabin. I'm estimating a nominal dry weight of 1200 tonnes. The body encloses enough space for 9800 cubic meters of tankage; using a dense propellant like hydrazine, this corresponds to 9,900 metric tonnes of fuel. With pebble-bed reactors giving a specific impulse of around 520 seconds, you get a lovely 11 km/s of dV. Enough to reach orbit as SSTO (if you use LOX-injection augmentation). Also enough to fly to the moon, land, take off, and come back.

Now to build the damn thing...

Yeah, the takeoff almost certainly lacks enough thrust. As I said earlier, if your interplanetary spacecraft has enough thrust to take off by itself, it has too much thrust. This situation is worse because NTRs are lower thrust and MUCH heavier than Chemical engines.

The crew cabin is likely too small in diameter, and the Coriolis effect would be a major problem to its astronauts. Good luck using the upper ring as a LAS. Also, the NTR engines are really too small- remember that you need to carry a nuclear reactor in each of them.

 

From my point of view, THE JEDI ARE EVIL you should design this to transfer crew only, and use a seperate vehicle to get the crew and cargo into orbit. I would even just use a separate lander to land at the destination, unless the destination is low-G (like the Moon).

Share this post


Link to post
Share on other sites

You don't want the nuclear power source in the middle.

 

My concept for one is basically a hybrid.  It's got a long central truss it's all built around.  Nuclear engine in the back. Some of the fuel tanks are along the truss.  The rest of the fuel is wrapped around the life support modules, which spin inside their own lightweight pressurized housing (could be made of fabric) shielded by the fuel tanks.  

Basically you try to distribute the rad shielding in such a way that the shieldable dose from all directions is roughly equal.  And you use mainly fuel as the rad shielding, and secondary machinery, and other things that are not the dead weight of a lead plate.

Share this post


Link to post
Share on other sites
6 hours ago, SomeGuy123 said:

You don't want the nuclear power source in the middle.

Right, that's why I had it at the edge.

6 hours ago, SomeGuy123 said:

My concept for one is basically a hybrid.  It's got a long central truss it's all built around.  Nuclear engine in the back. Some of the fuel tanks are along the truss.  The rest of the fuel is wrapped around the life support modules, which spin inside their own lightweight pressurized housing (could be made of fabric) shielded by the fuel tanks.  

Basically you try to distribute the rad shielding in such a way that the shieldable dose from all directions is roughly equal.  And you use mainly fuel as the rad shielding, and secondary machinery, and other things that are not the dead weight of a lead plate.

The long, skinny axially-rotating ship suffers from two issues: first, it requires a high RPM to produce even relatively low gravity, and second, you introduce asymmetric bending moments in the axial linkage that can rip your ship apart.

Share this post


Link to post
Share on other sites

Landing this thing on any celestial body would be no trivial task...

You would have to rely entirely on thrust vectoring for lateral control during the final descent unless your attitude thrusters are SSME's 

Share this post


Link to post
Share on other sites
9 hours ago, fredinno said:

Yeah, the takeoff almost certainly lacks enough thrust. As I said earlier, if your interplanetary spacecraft has enough thrust to take off by itself, it has too much thrust. This situation is worse because NTRs are lower thrust and MUCH heavier than Chemical engines.

Using LOX augmentation with an NTR in a LANTR design can approximately triple the T/W ratio while only reducing the specific impulse by around 30%. Moreover, the low T/W ratios of NTRs are typically the result of using liquid hydrogen; a denser fuel (necessary in order to fit inside the internal fuel tanks) reduces ISP but does wonders for thrust. A pebble-bed NTR has a T/W ratio of around 20 using liquid hydrogen. Running the NTR on hydrazine instead should yield performance approximately comparable to using liquid ammonia, for a T/W increase of 252%.

Injecting LOX downstream should be done with a hydrogen:oxygen mass ratio of 1:4.83. Since hydrazine is 12.5% hydrogen by weight, the hydrazine:LOX ratio needs to be approximately 3:2, so this would mean the propellant flow would be 45% H2/LOX. Each kilogram of H2/LOX has a kinetic energy of 9.84 MJ, while each kilogram of propellant coming out of the NTR has a kinetic energy of 13 MJ. Do the math, and optimized LOX injection ends up increasing your T/W ratio by an additional 168% while only decreasing specific impulse by about 8%. So we're looking at an engine T/W ratio of around 82:1.

At that specific impulse, SSTO requires a GLOW of 6,335 tonnes. Lifting off vertically, with a gee of acceleration, is going to require 124 MN of thrust, corresponding to a mass of 26 tonnes for each engine. More than I would want; doable, but overkill.

The ship would likely need a reaction wheel in order to spin up without burning propellant. I wonder...if you used an electromagnetically-powered centrifugal impeller located inside the central column (near the bottom) to suck in and compress air and then add it to your propellant stream, would the added weight cost of the ducting be balanced out by sufficiently greater thrust augmentation and improvement in specific impulse? The impeller could serve as the reaction wheel in space.

Another option would be to put a fat booster inside the central column and use it to zero out gravity drag while (smaller) main engines did the horizontal impulse burn; the booster would drop out and return to the launch site on its own.

11 hours ago, fredinno said:

The crew cabin is likely too small in diameter, and the Coriolis effect would be a major problem to its astronauts. Good luck using the upper ring as a LAS.

From my point of view, THE JEDI ARE EVIL you should design this to transfer crew only, and use a seperate vehicle to get the crew and cargo into orbit. I would even just use a separate lander to land at the destination, unless the destination is low-G (like the Moon).

The crew cabin is 27 meters in diameter; you can get over half a gee with a nice leisurely 2 RPM. Nothing to worry about there. Using the command ring as a LAS is a stretch, but that's hardly a necessary element.

The purpose of enabling Earth launch is that I just don't want to go through the painful process of trying to assemble this thing in space. I suppose that strap-on boosters could be used for the initial Earth launch, although the vehicle won't mate to them very easily in any sort of aerodynamic configuration. If you eliminate the requirement of Earth launch, then your thrust requirements drop pretty substantially.

For landing and return, you don't really have to worry about the landing side of things because the landing thrust requirements are far lower than the launch requirements. Takeoff from Mars requires a GLOW of 2,900 tonnes, corresponding to a liftoff thrust of just 37 MN; with this arrangement each nuclear reactor only weighs 7.6 tonnes. And that's plenty of thrust to make a propulsive Earth landing with fuel = bingo; it's also more than enough thrust to fly directly from the lunar surface to LEO.

1 hour ago, KerbonautInTraining said:

Landing this thing on any celestial body would be no trivial task...

You would have to rely entirely on thrust vectoring for lateral control during the final descent unless your attitude thrusters are SSME's 

Well, yes. Thrust vectoring FTW.

Share this post


Link to post
Share on other sites
4 hours ago, sevenperforce said:

Right, that's why I had it at the edge.

The long, skinny axially-rotating ship suffers from two issues: first, it requires a high RPM to produce even relatively low gravity, and second, you introduce asymmetric bending moments in the axial linkage that can rip your ship apart.

Yeah, and a spinning ship with a centrifuge inside would need to be massive in order to prevent the coriolis effect from causing disorientation.

46 minutes ago, sevenperforce said:

Using LOX augmentation with an NTR in a LANTR design can approximately triple the T/W ratio while only reducing the specific impulse by around 30%. Moreover, the low T/W ratios of NTRs are typically the result of using liquid hydrogen; a denser fuel (necessary in order to fit inside the internal fuel tanks) reduces ISP but does wonders for thrust. A pebble-bed NTR has a T/W ratio of around 20 using liquid hydrogen. Running the NTR on hydrazine instead should yield performance approximately comparable to using liquid ammonia, for a T/W increase of 252%.

Injecting LOX downstream should be done with a hydrogen:oxygen mass ratio of 1:4.83. Since hydrazine is 12.5% hydrogen by weight, the hydrazine:LOX ratio needs to be approximately 3:2, so this would mean the propellant flow would be 45% H2/LOX. Each kilogram of H2/LOX has a kinetic energy of 9.84 MJ, while each kilogram of propellant coming out of the NTR has a kinetic energy of 13 MJ. Do the math, and optimized LOX injection ends up increasing your T/W ratio by an additional 168% while only decreasing specific impulse by about 8%. So we're looking at an engine T/W ratio of around 82:1.

At that specific impulse, SSTO requires a GLOW of 6,335 tonnes. Lifting off vertically, with a gee of acceleration, is going to require 124 MN of thrust, corresponding to a mass of 26 tonnes for each engine. More than I would want; doable, but overkill.

The ship would likely need a reaction wheel in order to spin up without burning propellant. I wonder...if you used an electromagnetically-powered centrifugal impeller located inside the central column (near the bottom) to suck in and compress air and then add it to your propellant stream, would the added weight cost of the ducting be balanced out by sufficiently greater thrust augmentation and improvement in specific impulse? The impeller could serve as the reaction wheel in space.

Another option would be to put a fat booster inside the central column and use it to zero out gravity drag while (smaller) main engines did the horizontal impulse burn; the booster would drop out and return to the launch site on its own.

The crew cabin is 27 meters in diameter; you can get over half a gee with a nice leisurely 2 RPM. Nothing to worry about there. Using the command ring as a LAS is a stretch, but that's hardly a necessary element.

The purpose of enabling Earth launch is that I just don't want to go through the painful process of trying to assemble this thing in space. I suppose that strap-on boosters could be used for the initial Earth launch, although the vehicle won't mate to them very easily in any sort of aerodynamic configuration. If you eliminate the requirement of Earth launch, then your thrust requirements drop pretty substantially.

For landing and return, you don't really have to worry about the landing side of things because the landing thrust requirements are far lower than the launch requirements. Takeoff from Mars requires a GLOW of 2,900 tonnes, corresponding to a liftoff thrust of just 37 MN; with this arrangement each nuclear reactor only weighs 7.6 tonnes. And that's plenty of thrust to make a propulsive Earth landing with fuel = bingo; it's also more than enough thrust to fly directly from the lunar surface to LEO.

Well, yes. Thrust vectoring FTW.

Yeah, as I said earlier, if your interplanetary spacecraft can land and take off, you're over-engineering your spacecraft. It's really like SSTOs in that way, that it's not wholly practical.

Share this post


Link to post
Share on other sites
33 minutes ago, fredinno said:

Yeah, and a spinning ship with a centrifuge inside would need to be massive in order to prevent the coriolis effect from causing disorientation.

Yeah, as I said earlier, if your interplanetary spacecraft can land and take off, you're over-engineering your spacecraft. It's really like SSTOs in that way, that it's not wholly practical.

RPM is the primary factor in play for whether a crew will become disoriented in centrifugal gravity, and an RPM of 2 is well within acceptable limits.

There are a lot of logistics considerations surrounding the design of an interplanetary spaceship. There are lot of variables that depend heavily on the mission profile, and if you want to build an extended-persistence reusable spaceship, it needs to be capable of accommodating a wide range of mission profiles.

Unless you have a fully operating orbital spaceport at every possible destination (which won't happen for a LONG time, if ever), then you will need to bring along a lander. Perhaps you can expect that certain destinations, like Mars, will eventually be able to provide launch services to return a crew capsule to orbit; if this is the case, then you can get away with bringing along nothing more than an aerobraking descent capsule.

But aerobraking is only useful on Mars, Earth, and Titan, so a mission to any other world will need to bring along its own descent engine and fuel supply. The lander will probably need enough fuel for ascent as well, since you can't really expect to have refueling capacity on most worlds. This means your mothership needs additional fuel to tow the lander, plus heavier engines to compensate for the increased mass. If the lander is going to stay down for any appreciable period of time, it also needs its own separate hab and life support, which means more dead weight, more fuel, and even bigger engines on your mothership. Since resupply would primarily be via Earth launch, this has a tendency to make your costs skyrocket, since the landers and their engines and their fuel PLUS the added fuel for the transfer vehicle all have to be orbited before each mission. To save immediate launch costs, you'll design each lander for each specific mission, which means they might as well be expendable, which means more down time and fewer overall missions.

It rapidly becomes apparent that for the case of something like a mission to Ceres or the Moon, you're really going to come out a lot better if you use just slightly larger engines and simply land the whole ship. Do that, and you no longer have to worry about the lander with its separate engine, separate fuel supply, separate life support system and hab, and separate structure. Resupply is simply a matter of orbiting fuel and supplies, standardizing and streamlining the resupply process.

This is good, because larger engines are also important if you have even the slightest inclination of ever doing extended missions. One major point of an interplanetary spaceship with artificial gravity is to allow missions beyond the asteroid belt; this means it will need to carry more supplies and more fuel (all probably mated externally), all requiring more thrust if you want to get moving with any sort of haste.

 

But here's where designing for mission flexibility starts to drive you toward something like my design. If your mission profile for a lunar landing requires you to land and take off, then you really need enough takeoff thrust to be able to make the trip all the way back to LEO. With your "high-gear" specific impulse of 520 seconds, getting the 4.8 km/s of dV for the LLO-LEO transfer will require a propellant mass fraction of 61%. Getting from the lunar surface to LLO with your low-gear specific impulse of 478 s, then, requires a propellant mass fraction of 29%. Since the dimensions of the vehicle are set by the need for artificial gravity, and these generally dictate dry mass, getting off Luna with a standard gee of acceleration will require a liftoff thrust of 49.3 MN.

Since it turns out that 49.3 MN is more than enough for a powered takeoff from the Martian surface into LMO, it only makes sense to give it an aerodynamic shell and aerobraking capacity, even if that wouldn't be the typical mission profile. And a craft with an aerodynamic shell and aerobraking capacity which can SSTO from Mars can most certainly manage an empty-tank propulsive landing on Earth.

Share this post


Link to post
Share on other sites

I'm fine with flying saucers, but I think the mothership philosophy works best, here.

My question is if we can have separate landers. 

Basically make this an entirely vacuum ship, except for maybe aero braking...

Share this post


Link to post
Share on other sites
9 hours ago, Bill Phil said:

I'm fine with flying saucers, but I think the mothership philosophy works best, here.

My question is if we can have separate landers. 

Basically make this an entirely vacuum ship, except for maybe aero braking...

The mothership can certainly tow along any number of other vehicles. Of course, those vehicles will need to be launched separately from Earth, which is a substantial cost.

Logistics starts to play a part, however, when you look at the necessary requirements for your mothership. It already needs to have enough structural integrity to provide artificial gravity by rotation. Thus, slapping a lightweight aeroshell around it so that it can aerobrake for orbital insertion from a transfer won't really incur much additional weight cost.

While towing along a separate descent/ascent vehicle is fairly reasonable, it doesn't really make sense with landings on airless worlds. It is cheaper to simply carry extra fuel to deorbit/reorbit the entire ship them it would be to carry a separate landing vehicle and ascent vehicle. So it only makes sense to equip the mothership with powerful enough engines to support descent and ascent on airless worlds all by itself.

And once you have both aerobraking capability and lunar SST-LEO capability, you might as well give it Martian landing/ascent capability. 

Edited by sevenperforce

Share this post


Link to post
Share on other sites
6 hours ago, sevenperforce said:

The mothership can certainly tow along any number of other vehicles. Of course, those vehicles will need to be launched separately from Earth, which is a substantial cost.

Logistics starts to play a part, however, when you look at the necessary requirements for your mothership. It already needs to have enough structural integrity to provide artificial gravity by rotation. Thus, slapping a lightweight aeroshell around it so that it can aerobrake for orbital insertion from a transfer won't really incur much additional weight cost.

While towing along a separate descent/ascent vehicle is fairly reasonable, it doesn't really make sense with landings on airless worlds. It is cheaper to simply carry extra fuel to deorbit/reorbit the entire ship them it would be to carry a separate landing vehicle and ascent vehicle. So it only makes sense to equip the mothership with powerful enough engines to support descent and ascent on airless worlds all by itself.

And once you have both aerobraking capability and lunar SST-LEO capability, you might as well give it Martian landing/ascent capability. 

Except that building the mothership is already a substantial cost of both of our plans, and mine is cheaper than yours, since the mothership never touches an atmosphere. Of course we could launch it from Earth, if we design it right. But the great thing about space is that you can have huge mass ratios. The aero shell limits your geometry, and if you want to take full advantage of a vacuum, you can design your ship to survive a few gs, and only put a fraction of a g on it at any time. This increases mothership delta V. Now you can just about go anywhere in the solar system. At least, planets. 

Carrying separate landers just makes sense, especially for larger vessels. Then less energy is spent on a landing. Sure, low gravity bodies can be landed on, but there's still very little reason to do so with a huge mothership. It takes even more energy to do that.

Okay, let me put it like this: the extra fuel to land on those bodies will be more massive than the individual landers you can use, and you won't use that Delta V until landing. So why waste that mass?

Share this post


Link to post
Share on other sites
On 2/24/2016 at 1:43 PM, sevenperforce said:

If you're going to build an actual spaceship -- a craft you can take to another planet as easily as a pirate could sail across the Mediterranean -- there are a few things to keep in mind.

  • Gravity. You're going to need artificial gravity if you want to be able to manage long trips, so you'll need to either have a spinning hab, or you'll need to spin the whole ship.
  • Power. You need high-thrust engines to get on and off of planets, high-impulse engines to make your transfer burns propellant-efficient, and energy to run your ship in transit. Trimodal nuclear thermal engines are your only real choice (a trimodal NTR has three modes: high-impulse, where low-density propellant is heated and ejected by the nuclear core, high-thrust, where LOX is injected into the propellant stream to increase thrust at the expense of impulse, and thermal-electric, where the circulation of coolant generates electricity).
  • Shielding. Your hab needs to be shielded from both solar radiation and the nuclear radiation of your engine(s).
  • Volume. You need a large internal volume to carry an enormous amount of fuel if you're going to be able to make a transfer, land, take off again, and head back. Granted, you'd use in-orbit refueling wherever possible, but you need the flexibility to make a round-trip to unvisited worlds.
  • Surface area. Although you need a large volume, you also need a form factor with a blunt-body surface area, allowing re-entry heating to be as minimal as possible. Further, large surface area will help with radiating heat away in space. Finally, a lifting-body shape will make launch and re-entry a bit nicer.

What, then, is the optimal shape and configuration?

Here you go.

perspective.png

Yep, it's a flying saucer.

The hab is located in the center column and is a single floor, preventing any unpleasant gravity gradients. Because the center is open, it allows windows to be embedded in the ceiling, which will feel more natural.

hab_internal.png

The hab is far more "open" overall than most designs, while still being well-protected from radiation and micrometeoroid strikes due to its location in the center of the ship. It will feel very natural to have the sky "up" and the ground "down".

hab_external.png

The hab is shielded by wrapping the ship's tankage completely around it. The triangular cross-section maximizes internal volume while also having the optimal shielding profile. Obviously, the entire ship rotates.

The ship is powered by six small nuclear reactors, feeding three linear exhaust nozzles:

reactors.png

Each of the three exhaust nozzles is capable of running on a single reactor, so you still have maneuverability even if you need to scram one or two of your reactors. The coolant cyclers and generators are also in this area. Only minimal shielding is necessary, due to the placement of the large internal tanks.

For on-orbit burns, the three nozzles all fire together, providing moderate thrust even at the highest operating impulse:

on_orbit_burn.png

During any such burn, there will be a slight misalignment of the apparent gravitational field, but it will likely be no more disorienting than standing on a train while it starts to move. It is also likely that in most cases, the ship will only "spin up" after its transfer injection burns, relieving this issue entirely.

Takeoff and landing use the same orientation as on-orbit burns, but with the injection of liquid hydrogen or another oxidizer to dramatically increase thrust at the expense of specific impulse:

vertical_burn.png

During takeoff, however, maintaining this thrust orientation would make drag losses altogether unmanageable. For this reason, the nozzles are able to change orientation in order to thrust backward during in-atmosphere climbs (for reaching orbit from Terra, Mars, etc.):

gravity_turn.png

Because the "flying saucer" shape is one of the only shapes which is capable of achieving reasonable lift in subsonic, supersonic, and hypersonic flight, this allows for the higher-specific-impulse burn to be used (when applicable) to achieve orbital velocities. Obviously, on worlds without atmospheres, liftoff would be purely vertical without any intermediate horizontally-oriented burn stage.

Re-entry uses the large blunt surface area underneath to dissipate heat passively (although active cooling could be used, in principle):

re_entry.png

Passengers are seated in the upper ring during takeoff and landing, both for gravitational orientation and for safety reasons. Launch abort escape would be achieved either through individual ejection seats or through the ejection of the entire upper ring using built-in thrusters. It is possible that the upper ring could also be configured to serve as an ejectable lifeboat in the case of an on-orbit accident.

control_ring.png\

Attitude control could be achieved either by vectoring the thrusters, or by venting coolant.

Placing the heavy engines at the outside isn't ideal, but given that this is going to be designed with enough structural integrity for powered landing, it shouldn't prove too problematic.

The hab would have nearly 8500 square feet of floor space under artificial gravity, with more than 30 times the pressurized volume of the Space Shuttle crew cabin. I'm estimating a nominal dry weight of 1200 tonnes. The body encloses enough space for 9800 cubic meters of tankage; using a dense propellant like hydrazine, this corresponds to 9,900 metric tonnes of fuel. With pebble-bed reactors giving a specific impulse of around 520 seconds, you get a lovely 11 km/s of dV. Enough to reach orbit as SSTO (if you use LOX-injection augmentation). Also enough to fly to the moon, land, take off, and come back.

Now to build the damn thing...

We  already have a thread open on this topic, why do we need another one. Merge.

Share this post


Link to post
Share on other sites
2 hours ago, PB666 said:

We  already have a thread open on this topic, why do we need another one. Merge.

Because I was greedy and wanted a thread specific to my design. =P

2 hours ago, Bill Phil said:

Except that building the mothership is already a substantial cost of both of our plans, and mine is cheaper than yours, since the mothership never touches an atmosphere. Of course we could launch it from Earth, if we design it right. But the great thing about space is that you can have huge mass ratios. The aero shell limits your geometry, and if you want to take full advantage of a vacuum, you can design your ship to survive a few gs, and only put a fraction of a g on it at any time. This increases mothership delta V. Now you can just about go anywhere in the solar system. At least, planets. 

I'm not sure yours will be cheaper. Your mothership still needs a superstructure in order to support centrifugal gravity, and it needs some sort of basic shielding to protect the fuel tanks and the hab from micrometeoroid strikes. It won't cost significantly more to design your ship with the capacity to aerobrake.

The aeroshell only limits geometry if you're intending to enter an atmosphere during a given mission. If your specific mission is limited to vacuum operations, then you can add whatever geometry you want -- auxiliary tankage, solar sails, massive amounts of cargo -- to increase delta V or to tow a new space station into orbit or whatever else you want to do. And once you reach your destination, you retain the flexibility to leave the vacuum-limited components in a high parking orbit and aerobrake your main ship.

3 hours ago, Bill Phil said:

Carrying separate landers just makes sense, especially for larger vessels. Then less energy is spent on a landing. Sure, low gravity bodies can be landed on, but there's still very little reason to do so with a huge mothership. It takes even more energy to do that.

Okay, let me put it like this: the extra fuel to land on those bodies will be more massive than the individual landers you can use, and you won't use that Delta V until landing. So why waste that mass?

For each mission, the optimal landing configuration will be a function of the destination's surface gravity, escape velocity, atmospheric properties, and surface resources, plus the intended duration of your surface mission. There will be missions where it's cheaper overall to land the whole ship, there will be missions where it's cheaper to bring along a separate but reusable descent/ascent vehicle, and there will be missions where it's cheaper to bring along an expendable descent vehicle and a reusable ascent vehicle.

And while the extra fuel to land on certain bodies might mass more than individual landers, fuel is cheap and simple and may be accessible on-orbit, while landers may not be any of those things.

 

Share this post


Link to post
Share on other sites
20 hours ago, sevenperforce said:

The mothership can certainly tow along any number of other vehicles. Of course, those vehicles will need to be launched separately from Earth, which is a substantial cost.

Logistics starts to play a part, however, when you look at the necessary requirements for your mothership. It already needs to have enough structural integrity to provide artificial gravity by rotation. Thus, slapping a lightweight aeroshell around it so that it can aerobrake for orbital insertion from a transfer won't really incur much additional weight cost.

While towing along a separate descent/ascent vehicle is fairly reasonable, it doesn't really make sense with landings on airless worlds. It is cheaper to simply carry extra fuel to deorbit/reorbit the entire ship them it would be to carry a separate landing vehicle and ascent vehicle. So it only makes sense to equip the mothership with powerful enough engines to support descent and ascent on airless worlds all by itself.

And once you have both aerobraking capability and lunar SST-LEO capability, you might as well give it Martian landing/ascent capability. 

xpbjwh.jpg

Compare the Saturn 5, which used separate lunar landers, and Saturn 8, which landed the entire mothership on Luna. Saturn C8 here has 90 T more capacity to LEO. Landing the mothership simply requires much more fuel- even with ISRU, which is why designs like Mars Semi-Direct launched a separate crew transfer vehicle.

10 hours ago, sevenperforce said:

Because I was greedy and wanted a thread specific to my design. =P

I'm not sure yours will be cheaper. Your mothership still needs a superstructure in order to support centrifugal gravity, and it needs some sort of basic shielding to protect the fuel tanks and the hab from micrometeoroid strikes. It won't cost significantly more to design your ship with the capacity to aerobrake.

The aeroshell only limits geometry if you're intending to enter an atmosphere during a given mission. If your specific mission is limited to vacuum operations, then you can add whatever geometry you want -- auxiliary tankage, solar sails, massive amounts of cargo -- to increase delta V or to tow a new space station into orbit or whatever else you want to do. And once you reach your destination, you retain the flexibility to leave the vacuum-limited components in a high parking orbit and aerobrake your main ship.

For each mission, the optimal landing configuration will be a function of the destination's surface gravity, escape velocity, atmospheric properties, and surface resources, plus the intended duration of your surface mission. There will be missions where it's cheaper overall to land the whole ship, there will be missions where it's cheaper to bring along a separate but reusable descent/ascent vehicle, and there will be missions where it's cheaper to bring along an expendable descent vehicle and a reusable ascent vehicle.

And while the extra fuel to land on certain bodies might mass more than individual landers, fuel is cheap and simple and may be accessible on-orbit, while landers may not be any of those things.

 

And the missions where it'll be cheaper to land a whole ship will likely have negligible gravity, and can be done with almost any spacecraft design.:P

Share this post


Link to post
Share on other sites
On 2/27/2016 at 11:01 PM, fredinno said:

Compare the Saturn 5, which used separate lunar landers, and Saturn 8, which landed the entire mothership on Luna. Saturn C8 here has 90 T more capacity to LEO. Landing the mothership simply requires much more fuel- even with ISRU, which is why designs like Mars Semi-Direct launched a separate crew transfer vehicle.

One little difference: the Saturn rockets were ELVs.

If your vehicle is designed to be expendable, then you'll make a lot of really specific choices. You're going to have a LOT of engines, each designed for a specific thrust profile and throttleability and burn time and fuel type, and you're going to throw them away at the drop of your hat to save every drop of fuel.

With a reusable, on-orbit-refuellable vehicle, things will be very different. For example, the Apollo missions used two separate engines for the landing: one for descent and one for ascent, each with their own tanks, even though they used the exact same fuel mixture. That meant the 180-kg, 25.6 T/W ratio descent engine had to carry the dead weight of the 82-kg, 19.44 T/W ratio ascent engine down to the surface. But it turned out to be a better configuration than using a single engine. With reusability, on the other hand, you'd never dream of dumping your descent engine on the surface, so you'd bring a completely different design philosophy to the table.

On 2/27/2016 at 11:01 PM, fredinno said:

And the missions where it'll be cheaper to land a whole ship will likely have negligible gravity, and can be done with almost any spacecraft design.:P

Maybe, maybe not.

Using lunar landings as a test case but requiring full reusability and an artificial-gravity-equipped hab, and assuming Earth-orbit staging/refueling, you have a few different configurations. You need to get through the Van Allen belts quickly, keep your crew happy and gravitated during the trip, get them down to the lunar surface, keep them alive on the lunar surface, get them off of the lunar surface and up to LLO, then send them back to Earth alive and gravitated, then brake quickly since you have to pass through the Van Allen belts at high speed.

How many engines do you use? How many fuel tanks? How many habs?

Share this post


Link to post
Share on other sites
12 minutes ago, sevenperforce said:

One little difference: the Saturn rockets were ELVs.

If your vehicle is designed to be expendable, then you'll make a lot of really specific choices. You're going to have a LOT of engines, each designed for a specific thrust profile and throttleability and burn time and fuel type, and you're going to throw them away at the drop of your hat to save every drop of fuel.

With a reusable, on-orbit-refuellable vehicle, things will be very different. For example, the Apollo missions used two separate engines for the landing: one for descent and one for ascent, each with their own tanks, even though they used the exact same fuel mixture. That meant the 180-kg, 25.6 T/W ratio descent engine had to carry the dead weight of the 82-kg, 19.44 T/W ratio ascent engine down to the surface. But it turned out to be a better configuration than using a single engine. With reusability, on the other hand, you'd never dream of dumping your descent engine on the surface, so you'd bring a completely different design philosophy to the table.

Maybe, maybe not.

Using lunar landings as a test case but requiring full reusability and an artificial-gravity-equipped hab, and assuming Earth-orbit staging/refueling, you have a few different configurations. You need to get through the Van Allen belts quickly, keep your crew happy and gravitated during the trip, get them down to the lunar surface, keep them alive on the lunar surface, get them off of the lunar surface and up to LLO, then send them back to Earth alive and gravitated, then brake quickly since you have to pass through the Van Allen belts at high speed.

How many engines do you use? How many fuel tanks? How many habs?

Reusable vessels would be even larger, unless ISRU comes into play- even then, NASA had severe doubts about the MAV actually being able to get to Earth on Mars Direct, so they split the fuel.

The entire point of LOR and separate landers is the same logic to why you refuel at gas stations to get to point A to B. The unneeded resources and extra fuel can be carried in a separate reusable lander. It's not like you're prevented from picking up some fuel from your target body while you're there using your reusable SSTO lander. https://en.wikipedia.org/wiki/Lunar_orbit_rendezvous

Share this post


Link to post
Share on other sites
10 minutes ago, fredinno said:

Reusable vessels would be even larger, unless ISRU comes into play- even then, NASA had severe doubts about the MAV actually being able to get to Earth on Mars Direct, so they split the fuel.

The entire point of LOR and separate landers is the same logic to why you refuel at gas stations to get to point A to B. The unneeded resources and extra fuel can be carried in a separate reusable lander. It's not like you're prevented from picking up some fuel from your target body while you're there using your reusable SSTO lander. https://en.wikipedia.org/wiki/Lunar_orbit_rendezvous

I'd still maintain that once you relax the requirement of direct launch from Earth, a lunar mission with full reusability is not necessarily an open-shut win for LOR.

With LOR and a separate reusable lander, you'll need to leave LEO with:

  • Earth orbit boost engines (to get you out of and into LEO quickly enough to avoid the Van Allen belts)
  • Fuel tanks for Earth orbit boost engines
  • Transfer engines
  • Fuel tanks for transfer engines
  • Transfer hab and control center
  • Supplies for transfer hab
  • Descent hab and descent control center
  • Supplies for descent hab
  • Descent/ascent engines
  • Fuel tanks for descent/ascent engines
  • Two separate bodies, one with landing legs and one without

With a single craft, you need to leave LEO with:

  • A single hab and control center
  • Supplies for the hab
  • A single engine cluster
  • A single rather large fuel tank
  • A single body with landing legs

With LOR, you have the advantage of leaving some stuff in orbit...but when it all boils down, the only things you actually end up saving on are the fuel for the return trip and half the supplies for the transfer hab. Everything else you either bring down with you, or is an extra requirement of LOR.

On the one hand, you have to build two separate habs with separate life support systems, two separate crafts, and a maze of fuel tanks of varying sizes. On the other hand, you have more fuel in one tank.

 

Share this post


Link to post
Share on other sites
This thread is quite old. Please consider starting a new thread rather than reviving this one.

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Sign in to follow this