Cheat Sheet

For some good general notes on designing spacecraft in general, read Rick Robinson's Rocketpunk Manifesto essay on Spaceship Design 101. Also worth reading are Rick's essays on constructing things in space and the price of a spaceship.

For some good general notes on making a fusion powered spacecraft, you might want to read Application of Recommended Design Practices for Conceptual Nuclear Fusion Space Propulsion Systems. There are also some nice examples on the Realistic Designs page.

For less scientifically accurate spacecraft design the Constant Variantions blog has a nice article on historical trends in science fiction spacecraft design.

Like any other living system, the internal operations of a spacecraft can be analyzed with Living Systems Theory, to discover sources of interesting plot complications.


The improvised space warcraft are the type that seems to hold the most story potential.  These would, as mentioned, likely be built by colonies that are in conflict.  As they do not have to operate in an atmosphere, and are built by relatively poor colonies, they are likely to be rather crude.  The basic components required are structure, propulsion, weapons, life support, power, sensors, control, and communications, and each will be briefly discussed in turn.

There are two methods of assembling an improvised warcraft, either adapting an existing vessel, or constructing a new one from parts.  The use of an existing vessel removes the need for some, though not all, of the various components.  The structure will obviously be preserved, and propulsion and life support are almost certain to remain unchanged as well.  Weapons would obviously be added, along with their associated control equipment.  Existing power supplies might or might not be sufficient, depending on the weapons fit chosen.  Sensors and communications are gray areas, depending on the tasks required of the vessel, and the existing fits in these areas.

Structure is one of the easiest components to create.  So long as the builder does not mind the craft being somewhat heavy, slapping a few beams together should be sufficient.  Existing structures, such as cargo containers, could easily be modified, or simple new ones fabricated.  In any case, this is not likely to be a driving factor in construction or conversion of vessels.  Any group incapable of creating basic structure is also almost certainly incapable of surviving if it were to win a rebellion.

Propulsion for improvised craft is likely to be chemfuel due to the fact that it is by far the simplest to implement, and has sufficient delta-V for any operations that do not involve transiting deep space.  It is entirely possible that a colony will have standard chemfuel engines used in various places, and one of them, with appropriate fuel tanks, would be fitted to the vessel.  Nuclear propulsion is much more expensive, and might well involve detailed engineering to avoid killing the crew.  The performance advantage over chemfuel for nuclear-thermal is probably not significant enough to justify the problems involved, and nuclear-electric is only practical for vessels intended for deep space use.  

Weapons are a tricky issue. These are likely to be improvised as well, and would fall into the same categories already discussed.  Improvised lasers are highly unlikely.  Industrial lasers lack the optics trains required for weapon use, while any optical trains available (probably from astronomical or other scientific sources) are unlikely to be able to handle the high powers output by the lasers.  With some time, an appropriate optical train could be designed and mated to an industrial laser, and it is even possible that colonies could design and test such things in case of war.  Kinetic projectors are in largely the same boat.  While small mass drivers could be adapted to such a task, it is difficult to see a role for such devices on a colony.  There is also the issue of targeting, which, while not insoluble, requires good pointing accuracy and possibly the creation of guided projectiles, which have even less peacetime use then the launchers themselves.  

This leaves three options, missiles, lancers, and unguided kinetics.  Unguided kinetics can be as simple as junk thrown out of the airlock, but they are of very limited effectiveness, as shown in Section 8.  Missiles and lancer projectiles face many of the same issues, and the only practical difference is the motor, which should be relatively easy to improvise.  A missile or lancer will require sensors, thrusters, and guidance logic of its own.  As this force is presumably facing another more or less improvised one, complicated guidance logic is probably unnecessary, and proportional navigation is quite easy to implement.  The sensor might well be adapted from another role, which means that the likely problem is in the thrusters.  These must be well-balanced and integrated with the guidance logic.  Depending upon the materials available, this could range from very easy (if there are large numbers of small, self-propelled objects that can be used as warheads lying around) to extremely difficult (if the entire system must be designed from scratch).  Small thrusters themselves are an unknown.  There are some systems that might use small thrusters, such as thruster packs for spacewalkers, in which case the actual integration is all that is required.  However, the number available might well be strictly limited, forcing the builder to start from scratch.  Note that this is not as easy as it seems.  While a primitive kinetic could probably be built with the simplest of systems, it would be inefficient, of dubious reliability, and probably quite large.  In the end, this is an area in which the specific situation plays a very large role, leaving us unable to anticipate exactly what might occur.  

Life support should be straightforward to build into a vessel.  Any space colony will undoubtedly have small, portable habs that can be used for surface expeditions or what have you.  Mounting one of those would be relatively simple, and the actual mechanisms for short-term life support are fairly rudimentary, easing implementation if for some reason a hab had to be constructed from scratch.

Power is a rather tricky proposition.  Unless a nuclear propulsion system is used, power is likely to be at a premium.  Most non-nuclear power studies assume that solar panels will be used, but these have significant drawbacks for space warfare.  The biggest problem is that solar panels are vulnerable to damage from opponent’s lasers or powder weapons, and cannot be angled for protection, unlike radiators.  Radiators, discussed in Section 7, are both somewhat less vulnerable to damage, and can be kept edge-on to the enemy.  A clever opponent could manage to create a dilemma between getting power and preserving the panels from damage.  Alternatives include fuel cells and batteries.  Fuel cells are the current solution for short-duration spacecraft, due to their possessing higher specific energy than batteries.  The problem is that fuel cells are somewhat involved to manufacture and are not likely to be common on space colonies, unless the colony is far enough from the Sun that solar panels are not effective.  Batteries are somewhat more likely to exist, but are heavy for their power output.  The only bright side is that a truly improvised spacecraft is unlikely to need much in the way of power, particularly when compared to a nuclear-electric laserstar.

Sensors are probably the biggest unknown.  A proper space warcraft needs some form of active sensor to localize the enemy, although it is possible that a simple passive sensor would be adequate for simple missiles.  The sensors might or might not be readily available.  Sensor suites for existing spacecraft are the most likely source, although cobbling sensor suites together from other uses might be possible.

Control is mostly a matter of systems integration.  Depending on the nature of the systems involved, control setup could range from simple running of cables and plugging together a few modules, to having to write all of the code to make everything talk, or simply doing without an integrated control system.  While it is certain that some systems will have to talk (sensors and weapons spring to mind) a large portion of the integration could be skipped, with a resulting loss of efficiency due to the crew having to move around.

Communications is fairly simple, as one thing people in space will have to do is talk.  This should ensure the availability of communications modules, which can be attached to the vessel.  The most likely cause of problems is lack of strong encryption and particularly electronic warfare capability in such modules.  Depending upon the capability of the opposition, this may or may not be a serious hindrance.  The encryption capabilities should be a reasonably simple fix, involving mostly software updates.  The EW work will be harder, as there are likely to be physical changes required to ensure freedom from interference by enemy radio transmissions.  This is not likely to be a problem if the communications module is radio-based.

by Byron Coffey (2016)


RocketCat sez

This is the living breathing core of all rocket design. Delta Vee equals Vee Ee times Natural Log of Arr. This is the secret that makes rocket design possible. Now it is time to see the practical application of the key to rocketry.

Everything about fundamental spacecraft design revolves around the Tsiolkovsky rocket equation.

Δv = Ve * ln[R]

The variables are the velocity change required by the mission (Δv or delta-V), the propulsion system's exhaust velocity (Ve), and the spacecraft's mass ratio (R). Remember the mass ratio is the spacecraft's wet mass (mass fully loaded with propellant) divided by the dry mass (mass with empty propellant tanks).

The point is you want as high a delta-V as you can possibly get. The higher the delta-V, the more types of missions the spacecraft will be able to perform. If the delta-V is too low the spacecraft will not be able to perform any useful missions at all.

Looking at the equation, the two obvious ways of increasing the delta-V is to increase the exhaust velocity or increase the mass ratio. Or both. Turns out there are two more sneaky ways of dealing with the problem which we will get to in a moment.

Historically, the first approach has been increasing the exhaust velocity by inventing more and more powerful rocket engines. Unfortunately for the anti-nuclear people, chemical propulsion exhaust velocity has pretty much hit the theoretical maximum. The only way to increase exhaust velocity is by using rockets powered by nuclear energy or by power sources even more frightful and ecologically unsound. And you ain't gonna be able to run a large thrust ion-drive with solar cells.

The second approach is increasing the mass ratio by reducing the spacecraft's dry mass. This is the source of the rule below Every Gram Counts. Remember that the dry mass includes a spacecraft's structure, propellant tankage, lifesystem, crewmembers, consumables (food, water, and air), hydroponics tanks, cargo, atomic missiles, toilet paper, clothing, space suits, dental floss, kitty litter for the ship's cat, the ship's cat itself, and other ship systems. Everything that is not propellant, in other words. All of it will have to be trimmed.

To reduce dry mass: use lightweight titanium instead of heavy steel, shave all structural members as thin as possible while also using lightening holes, make the propellant tanks little more than foil balloons, use inflatable structures, make the floors open mesh gratings instead of solid sheets, hire short and skinny astronauts, use life support systems that recycle, impose draconian limits on the mass each crewperson is allowed for personal items, and so on. Other tricks include using Beamed Power so that the spacecraft does not carry the mass of an on-board power plant, and avoiding the mass of a habitat module by hitching a ride on an Aldrin Cycler. Finally the effective mass ratio can be increased by multi-staging but that should be reserved for when you are really desperate.

The third approach is trying to reduce the delta-V required by the mission. Use Hohmann minimum energy orbits. If the destination planet has an atmosphere, use aerobraking instead of delta-V. Get more delta-V for free by exploiting the Oberth Effect, that is, do your burns while very close to a planet. Instead of paying delta-V for shifting the spacecraft's trajectory or velocity, use gravitational slingshots. NASA uses all of these techniques heavily.

If your technology is high enough, use space tethers, launch catapults, and MagBeams.

The fourth and most extreme approach is to cheat the equation itself, to make the entire equation not relevant to the spacecraft. The equation assumes that the spacecraft is carrying all the propellant needed for the mission, this can be bent several ways. Use Sail Propulsion which does not use propellant at all. Use propellant depots and in-situ resource utilization to refuel in mid-mission. The extreme case of ISRU is the Bussard Ramjet which scoops up propellant from the thin interstellar medium, but that only works past the speed of 1% lightspeed or so.

In our Polaris example, given the mass ratio of 3, we know that the Polaris is 66% propellant and 33% everything else. Give the total mass of 1188.9 tons means 792.6 tons of propellant and 396.3 tons of everything else. Since each GC engine is 30 tons, that means 150 tons of engine and 246.3 of everything else.

Every Gram Counts

RocketCat sez

Listen up, rocket designers. Write these words in letters of fire on your cerebellum. Every Gram Counts! Add an extra gram and you will pay for it with extra propellant as if the Mafia loan shark wants you to pay up with liquid hydrogen. With interest compounded hourly.

The only exception is if you are dealing with an Orion drive spaceship or other torchship.

The most fundamental constraint on designing a rocket-propelled vehicle is Every Gram Counts.

Why? Short answer: This is a consequence of the equation for delta-V.

Why? Slightly longer answer: As a general rule, a rocket with the highest delta-V capacity is going to need three kilograms of propellant for every kilogram of rocket+payload. The lower the total kilograms of rocket+payload, the lower the propellant mass required. This relates to the second strategy of rocket design mentioned above.

Why? Long Answer:

Say the mission needs 5 km/s of delta-V. Each kilogram of payload requires propellant to give it 5 km/s.

But that propellant has mass as well. The propellant needed for that original kilogram of payload will require a second slug of propellant so that it too can be delta-Ved to 5 km/s.

And the second slug of propellant has mass as well, so you'll need a third slug of propellant for the second slug of propellant — you see how it gets expensive fast. So you want to minimize the payload mass as much as possible or you will be paying through the nose with propellant.

This is called The Tyranny of the Rocket Equation.

Even worse, for a given propulsion system, the easiest way to increase the delta-V you can get out of that system is by increasing the mass ratio. It probably is not economical to push the mass ratio above 4.0, which translates into 3 kg of propellant for every 1 kg of rocket+payload. And it is nearly impossible to push the mass ratio above 20. Translation: spacecraft with a mass ratio of 20 or above are basically constructed out of gossamer and soap bubbles.

This is why rocket designers are always looking for ways to conserve mass.

Orion drive spacecraft and other torchships are not subject to this constraint, because they are unreasonably powerful.

It also does not apply to "stationary" items such as space stations and planetary bases, since they do not move under rocket propulsion. In fact, the added mass might be useful to stablize a space station's orbit, or as additional radiation shielding. Rocket vehicles might use aluminium, titanium, magnesium, or other lightweight metal as their structural material; but a space station would be better off using heavy iron or Invar.

The only consideration is if the station or base components have to be transported to the desired site by a rocket-propelled transport. Then it makes sense to make the components low mass, because then the station bits are payload. It makes even more sense to construct the space station or base on site using in-situ resources, so you don't have to eat the transport costs at all.

Everything Is Connected

Like aircraft and sea-going warship design, one soon discovers that everything is connected to everything else. When the designer changes one aspect of the design this causes a series of related changes to ripple through the rest of the design.

For instance, if the designer reduces the propellant tank capacity by 5% this has implications for the spacecraft's mass ratio. If it is important for the spacecraft's delta V to stay the same, the payload will have to be reduced by the same amount. This might cut into the amount of life support consumables carried, which will reduce the number of days a mission can last. If the same amount of scientific observations have to be done in the reduced time, another crew member might have to be added. This will decrease the mass available for consumables even more. And so on.

The technical term is "cascading changes." The only thing worse is cascading failures.

As an example, I have some notes of graphing various space warship optimizations with examples of how various warship classes map onto the graph.


(ed note: Captain Randall, his crew, and his submarine were inadvertently put into suspended animation in 1945. They have been awakened one thousand years later, and must go into battle with alien invaders. Captain Randall is talking over the refurbishment of his submarine with a robot named Austin. The discussion is about limitations in submarine design, but the "everything is connected to everything" principle still holds true.)

     "I'll take your word for it, Austin. At the moment there are more important things. What is the state of the ship?"
     "As on the day of trials, sir, without fault, and completely refitted. The experts have many ideas for improvements, sir, which will be submitted to you in due course."
     "And the munitions?"
     "None, sir. They had deteriorated beyond safe use. However, they can be replaced or, if you prefer it, something more violent substituted."
     "One thing at a time. At the moment, have them replaced. Does this age understand explosives?"
     "Chemical history does, sir. Your ammunition holds can be filled in approximately half an hour and that includes the torpedo tubes. Will you continue with ancient compressed air, underwater missiles, sir? We have some smaller and far more efficient."
     "Normal torpedoes with a normal warhead, please, Austin."
     "As you wish, sir."
     Randall smiled faintly. "You sound disappointed, Austin. So, since you have a logical mind, I will be logical. You probably have in mind more advanced and far more destructive missiles which came into being just before man decided on peace for all time. No doubt these missiles are faster, have a greater range, are self-seeking and are several times more destructive. How big are they, Austin?"
     "About the quarter of the size of yours sir. I estimate you can carry—"
     "They won't fit our torpedo tubes, will they? This will mean interior adaptation, rebuilding, all of which will have to be related to buoyancy, speed—above and below surface—diving and surfacing. Which method is going to be the quickest?"
     "I had overlooked that angle, sir." There was considerable respect in the robot's voice.
     "That is because you are not yet a sailor—and let me assure you. You cannot become a seaman from text books or even convenient memory banks: It's a question of experience. No doubt, Ordinary Seaman Austin, your mind is brimming with ideas for replacing the ship's motors with something modern, smaller, a hundred times more efficient and one tenth the weight. This would play hell with our diving, upset our displacement and require the alteration of our tank capacity to meet our increased buoyancy. Am I making myself plain?"
     "Too plain, sir. With respect, I am learning a great deal."
     "Good. Don't misunderstand me, Austin. I want your ideas and I shall ask for them, but only when you, yourself, can relate them to experience."

From THE TIME MERCENARIES by Philip E. High (1968)

Design Humor

Fundamental Design

As mentioned in Rick Robinson's Spaceship Design 101, all spacecraft are composed of two sections: the Propulsion Bus and the Payload Section.

The Propulsion Bus has the propulsion system, propellant tankage, fuel container (if any), power plant, power plant heat radiator (if any), anti-radiation shadow shield (if any), and a keel-structure to hold it all together. Sometimes the keel is reduced to just a thrust-frame on top of the engine, with the other components stacked on top.

The Payload Section is what the propulsion bus is pushing from planet to planet. It can include crew, flight control station, propulsion/power plant control station and maintenance center, astrogation station, detection and communication equipment, habitat module with life support equipment (including environmental heat radiators) and consumables (air, food, water), space taxis, space pods, and docking ports.

But most importantly, the payload section must contain the reason for the spacecraft's existence. This might be organized as a discrete mission module, or it might be several components mounted around the payload section.

The actual engines proper of the propulsion bus push against the Thrust Frame. The Thrust Frame pushes against the space frame, which is sort of the ship's skeleton. The various components of the propulsion bus and the payload section are attached to the space frame, like ornaments on an xmas tree. The space frame drags all the components along.

The thrust frame is sort of the heart of the ship's body. So in the future, instead of starting sea-going vessel construction with laying the keel you may have starting space-going vessel construction with "assembling the thrust frame" or something like that.


I will argue that deep space craft have essentially two sections that can largely be treated separately from one another. One section is the propulsion bus — drive engine, reactor if any, solar wings or radiator fins, propellant tankage, and a keel structure to hold it all together. The other is the payload section that it pushes along from world to world.

There are both conceptual and economic reasons to treat them separately. Conceptually, because a propulsion bus might push many different payloads for different missions, such as light payloads on fast orbits versus heavy payloads on slow orbits. A little noticed but important feature of deep space craft is that you cannot overload them. They do not sink, or crash at the end of the runway, or even bottom out their suspension. They merely perform more sluggishly, with reduced acceleration and (for a given propellant supply) less delta v.

Conceptual logic is also economic logic. The outfits that build drive buses would like to sell them to lots of different customers for a broad range of assignments.

This is not necessarily an argument for true modular construction, with drive buses hitching up to payloads on an ad hoc basis like big-rig trucks and trailers. Building things to couple and uncouple adds complexity, mass, and cost — plug connectors, docking collars, and so forth. Moreover, drive buses intended for manned ships need to be human-rated, not just with higher safety factors but provision for supplying housekeeping power to the hab, etc. But these things, along with differing sizes or number of propellant tanks, and so forth, can all be minor variations in a drive bus design family.

The payload we are most interested in is, naturally, us. The main habitat section of a deep space ship closely resembles a space station. It is likely that habs intended for prolonged missions will be spun, for health, efficiency, and all round convenience. (Flush!) The design of a spin hab is dominated by the spin structure and — unless you spin the entire ship — the coupling between the spin and nonspin sections.

Because ships' spin habs have the features of stations they may be used as stations, and again we can imagine design families, with some variants intended for ships and others as orbital platforms having only stationkeeping propulsion. Habs are the one major part of a deep space ship that correspond fairly well to our concept of a hull. Spin habs are entirely different in shape, but the shape is constrained; once you build it you can't easily modify it, beyond adding another complete spin section.

For those with bank cards at the ready, buying a deep space ship might be not unlike buying a computer. If your mission needs are fairly standard, you check off options on a menu. Those with more specialized requirements can select major components — perhaps a drive bus from one manufacturer, a main crew hab from another, along with custom payload sections, service bays, and so forth, assembled to your specifications.

In fact, both technology and probable historical development suggest that fabrication and overall assembly will be two distinct phases, carried on in different places, quite unlike either shipyard or aircraft assembly practice. In the early days, large deep space craft will be built the way the ISS was, assembled on orbit out of modules built on Earth and launched as payloads. In time fabrication may move to the Moon, or wherever else, but final assembly (at least of larger craft) will continue to be done at orbital facilities. I call them cageworks, on the assumption that a cage or cradle structure provides handy anchoring points for equipment.

For game or sim purposes, my advice would be to treat drive buses and hab sections as the primary building blocks for ships, whether these components are permanently attached to each other or simply coupled together. Both approaches might be in use.

A couple of provisos. All of the above applies mainly to deep space craft, especially with high specific impulse drives. Ships for landing on airless planets have some similar features. Ships that use rapid aerobraking, however, are aerospace craft and broadly resemble airplanes, even if they never land or even go below orbital speed.

And I have said nothing of warcraft. Kinetics are essentially just another payload. Lasers, and other energy weapons such as coilguns, probably draw power from the drive reactor, calling for some modifications in the drive bus. These things don't much affect the overall configuration. Armor protection would, but discussions here have left me doubtful of its value against either lasers or kinetics. Laser stars and other major warcraft may not be dramatically different in appearance from civil craft of similar size.

Nyrath: I'm reminded of Sir Arthur C. Clarke's early space science books. He noted that a nuclear powered spacecraft would probably resemble a dumb bell, that is, two spheres connected by a stick. The hab module is the forward sphere, the nuclear drive is the rear sphere, the stick is long to provide some inverse square protection from radiation, and the propellant tanks would be on the stick, probably clustered near the nuclear drive.

One can also imagine modules designed by diverse corporations being incompatible with others on purpose. "Not invented here" syndrome.

One can also imagine a tramp freighter composed of incompatible modules, being held together with bailing wire and spit.

Qwert: The compatibility between modules will mostly depend on how the market develops. One extreme maybe Microsoft, a monopoly that basically sets the standard, as everything has to adhere to it. The other one extreme is the current hardware industry. Your memories and components have to fit on every motherboard if you want to sell them.

On the field of big aircraft manufacturing, standardisation dominates almost everything... excerpt the end product. If you manufacture engines, they have to fit on Airbus as well as on Boeing's aircraft. The components industry is dispersed and competition is intense.

On the other side, a pilot trained to flight with a Boeing can´t immediately switch to an Airbus without some training. Competition centers around two big players and nobody is interested in making life easier to the other.

In short: how spaceship components will be build, will mostly depend on how the industry evolves. A monopoly, strict government regulation, competition between many small producers or a highly dispersed specialised components industry may benefit a system of standards. On the other side competition between a small number of giants, may produce different incompatible systems.

Rick Robinson: True modularity is by no means a given. But some features of modularity, call it demi-modularity, are inherent to deep space technology.

You probably want to keep your propellant tanks separate from the corrosive, explosive stuff we breathe. Drive engines are essentially bolted onto the tail. Generally the major parts of a deep space ship don't have to fit together snugly. If you want to hang something out on a bracket you probably can.

Ferrell: I think that the engine package, mated to a suitable tank, mated to a hab module, mated to a mission module (a seperate entity from the hab module) would (as the last step in design) incorporate the heat management system suitable to the final design. Then construction/assymbly would occure.

If there aren't landers/shuttles at the final destination and you need them, then you can carry landers/shuttles in place of cargo. A mission module may be an extended docking module that a number of small modules 'plug' into, (or your transfer craft).

Manuvering thrusters should be at mutliple points/modules (distributed from the 'nose' to the 'tail').

Of course, you could stand it all on it's ear and have the mission module be on the inside of the ship, the hab ring be around the middle (with its radiators in arcs between its connecting pylons),with the engines, tanks, powerplants, radiators, and nav sensors clustered around both ends of the mission module; and any docking would be at the tips of the mission module, or on the inside walls of the mission module.

Nyrath: There might be a brisk trade in "interface modules", that would connect modules made by different manufacturers. I'm reminded of the Apollo ASTP Docking Module used in the Apollo-Soyuz mission. This was a tiny airlock module with a NASA style docking collar on one end, and a Soviet style docking collar on the other.

Not to mention the International Space Station Pressurized mating adapters.

Rick Robinson: Sabersonic — Yes, I'm gliding over a host of devils in the details. The payload section will surely have attitude thrusters, for example, and these must coordinate with attitude thrusters on the drive bus end.

Nyrath — 'Standards,' indeed! Again this is a promo for full service cageworks that will provide things like docking adapters.

Jean Remy: On standards:

Currently the main type of freight vessel is the container-carrier. They are favored because the containers can be loaded/unloaded easily, then simply popped on a freight train or a truck. I can see the same things for a cargo spaceship. There won't be a cargo "module" but rather anchoring hard points for standard containers. Those containers will be loaded on a booster to orbit, transferred onto a ship, then at the other end the container is loaded into a simple remote-controlled lifting body (for planets with an atmosphere, say, Mars) or just a simple frame with thrusters (for the Moon) where they will be loaded into maglev trains if needed, etc.

The command post/bridge:

I'd still put in in the non-rotational part of the ship, certainly on warships (you want it as deep inside the ship as you can for protection) Not only that, but I would stop rotation in combat: precession might alter your maneuverability, and damage might weaken the structure or cause a wobble which would rip your ship apart. I would also put it in the middle for a civilian vessel, as then the command module can double as storm cellar, or if there's a separate storm cellar, you'll still have access to the command post during a solar flare. It's also a good defense against micrometeorites. The command center is just too important to risk placing it on the outer rim of a ship, no matter how convenient or comfortable it would be to the crew.

Sabersonic: Even so, it would probably be prudent to not have all command, or at least navigation control, be monopolized by the CIC. At minimum and barring mass budgeting, there should be two for overall spacecraft control: One that is the CIC/Navigation primary with the second being the engine room for mostly emergency purposes. If only because for something as complex and (for the first few decades if not centuries of interplanetary travel, let alone interstellar) inevitably as fragile as a spacecraft, one should avoid "putting ones eggs into one basket" and to always "have a plan B". Space and interplanetary/interstellar travel is not a place for the ill-prudent, and that's just the natural dangers.

As for standards, well, before standards could ever become "standard" for lack of a better word, there lies the inevitable "Format War" in one form or another that would occur when one believes that their system is more efficient and reliable than the other or worse: a new industry/business standard that has the potential to supplant the jobs of numerous Dock workers or in this case "Cage Workers" that could drive hostility and perhaps a little bit of political pressure before the whole matter is settled in one way or another. It would not be a full parallel to the Teamsters Strike that is basically Teamsters vs Trucks, but it would not be a completely quiet deal when paychecks are involved.

Rick Robinson: My division of ships into drive buses and payload sections is more because of operational factors than manufacturing considerations.

For example, ships inherently have (at least) two big 'hull' structures, the crew hab and the main propellant tank. These are probably at very different temperatures, which right away is a big reason to keep them physically separate.

Using the propellant tankage as shielding is appealing, but even with a vacuum separation it means that your 290 K hab shell is dumping 350 W/m2 of waste heat right into your 20 K liquid hydrogen tanks.

Jean — The containership principle seems highly likely for most space cargo, with the standard pod being defined originally to fit Earth orbital shuttle bays.

In this case, your typical space freighter is a drive bus pushing a rack structure with clamps for pods. The rack might be configurable so that you can also carry 'oversize' loads.

Jean and Sabersonic — In a parallel discussion at SFConsim-l, the question was raised whether civil ships need a 'control room' at all, or whether people could just stand watch from their regular work stations.

I think that any spacecraft with a fair number of passengers or crew will have a watch at the main life support panel, because life support is always running, has constantly changing loads, and things can go very bad very quickly. The life support panel will almost certainly be in the spin section, because that is where the life support is.

You may as well put the engineering panel here as well. There's no reason for an 'engine room' in the maritime sense, since the drive is mounted externally, and if it is nuclear you don't want to go anywhere near it.

No doubt you could maneuver and navigate the ship from here as well. But en route there is very little of this to do.

The only time you really conn a (civil) spacecraft is during rendezvous and docking, or similar evolutions. At these times you surely de-spin, but you might want a separate control station next to the main airlock, with viewports for maximum situational awareness.

Because of its location, this station would also naturally serve as the ceremonial 'quarterdeck' where VIPs are greeted, and ordinary mortals report aboard.

Warcraft are a whole 'nother matter, with protective considerations arguing for a control room at the center of the hab section.

Jean Remy: Multiple Command posts: Yes. That's why I used the term command post rather than CIC, Bridge, DCC, Main Engineering. It was meant to be generic and refer to all these. There ought to be at least CIC and DCC, if not a redundant third post, all of them capable of doing every job, but each ideally suited to one. The ship is more efficient with all of them, but can still work with the loss of one or two.

Decentralized Command posts: Possibly, but I don't think it likely. First of all we could do that now, but we don't. Psychologically, I think crews want to have the "Captain on Bridge" so to speak. It is generally recognized that good officers are the ones who stick by their men when the going gets tough. Caesar rode into battle at Alesia, rallying his troops to victory when it looked uncertain. Washington rode with his men. Patton led from a tank. Now granted the captain of a ship can't go anywhere, but seeing him on the Bridge will be a boost to morale.

Rick Robinson: The question of having a control room at all was in the context of civil spacecraft. If they have an sudden emergency it is most likely to be a life support crisis such as fire, for which the classic 'bridge' functions are fairly irrelevant.

Military craft are a special case, and I'd certainly expect them to have a control center.

And maybe all ships, because to extend a point Jean makes, existing space programs are quasi-military in origin. The military outlook toward emergency response is coded into their DNA, so to speak.

All human carrying deep space ships will need a storm shelter in any case, and it would be fairly natural to configure this as an emergency control center.

Luke Campbell: Re: Shielding. It now seems likely that a plasma magnet generated by a low mass antenna could deflect any charged solar radiation, so the crew would be safe from flares and CMEs. It does not seem like a plasma magnet could stop galactic cosmic rays, GCRs are a steady source of background radiation, not the sort of thing that a "storm cellar" would help with. A plasma magnet also would not protect against neutral particle radiation, but the only neutral particle radiation likely to be a threat is man-made: neutron radiation from nukes and possibly high energy photons from x-ray or gamma ray lasers.

For those not in the know, a plasma magnet uses low frequency radio waves to produce a rotating field that induces a current in the surrounding solar wind plasma. This current forms a dipole magnetic field that deflects and reflects charged particles. The field is not strong enough to deflect solar wind protons, but it does deflect the electrons, leading to charge separation that pulls the protons back to the electron cloud before they reach the section being protected.

Yes I was trying to stop GCRs (but calculations showed it would need an unreasonably huge magnet). The plasma magnets wouldn't stop the solar wind protons either, when considered as individual particles — you need the plasma effects of the electrons to stop the solar wind. This lets you get by with a much smaller magnet.

However, there is one possible additional method of mitigating the GCR dose - medication. As we learn more about cellular repair and cell "suicide", new treatments may become possible for both chronic and acute radiation poisoning (and oddly, you are likely to want the opposite reaction in these two cases - for chronic exposure, you want the damaged cells to destroy themselves to prevent cancer; for acute exposure you want the damaged cells to repair themselves to prevent anemia, hemophilia, a compromised immune system, and digestive difficulties). Incidentally, it has been shown that vitamin D helps with chronic radiation exposure, although the mechanism is not clear.

Rick Robinson: To keep cryogenic propellant from boiling off on long missions you will need active refrigeration, pumping heat out of the tank. Otherwise heat buildup will be inexorable.

The ship will be designed to keep the propellant tanks away from the main radiator fins and such, and generally minimize heat absorption from the rest of the ship, so the heat you mainly have to deal with is from sunlight.

The tanks will be painted white or silvery to reflect away most sunlight. I assume that you can reflect about 90 percent. For a spherical tank at 1 AU, that means about 35 W/m2 of absorbed solar radiation that you'll have to pump out of the tank.

A 20 meter diameter tank holds about 250 tons of hydrogen, or 1500 tons of methane. Surface area is 1250 m2, so at 1 AU you'd need 44 kW of refrigerating capacity, i.e. heat extraction, to keep propellant cold.

At Jupiter distance, 5 AU, solar flux is reduced by 96 percent, and you only need 2 kW of refrigeration for hydrogen - none for methane, which will stay liquid or even tend to freeze.

Suppose you put a 10 meter diameter hab inside the tank {to give the hab some radiation protection}. (This only reduces tank capacity by 1/8.) But even with a vacuum layer you will have IR heating from the hab surface, at room temperature: 400 W/m2 * 314 m2 = ~125 kW.

So in this case putting the hab inside the fuel tank multiplies your propellant refrigeration bill by 4x. Which is a lot, but not horrible; the shielding might be worth it. But wrapping propellant around a spin hab is tougher.

Rick Robinson: Modular design is always a tradeoff. You get more operational flexibility, at cost of more complicated/heavier/weaker connections. Integral designs will be favored when the components will consistently be used together.

Much will depend on tech. Torch type drives and even 'conventional' nuke electric drives pretty much have to be mounted on a pylon, which sort of invites the option of unbolting it from the rest. OTOH, as you note, the drive section may well have its own control center. And since the rest of the ship sits on top of the pylon, it's a fine line between 'pylon' and 'chassis.'

On naming, I could also make a case that the crew hab compartment is the main component, and so would be named. Especially if it is a spin gravity structure. And 'spaceships' may end up having more than one name, just as a named train might included Pullman cars with names of their own.

And if ships are highly modular, some terms might be borrowed from railroading. For example, 'consist' as a noun (pronounced CON-sist) for the whole assemblage. Thus, 'The Ty Cobb departed Mars with a consist of [such and such modules].'

Amusing side note: modular spacecraft reverse the order of trains: the 'locomotive' or drive engine is at the back (more precisely the base), while the 'caboose' or control cabin might well be at the front/top.

From SPACESHIP DESIGN 101 by Rick Robinson (2009)

This section is intended to address some gaps in available information about spacecraft design in the Plausible Mid-Future (PMF), with an eye towards space warfare.  It is not a summary of such information, most of which can be found at Atomic Rockets.  The largest gap in current practice comes in the preliminary design phase.  A normal method used is to specify the fully-loaded mass of a vessel, and then work out the amounts required for remass, tanks, engine, and so on, and then figure out the payload (habitat, weapons, sensors, cargo, and so on) from there.  While there are times this is appropriate engineering practice (notably if you’re launching the spacecraft from Earth and have a fixed launch mass), in the majority of cases the payload mass should be the starting point.  The following equation can be used for such calculations:

Where P is the payload mass (any fixed masses, such as habitats, weapons, sensors, etc.), M is the loaded (wet) mass, R is the mass ratio of the rocket, T is the tank fraction (or any mass that scales with reaction mass) as a decimal ratio of such mass (e.g., 0.1 for 10% of remass), and E is any mass that scales with the overall mass of the ship, such as engines or structure, also as a decimal.  

This equation adequately describes a basic spacecraft with a single propulsion system.  It is possible to use the same equation to calculate the mass of a spacecraft with two separate propulsion systems.

The terms in this equation are identical to those in the equation above, with R1 and T1 representing the mass ratio and tank fraction for the (arbitrary) first engine, and R2 and T2 likewise for the second.  Calculate both mass ratios based on the fully-loaded spacecraft.  If both mass ratios approach 2, then the bottom of the equation will come out negative, and the spacecraft obviously cannot be built as specified.  Note that when doing delta-V calculations to get the mass ratio, each engine is assumed to expend all of its delta-V while the tanks for the other engine are still full.  In reality, the spacecraft will have more delta-V than those calculations would indicate, but solving properly for a more realistic and complicated mission profile requires numerical methods outside the scope of this paper.

One design problem that is commonly raised is the matter of artificial gravity.  In the setting under discussion, this can only be achieved by spin.  The details of this are available elsewhere, but these schemes essentially boil down to either spinning the entire spacecraft or just spinning the hab itself.  Both create significant design problems.  Spinning the spacecraft involves rating all systems for operations both in free fall and under spin, including tanks, thrusters, and plumbing.  The loads imposed by spin are likely to be significantly larger than any thrust loads, which drives up structural mass significantly.  This can be minimized by keeping things close to the spin axis, but that is likely to stretch the ship, which imposes its own structural penalties.  A spinning hab has to be connected to the rest of the spacecraft, which is not a trivial engineering problem.  The connection will have to be low-friction, transmit thrust loads, and pass power, fluids, and quite possibly people as well.  And it must work 24/5 for months.  All of this trouble with artificial gravity is required to avoid catastrophic health problems on arrival.  However, there is a potential alternative.  Medical science might someday be able to prevent the negative effects of Zero-G on the body, making the life of the spacecraft designer much easier.  

When this conclusion was put before Rob Herrick, an epidemiologist, he did not think it was feasible.

     “The problem is that they [the degenerative effects of zero-G] are the result of mechanical unloading and natural physiological processes. The muscles don't work as hard, and so they atrophy. The bones don't carry the same dynamic loads, so they demineralize. Both are the result of normal physiological processes whereby the body adapts to the environment, only expending what energy is necessary. The only way to treat that pharmacologically is to block those natural processes, and that opens up a really bad can of worms. All kinds of transporters would have to be knocked out, you'd have to monkey with the natural muscle processes, and God knows what else. Essentially, you're talking about chemically overriding lots of homeostasis mechanisms, and we have no idea if said overrides are reversible, or what the consequences of that would be in other tissues. My bet is bad to worse. As the whole field of endocrine disruptors is discovering, messing with natural hormonal processes is very very dangerous.

     Even if it worked with no off-target effects, you'd have major issues. Body development would be all kinds of screwed up, so it's not something you'd want to do for children or young adults. Since peak bone mass is not accrued until early twenties, a lot of your recruits would be in a window where they're supposed to still be growing, and you're chemically blocking that. Similarly, would you have issues with obesity? If your musculature is not functioning normally (to prevent atrophy), how will that effect the body's energy balance? What other bodily processes that are interconnected will be effected? Then you get into all the effects of going back into a gravity well. Would you come off the drugs (and thus require a washout period before you go downside, and a ramp-up period before you could go topside again)?

     Spin and gravity is an engineering headache, but a solvable one. Pharmacologically altering the body to prevent the loss of muscle and bone mass that the body seems surplus to requirements has all kinds of unknowns, off target-effects and unintended consequences. You're going to put people at severe risk for medical complications, some of which could be lifelong or even lethal.”

This is a compelling case that it is not possible to treat the effects of zero-G medically.  However, if for story reasons a workaround is needed, medical treatment is no less plausible than many devices used even in relatively hard Sci-Fi.

The task of designing spacecraft for a sci-fi setting is complicated by the need to find out all the things that need to be included, and get numbers for them.  The author has created a spreadsheet to automate this task, including an editable sheet of constants to allow the user to customize it to his needs.  The numbers there are the author’s best guess for Mid-PMF settings, but too complicated to duplicate here.

Rick Robinson’s general rule is that spacecraft will (in the sort of setting examined here) become broadly comparable to jetliners in cost, at about $1 million/ton in current dollars.  This is probably fairly accurate for civilian vessels, at least to a factor of 3 or so.  Warships are likely to be more expensive, as most of the components that separate warships from civilian ships are very expensive for their mass.  In aircraft terms, an F-16 is approximately $2 million/ton, as is the F-15, while the F/A-18E/F Super Hornet is closer to $4 million/ton.  This is certainly a better approximation than the difference between warships and cargo ships, as spacecraft and aircraft both have relatively expensive structures and engines, unlike naval vessels, where by far the most expensive component of a warship is its electronics.  For example, the ships of the Arleigh Burke-class of destroyers seem to be averaging between $150,000 and $250,000/ton, while various cargo ships seem to hover between $1000 and $5000/ton.

As mentioned in Section 5, some have suggested that the drive would be modular, with the front end of the ship (containing weapons, crew, cargo, and the like) built separately and attached for various missions.  This is somewhat plausible in a commercial context, but has serious problems in a military one.  However, the idea of buying a separate drive and payload and mating them together is quite likely, and could see military and civilian vessels sharing drive types.  (This is not as strange as present experience would lead us to believe.  It was only during WWII that military aircraft clearly separated from civilian ones in terms of performance and technology.)  This simplifies design of spacecraft significantly, as one can first design the engine, and then build payloads around it.

One common problem during the discussion of spacecraft design is the rating of the spacecraft.  With other vehicles, we have fairly simple specifications, such as maximum speed, range, and payload capacity.  However, none of these strictly applies in space, and the fact that spacecraft are not limited by gravity and movement through a fluid medium makes specifying the equivalents rather difficult.  Acceleration and delta-V obviously depend on the masses of the various components, which can be changed far more readily than on terrestrial vehicles, and cargo capacity is limited only by how long you’re willing to take to get where you’re going.  A replacement might be a series of standard trajectories, and the payload a craft can carry on them.  This works well if all of the spacecraft being rated are generally similar in terms of performance, and take similar trajectories in reality.  However, it does not work as well in a scenario where different types of ships take wildly different trajectories with different amounts of cargo.  In that scenario, ships might be rated by the minimum time for certain transfers (Earth-Mars at optimum, for instance) with a specified payload, either a fixed percentage of dry mass, or a series of specified masses for various sizes of ships.  This allows a comparison between ships of different classes, but within a class (liner, bulk cargo, etc.) the first method would probably be preferred.

A related problem is the selection of an appropriate delta-V during preliminary design.  In some cases, this is relatively easy, such as when a spacecraft is intended to use Hohmann or Hohmann-like trajectories, as numbers for such are easily available.  But such numbers are inadequate for a warship, or for any ship that operates in a much higher delta-V band, and unless the vessel has so much delta-V and such high acceleration that Brachistochrone approximations become accurate (and even then, if the vessel is not using a reactionless drive, the loss of remass can throw such numbers off significantly, unless much more complicated methods are used, numerical or otherwise).  The author has attempted to fill this gap by creating a series of tables of delta-Vs and transit times between various bodies, with the tables giving the percentage of the time that a vessel leaving one body can reach another within a specified amount of time with a given amount of delta-V.  The tables can be found at the end of this section.  Table 9 covers Earth-Mars transits, while Table 10 describes Earth-Jupiter transits.

The tables are generated in MATLAB by solving Lambert’s problem for a large number of departure days and transit times, and calculating the delta-V to go from stationary relative to the departure planet to stationary relative to the destination planet.  This involved assuming that there was a single instant delta-V burn at each end, which is a good approximation if the burn time is short compared to the transit time, as it would be for chemical or most fission-thermal rockets.  For systems which burn a significant amount of the time, this approximation is not as good, and the tables should only be used as a general guide to the required delta-V.  

Each table is the composite of 16 different tables generated with different starting geometry, and with each table containing data from at least one synodic period.  Note that this was all done in a sun-centric system, and that the delta-V necessary to deal with either planet’s gravity well was not included.  This will add some extra delta-V, the necessary amount shrinking in absolute terms as the overall delta-V is increased due to the Oberth effect.  The decision to not include escape and capture delta-V was made because to do otherwise would have involved specifying reference orbits to escape from and capture to, and would have added significant complexity to the program at a minimal gain in utility for most users.  

One thing that is apparent from these tables is the degree to which Jupiter missions are more hit-or-miss than Mars missions.  For Jupiter transfers, 84% of the options are either going to be viable all of the time, or not going to be viable at all.  For Mars, the equivalent value is only 56%.  Some of this is due to the much larger and more variable time increments used in the Jupiter calculations, but much of it is due to the fact that the geometry changes significantly less between Earth and Jupiter than it does between Earth and Mars.

It should also be noted that these tables are an attempt to find an average over all possible relative positions of the two bodies.  For the design of an actual spacecraft, analysis would instead start with modeling of geometries over the projected life of the spacecraft.  The approximations given here are reasonably close for theoretical use, but should not be used to plan actual space missions.  

Heat management is a vital part of the design and operation of a space vessel, particularly a warcraft.  Section 3 mentioned some of the issues with regards to stealth, but a more comprehensive analysis is necessary.  There are two options for dealing with waste heat in battle: radiators and heat sinks.  If the waste heat is not dealt with, it would rapidly fry the ship and crew.  

All space vessels will need radiators to disperse the heat they produce as part of normal operations.  If using an electric drive, power (and therefore waste heat) production will be no higher in battle then during cruise.  This would allow the standard radiators to be used indefinitely during battle without requiring additional cooling systems.  The problem with radiators is that they are relatively large and vulnerable to damage.  The best solution is to keep them edge-on to the enemy, and probably armor the front edge.  The problem with this solution is that the vessel is constrained in maneuver, and can only face one (or possibly two) enemy forces at once without exposing the radiator.  If the techlevel is high enough to make maneuver in combat a viable proposition, then radiators are of dubious utility in combat.  On the other hand, the traditional laserstar battle suits radiators quite well.  The faceplate and the forward edge of the radiators are always pointed at the enemy, and almost all maneuvers are made side-to-side to dodge kinetics.  The only problem is vulnerability to a direct kinetic hit.  If a projectile were to arrive precisely edge-on, it could tear the entire radiator in two.  Bending the radiator slightly would eliminate this vulnerability, but would also increase armor requirements.  However, even a bent radiator would still have issues with grazing impacts.  A projectile coming in very close to parallel with the radiator’s surface would tend to tear a long hole in it, as opposed to the small hole left by a projectile traveling perpendicular to the surface.  However, the low-incidence projectile would have to penetrate much more material, so kinetics designed for such attacks would naturally have more mass or fewer pieces of shrapnel than one designed for normal attacks.

Heat sinks avoid the vulnerability to damage of radiators, but have a drawback of their own.  By their very nature, they have a limited heat capacity, which places a limit on how much power a ship can produce during an engagement, and thus on the duration of an engagement.  If the heat sinks fill up, the ship would begin to fry unless the radiators were extended immediately.  In the game Attack Vector: Tactical, extending the radiators is used to signal surrender.  Obviously, the heat clock is a major disadvantage, but it is necessary when the vessel expects to expose several aspects to the enemy.  

One topic that briefly needs to be addressed is electric propulsion. In discussions, VASIMR-type engines are usually considered the baseline.  However, Dr. Joshua Rovey of Missouri S&T told the author that Hall Effect thrusters today are capable of the sort of performance that VASIMR is currently promising after development is finished.  VASIMR is apparently getting attention due to good marketing people.  

Another topic that deserves discussion is the effect of nuclear power on spacecraft design.  For large warcraft, nuclear power, both for propulsion and for electricity is a must-have.  Even if the design of solar panels advances to the point at which they become a viable alternative for providing electrical propulsion power in large civilian spacecraft, there are several major drawbacks for military service.  The largest is that solar panels only work when facing the sun, unlike radiators, which work best when not facing the sun.  The distinction between the two is important, as it is nearly always possible to find an orientation which keeps the radiator edge-on to the enemy and still operating efficiently, while a solar panel must be pointed in a single direction, potentially exposing it to hostile fire.  A solar panel is particularly vulnerable to laser fire, as it is by nature an optical device.  While hard numbers on this are surprisingly difficult to find, it appears that damage will probably occur to photovoltaics when exposed to intensities of around 300 KJ/m2, for short pulses (<10-4 seconds), with threshold requirements increasing from there as the pulse length increases.  For a CW laser (>1 second), the power flux required for damage is approximately 10 MW/m2.  Photovoltaics can also be attacked using small particles such as sand, as described in Section 7 for use against lasers.  While a full analysis of the potential damage is beyond the scope of this section, it appears that sand would be a reasonably effective means of attacking photovoltaics, particularly given the large area involved.  The size of a solar array also complicates maneuvering the panels edge-on to the incoming particles, and could potentially raise structural concerns.

Radiators, on the other hand, are more resistant to damage.  Firing lasers at them will only decrease the thermal efficiency of the reactor slightly, as the radiator is designed to disperse heat.  Particle clouds that are designed for surface effects would be ineffective against a properly-designed radiator, or at very best reduce the emissivity by a small amount.  Small pieces of shrapnel designed to pierce the radiator entirely would be the best means of attack (described above), as fully armoring a radiator is likely to be impossible because of the mass requirements.  

However, it could be argued that this ignores the vulnerability of the reactor itself to damage.  While in Hollywood, “They’ve hit the reactor!” is usually followed by a massive explosion, that is not the case in reality.  First, the reactor is a very small target, usually shielded by the bulk of the ship, so it’s unlikely to be hit in the first place.  Second, nuclear reactors simply do not turn into bombs under any circumstances, and particularly not random damage to the core.  The few cases in history in which a reactor has gone prompt critical (SL-1 and Chernobyl being the best-known) were caused by poor procedure, and are vanishingly unlikely to happen due to random damage.  

That said, it still seems a potentially bad idea to put all of one’s eggs in a single basket.  Solar panels are highly redundant, but the reactor could still be put out of action with a single hit.  The response to this is fairly simple.  First, there are reactor designs using heat pipes that have sufficient redundancy to continue operation even if the reactor core itself is hit.  The specific heat pipe will be put off line, but if the design has 150, that’s not a great worry.  Second, the reactor and associated gear (power converters and such) are buried deep in the ship, where they will be difficult to get at, and the power converters can be duplicated for redundancy.

One last concern is the ejection of reactor core material after a hit, and the potential for said material to irradiate the crew. This is also probably minor, as the crew is still being shot at, and the spacecraft will have some shielding against both background radiation and nuclear weapons. (Thanks to Dr. Jeffrey King of the Colorado School of Mines for providing much of the material on space nuclear power and propulsion.)

A couple of other issues with nuclear power are relevant and of interest.  The first is the choice of remass in nuclear-thermal rockets.  While hydrogen is obviously the best possible choice (the reasons for this are outside the scope of this paper, but the details are easy to find), it is also hard to find in many places.  With other forms of remass, the NTR does not compete terribly well with chemical rockets, but it can theoretically use any form of remass available.  The biggest problem with alternative remasses is material limits.  With most proposed materials, oxidizing remasses will rapidly erode and destroy the engine.  The alternatives to avoid this are rhenium and iridium, which are both very expensive, explaining why they are not in use today.  However, both elements are common in asteroids, making them viable choices in a setting with large-scale space industry.

As discussed in Section 7, vibration is a serious issue for laser-armed spacecraft.  Any rotating part will produce vibrations, and minimizing these vibrations is of interest to the designer.  While there is undoubted a significant amount that could be done to reduce the vibrations produced by conventional machinery (the exact techniques are probably classified, as their primary application is in submarine silencing), it seems simpler to use systems with no moving parts, which should theoretically minimize both vibration and maintenance.  Heat pipes, as mentioned above, are an entirely passive means of moving heat around, both from a reactor to an energy converter (which could mean a turbine, a thermocouple, or any of the other wonderful things engineers can think of) and from the energy converter to the radiator.  There are also electromagnetic pumps for liquid metal which while not entirely passive, but will cut down on the vibration load.

There are even proposed systems of energy conversion which are reasonably efficient and involve no moving parts.  The best-known of these proposed systems is probably the Alkali Metal Thermal-to-Electric Converter (AMTEC), which has been extensively studied.  However, a recent effort by NASA to bring the technology into deployment failed, giving the technology a bad name.  There are some, however, who believe it still holds promise.

If systems like AMTEC are not available, the spacecraft will have to use conventional hat engines.  These are likely to use one of the two standard thermodynamic cycles, the Brayton cycle (gas turbine), and the Rankine cycle (steam turbine).  The primary difference between the two is that in the Brayton cycle, the working fluid remains a gas throughout, while in the Rankine cycle, it moves from liquid to gas and back again.  In theoretical design, radiators are normally sized assuming constant temperature throughout, which is true for most Rankine cycle systems (as the radiators are where the fluid condenses at a constant temperature) and produces the well-known result that radiator area is minimized when the radiator temperature is 75% of the generation temperature.  However, this is not true for Brayton cycle radiators.  There is no convenient mechanism to release the necessary heat at a constant temperature, so the radiator performs differently as the gas cools.  There is not a simple formula here, but an iterative procedure can be used to minimize radiator area for an ideal system (which is close enough for our purposes).  Use of this method does require some knowledge of the basics of gas turbine propulsion, but it is not terribly esoteric. (Thanks to Dr. David Riggins of Missouri S&T for presenting this material in class. Those who have had more experience in propulsion and fluid dynamics might recognize some simplifications of the explanatory material, and some nomenclature changes. This was intended to hold down the length of this section and clarify it without sacrificing accuracy of the results.)

An ideal gas turbine can be thought of as being made of 4 separate stages.  First, isentropic compression, which means that there is no heat transfer and all energy put into the system by the compressor, is used to compress the gas instead of heating it.  Second, isobaric (constant-pressure) heat addition, which occurs in the reactor.  Third, isentropic expansion through a turbine, which outputs mechanical work (the goal of this whole process).  Lastly, isobaric heat rejection, through the radiator, which returns the working fluid to the condition it was at before entering the compressor.  The compressor and turbine are defined by their pressure ratios, written as πc and πt respectively.  The pressure ratio is the pressure of the fluid after the component divided by the pressure ahead of the component.

For this method, values for Cp, γ, ηc, ηt, and T3 must be selected.  Cp is the constant-pressure specific heat capacity of the fluid, while γ is the ratio of specific heats.  Definitions and values for various fluids can be found online.  ηc and ηt are the efficiencies for the compressor and turbine respectively.  Values between 0.8 and 0.9 are probably reasonable.  T3 is the temperature at the outlet of the heat addition stage, and is normally set by the design of the reactor itself.  Representative values might be 1600-1700 K for a conventional nuclear reactor, although higher values are possible.  (All temperature values throughout should be in Kelvin, not Celsius.)  

The first step is to select a value for πc, with anything from 2 to 10 being plausible.  Because it is a closed system, πt will be equal to 1/ πc.  Once this is known, it is possible to calculate T4 (temperature downstream of the turbine) using .

At this point, a value for T1 must also be selected.  This is the temperature at the entrance to the compressor.  Using this, the value for T2, the temperature at the compressor exit, can be calculated using .  This allows the overall efficiency of the power-generation system, &eta (work output/heat input), to be calculated using .  All of this information can then be used to find the radiator area per unit work output (A’, m2/W) with where σ is the Stefan-Boltzmann constant (5.670373×10-8) and ε is the emissivity of the radiator (0.9-1.0). Once you have this value, select a different value of T1, and repeat the rest of the paragraph.  When a minimum has been found, select a different value for πc and repeat the entire procedure until a global minimum has been found.  It would probably be a good idea to use a spreadsheet to automate this.

Table 9: Earth-Mars Transits
Maximum Transit Time (Must Arrive No Later Than, Days)
Table 10: Earth-Jupiter Transits
Maximum Transit Time (Must Arrive No Later Than, Days)
by Byron Coffey (2016)

Well, I was going to post the second part of The Shipping Trade today, except that writing it didn’t happen because of day job, and so forth. Then, I thought I might post a sketch of the ship involved, just to give y’all an idea of what you’ll be looking at, but then that would require me to go out and hire a scanner. That, and I made said sketch, and then looked at it, and then concluded that I couldn’t possibly inflict such a terrible picture on my readers…

So permit me, please, instead to sketch a verbal picture for you of the

CMS Greed and Mass-Energy

To start with, Greed and Mass-Energy is atypically large for a free trader; in those leagues, which principally deal in small, high-value-to-mass/volume cargoes, lugging around 40,000 tons displacement of cargo is huge. (It’s still not in the major freight line league, though; those guys can use freighters that are million ton-displacement behemoths.) Thus, the shipcorp that owns her (it’s essentially a syndicate of officers, crew, and former crew, with executive power vested in the captain-owner) is pretty prosperous to be able to cover her running costs. Dealing in brokered cargo actually isn’t her main business – she specializes in contracts like the RCS-assembly charter from Kerbol to Kythera she just left, but an empty hold is a hole that drinks money, so you take the cargo when you can get it.

Also, obviously, at a size like that, she’s not streamlined, or built to land planetside (gravity wells being acutely expensive); and is even rather more massy than anything that most stations like to have dock directly to them. Her cargo’s generally ferried to station, or upwell and downwell, by local lighters at each end of the trip. Rather, she’s built very much in the classic mode; a long, relatively thin, open-frame truss structure. Attached to that, going from fore to aft, we find these different sections of the ship:

Right at the bow, sitting on the end of the main truss, is the command capsule, an ellipsoid slightly stretched along the ship’s main axis, relatively tiny compared to the rest of the ship, and containing, for starters, the bridge and associated avionics systems. (The bridge is actually buried in the center of the capsule, for its protection; it’s displaced off to the front end of the ship, however, because the command capsule is also where the primary sensors are housed to keep them out of the way of cargo, fuel, and drive radiation, and this positioning cuts down on sensor lag. It’s still pretty safe; it’s not like anyone’s going to be shooting at them.) The first of the other two notable features it houses is docks and locks, right for’ard on the axis where it’s easiest to match thrust and spin, which usually houses a couple of cutters used for taking the crew ashore and for occasional maintenance, and a skimmer for in-field refueling. (The fuel itself doesn’t pass through here – the skimmer docks aft to offload what it scoops. No fuel for’ard of the support plate, that’s the general rule.) The second, aft by the truss, is the robot hotel for all the little space-rated utility spiders you may see now and them crawling about the structure doing maintenance, thus saving the engineering department any need to get suited up and go outside for routine work, although they still may need to do so from time to time.

Just aft of that, accommodations and secondary systems are housed in a toroidal gravity wheel. This is actually a very unusual design feature in an Imperial ship-class; just about everyone and especially the spacer-clades are genetically adapted to microgravity, and the spacer-clades prefer it, as a rule; but the Cheneos-class architects originally designed her class for near-frontier work, and included this for occasional passenger service. Greed and Mass-Energy only rarely carries passengers, so they keep it geared all the way down, producing only a tenth of a standard gravity, which doesn’t offend the spacer-clades all that much. There’s a second, smaller wheel rotating inside it to null out the gyroscopic effects; it’s used to house some other equipment that likes a little gravity, but for the most part, this one’s just a countermass.

(The wheel does, however, provide enough gravity to let the CELSS Manager run a pretty decent microbrewery in the spare volume, and perhaps more importantly, provides a place where you can drink it off-shift without suffering from a nasty case of the zero-g bloat. [Remember, folks, bubbles don’t rise in microgravity!] And apart from crew morale, having decent beer makes for good PR when traders meet.)

These areas, incidentally, are one of the few places on board where the really high-tech ontotechnological stuff makes an appearance, in the form of inertial damping. The people who built her liked microgravity, and weren’t all that keen on losing that while under thrust, especially since she was built to fly brachistochrones or near-brachistochrones (bulk tankers and ore freighters, etc., are usually built to fly economic minimum-delta/Hohmann transfers; no-one else wants to wait that long for their cargo) and so would be spending most of her time under thrust. The job of the inertial dampers is to apply the thrust of the drives evenly across the entire area’s structure and everything in it, thus ensuring that no-one actually feels any acceleration, and the lovely microgravity environment is preserved. (It also avoids having to come up with some wretchedly complicated gimbal arrangement for the already wretchedly complicated seals-and-bearings for the gravity wheel, no longer having to do which is something that made architects particularly grateful for this innovation.)

Behind this, the cargo. ‘Way back along the truss there is a very large, solid plate, the support plate. The cargo containers are simply stacked “atop” – by which we mean for’ard – of it, in six big blocks arranged around the axis with sixfold symmetry (this arrangement being a reasonable compromise between use-of-volume and convenient straight lines), and are designed to lock to the plate, the truss, and each other to form a solid interlocked structure. There’s no hold or other walls around the cargo; the containers are themselves spacetight when they need to be, and so lighters can just drop them into place and pick them up freely while in port.

The breakbulk cargo, on the other hand, is messy. It has to be podded up individually when not spacetight, and then individually lashed down and made secure atop the cargo container stacks. This annoys the cargomaster, which is why breakbulk is unpopular these days despite the fact that breakbulk shippers usually pay a premium in exchange for you having to do this (the “lash comp”). Actually, what really annoys the cargomaster is that she can punch a button and have the ship automatically query the v-tags on the container cargo for its mass stats, and so forth, whereas for breakbulk she’s got to recall her Academy training, dig out the spreadsheets, and work out the corrections to the center-of-mass-and-moment-of-inertia chart by hand. Well, still by computer, but you know what I mean.

Aft of the support plate, still in sixfold symmetry, you have the bunkerage – fuel tanks, stacked three deep in multiple rows, all filled with slush deuterium, running right to the stern, where they surround the cylindrical shroud of the mostly-unpressurized engineering hull (you can take a crawlway right back along the truss to the small, pressurized maneuvering room back this far, should you need to examine the drives close-up in flight, but the actual machinery space isn’t), which contains the interlinked systems of the main power reactors and the fusion torches themselves, strapped to the aftmost extent of the main truss.

And there are lots of fuel tanks. Even though said fusion torches are miracles of a mature nuclear technology, capable of achieving near-theoretical efficiencies and outputs and delta-v per unit fuel that routinely makes naval architects from less advanced civilizations throw down their slide rules in despair and weep into their terrible coffee-equivalents, the one unchangeable rule of space travel is that your mass ratio is always much, much less favorable than you might want it to be.

Good thing deuterium’s so cheap, isn’t it?

…and most prominently of all from a distance – dominating the entire view of the ship from a distance, by area as well as by temperature – sweeping out from among the fuel tanks (although comfortably retracted to sit alongside them, leaving approximately a sixth of their radiative area useful, while idling in dock – the vast panels and pipework of the heat radiators. Because the other one unchangeable rule of space travel is that you always have waste heat, too damn much waste heat, and you’ve got to get rid of it somehow. Especially once you fire up those fusion torches. (The radiators, however, unlike the rest of the ship, have only fourfold symmetry – so that they can be perpendicular to each other when unfolded, because there’s very little point in radiating heat right back at your own radiators.)

Rockets Are Not Hotels

RocketCat sez

See that space fantasy at the top? Yep, the good ol' Starship Enterprise. There are two glaring thing wrong with it, right off the bat.

First off, the direction of "down" is almost but not quite totally FUBAR. We'll get into that later.

But secondly, which of the other two ships look most like Enterprise? In terms of blue pressurized habitat module. Yep, the freaking Queen Mary, an ocean liner. Not the Lewis design, a nuclear thermal rocket spacecraft created for a Mars mission. I hope you see the problem.

If you look at most blueprints for the various iterations of the Starship Enterprise, you will notice that every single part of the spacecraft interior is pressurized, with doors, rooms, and toilets. The corridors are wide enough for five people to walk abreast on nice carpeted floors with indirect lighting.

This is ludicrously wrong. And it is not just Star Trek that does this, pretty much all of media science fiction has ships like this. TV Tropes calls this fallacy "Starship Luxurious".

This is an extension of the "Rockets are Boats" fallacy. Passenger aircraft and luxury liners have their entire interior pressurized because so is everything else at sea level on a planet with a breathable atmosphere. For free. So careless starship designers, without a thought, made the unconscious assumption that spacecraft would be totally pressurized as well.

Wrong. Tain't no air in space, and atmosphere is expensive when you have to cart it up out of Terra's gravity well. Not to mention the expensive pressurized hull that has to encase it.

And it is not just the cost of hauling it up the gravity well, the spacecraft's engine has to accelerate the mass of all that junk. Every Gram Counts, so every gram of carpeting, atmosphere, and pressure hull is one less gram of payload, i.e., the reason the spacecraft was created in the first place. Payload is what you are being paid to load, less payload means less pay. See The Tyranny of the Rocket Equation.

In the real world, spacecraft will be mostly tanks of propellant, propulsion system, payload bays, and a lacy lattice-work of support struts holding everything together. The part the people live in will be a tiny pressurized habitat module tucked away somewhere.

Ignorant starship designers have the unconscious assumption that the important part of a spacecraft is the crew, so they designed ships with their priorities reversed. Their ships were mostly gigantic habitat modules with a tiny engine stuck to the rear. Their ships are also ludicrously wrong. If the designers thought about it at all, they might grudgingly include a tiny fuel tank. Which is like the cherry on top of their big icecream sundae of Fail.

So quit drawing ship blueprints with every square inch pressurized and human-accessible. On a real spacecraft if the ship's engineer has to repair the propulsion system, heat radiators, power plant, propellant tanks, or anything like that, they will have to put on their space suit. They will not have the luxury enjoyed by Scotty the engineer, waltzing down a carpeted floor in a shirt-sleeve atmosphere.


The Enterprise's corridors seemed awfully roomy, they were about twice as wide as they should have been. In fact, the whole ship was too roomy. Space is at a premium in any kind of enclosed environment. Anyone who's ever been aboard a submarine—or even an aircraft carrier, for that matter—knows that they are designed for the maximum utilization of their volume. Efficiency is a necessity, and a spaceship is going to have to be designed the same way.

In fact, the requirements of a spaceship are much more stringent—for instance, the interior atmosphere must be maintained with the correct combination of gases, at the right temperature, pressure, humidity, and ionization, to maintain not just the lives, but the comfort as well, of the crew. The margins for deviation are narrow; therefore, every cubic inch of interior volume means airspace that must be maintained—and maintenance requires the expenditure of energy. When you have to conserve your ship's power, you don't waste airspace.

The reason for such broad corridors? They had to be wide enough for a camera dolly, cables and a film crew.

To attempt to show that the ship was cramped would have required the construction of cramped sets—which are harder to work with and would have meant much more in the way of production time.

(So, instead, we're told that the ship has power to waste—it's implied, not specifically said. But if that's so, then we should never see a story in which maintenance of life-support functions become a critical factor for building suspense. A ship can't be both wasteful and limited.)

Another example: the turbo-elevators. These were the machines that took the various crew members from one part of the ship to another. Cood idea; especially as we are told that the thickest part of the Enterprise‘s disc is twelve stories thick.

But—the elevators seemed to be the only way to get from one deck to another. If the ship's power supply were cut off, every deck would be separated from every other. Oh, well, not really—somebody could always crawl through the air vents, or through the Jefferies Tube, or down one of the ladders which we saw very infrequently. All except for the bridge. Cut off the turbo-elevators and you isolate the bridge. Tsk. That's bad designing. Illogical.

Another one: the Captain's cabin. Or anybody’s cabin for that matter. They were all redresses of the same set. If any of those cabins had a bathroom, it was never shown. There weren't even any doors to imply a bathroom. We were never shown the cleanup facilities on the whole ship—not even the sick bay. And the Enterprise was on a five-year mission—isn't that a long time to hold it? Isn't that carrying it a bit too far?

Also, about those cabins—all of the major officers aboard the ship had their own cabins, and roomy places they were too. No complaint here, but what about the crew's quarters? Those were never even shown or suggested. Did each member of the crew have a cabin too? That would have made the Enterprise more of a hotel than a starship. Or did they have bunkrooms?

If they did have bunkrooms, how come they were never mentioned or shown? How come we never got into the crew's lives?

Or was the crew just a collection of some 400-odd androids to walk up and down the halls—scenery behind the main characters, to be moved around as necessary, but not really important to the story except as another part of the background to support the overall illusion?

From THE WORLD OF STAR TREK by by David Gerrold (1973)

The ships above are tail-sitters, so properly avoid the "wrong way is down" problem. But the artist made the second problem much worse. Apparently they figured the entire interior of the spacecraft was for habitable volume. Notice what they got wrong? Well, where the heck is the space for the rocket engines? I lay the blame for this at the artist, I know from experience that writer Jack Williamson knew better.

Modular Construction

An attractive notion is the practice of constructing one's spacecraft out of mix-and-match replaceable components. So if your spacecraft needs to do a planetary landing you can swap the low thrust ion drive for a high thrust chemical rocket. In Charles Sheffield's The MacAndrews Chronicles, the protagonist just calls her ship "the assembly", customized out of whatever modules it needs for the current mission contract.

For detailed examples, see the Boeing Space Tug, NASA Space Tug, JPL Modular Hab System, Minimal Volume Spacecraft Cabin, and the Flyaway Engine.

This will also make ships basically immortal. It will also make it really easy for space pirates to fence their captured prize ships. All they have to do is get the prize ship to the spacecraft equivalent to an automobile chop-shop. There the ship vanishes as an entity, becoming an inventory of laundered easily sold anonymous ship modules with the serial numbers filed off.

One can also imagine junker spacecraft, lashed together out of salvaged and/or junk-heap spacecraft modules by stone-broke would-be ship captains down on their luck.

Or mechanically inclined teenagers who want a ship. This would be much like teens in the United States back in the 1960's used to assemble automobiles out of parts scavenged from the junkyard, since they could not afford to purchase a new or used car. Such teens would gain incredible practical skills as spacecraft mechanics. I wonder if this is how Kaylee from Firefly learned her trade.

Yet another scenario is Our Hero stranded in the interplanetary Sargasso Sea of lost spacecraft, trying to scavenge enough working modules from three broken spacecraft in order to make one working spacecraft.

Rick Robinson notes that attractive as the concept is, there are some practical drawbacks to extreme modularity:

Rick Robinson: This is not necessarily an argument for true modular construction, with drive buses hitching up to payloads on an ad hoc basis like big-rig trucks and trailers. Building things to couple and uncouple adds complexity, mass, and cost — plug connectors, docking collars, and so forth. Moreover, drive buses intended for manned ships need to be human-rated, not just with higher safety factors but provision for supplying housekeeping power to the hab, etc. But these things, along with differing sizes or number of propellant tanks, and so forth, can all be minor variations in a drive bus design family.

Rick Robinson: True modularity is by no means a given. But some features of modularity, call it demi-modularity, are inherent to deep space technology.

You probably want to keep your propellant tanks separate from the corrosive, explosive stuff we breathe. Drive engines are essentially bolted onto the tail. Generally the major parts of a deep space ship don't have to fit together snugly. If you want to hang something out on a bracket you probably can.

Nyrath (me): There might be a brisk trade in "interface modules", that would connect modules made by different manufacturers. I'm reminded of the Apollo ASTP Docking Module used in the Apollo-Soyuz mission. This was a tiny airlock module with a NASA style docking collar on one end, and a Soviet style docking collar on the other.

Not to mention the International Space Station Pressurized mating adapters.

Rick Robinson: Modular design is always a tradeoff. You get more operational flexibility, at cost of more complicated/heavier/weaker connections. Integral designs will be favored when the components will consistently be used together.

Much will depend on tech. Torch type drives and even 'conventional' nuke electric drives pretty much have to be mounted on a pylon, which sort of invites the option of unbolting it from the rest. OTOH, as you note, the drive section may well have its own control center. And since the rest of the ship sits on top of the pylon, it's a fine line between 'pylon' and 'chassis.'

On naming, I could also make a case that the crew hab compartment is the main component, and so would be named. Especially if it is a spin gravity structure. And 'spaceships' may end up having more than one name, just as a named train might included Pullman cars with names of their own.

And if ships are highly modular, some terms might be borrowed from railroading. For example, 'consist' as a noun (pronounced CON-sist) for the whole assemblage. Thus, 'The Ty Cobb departed Mars with a consist of [such and such modules].'

Amusing side note: modular spacecraft reverse the order of trains: the 'locomotive' or drive engine is at the back (more precisely the base), while the 'caboose' or control cabin might well be at the front/top.

From SPACESHIP DESIGN 101 by Rick Robinson (2009)

The ship had several “holds,” actually just enormous, detachable cylinders adapted to carry cargo or passengers. Some of these were sealed and Shaw was reluctant to reveal what was in them. For an unabashed smuggler, that suggested to Thor that some things were unacceptable, even in the freewheeling society of the space settlers. Drive, holds and control were all in separate modules, connected by struts and passage tunnels. It was a common system for ships never intended to make planetfall, allowing great flexibility of size and function. “Also,” Shaw told Thor with a sharklike grin, “it makes it very difficult to keep up with how many and what type of ships are out here. If the authorities were looking for Spartacus, I'd break her up and rearrange her modules with other ships. You can have as many ships as you have command and drive modules.”

It must be a nightmare for customs authorities,” Thor observed.

“We do our humble best. Hijacked ships are never found again because they’re broken up and utilized or sold off as modules. You’ll have to go to a ship sale some time. There’s no pirate hangout like in the holos. Word just gets passed that there’s going to be ship hardware for sale and everybody just sort of congregates at a certain set of coordinates that all the bartenders seem to know about. I've seen whole government military vessels broken up and sold, weaponry and all.”

“Military!” Thor said, aghast. “I thought that was supposed to be impossible. Are there hijackers powerful enough to attack a Space Service ship?”

“Who attacks?” Shaw said. “Usually, it’s just a matter of paying someone to look the other way. The degree of corruption in the higher echelons of the military is immense and has increased tremendously in the last fifteen years. It was historically inevitable. I’ll let you read my monograph on the subject. There are other ways that service vessels make it onto the black market. Sometimes, a whole crew will decide to take early retirement from the service and bring their ship along with them.”

“I think that society out there will be quite different from what I anticipated,” Thor mused.

“I can guarantee it,” Shaw said.

From THE ISLAND WORLDS by Erick Kotani and John Maddox Roberts (1987)

Here's a fat-ass merchant ship that prefers to hand off its cargo in orbit but of course needs to land now and then. It would be a standard hull in Cepheus Engine or MgT First Edition or a partially streamlined hull in CT. It has no business landing anywhere without a beacon, landing lights, and a level field. But of course sometimes it has to. To facilitate docking the original ship has a docking module stuck on the top.

Whether you have rockets, reactionless drives (boo!) or warp drive handwaving docking is a maneuver that gives pilots grey hairs early on. The docking module bears the brunt of it. It's easier and faster to replace a module than an airlock that requires welding a heat resistant hull.

Big companies usually stocked a few modules at starports so if their merchant ships called and were in a hurry or banged their module up they could just swap modules and be on their way to make that hot delivery by the contract deadline. Swap modules and jump the hell out.

As an aside the module also held 10 dTons of cargo for really fast transfer of priority deliveries. If you were lucky the shipment was throw pillows. If you were unlucky munitions but hey no pressure there.

Then some ship builder realized there were a bunch of these 'kegs' just dumped in starports awaiting repair or surplussed. He bought one, stuck a drive and power plant on it and voila! a 20 dTon launch!

The damned things proved popular. They could aid in docking maneuvers since their little bitty engines had more fractional control than the big ship engines. But wait there's more!

The 'caps' at both ends on the keg came off easily (for repairs ... not while docking, that would be a bad thing masquerading as a design feature).

People began modding the modules.

In the 'standard' configuration this keg has the top module fitted with proper landing legs. It docks "nose in" the the ship and of course she can't be used to dock with this feature. If you expect to land her on rough terrain, say in the case of a delivery to a mining outpost on a rocky moon, they provide more stability than those stubby docking clamps.

More modules will be forthcoming. Suggestions will be entertained. I'm pretty sure there's a market for a fuel scoop version with a streamlining module (I know it breaks the at-the-time-of-construction rule but I feel it's justified.) There will be other modules with waldos and power tools for salvage and mining.

(ed note: you might find some inspiration in the modification kits for the Boeing space tug and the NASA space tug)

From SORRY BUT ... by Rob Garitta (2016)

Ship of Theseus Paradox

I thought of a problem with modular designs, based on the ancient Ship of Theseus paradox.

This is my Grandfather's ax.
This is my Grandfather's ax.
My Father replaced the handle.
I replaced the ax-head.
This is my Grandfather's ax.

Is it really still Grandfather's ax or not?

Plutarch first wrote about the paradox in 75 CE. But it was that 17th-century smart-ass Thomas Hobbes who slipped the exploding cigar into the box. He asked the question: what if somebody saves the original discarded handle and ax-head, then assembled them into a second ax. Which one of the two axes is Grandfather's ax? Both, neither, the new one, the old one?

This sounds academic, until you apply it to modular spacecraft.

For purposes of insurance, liability, national registration, contract penalties, mortgages, and a host of other expensive issues; it is crucially important to know the identity of the spacecraft in question. Which ship exactly is being referred to in all those legal documents?

But what if the SS SkyTrash's modules are replaced and the old modules used to make a new ship? Legally which one is the SkyTrash? For that matter, intentionally making a stolen ship vanish by passing it through a spaceship chop-shop can make another set of legal headaches.

The problem of spacecraft identity has got to be legally nailed down.

Don't look to the Theseus Paradox for a solution. The problem was stated almost two thousands years ago and they are still arguing about it

Off-hand I'm not sure what a fool-proof solution would be. My first thought was to attach the identity of the spacecraft to some sine qua non "must-have" ship module. Unfortunately there does not seem to be any. Not all ships are manned, so the habitat module won't work. The only must-have module I see is the propulsion bus (otherwise you have a space station, not a spacecraft). However Captain Affenpinscher might find it strange that the identity of her ship has changed just because she swapped out the propulsion module.

I had a discussion on Google Plus with some of my brain-trust:

Winchell Chung (me): I was mulling over modular spacecraft design, when I suddenly realized I had a "Ship of Theseus" paradox on my hands. Does anybody have any bright ideas about where the legal identity of a spacecraft resides in?

Ray McVay: Keel? Drive core? The route, like railroad trains? In Black Desert, it would be with the AI (spacecraft's installed artificial intelligence)... Annabelle Li (a ship AI) has been two kinds of Heinlein and a CASSTOR, but kept her name. Possibly valid inspection certs for the given configuration is the legal identity of a ship... After all, no transit authority will let one boost in a franken-rocket without giving it a once-over...

John Reiher: I had a thought. There's one component that never changes on any ship. It can be updated, but it can't be replaced, otherwise the Insurance companies will void your policies.

What is it?

The Vehicle Identification Number (VIN) Box. It's a transponder with a unique ID and call sign. It can be customized at time of purchase, but it is the "ship". Anything attached to it becomes the ship.

No VIN Box, no ship.

Two or more VIN Boxes on a single ship, you're breaking the law and you have to inactivate all but one of them.

This is something the Banking and Insurance companies would come up with. They need something that can be unique to each ship, and nothing is more unique than a government issued, sealed, black box VIN Box.

William Black: Modular freighters in my future history setting the Command Module (CMOD) is what the freighter captain actually owns — in a single owner operator context, a transportation company being a single individual or corporate entity that owns many individual CMOD's. However I like +John Reiher's suggestion that the VIN box be the source of identity, and +Raymond McVay's suggestion that identity is locked to the ships AI has obvious merit especially in regards to spacecraft-as-characters. I may steal borrow either or both concepts with a note of attribution.

Raymond McVay +William Black, I proposed something similar in terms of Command Modules myself back in the day. The article actually illustrates +Winchell Chung 's point rather well.

John Reiher Nice thing about a VIN Box is that the issuing authority can put a time limit on how long it's good for before you have to go and get it renewed. Think of it as license plates for spaceships.

Winchell Chung Wow, lots of good ideas here.

I had dismissed the idea of using the habitat module for ship ID because a robot unmanned ship would not have one. But +William Black idea of command module has some appeal. If you narrow it to the module that has actuators and computers controlling the various other modules. So the command module is a box with cables or BlueTooth connections to control all the other ship modules. Plus an I/O port for the captain to issue commands to the command module, and so the CMOD can report status reading to the captain.

This would indeed make the CMOD the sine qua non of a spacecraft, worthy to be indelibly embossed with a serial number usable as the ship's identity.

Plug in a human usable control panel into the CMOD I/O for human manned ships.
Plug in an AI interface connecting the AI computer to the CMOD I/O for AI manned ships.
Plug in a sequencer interface connecting a moron computer to the CMOD I/O allowing the moron computer to execute a pre-programmed set of commands as if it were a space-going player piano.
Plug in a radio interface connecting a radio to the CMOD I/O for a remote-controlled drone ship.
Or any combination of the above

+Raymond McVay's idea of the ID of the ship tied to its AI has merit. The only thing is AIs are absurdly easy to clone (once you crack the DRM copy protection). It is hard to stamp a serial number on software running in a computer. Especially if the AI software is self-modifying, as human being are. How does one distinguish one AI from another?

+John Reiher idea for a VIN Box is probably my favorite. It is very much the sort of thing that banking and insurance companies would come up with. And the requirement for renewability makes perfect sense. However it must always contain a legal ID, for liability purposes. Much like automobiles. If somebody crashes their car into a building or something else expensive, then flees the scene on foot, the car's license plate may be expired but it still allows the police and building owners to discover who is liable for the damages.

This also vaguely reminds me about ship transponders in the Traveller role playing game. They constantly broadcasts the ship's unique ID and location. Civilian starships are legally required to have transponders always turned on, unless there are extenuating situations. Such as pirate corsair ships in the area, using the transponder to home in on their prey.

William Black I was thinking about that very same problem with AI's last night and this AM, as Winchell points out AIs are absurdly easy to clone. It is hard to stamp a serial number on software running in a computer. How does one distinguish one AI from another?

All I could come up with is this: A ship's AI is hardware configured so its inputs and outputs must plug in through an interface that is part of the VIN box, and these are highly tamper resistant. Both the AI and the VIN box are hardened in various ways, with both physical and with hardware/software safeguards against tampering.

In other words a ship's AI cannot function without a registered spacecraft with legal ownership.

Under the heading of future crime: Cracking the AI-VIN box security feature.

It's probably not impossible to do, but it might be very, very, difficult to do — a man with the right skill set could sell his labor at a high price on the black market. Skills would be comparable to a high-level safe-cracking expert combined with high level hardware/software expertise.

I was thinking of similar safeguards in relation to nuclear pulse systems in my setting as well.

John Reiher Actually, the current lines of research into AIs indicate that they will be very hard to "clone" as they will be as much hardware as human minds are. You can copy the data, but not the mind. (Unlike human brains, which have no I/O ports.)

But I do like +William Black's idea of tying the AI to the ship's VIN Box. That way the ship's AI is really part of the ship. But that lasts until AI's get equal rights. Then they will operate independently of a ship's VIN Box. Of course if they do get equal rights, there's nothing stopping one from buying their own VIN Box and leasing their "ship" to whoever can afford their terms.

Alistair “Cerebrate” Young I'm pretty sure in my 'verse the legal identity of a starship is vested in the leatherbound data rod/smart-paper folio in a safe in the captain's office (or welded in a suitable location on a drone ship), which is to say, the certificate of registry.

Components may come and components may go — and the Flight Administrator will faithfully update said folio's documentation of said components[1] each time or be hauled up in front of an Admiralty Court for seriously violating the Imperial Navigation Act — but CMS Gorram Freebies, Hull Number Eleven-Oh-Seven-Four-Two-Niner remains CMS Gorram Freebies, Hull Number Eleven-Oh-Seven-Four-Two-Niner whatever gets replaced up to and including said hull as long as it's operating under the same non-decommissioned certificate of registry.

(Of course, the bank that holds the note on your starship may have its own documented ideas on what exactly it holds the note on, and should the registry and the mortgage get out of sync on this point, your life may become... interesting.)

(ed note: a visit from the Repo Man)

((Equally of course, the spec plat in the engineering computers will also have its own documented ideas about what the ship's made of and can be expected to do, and when it gets out of sync, your life may also become interesting.

Also, depending on how out of sync it is with reality, potentially hot, noisy, and short.))

(ed note: spacecraft undergoes rapid explosive disassembly because the engineering computer's mathematical model of the spacecraft did not match reality)

[1] I like to imagine this as a nice hierarchically-organized document that begins with: Starship, free trader, Kalantha-class: 1
And ends with something like: Rivet, 4mm: 18,297

John Reiher Papers can be forged, data copied and manipulated. You need something that breaks if someone tries to open it. That's where the VIN Box comes in. Trying to crack one open is a operation in futility. The ID in the VIN Box is one half of a public encryption key that mates with a governmental key to validate your ship. If the two don't mate, you have a hacked box.

The Key that produces the public keys is private and in its own black box. The guys in the Ship Registry office just know that they have theses boxes that need ship's names, and that they already have IDs ready to go.

As an aside, you realize that we've been using modular space ships ever since we started building them. What do you think the Saturn V is? It's a one use modular rocket that you can put anything you want on top of the booster section and throw into orbit.

So that implies that a modular ship will have very good connections between the parts. On par with what a multi-stage rocket uses. Docking connections are just that, docking connections, and may or may not have the necessary structural strength for a space ship.

The type of connections that make up a modular ship most folks would call "permanent" but to folks in the business, they are temporary.

Alistair “Cerebrate” Young Side note on registries and transponders: of course, the Worlds being a not-exactly-unified group of polities, actual requirements on these points vary widely.

Even leaving aside the anarchic Rim Free Zone (the entirety of whose admiralty law could be summed up as "try not to hit anything that might complain"), the Accord on the Law of Free Space leaves it at the minimal "you should have a certificate of registry and a transponder that will squawk it out when queried". Local regulations, on the other hand...

(The Empire, for example, which holds that sovereignty begins with the individual, will happily accept self-signed registries and doesn't require much transponding, although lacking functionality in this area may leave you restricted to operating VFR and staying outside all regions of controlled space.

The other end of this particular bell curve is the Hope Hegemony, which wants transponders willing to disgorge pretty much any information you can think of on demand, including your code-signed visa from the Hegemony Bureau of Navigation and remote-slave ackles for your starship, and legally defines any vessel without such as "debris, subject to salvage and/or destruction at discretion".)

(ed note: "ackles" is Access-Control List. "Remote-slave ackles" means to grant the Hope Hegemony government the ability to seize command of your ship and fly it by remote control, at their whim)

John Reiher So +Alistair Young in this setting, if you want to be safe, you register your ship in the Hope Hegemony, which is the most restrictive polity and go elsewhere to do your business. (Or at least get a HH compatible VIN Box). 

Alistair “Cerebrate” Young Well, it depends on what you mean by "safe"...

Flying around most of the Worlds with an HH registry is kind of like sailing around Earth in a North Korean-flagged ship. It may make you welcome in Pyongyang, but the association with the local crazies may not do you any favors elsewhere...

(Common wisdom would probably suggest either a registry from one of the inoffensive, minor, single-system polities that hasn't really had the chance to offend anyone yet, or possibly an Imperial registry on the grounds that while they have offended lots of people, they're also notorious for sending gunboats after people who trouble them and theirs, so...

Neither of those is safe everywhere, though, which is why the Starfall Arc Free Merchant Confraternity suggests that the wise smuggler free trader goes to the trouble of discreetly procuring a number of suitable registries...)

+John Reiher I'll admit to being rather cynical where uncrackable devices are concerned, mostly because over my past IT career I've spent an awful lot of time watching notionally uncrackable software and hardware both be cracked, commonly within about a week of shipping...

So, the way I look at it is the standard software security way — never trust the client. If the papers and their cryptosignatures and the ship's "biometrics" and whatever other information you can gather match a recent update of the issuer's copy of the database, you can probably be sure that they are who they say they are, or at least who the issuer thinks they are. If you don't have that handy to verify against... well, it's time to take your best guess. 

Winchell Chung +John Reiher said: of course if they (AIs) do get equal rights, there's nothing stopping one from buying their own VIN box and leasing their "ship" to whoever can afford their terms.

Hmmmmm, interesting. I am reminded of the Spline aliens from Stephen Baxter's Xeelee novels. They are whale-like aliens who genetically engineered themselves to be exceptionally good spacecraft. They then rent themselves out to other races as spaceships for hire.

This also reminds me of the brain and brawn ships from Anne McCaffrey's The Ship Who Sang. The parents of severely deformed babies are given the option of having the baby engineered into becoming a shell person. They are encased in titanium shells and given a brain-computer link. Among other things they can be plugged into a starship, making a living ship. The process is expensive so the shell people come of age with heavy debts which they must work off in order to become free agents.

McCaffrey said: "I remember reading a story about a woman searching for her son's brain, it had been used for an autopilot on an ore ship and she wanted to find it and give it surcease. And I thought what if severely disabled people were given a chance to become starships? So that's how The Ship Who Sang was born."

John Reiher +Alistair Young it's a given that most hackers can crack the security of simple systems, but I'm talking about an ID that's at least 128 characters long, using the entirety of the UNICODE font character set, you have almost 39,000 glyphs. It will take the heat death of the universe to crack that code.

It would be easier to get a bootleg one that's already registered and in the system.

Also, I'm using a setting where while there are multiple governments, they do have treaties with each other and have agreed upon a common ship's registry system just to keep the confusion down and to prevent what you proposed: Smugglers with multiple transponders.

+Winchell Chung I'm also reminded of Eric the brain in the jar from Niven's Becalmed In Hell and The Coldest Place. He was the ship and could be put into any vessel as best as I can remember. 

Alistair “Cerebrate” Young +John Reiher Well, that depends on your methods. 128 UTF-16 characters — well, that's basically a 2048-bit key, and on average, sure, an RSS key of that length will take 6.4 quadrillion years to crack by brute force .

But we've cracked lots of them in reality by other methods: usually involving taking advantage of things like patterned data (even using only printing characters, rather than the entire code space, will halve the effective entropy of the key); flawed algorithm implementations; finding the government's or corporation's back-door they so often have left for themselves; or information leaks because there's always good old rubber-hose (torture) or brown-envelope (bribery) cryptanalysis.

(It also assumes no-one's invented quantum computing or found a constructive proof that P=NP.)

One advantage of checking against an external database is that, in theory, cracking the client won't do you any good because the database still won't match and cracking the db should be much harder without physical access, etc. It's just that that then begs the question of why it's worth bothering to secure the client in the first place.

...of course, verifying against an external db has its own issues of synchronization and light-lag, such as when Cap'n Harbatkin squawks an id that isn't in your local database and claims, on asking, that he updated his registry back on Flern and it's not his problem that you haven't got the updates yet. Is he a smuggler with a random number generator, or is he a legitimate trader whose lobby group will be screaming for your head on a spike if you hold him while you query Flern for verification and wait for an update to come back...?

John Reiher True, there is that, but still, you'd have to be a government or a corp to afford the computing power to crack one... OK, or have a botnet that doesn't go down because someone starts downloading pr0n.

But now we're falling into one-upmanship and that game never ends well. So I'll concede that there ain't no such thing as perfect security. Just good enough that only professionals and governments will bother to try to crack it.

Well, one solution is that your external db contains registry IDs that haven't been issued yet, but has a VIN Box waiting to be issued.

Of course that means if you can get one of those VIN Boxes illegally, then you got a valid ID. Or better yet, bribe the registrar of some backwater world to issue you three or four for use as you see fit.

From private conversation on Google Plus (2015)

If modular design is taken to its limit, "ships" will have no permanent existence. Instead they will be assembled out of modules and pods specifically for each run, much like a railroad train.

In that case, a ship's identity is attached to a service, not a physical structure. Example: the Santa Fe "Chief" was identified by a timetable and reputation, not a particular set of locomotive and cars.

From INTERSTELLAR TRADE: A PRIMER by Rick Robinson (2001)

(ed note: Veyndayk, Velmeran, Dveyella, and Keth are Starwolves. They are in the business of capturing warships of the stodgy bureaucratic interstellar Union and selling said warships back to the Union.)

     Soon they saw that it was Veyndayk, the cargo supervisor.
     "Business done," he said, stepping up to join Velmeran and Dveyella at the rail where they had been watching traffic pass on the level below.
     "Did you sell Keth back to the Sector Commander?" Velmeran asked.
     Veyndayk laughed. "No, although that might be a good use for old Starwolves. Farstell Freight and Trade bought back a shipment of clothing, conveniently packed in their own shipping containers. And (Union) fleet ordnance has just now payed us a finder's fee for an intact cutter."
     "A cutter?" Velmeran asked. Cutters were the smallest of the military ships, hardly bigger than a transport, and generally used only for police work.
     "My little joke," Veyndayk explained. "We took two intact cutters as riders on salvaged battleships, and one we have had sitting in a forward bay for the last year. We took them apart down to the smallest bolt and rebuilt the ships by taking parts at random. Now I am going to collect finder's fees on those ships in three different ports. That should give the boys in fleet ordnance fits, when they cross-check serial numbers of those parts."
     That appealed to Dveyella, who liked frustrating Union officials best of all. "You know, they will not be able to use those ships until they take them apart and rebuild them as they originally were."
     "You laugh, but that is probably the truth," the cargo officer said.

From THE STARWOLVES by Thorarinn Gunnarsson (1988)

Specialized Ship Types

This section has been moved here

Bare Bones Examples


      Now, to regress to the region of solar system transportation. I'm going to confine my discussion almost completely to gaseous fission rockets. The reason is not that I have decided that they are the things to be used rather than Orion or electrical rockets. The reason is simply that I have done more thinking about them. I think that I can see how to combine gaseous fission engines with advanced vehicles more efficiently than is possible with the other engine types. I believe that not enough thought has been given to the engine/vehicle interaction. There was a statement in the letter of invitation to this symposium to the effect that everybody knows what to do with 1800 seconds specific impulse. I happen to disagree. I don't think anybody knows what to do with 1800 seconds specific impulse, and I think we wouldn't know what to do with a good space engine if it walked up and bit us. I will try to prove those opinions as I go. I've used the "we" advisedly because I don't think I know either. Rather than getting into a big mish-mash by attempting to cover all various forms of propulsion, I will just stick with some mythical gaseous fission rockets. Presumably in the next 3 days we will discuss which engines you really should do this with, if you should do it at all.

     Figure 6 is a presentation of operating cost in dollars per pound of payload versus total velocity capability for single-stage vehicles with chemical, nuclear solid-core, and two different kinds of gaseous fission rocket engines. These curves were calculated four years ago when I was at Douglas for a paper by myself, Bill Mathiesen and Bob Trapp that was given at the I.A.F. Congress in Stockholm. We thought this was pretty interesting, but other than shocking an occasional person here and there, not very much has happened as a result. The thing that has been interesting to me is the fact that here was a clear indication that one could get out to extremely high velocities for a very low transportation cost. Velocities so high that you could open up the whole solar system for exploration with reasonable costs. Yet almost everyone believes it is extremely hard to do a little bit of space flight down in the low velocity region where we talk about just barely going to the moon.

     Figure 6 also is an interesting indication of the fact that you shouldn't drive a rocket faster than it wants to be driven. This is something that apparently a lot of people are forgetting. Theoretically, a rocket can go up to any velocity, not counting Einstein. But the way to make an inefficient rocket go to very high velocity is to stage it, and stage it, and stage it. One carries fuel to carry the fuel to carry the fuel that's going to be used later. Once the weight starts to pyramid, it's a logarithmic function and it just plain gets ridiculous in a hurry. So one has to be very careful about taking something like a solid-core nuclear rocket and deciding to perform missions at 200,000 fps. You can stack tip all that equipment if yqu want to do it, but it's a horrendous thing.

     A question that has bothered me quite a bit is why more attention was not paid to these curves. Consequently, I'm going to break down some of the assumptions used in these curves, then build new curves back up with this year's assumptions. Perhaps I can make the story more believable.

     Now, there are two big ringers in the curves of Figure 6. One is the obvious one. At that stage of the game, nobody had the foggiest idea of how to build a gaseous fission engine at all, let alone one with 5,000 to 20,000 seconds specific impulse. That, right off the bat, caused everybody to throw up their hands and forget it. The second ringer is that we used transport airplane operating cost assumptions. To put it mildly, we used recovery and reuse assumptions which were not the standard thing in rocket work. I'm going to examine both of these assumptions in today's light.

     Since so little was known of gaseous fission engines four years ago, the previous study assumed a thrust/weight ratio of 30 independent of specific impulse. This was recognized to be a very sporty assumption since, even if the containment problem coUld be solved, the achievement of specific impulses beyond about 3,000 seconds requires the use of a radiator to reject excess heat which cannot be handled by the thermal capacity of the propellant utilized (excess heat above what can be handled by open-cycle cooling). Although it was originally felt that such radiators would represent an intolerable decrease of thrust/weight ratio, it has since been pointed out that this is not true if high temperature radiators are used. Figure 7 shows a current estimate of the variation of the thrust/weight ratio with Isp achievable for a gaseous nuclear rocket system with radiator using as a basis an assumed thrust/weight ratio of 20, at an Isp of 2,500 seconds. The values fall off substantially at high specific impulses compared to the assumptions of 4 years ago, but are still greater than one to beyond 10,000 seconds specific impulse.

     It should be pointed out that the use of water, ammonia, methane, or other non-hydrogen working fluids should be seriously considered in gaseous fission engines from the start. Not only are better ship designs permissible due to small tankage sizes and ease of propellant storability, but the use of a higher density propellant might well ease the fuel containment problem if a vortex system is used. If so, it could result in smaller, lighter engines. It might also result in earlier development programs if the ability to prove adequate containment occurred at an earlier time.

(ed note: In thermal rockets, hydrogen produces the best specific impulse. But on the minus side: hydrogen tanks have to be huge because of low density, and it is a pain in the ass to store because is is cryogentic and needs lots of refrigeration)

     An interesting result in Figure 7 is that the value of thrust/weight ratio as the specific impulse approaches 10,000 seconds is independent of propellant used. This is because the reduced propellant flow at such a high specific impulse results in such small thermal capacity in the incoming fuel that the engine must be almost completely cooled by the radiator system. The propellant to fuel burned ratio required for a given specific impulse is independent of propellant used. Hence, the engine uses the same amount of energy to generate a given specific impulse, the same fraction of energy must be rejected by the radiator, and the radiator area is unaffected by the type of propellant used.

     It seems clear that an engine design cooled by radiator alone should be investigated. Such an engine might be easier to develop since a major interaction between propellant and cooling system would be severed. Furthermore, such an engine could more easily use a variety of propellants. This could be very helpful in early planetary exploring.

     A limitation on specific impulse of 10,000 seconds has been shown tentatively in Figure 7, assuming that the engine would be of the type which transfers heat from the fission plasma to the propellant by radiation. This is due to an unfortunate tendency of the propellants examined to date. Although adequately opaque to absorb the radiant energy at medium-high temperatures, they apparently become transparent at very high temperatures. At the moment seeding the flow, which is very effective at low temperatures, does not look promising at high temperatures.

(ed note: "seeding the flow" means putting tungsten dust in the propellant. Otherwise the heat radiated by the nuclear reaction fails to heat up the transparent propellant, and instead the rocket engine is destroyed)

     One other point of interest in connection with the thrust/weight ratios of gaseous fission engines is the power conversion weight thus achieved. Electrical propulsion enthusiasts feel extremely optimistic when power conversion weights of the order of 10 pounds per kilowatt are mentioned. A gaseous fission engine of 2,500 seconds Isp and T/W of 20 achieves about one-thousandth of a pound per kilowatt. In other words, gaseous fission engines are almost certain to be 10,000 times better than electrical rockets in power conversion weights. The fabulous effect of this number on spaceship design must be understood if anyone expects to make rational development decisions on future propulsion systems.

     I can defend the 1963 curves of Figure 7 today. Not very well, but at least I can begin to defend them. Four years ago, the 1960 assumption was nothing that could be defended at all. However, I have always liked the calculations we made then. It influenced me in feeling strongly that, at least theoretically, there was a great deal more which could be done with nuclear rockets than anyone realized, far more than just trivial improvements in our current systems.

     As part of the process of understanding spaceship operating costs, it is instructive to consider first only the fuel and propellant cost. This is true because this cost represents the minimum achievable. It is important to understand the mechanics of achieving a low fuel and propellant cost, particularly when truly reusable ships are used. In transport aircraft practice, the amount of reuse is so high that initial airframe costs are only a small fraction of the operating cost, and fuel costs represent about one-half the total. Thus, we shall examine fuel costs for their basic limitations on performance, and then see how closely these limits can be approached with reusable ships.

     Fuel and propellant costs as a function of total velocity increment for chemical, solid core nuclear, and gaseous nuclear rockets are shown in Figure 8. Compared to the assumptions of Figure 6, the specific impulse of the high energy chemical has been increased to represent a modern, high-pressure system: the solid core nuclear has been decreased in view of current development difficulties: and a number of different propellants and degrees of containment are shown for gaseous fission engines. All curves are for single-stage ships with structural assumptions more conservative than those of Figure 6 and each specifically sized for the velocity shown.

     It is evident that, on this basis alone, a gaseous fission engine without radiators and with separation ratio of 10-3 is not significantly better than a solid core engine. Gaseous engines with better containment would be much better. It is also evident that gaseous engines with space radiators, but with specific impulse limited to 10,000 seconds, can drive ships up to about one-half million feet per second and still maintain reasonable fuel cost. The attainment of a fuel separation ratio of 10-4 is almost as effective as perfect fuel containment.

     The optimum fuel cost curves for gaseous fission engines with radiators were obtained by determining the optimum specific impulse for each velocity and separation ratio. This is necessary since too low a specific impulse will result in excessive propellant cost while too high a specific impulse will result in excessive fuel cost. The optimum specific impulse is much higher than 10,000 seconds for all velocities beyond a few hundred thousand feet per second. Hence, these curves represent a future capability presently unattainable due to the propellant transparency problem at high temperatures previously mentioned. If it were not for this, gaseous fission ships could be driven to almost one million feet per second before fuel costs became a limitation.

     Under certain circumstances, a great deal can be learned about spaceship design without a detailed knowledge of the missions to be performed. In recent years, there has been a tendency to become so detail-mission-oriented that ship design is not even attempted until the exact mission is clearly understood. This may be a valid procedure when only one mission is in sight, although even then the inevitable lack of versatility usually leads to needless redesign much earlier than anticipated. In transport airplane design, the basic design procedure usually centers around the calculation of airplane operating characteristics as a function of range. The maximum range required comes from a knowledge of the total mission complex, but the airplane design is refined primarily by using general curves as a function of range, rather than by a detailed series of specific mission analyses.

     When considering total solar system transportation as we are, it is. clear that we face a variety of missions. It is also very unclear as to which of these will be paramount. One way of approaching the problem is to present the characteristics of the vehicle as a function of total velocity increment which the ship can achieve. This is exactly analogous to the use of range in aircraft design practice. In this way, an understanding of the ship's basic ability to deliver payload to a certain speed economically can be rather easily understood. The complex mission analyses, then, can be made to reflect the maximum design velocity increment required.

     This approach was actually used in the 1960 study, and Figure 6 represents one of the results. Figure 6, however, contains assumptions as to degree of reuse achieved by the vehicle which, although consistent with transport aircraft practice, may not apply to space transportation. At least, if they do, their application must be better documented.

     The 1960 study assumed a large number of reuses per vehicle, somewhat analogous to the number of times a transport airplane is reused. A transport aircraft is actually utilized about 50 percent of the time, and average flight durations are less than 4 hours. It is clear, therefore, that such vehicles are used over one thousand times per year. However, space travel durations are much longer, and it is obvious that the interaction between travel duration and number of reuses must be considered.

     For the lunar mission, it is clear that large numbers of reuses are feasible. Typically, 100 flights per year (50 each way) can be envisioned on the basis of 2-day travel times, one day turn around time at each terminal, with Sundays and 2 weeks off for vacation. Over a 10-year ship lifetime, 1,000 uses will be achieved.

     One can get a feeling for the number of interplanetary uses by assuming a certain ship total life. Typically, transport aircraft are designed for 40,000 hours (4.6 years) total life. On the basis of slightly less than 50 percent utilization, such a vehicle would last for 10 years. They always last much longer, but the amortization time of the airframe is usually about 40,000 hours, since new equipment always becomes available in even shorter time.

     Selecting a suitable lifetime for a spaceship presents a considerable technical dilemma. One viewpoint would simply take 10 years as above. An even shorter lifetime might technically be justified due to the severe aerodynamic environments associated with atmospheric entries, and the generally unknown operational environment of space. This type of assumption has become standard in this country recently. If you don't understand the problem, assume it's horrible.

     It may well be, however, that spaceships will last much longer than transport aircraft. The transport has its main propulsion system operating continually during flight, and is also continually facing the temperatures and gust loads within our atmosphere. The question is whether spaceship operating life should be determined by the total time of operation, or only by the times during which the main engines operate and/or it is within an atmosphere. In other words, is a spaceship coasting between planets actually operating in the aircraft transport sense, or is it merely parked in space, breathing quietly, waiting for its next mission.

     One can make an excellent case for the latter point of view in terms of the general environment that the ship faces, either from space or its own propulsion systems, while coasting. The ship would have to be on interplanetary runs for several centuries in order to build up 40,000 hours of engine and atmospheric operation. It is, however, bound to be replaced by better equipment within a few decades. As a base for calculations, this report assumes 25 years ship useful lifetime.

     The variation of various weights as a function of velocity is shown in Figure 9 for both specific impulse limited to 10,000 seconds and for the optimum specific impulse. These curves are for ships designed for 20 percent payload, then operated at lower velocities by off-loading propellant and at higher velocities by off-loading payload. Thus, these curves represent a penalty for using a single ship for multiple missions, just as in other forms of transportation.

     By using suitable planetary travel time data, which is not yet easy to come by, the weight data of Figure 9, the same fuel plus propellant cost assumptions as in Figure 8, and assuming a vehicle cost of $100 per pound, the curves of Figure 10 were obtained. One hundred dollars per pound is the currently estimated cost of a supersonic transport. These curves show fuel cost plus amortized airframe cost as a function of design velocity increment for the missions selected.

     The lowest curves on Figure 10 are fuel cost only. Comparing them with the other curves show that for operations as far out as the planet Saturn, the structural costs are comparable to fuel costs. Further improvements in convenience of operation can be achieved with engines not limited to 10,000 seconds specific impulse. In that case, velocity increments beyond a half-million feet per second are economically reasonable.

     The average travel time between planets corresponding to the velocities of Figure 10 are shown in Figure 11. These two Figures taken together give a better feel for solar system transportation than Figure 6 alone. With specific impulse limited to 10,000 seconds, the solar system as far as Jupiter is available with travel times not exceeding 4 months. Inner solar system travel times need not exceed 2 months. The advantage of optimum specific impulse becomes more evident at Saturn and beyond.

     The curves of Figures 10 and 11 apply for a given ship design velocity only if the ship can be refueled at each terminal. If it must carry its own fuel for the return journey, then it must operate at half the total velocity shown. Except for Pluto, refueling bases at the major planets are much more needed than at the minor ones, as can be seen by Figure 10. Refueling bases could be expected to be located on the surfaces of all the minor planets, although it may require some design effort in the case of Venus and Pluto.

     The major planets are a different situation. Their surfaces are extremely forbidding as far as we know, to the extent that we are not even sure they have solid surfaces. It makes sense in that case to establish bases on one of the satellites of each of the four major planets. The curves are drawn with that assumption. If we do decide to penetrate to the surface of these planets, then the velocity requirements for doing this when operating from one of the satellites is a reasonable number. Thus, bases on the larger planets' satellites not only greatly facilitate the convenience of transportation, but also present a reasonable base for surface exploration, if required.

     These particular curves are also calculated for the average flight times involved in year around operations between all planets. There are no launch window restrictions. If you're ever going to have a transportation system, you're going to have to be able to go when you want to. You cannot spend most of the time waiting. A rough averaging process between the best and worst times of the year was used in. an attempt to make this a realistic transportation assumption.

     A few words about perspective on these curves are in order. There is nothing magic here except a ridiculous willingness to plot curves wherever the data is leading, rather than stopping somewhere. Both better and worse situations may well occur. Even the case of specific impulse limited to 10,000 seconds requires gaseous fission engines with radiators, and most people today would rather agree to engines without radiators. In that case, the velocity increment achieved will be only about 25 percent of the curves shown. Furthermore, the economic penalty of, if necessary, ejecting a critical mass of fuel in the process of shutting down the engine has not been included. This will be on the order of $100,000 per shutdown ($778,160 in 2017 dollars).

     On the other hand, perfect containment might be achieved. We might design ships for each velocity increment, rather than use the single design assumed here. Furthermore, one can get a greater utilization of vehicles by the expedient of refueling the vehicles which go on deep space missions. This is preferable to multi-stage vehicles, since a fleet of ships used for refueling can also be used for other missions. No attempt will be made here to present detailed curves showing the effects of refueling. Cursory checks show that over 200,000 fps can be added for reasonable cost with only two refuelings.

     The greatest conservatism of all in Figures 10 and 11 is, of course, in the magnitude of the ordinate scale. Costs beyond $12 per pound have not been plotted so that the entire set of curves is about 100 to 1,000 times lower than virtually all space cost analyses to date. This must be clearly remembered as we discuss the performance of these ships.

     I can't resist making one more solar system point here. So far, only travel between Earth and the other planets has been discussed. There is also the question of travel between planets other than Earth. The use of bases in other parts of the solar system to aid in the exploration of the even more remote portions should be considered. In fact, such considerations might well dictate the strategic location of bases.

     At first thought, it would seem to be a good idea, for instance, to use a base on one of the farther planets, say Saturn, to permit further exploration of the more remote planets like Pluto. Although this is an intriguing thought, such deep bases will have only limited utility. The reason is the extremely long synodic periods which exist among the outer planets since they move so slowly around the Sun. In the worst case of all, the synodic period between Neptune and Pluto is slightly over 500 years. In addition to the long synodic period, the difference between travel at the optimum time of the year and the worst time of the year becomes more extreme the farther the planet is located from the Sun.

     One way of illustrating this is shown in Figure 12, where the effects of basing on selected planets is shown for a constant ship velocity. It is true that a deep space base will be closer than Earth to the other deep space objects when in favorable position, but equally true that it will be much farther away during the worst conditions. Surprisingly enough, the base wants to be reasonably close to the Sun, once again emphasizing that the Sun is the center of the solar system. Although Mercury might be the best planetary base of all, the Earth is still sufficiently close to the Sun that it represents a pretty good compromise. Thus, the major space logistics support operations could, from a celestial mechanics viewpoint, be located efficiently on the Earth or its Moon. This is very convenient since the known industrial and research bases of the solar system also happen to be located in that vicinity.

     The slow movement of the outer planets leads to some interesting paradoxes. One would naturally assume that a base on Triton would be an excellent place from which to explore Pluto, since Neptune is at 30.09 A.U. from the Sun, while Pluto is 39.5 A.U. However, it turns out that Neptune at the moment is already leading Pluto around the Sun, and pulling away. In fact, in approximately 9 years, Neptune will be farther away from Pluto than Earth ever is. Furthermore, due to the long synodic period of Neptune and Pluto, that statement will be true for somewhat over the next 300 years. It would be nice to be sure that every other statement in this discussion will be true for that duration.

     I threw that in as a bit of tidbit. I' m gradually working back down to our more normal systems, and there's a point that I want to harp on further — that is, this whole question of structural reuse. As I indicated before, Figures 10 and 11 show costs that are less than $10 per pound throughout the entire solar system. Yet large and elaborate studies are made these days proving that it's going to take many hundred dollars per pound to go to the moon, no matter what we do with recovery, reuse, or anything else. It1 s quite clear that either I'm insane, or a lot of other people are, or we have to have an explanation. It was easier to give an explanation than to prove everyone else insane.

     Figure 6 was drawn with what I like to refer to as "transportation" type assumptions for operating cost. Maintenance costs, for instance, were taken from normal air transport practice. The vehicles were assumed to be as reusable as transport aircraft. This is not the fashion when calculating rocket operating costs today. Almost all of our rocket builders, including myself, have been doing nothing but "ammunition" work for a large number of years. It is axiomatic that, if you want to talk about recovery and reuse, you should not talk to an ammunition builder.

     To illustrate what can happen with different classes of maintenance and reuse assumption, I will shift gears drasticaily and discuss merely placing objects on Earth orbit with chemical propulsion. In Figure 13, I assumed, arbitrarily, about $300.00 per pound of payload as typical of current day orbital transportation systems with no reuse at all. It so happens, however, that if you consider the actual price of high energy fuel needed for orbital velocities with advanced rockets, it is only on the order of $1.00 per pound of payload. Figure 13 is simply a plot of operating cost as a function of recovery reliability and refurbishment cost with these assumptions.

     It is quite fashionable, whenever over-all system analyses for recoverable space vehicles are performed, to assume that recovery reliabilities will be around 75 percent. After all, that is the recovery experience to date. Also, refurbishment costs around 25 percent are quite likely to be used. The shaded region brackets these assumptions, and is typical of a good, solid, rational ammunition type analysis. It is evident that after spending that much money on refurbishment between flights with that low a recovery reliability, an improvement of at most two in over-all cost performance is the best to be expected.

     Also shown on Figure 13 is what had already been achieved many decades ago in air transportation. This is what happens when you think like a transportation man. The recovery reliability is so close to 1.00 that you can't possibly see it on this scale. The same is true of the maintenance cost, which is on the order of 0.04 percent. If anyone here thinks that a DC-8 is less complicated than a Thor, just take a good look at the inside workings of a DC-8 some day. Wonder, then, at the fact that a few people turn it around, give it some fuel, pat it on the head, and it takes off again. This is what we should be trying for in future spaceships. There is an improvement of a factor of 100 over current operations to be made. I do not want you to get the impression that I am all for airplane designers. I think they, too, are irrational conservatives. But if useful design techniques have been developed, I think they should be used in space.

     As a matter of fact, you can get rougher with this. You can make a calculation on what would have happened in our air transport system last year if the philosophy of our ammunition people had been used in running it. If you do that, you'd find that we would have killed 4 million people last year. You will also find that the attrition of equipment and refurbishment costs are so high that it would cost you $10,000 for a ticket to anywhere. The only way out of the ammunition dilemma, surprisingly enough, is the kind of advance propulsion we're planning to talk about the next few days. You have to get enough margin into the propulsion so that you can have extra weight available both for use and reuse.

     An interesting interaction exists between the containment capabilities of gaseous fission systems and the cost of boost to orbit. Attempts are frequently made to show that high fuel consumption gaseous fission systems (either Orion or co-axial systems) would be acceptable after all, since the extra economic penalty which they incur compared to the cost of chemical boost to orbit is relatively small. This conclusion would obviously be strongly influenced by the wide spread of orbital costs mentioned.

     Figure 14 shows typical interactions between containment of fuel and economics of boost to orbit. The point is obvious. Chemical take-off is not too bad. If chemical boost to orbit is tolerated at all, however, it must be of the economical "transportation" variety, or it will completely cripple the ability of gaseous fission engines to explore economically the solar system. This to me is the real challenge of advanced propulsion. This is also why I think there's a tremendous interaction between the engine and vehicle. It is not just a matter of sitting down with a specific impulse, and making one simple performance calculation. The big gain is made by the interaction of the engine and the development of transportation techniques.

     Figure 15 is a sketch of one result of using a gaseous fission engine to power a reusable spaceship. We all, by now, expect manned rockets to be hundreds of feet long. If drawn to the scale of Figure 15, Saturn V would be two pages long. It would have a little bit of payload on the front. If, however, we were to combine the kind of nuclear rocket engine we would like to have (running on water or ammonia rather than hydrogen) with a reusable structure for the entire ship, a possible result would be the ship shown. This is a typical case of about a million pounds gross weight with cargo weight on the order of 200,000 pounds.

     It turns out, not surprisingly, that for reasonable economy, large payload fractions, perhaps even more than 20 percent, are required. Note that 20 percent cargo at a density of 10 pounds per cubic foot (standard transport airplane practice) when combined with the required propellant results in a rocket vehicle with 60 percent of its length devoted to cargo and crew. The engine and propellant take up only a small portion at the rear, just like "Buck Rogers" has said it should all along.

     An interesting example of the change in design philosophy with such ships is in the matter of shielding. Indications are that about 20,000 pounds of shielding weight would be required. This is a severe penalty for most rockets. Since cargo itself is effective shielding material, however, by the simple expedient of never flying this ship with less than 10 percent cargo aboard, properly packaged, the shielding penalty is reduced effectively to zero.

     If such a ship were used without radiators on the engine but with hydrogen propellant, it would be able to generate about 80,000 feet per second. This is more than needed for a lunar round trip. We can hence examine the effect of this ship on a lunar run, performing like a normal transport airplane. Logical assumptions, as previously discussed, would lead to 50 flights a year, and the ship can carry 100 tons per flight. By maneuvering a little bit with that number, I concluded that one ship like this is equivalent to 300 Saturn V launches per year.

     This is the kind of thing that we're driving at. Incidentally, in our scientific operations in the Antarctic, we deliver to the Antarctic about 50,000 tons a year. Ten such ships, shuttling back and forth to the moon, could mount the same magnitude of operation on the moon as we mount in Antarctica. This is, to say the least, an interesting capability.

  • Fission Products Always in Vapor Form
    • No Burn-Up Problem as in Rover or Power Supplies
  • No Fission Products aboard upon Return
    • Never Overfly with Fission Product Load
    • Servicing Problem Not Great as in ANP Application
  • Fission Product Load small in case of Accident
    • Megapound Thrust is Only Kilo (Not Mega) Ton Equivalent
    • Only Millirem Exposure if Accident above 5000 Ft.
  • High Velocity Exhaust Jet
    • Velocity of at Least 100,000 ft/sec exceeds Earth Escape Speed
      • Most Products actually Ejected from Solar System
  • Characteristics also affect Development Procedures
    • Time between Static Tests Reduced
      • Engine does not contain Products after Test
      • Fuel Element Fabrication Delays Avoided
    • May Not be a Logical Extension of Solid Cores
      • Progress is Rarely Logical
      • ICBM was not Extrapolation of Winged Aircraft

Figure 16

     Some points should be made about the safety of gaseous fission spaceships. Contrary to most opinion, a gaseous fission rocket is probably a lot safer to use than a solid core rocket. Several reasons for this are listed on Figure 16. The fission products are always in vapor form, so there is never a fuel element burnup problem if emergency atmospheric entry is necessary. On return trips, one need never overfly a city with a fission product load, since the products can be ejected into space and the landing made aerodynamically. After landing, of course, the ship is radioactive only to the extent that any material has been locally activated. This will be very small with proper material selection and is certainly far lower than when the fission product load is a permanent feature of the structure. Hence, the servicing problem would be nowhere near as great as with the case of aircraft nuclear propulsion.

     Furthermore, in case of an accident, the fission product load is always small. A million pounds of thrust is, after all, only 1/2 kiloton of thrust, and the actual fission products created are comparable to those from kiloton, not megaton, bombs. As a matter of fact, some interesting calculations indicate that an accident as low as 5,000 feet in the air yields almost no exposure on the ground due to the effectiveness of atmospheric dispersal of the small fission product load on board.

     In addition to the fact that the fission product load is small, one can do interesting things by recognizing the fact that the exhaust jet velocity is actually higher than solar system escape speed. If trajectories are properly programmed, once out of the Earth's atmosphere, most of the fission products ejected with the exhaust will be thrown completely out of the solar system. This is the one way of not contaminating space. Space, incidentally, is a doggone big place, and the contamination of small local radiation belts, atmospheres, or planets should not be confused with all of space. Even our Sun, which is continually making a real attempt at space contamination compared to any puny spaceship, has not succeeded to any great distance.

     Some of the characteristics which make a gaseous fission rocket different from solid core propulsion systems also result in development differences. There is a tendency for many people to believe that gaseous fission engines are a logical extension of Rover. At the risk of losing a number of friends, I would like to point out that they probably are not a logical extension of Rover. For instance, the time between reactor tests should be greatly reduced for gaseous engines. One need not fabricate fuel elements between tests, and does not have to live with fission products imbedded within the engine. The handling advantages in operation previously mentioned also extend to the engine development process and the development program should be a lot easier to run than that of a solid core engine. Gaseous fission engines may not be a logical extension of solid core engines at all.

     A good space engine can also have a profound effect on spaceship development cost. It is not just a question of a high degree of reuse, there is also the effect of making the ship abortable during any part of the flight.

     Transportation systems not only achieve very high reuse, they contain sufficient redundancy to permit flying with partial equipment failures, and also have the ability to abort successfully from any flight condition. It is this last capability which is very important to the development program of such ships.

     The savings in development cost of not losing ships continually is obvious, yet our ammunition thinkers are so used to the massive throwaway that they usually claim that recoverable equipment would be more expensive to develop since it is more complicated. This might be true of the recovery of marginal performing rockets, but would not be true of a properly designed reusable spaceship which would not be marginal with a gaseous fission engine.

     It simply is not possible to overemphasize this difference in development philosophy. Commercial transports are extensively tested, and much of the equipment refined by flight tests. They become reliable pieces of equipment for expenditures very small in space budget terms, because it is possible to test the equipment over and over for a reasonable expenditure. If a high performance propulsion system can permit us for the first time to pursue a space vehicle program with the very efficient development techniques of transport aircraft systems, we will be very remiss if too blind to even consider these techniques.

     The use of a reusable and abortable ship from the start of the development program can well have a profound interaction on engine development tests. This interaction will be enhanced greatly if the engine is a "tractable" engine. A "tractable" engine is one which has "benign failure modes." In other words, it does not explode catastrophically when it fails. There is an excellent chance that gaseous fission engines will tend to go out rather than explode when trouble occurs.

     If the engine is tractable, and the ship abortable, then flight failures consist mostly of unscheduled landings. The ship is then capable of testing the main engine without the extreme sensitivity to component malfunction which exists in ammunition development programs. This can be a very large leverage on total development costs. Clearly, there must be extensive ground testing of the engines. However, it may be much easier to arrange partial duration ground runs in enclosed areas than it is to arrange total duration ground runs. The total duration runs could then be performed in the ship.

     Many engine developments in the past have made extensive use of flying test beds when ground facilities were not adequate. The technique should not be ignored if the flight vehicles are able to reintroduce it. This is one of the examples of development techniques available with transportation devices which are not within the realm of experience of ammunition developers.

     The final point I wish to make is, I'm against logical progress.

     A common mistake in development thinking seems to be a tendency to relate the basic performance achieved by a device with its development difficulty. It seems so logical to assume orderly progress in development programs. Actually, many major programs are not a result of orderly progress. One of the most recent interesting examples is the development of the ICBM. These ballistic missiles penetrate to their targets at a Mach number of 25. Orderly progress would have dictated that we build first fleets of supersonic bombers, then fleets of hypersonic bombers. Only after that would we consider whether or not Mach number 25 penetrators were desirable.

     The fact is that Mach number 25 ballistic missiles are considerably easier to build than hypersonic bombers (and evidently Mach number 3 bombers). Their performance is attained in a different manner, with different engines (not breathing air) and in a different flight region (out of the atmosphere). They are not a Mach number 25 airplane.

     The gaseous fission spaceship has many analogous elements. It is easy to achieve 500,000 fps out in space, as long as the engine is capable of it. A ship which never carries fission products aboard need never fight the safety problems of solid core nuclear rockets, or even the analogous problems of nuclear airplanes. An abortable transport rocket is a different development job than the building of larger ammunition. We must look at the gaseous fission ship in terms of its difficulty of development, not in awe of its possible accomplishments.

Space Transportation by M. W. Hunter II (1965)

(ed note: This is an interesting design example from the always worth reading Bootstrapping Space blog by Chris Wolfe. It is mostly centered around estimating a mission delta V and sizing a propulsion system to fit, but his thought processes are interesting.)

All previous posts described all-chemical systems that could be built and operated profitably in the near term. This one focuses on electrical propulsion systems.
The defining features of most electric propulsion:
 - High efficiency (high Isp)
 - Low thrust
 - High power requirements
 - Long trip times
 - Long operating life

     I chose a specific paper (Frisbee, Mikellides) to examine since the authors thoughtfully included most of the interesting parameters for a reusable Nuclear Electric Propulsion (NEP) Mars cargo tug. I don't really dive into how to calculate this for yourself because the problem is quite difficult without modeling software.

     It all comes down to the details; the question of NEP vs. Solar Electric Propulsion (SEP) vs. Chemical depends on the specific mission goals and technologies used.

The summary:
23 tons dry mass for a nuclear-electric tug of ~6 MW thermal / 1.2 MW electric
64 tons cargo capacity from low Earth orbit to Phobos-Mars orbit
Just under 40 tons of water propellant for the outbound trip and another 7.2 tons acquired at Phobos for the return
2.2 years outbound, slightly less inbound
Two round trips between thruster refits, five round trips between reactor refits

(skip to the next section if you are already familiar with electric propulsion)

     The general idea is to use electrical power to dump energy into a propellant and then release it at very high speed.

     The simplest of these is Electrothermal.
     An electric current produces heat and the propellant is passed through it. First up is the resistojet, where a resistor somewhat like an incandescent lightbulb filament is heated and then the propellant is pumped past it. These are common devices in the RCS systems of satellites. Second is arcjet, which passes an electric arc directly through the propellant instead of through a resistor. These can reach higher Isp because they can heat the propellant beyond material limits for resistor elements.
     Efficiency is moderate (Isp of 500 to 1000, well above any realistic chemical system but easily an order of magnitude lower than the most efficient electrics). Preferred fuels are low atomic weight with no particulates (hydrogen, water, ammonia, hydrazine). Thermal power is most efficient when there are few degrees of freedom for the molecules so more of the energy can be applied to macroscopic motion, so hydrogen or hydrogen-rich propellants are ideal.

     Next up is Electrostatic.
     These are the 'traditional' ion thrusters, NSTAR, Hall effect, etc. The propellant is ionized by a strong electrostatic field (with some variations) and then the ions are accelerated by a negatively-charged grid or electrode. An electron gun is used to neutralize the electric charge of the ion beam and keep the spacecraft electrically neutral.
     Efficiency is high. with Isp ranging from 1000 to 5000 for most designs and a few reaching as high as 10,000. Preferred fuels are high atomic weight gases or elements with a very low first ionization energy; argon and xenon are frequently used. Sometimes iodine or metals like tin, magnesium or sodium are used, while caesium and rubidium can be used in a FEEP thruster.

     Lastly, Electromagnetic.
     These are the 'new generation' plasma thrusters like VASIMR, MPD, PIT, etc. Note the idea is not new, it's just that these technologies have been getting a lot of press lately. In fact, a pulsed plasma thruster (from this family) was the first electric thruster flown in space. In these devices the propellant is ionized by arc discharge, microwave heating or other means and accelerated by a magnetic field rather than an electrostatic grid.
 Efficiency is typically variable but can be very high, with an Isp of 1000 to 30,000 (most commonly about 1500 to 6000). Preferred fuels are the same as electrostatic thrusters for the most part, with lithium making an appearance in MPD thrusters.

     All three families have been in use for decades, while each family has relatively recent members pushing the limits. All share common fundamental physics with regard to their efficiencies. Sources of loss are in the power processing units, ionization energy of the propellant, dissociation energy of the propellant if it is a molecule rather than a pure element, ion impacts with the grid or body and in the charge density and geometry of the exhaust.


     Of key concern for a reusable vehicle is the propellant should be available in space. Xenon, argon and nitrogen are available in the atmosphere of Mars. Small amounts of nitrogen may be available in lunar cold traps or bound in soils of Ceres and many C-type asteroids. Water appears to be widely available. Alkali metals would be available on the Moon and in most asteroids.

     Another critical factor is that the thruster system should be reliable over the long term.  Electrostatic systems have already demonstrated very long operational lifespans in the range of 5 to 10 years of active thrust. Electromagnetic systems don't yet measure up in demonstrated lifespan, but that is mainly because most systems of this type are low Isp / high thrust RCS components designed for relatively short operating times rather than main drive units designed to last a decade. As a practical engineering concern, there are difficult challenges in improving lifespan of systems with exposed electrodes or grids. Designs without grids, whether electrostatic or electromagnetic, have the potential for decades-long operation.

     A practical system will be able to service a cargo route in a reasonable time. To illustrate this, let's dive into a paper on a pulsed inductive thruster proposal that includes payload transits to Mars, Saturn and solar escape. A reusable cargo tug with about 64 tons of payload and a 2.2-year Mars transit would require about 2 megawatts of electricity and about 275 tons fueled mass at departure. Propellant would be plain water, though ammonia works as well. If propellant supplies are available at Mars then we require only 1.2 megawatts and 165 tons fueled mass. Of that, the tug itself is just under 23 tons. Enough spare components would be included to make two round trips between thruster refits. A refit would mean swapping out the entire thruster pod with a fresh assembly, quick and easy.
     2.2 years is a long trip, but it is right at the synodic period for Earth and Mars. A cargo tug launched at one opportunity would arrive just in time to go / no go the crew launch at the next opportunity. Part of the problem is that a low-thrust vehicle has to produce about 16 km/s of dV for this trip, far more than a high-thrust chemical system requires thanks to the Oberth effect. Assuming an Isp of 6000 seconds the fuel mass ratio is 23.81% or 39.285 tons of water. An empty return trip would require 7.2 tons of propellant. Put another way, each ton of fuel delivers 1.6 tons of payload. Contrast that with my chemical tug's ratio of 0.8 tons of payload per ton of fuel and you can see the advantage; twice the mass delivered for the same quantity of fuel. Of course, the two approaches trade off costs between dry structure and fuel; electric propulsion is not automatically better but it certainly lets you do a lot more with the same starting mass.

     Power can come from two sources, solar or nuclear. The tug described above is nuclear; its reactor should be good for five or possibly six round-trips out of the box but could be designed for a much longer lifespan of 40 to 50 years with a bit more mass. This is partly because the same propulsion system is intended for exploration missions to the outer planets, where solar power is minimal.

     A Mars cargo tug could certainly use solar power instead, with a lifespan in the 20 to 30 year range and less complicated refitting / disposal. The ~19 tons of reactor, radiators and conversion hardware would be replaced by very large solar panel arrays of about 30 tons, 3.0 MW at Earth beginning-of-life yielding 1.2 MW at Mars end-of-life, with a whole-system specific power of 100 W/kg, 20-year useful life, 20% degradation and 50% of Earth-normal power available at Mars. If Photovoltaics (PV) refits are available every two or three trips then the allowance for degradation can be reduced to perhaps 10%, saving about 3.5 tons.

     Reactors also degrade over time as their fuel decays; some designs such as traveling wave, drum reflector and pebble bed can level the power output over time by only burning part of the nuclear fuel load at any one time. These designs can be life-extended by including more fuel during construction and / or by replacing fuel elements. Since the nuclear fuel is only a tiny fraction of a reactor's mass, this life extension adds very little mass to the overall system. This is the same reason why increasing a reactor's power takes less mass than increasing a solar panel array's power; for low power outputs the solar panels are nearly always lighter due to the reactor's heavy power conversion equipment and shielding, but as the power output grows the reactor eventually beats PV. As you can see from the above example, 1.2 megawatts at Mars after 20 years is firmly in nuclear territory given current state of the art solar performance. Even so, I would bet there is room in the design space for solar PV to be competitive at this power level, design life and solar distance.

Future work:

     I think the next step is to work up some EML2 to Phobos tether-capture cargo runs and see how they compare to the baseline LEO to LMO mission. Keeping used nuclear reactors out of LEO is a good idea, as is keeping large solar arrays out of the Van Allen belts. I'll try for an electric interplanetary tug with similar payload to my chemical tug. These would have the added bonus of providing abundant power while parked; there may be a case for a set of tugs such that one is always at Phobos providing megawatt-scale electrical power.


     I need to continue on the topic of electric propulsion. The previous post was a lot of words but not a lot of meat. I felt it was too weak to stand alone, particularly as a part of this series where I am trying to focus on a realistic near-term plan for cargo transport. If you are interested in more background information I'd start with the Wikipedia page on electric propulsion and follow up with a look at the Atomic Rockets engine page. Another good look in the context of interplanetary travel is this paper (Hellin), while a deep look at relevant equations can be had in this paper (Keaton).

     One interesting result is a general rule to find required thrust given average acceleration. Google failed me on finding an exact solution, but it looks like there is a simple approach that is within 1% of the target value.

     I eventually settled on a design massing 33.4 tons, 1.6 MW solar-electric, Isp 6,000 and 40 N thrust using PIT thrusters with water propellant.

     Let's look at an electric tug with payload comparable to my reference tug, both a solar PV and a nuclear version. The main routes for this vehicle will be between LEO, GEO, EML1/2 and Mars orbit. Unlike the chemical tug we can't get much out of the Oberth effect, so the delta-V requirements are higher. Just like the chemical tug, the LEO to EML1 leg has the highest dV requirements (about 7km/s), so if we design for that case then the other trips will be faster, carry more cargo or burn less propellant.

     A key design factor here is trip time. If we throw enough power at the problem we can get to EML1 in the same amount of time as a chemical rocket, but that is a poor use of the mass. We need to decide how long we are willing to wait for the cargo and design enough thrust into the ship to make the trip in that span. I'm going to suggest four weeks to EML1 as a reasonable compromise, so let's see the consequences of that choice.

Estimating thrust requirements for average acceleration

     To apply 7km/s of delta-V in 28 days we need to make an average acceleration of 2.89 mm/s. To allow some wiggle room let's assume we can only thrust 90% of the time, meaning now we need 3.22 mm/s. Since this is our average acceleration, we need to find either the initial or final acceleration to find the thrust of the propulsion system. To do that we will first need to know the vehicle's propellant mass fraction, so let's take a few test cases at Isp of 3000, 6000 and 10,000.

     Propellant mass fraction (Mf) is equal to 1 - e ^ (- dV / Ve), where dV in this case is 7000 m/s and Ve is Isp * g. See the rocket equation page for more details.
Isp 3000 -> Mf of 0.21175
Isp 6000 -> Mf of 0.11217
Isp 10000 -> Mf of 0.06890

     It would be nice if the average acceleration also matches up with the midpoint of fuel consumption, but somehow I doubt it. Let's find out.

     Given a dry mass of, say, 10 tons and an Isp of 3000, the fueled mass is ( 1 / ( 1 - Mf ) ) * dry mass, or 12.686 tons. When half of the fuel is burned the craft masses 11.343 tons. The target acceleration is 0.00322 m/s, so the required thrust is 36.52 newtons. Thrust is mass-flow (mdot) times exhaust velocity, so mdot is 1.2414 grams per second. That rate of propellant consumption would require 25 days to empty the tank, or 27.825 days after accounting for our 90% duty cycle.

     That surprises me. It's not exact but it is close enough for exploratory work. The case of 10,000 Isp works out to 27.947 days, so it looks like this general rule is valid across a fair range of Isp values. I also spot-checked some different mission dV values and found similar agreement, always within 1%. If anyone out there knows of an exact solution I would love to hear it.

     To calculate this yourself you need your mission dV, Isp, thrust duration and a test mass. If you set the test mass to 1kg (or 1t) then you can find a multiplier to use for different dry masses. The relationships are linear.

The required acceleration a will be dV in meters per second divided by thrust duration in seconds.
First, find fuel mass fraction, which is 1 - e ^ (- dV / Ve).
Convert to dry mass fraction Md, which is -Mf + 1
Convert to the 'gear ratio', which is simply 1 / Md
Multiply by dry mass M1 to get fueled mass M0 and note this value.
Find the fuel mass by taking M0 - M1 and note this value.
Find the 'halfway point', which is half the fuel mass plus M1; let's call this Mh.
Find the thrust F, which is Mh * a. This is the value you are looking for.
Find mdot, which is thrust divided by exhaust velocity, or F / (g * Isp ).
Find the real thrust duration, which is fuel mass divided by mdot. This should be within 1% of your stated thrust duration; if it is then the average acceleration value is accurate enough to use.

     If you have a known spacecraft (known dry mass and fuel mass, known thrust), you can use thrust divided by (dry mass plus half the fuel mass).

Electric tug design

     To align with the chemical tug, let's target a payload of 40 tons from LEO to EML1. Note that EML2 is a better target, but for purposes of comparison I'm using the LEO to EML1 trip as the most costly trip in the set. As mentioned above, we need to deliver in 28 days or provide an average acceleration of 3.22 mm/s. I don't have an exact solution, so I can't solve the problem in a single step. That's fine; spacecraft design is an iterative process.

     Let's assume an electric thruster at Isp = 6000 and mission dV of 7000 m/s. Also assume a one-way trip (meaning fuel is available at both endpoints). Power alpha is assumed to be 18 kg/kW, whether that be nuclear or long-life solar. Thrusters will be the NuPIT design shown in the last post, using the design values for the 5 N, 200 kW unit at 2.75 kg/kW (550 kg per thruster).
     As a first guess let's try eight thrusters, 1.6 MW. That's 33.2 tons, for a dry mass without tanks of 73.2 tons. We will need approximately 10 tons of liquid water propellant; using a tankage fraction of 2% would be reasonable in this case, so tack on 200kg for tanks for a total dry mass of 73.4 tons. Actual propellant load is 9,273 kg, so tankage is sufficient. The half-fueled mass is 78,037 kg and approximate average acceleration is 0.513 mm/s. We're not even close. Trip time would be 157.9 days, or 2.3 one-way trips per year.
     Maybe 20 thusters / 4 MW? Power alpha would improve to about 15, yielding 71 tons of power and propulsion. 40 tons of payload and perhaps 0.4 tons of tankage gives a dry mass of 111.4 tons, fuel mass of 14.1 t and average acceleration of 0.844 mm/s. This is clearly not going our way. Trip time would be 106.3 days, or 3.4 one-way trips per year.
     Let's aim much higher, 50 thrusters / 10 MW. Power alpha would continue to improve to about 14, yielding 167.5 t of power and propulsion. 40 tons of payload and 0.6 tons of tankage gives a dry mass of 208.1 tons, 26.29 tons of fuel and an average acceleration of 1.13 mm/s. Trip time would be 71.7 days or 5.1 one-way trips per year.

     Clearly, short trip times require increasingly absurd power levels. Matching the payload size of a chemical thruster with the 1.6 MW version means only making one round-trip per year. In fact, looking at that version of the ship, if we eliminate the payload entirely the highest acceleration the ship can make is 1.2 mm/s on its last gasp of propellant. Since fuel mass, dry mass, power and thrust are all linear relationships* that means no matter how we scale up the ship it can never get better than this. (The power system alpha does actually get better as we scale up, but moving a ship that masses several times your payload is inefficient and extremely expensive.)

     One thing we can do is increase the thrust of each propulsion unit, which usually means decreasing the Isp significantly. Let's look at a VASIMR thruster for comparison, since I have some data on performance at different Isp levels handy. A VASIMR thruster at 200 kW and 6000 Isp produces about 4.75 N of thrust, a fairly close match to the NuPIT. We need about six times that thrust (28.5 N), which occurs right at an Isp of 1000. That would bring the 8-thruster 1.6 MW vessel up to about 230 N of thrust. However, dropping the Isp so dramatically brings the fuel fraction just over 50%. That pushes our dry mass up to 75t, fuel mass to 78.1t and nets us only 2.0 mm/s average acceleration. It's a 40.5 day trip or 9 trips per year, but now we are burning more fuel than the chemical tug thanks to our drastically higher dry mass. Still no net benefit to be had.

Putting the tug to work

     Let's look at what an electric tug actually saves: propellant. In a fully functional ecosystem of cis-lunar services propellant is fairly plentiful. The speed, convenience and throughput of chemical vehicles far outweighs the efficiency of ion vehicles in this environment. Where an electric tug shines is in the buildup phase, where all of the propellant is coming from Earth. The tug would save money during a critical part of the project. What that means is we do not need to survive dozens of Van Allen belt transits over two decades, we just need to make a reasonable number of trips over two or three years. We also don't need to standardize on the same payload sizes as the chemical tug, nor do we need to make trips in 1 month. I would say that using the same power system alpha for the solar version as I do for the nuclear version is very pessimistic; these vessels would not need to function at Mars orbit, though they do need significantly thicker front-glass shielding on the panels than other craft.

     So, a lunar ISRU plan would still start with a single chemical tug / lander as described in part 1. Using performance for the detailed reference tug, a 15-ton package can be delivered from LEO direct to the lunar surface. This will be 12.4 tons of ISRU equipment and 2.6 tons of spares (2.1 year supply). Refilling tug 1 will take 6 months, after which it can deliver 33 tons to EML1.
     In the meantime, an electric tug (call it tug A) will deliver a 9-ton fuel depot (135 ton capacity) to EML1. Let's use our 40 N / 1.6 MW / 33.4t / 6000 Isp vehicle from above. It does the job in about 92 days, which means there is a window of three months after the launch of the first ISRU package to get the 33.4t tug, 5.36t propellant and 9t payload into LEO.
     At the first lunar launch, 33 tons are delivered to the EML1 depot. Tug A will collect this and head to LEO, taking 127 days and consuming 7.45t of water. During this trip a LEO depot is launched, identical to the one at EML1. The tug turns back around and heads for EML1, taking 4.22t of water for the return trip and leaving 21.33t of cargo in LEO. This could be a mix of surface samples and water as desired. Let's assume five tons are samples and the rest is fuel.
     The return trip takes 72 days, during which tug 1 will have delivered another 33 tons to the EML1 depot. Tug A repeats its performance, returning to LEO with a full load of 23.78t water and another 5t of samples. At this point we are at 598 days elapsed since start of ISRU operations, which should be enough time to settle on and construct additional hardware to expand the lunar surface capacity.
     This is significantly longer than the all-chemical scenario and has an IMLEO of 129.71 tons, within a few tons of all-chemical. Hardware costs are higher since more of the mass is spacecraft and much less of it is fuel. The main benefit is that schedule pressures are greatly reduced; final design, construction and testing of the second round of ISRU plant is allowed more than a year and a half of time rather than two months. More operational data is available and the tolerance for mistakes or inefficiencies is higher. Another benefit is that this profile includes depots in LEO and at EML1; even if things do not progress beyond the first ISRU package the infrastructure is still useful for this and future projects.
     This baseline hardware could continue to deliver 21 tons of cargo to LEO every ~200 days for about a decade, eventually reaching 426 tons over 20 trips at a cost of 141 tons of Earth mass or a leverage of about 3 to 1. Things improve if we continue to expand, since about 44% of that mass was fuel to get the first ISRU plant in position; additional ISRU hardware is delivered using lunar propellant.

     The next phase would be to send more ISRU hardware. Tug A can pick up 33 tons at EML1, deliver 19.18t of net payload to LEO over 127 days, pick up a 17-ton package and head to EML1 in 109 days. All of the required propellant is lunar and picked up at EML1. Round trip time is 236 days (a bit under 8 months). The harvesting process run by tug 1 has a shorter turnover time of 6 months, so on average an extra 19 tons is accumulated at the depot. That's not quite enough to provide for a cargo landing, so tug A may not always be bringing a full load of cargo to LEO (meaning shorter round trips in practice).
     An alternative might be to use 12-ton packages that will fit into a Falcon 9 for cheaper launch costs; the delivery time for that is 98 days. If less cargo is returned to LEO then that trip time can be shortened as well; for example, 6 tons of return cargo plus round-trip fuel would make each leg of the trip take 98 days, or 196 days round-trip. Each 6.5-month trip would deliver another 10 tons of ISRU with two years of spares. Two electric tugs could deliver 80 tons of ISRU capacity in 26 months, roughly a single Mars synodic period. That would place 95 tons of ISRU with expected output of 950 tons of propellant annually at a cost of 9.5 tons of spares. Net propellant delivered to EML1 would be 505 tons annually, or could be 168 tons to LEO annually with chemical tugs. The annual demand for spares (both ISRU and depots) can be met in a single hardware run with minimal fuel costs, leaving 3/4 of the electric tug schedule open for assignments like delivering new chemical tugs or GEO debris retrieval (a mission that avoids the majority of the radiation belts and prolongs the tug's useful life).
     The total phase 2 IMLEO would be 134.4 tons, all of it hardware. Lunar mass to LEO during this period would only be another 48 tons since capacity is focused on buildup. This phase would run for 26 months, or a total of 46 months since first launch.
     Ongoing maintenance would require approximately 12 tons per year. The initial depots would be insufficient, so we need another 27t of hardware for fuel storage. If we rate all flight and depot hardware with a 10-year lifespan and pro-rate the replacement mass then we need an additional 12.2 tons annually (24.2t total). Depending on how the output is allocated, this could be considered an ongoing leverage of 23.5 tons in EML1 per ton IMLEO or 7.8 tons in LEO per ton IMLEO. Another way to look at it would be as a fuel supply for three manned Mars missions covering four synodic periods (104 months), or a full ISRU program length of 150 months (12.5 years). Overall Earth mass to LEO is then 511.15t to harvest 5,501.7 tons of gross lunar propellant, yielding 2,924.6 tons net lunar propellant at EML1. That's a gear ratio of about 5.7 to 1. If you are only interested in delivering fuel to LEO then you can net 972.9 tons, still a favorable 1.9 to 1 mass ratio. 511 tons is a lot of mass to launch, but only three payloads require a heavy lift vehicle: the initial chemical tug stack (62t fuel and 22t hardware, split across two Falcon H) and the two electric tugs (33.4t each, also requiring a Falcon H unless they can be built in parts and flown on two Vulcan launches). The remaining 360 tons would be delivered by 30 Falcon 9 launches, or by some combination of any price-competitive launchers with at least 12 tons of payload.

     Launch costs would be roughly $2.1 billion. Hardware would run another $6.7 billion (at $15m per ton). Operations might cost $125-$250 million. Call it a total of $9 billion over about 15 years (12.5 years of operation plus 2.5 years of r&d, manufacturing and testing). Overall cost of fuel at EML1 would be $3,094.44 per kg, about $3.1 million per ton and expected to decline to $0.9 million per ton in the long run. Savings are about the same as the all-chemical approach, a bit over $4.5 billion vs. NASA baseline. Additional savings could be realized by using the chemical tugs as cargo haulers to and from Mars as described in part 2, resulting in excess capacity that could be sold or used for other purposes. One of those purposes might be ISS reboost and water supply for life support. Another might be developing a significant water supply on the Moon for growing food, in support of manned missions.

Ship mass and size

Full load mass and physical size depends upon assumptions about fuel mass ration, fuel bulk, etc.


Deadweight (inert mass)117%
Cargo (payload)233%
Fuel (propellant)350%

Note that total mass is three times the cargo capacity. As you can see, deadweight is the ship proper, structure, engines, anything that is not cargo or propellant.

With this assumption, the big freighters will have a fully loaded mass of 60,000 tons. The largest ships might be twice as big: 120,000 tons.

Our building cost is $500,000 per ton of cargo capacity, the mass assumption makes a building cost equal to $1 million per ton of deadweight. Annual service cost is $100,000 per ton of cargo capacity, the mass assumption makes the annual service cost equal to $200,000 per ton of deadweight. The starship hulls are not cheaper, but they can carry more cargo in proportion to their structural mass.

Type of shipCargo capacityPurchase price
Large20,000 tons$20 billion
Medium5000 tons$2.5 billion
Small1500 tons$750 million

At $500,000 per ton of cargo capacity, largest giant freighter cost $20 billion to build, but it it has a cargo capacity of 200 Boeing 747 jets, and accounts for over one percent of whole fleet's cargo capacity all by itself. Small freighter costs $750 million, and has seven time the capacity of 747.

With a 30 year service life, the combined shipbuilding yards of the 12 planet trade network will turn out about 25 ships per year.

Hulls will last longer than 30 years but the equipment wears out and has to be replaced. Ships go back to the yards for an overhaul every decade or so, but eventually the cost of stripping everything and replacing it will exceed the value of the ship. Depending upon overhaul costs the shipyards may make more money on rebuilding than on constructing brand new ships. Some ships will stay in service for many decades. Others will be retained as the futuristic equivalent of naval hulks or the old passenger equipment that railroads use as work trains. Every big commercial space station will have a bunch of these old ships in the outskirts.

If modular design is taken to its limit, "ships" will have no permanent existence. Instead they will be assembled out of modules and pods specifically for each run, much like a railroad train. In that case, a ship's identity is attached to a service, not a physical structure. Example: the Santa Fe "Chief" was identified by a timetable and reputation, not a particular set of locomotive and cars.

Starship Performance

The analysis up until now focused on money and economics. Businessmen only care about how long it takes to deliver the cargo and how much transport costs, they could care less about the scientific details of the ship engines. But authors care.

As with everything else, it all depends upon the assumptions. Your assumptions will be different, so feel free to fiddle with these and see what the results are.

Assumption: the time spent in FTL transit is zero (jump drive). For the FTL segment of the transit you can use whatever you want, as long as the details do not affect the analysis. The main thing is that the required time spent in FTL transit will add to the total trip time, and thus the number of cargoes a starship can transport per year.

Assumption: starships use reaction drives for normal space travel.

We know that the mass ratio is 2.0. So the Tsiolkovsky rocket equation tells us that the starship's total delta V will be the propulsion system's exhaust velocity times 0.69 (i.e., ln(2.0) ). Since starships accelerate to half their delta V, coast, then decelerate to a halt, their maximum speed is half their delta V, or exhaust velocity times 0.35 (i.e., ln(2.0) / 2). In practice you would accelerate up to a bit less than half their delta V in order to allow a fuel reserve in case of emergency.

It will be even less if the FTL drive happens to use the same type of fuel that the reaction drive does. Basically part of the fuel mass will have to be considered as cargo, not propellant, which will alter the ship's mass ratio.

Reaction driveExhaust velocity
general rule
Nuclear powered Ion~100 km/s
Fusiona few thousand km/s
Beam core matter-antimatterabout 100,000 km/s
( 1/3 c )

We have assumed that the ship spends 27 days in route (with an instantaneous FTL jump), so the outbound and inbound legs are 13.5 days each (1.17 million seconds).

Assumption: the acceleration on each leg is constant. In reality at the same thrust setting the acceleration will increase as the ship's mass goes down due to propellant being expended. The thrust will probably be constantly throttled to maintain a constant acceleration. Makes it easier on the crew and easier on our analysis. The implication is that obviously the average speed will be half the maximum speed (which is half the delta V)

Reaction driveExhaust velocity
general rule
leg distance
Advanced Ion
or Early Fusion
400 km/s130 km/s75 million km
(1/2 AU)
0.01 g
Advanced Fusion10,000 km/s5000 km/s20 AU
0.44 g
c0.3 c350 AU
(x5 Pluto's orbit)
8 g !!!

These figures will be lower if time is consumed in FTL flight, maybe be only Terra-Luna distance

Propulsion system's thrust power is thrust times exhaust velocity, then divide by 2. To get the thrust, we know that thrust is ship mass times acceleration. The ship mass goes down as fuel is burnt. As a general rule for ship mass, figure that it only has 2/3rds of a propellant load. That is, multiply the total ship mass by 0.83. So our 120,000 metric ton ship would have a general rule mass of 120,000 * 0.83 = 100,000 metric tons (100,000,000 kilograms).

Reaction driveExhaust velocity
general rule
ThrustThrust power
Advanced Ion
or Early Fusion
400,000 m/s
(400 km/s)
0.108 m/s
(0.011 g)
1.08×107 N2.16×1012 W
(2 terawatts)
Advanced Fusion10,000,000 m/s
(10,000 km/s)
4.3 m/s
(0.44 g)
4.3×108 N2.15×1015 W
(2,000 terawatts)
3.0×108 m/s
76.5 m/s
7.65×109 N1.15×1018 W
(1 million terawatts)

Where does fuel come from and who does it get into the ship's fuel tanks? Easiest if it is obtained locally at the destination's solar system. The economics of interplanetary transport is same as interstellar (since we did a lot of work making interstellar a cheap as interplanetary).

if fuel from a gas giant at a distance comparable to Terra-Jupiter and round trip is to only take weeks, interplanetary tankers will need speeds of around 1000 km/s. So tankers will be almost as expensive as starships. If tankers use low speed (to make them cheaper), the round trip balloons to a year or more. To service the starship fleet's thirst for fuel, tankers will need to be huge or there will have to be a lot of them. Either way, fuel shipped from gas giants ain't gonna be cheap.

If we forgo interplanetary tankers and instead have starships make extra leg to the local gas giant to refuel, it will cost you more than you will save.

The alternative is shipping fuel up from destination planet. Yes, we know about how surface to orbit is "halfway to anywhere" in terms of delta V cost. But in order to colonize space at all, surface-to-orbit shipping cost will have to be cheap anyway. The industrialization of space will start with using space based resources, but eventually surface-to-orbit will have to be cheap or there is no rocketpunk future. Laser launch, Lofstrom loop, space elevator, something like that.

Assumption: surface-to-orbit shuttle economics are equivalent to current day airliner economics. Round trip to LEO and back is about two hours (not counting loading/unloading). With loading/unloading and maintenance, figure 4 flights a day. Implication is that a round trip passenger ticket is $250 and round trip freight service is $1000/ton (which is +10% added to interstellar transport costs)

Fuel is not round trip, it only goes from surface to orbit, but shuttles have to go orbit to surface in order to get the next load. You will have to streamline the process. High capacity pumps to minimize load/unload times, crew-less shuttle. You might be able to squeeze fuel lift cost to $500/ton. So if starships carry 1.5 tons of fuel per ton of cargo, surface-to-orbit fuel lift costs adds $750/ton to interstellar shipping cost.

So total surface-to-orbit overhead is $1000/ton + $750/ton = $1750/ton or 17.5%. This is an ouch but not a show-stopper.

Back to starships. How big are they?

Present-day maritime tonnage rule: 1 registered ton = ~3 cubic meters.

Assumption: 1 ton = 3 m3 applies to fuel and hull (e.g., crew quarters, engineering spaces, etc) as well as cargo. Therefore, if the absolutely hugest cargo starship in service has a cargo capacity of 40,000 tons (twice that of a large cargo starship), then:

Wet Mass
Payload mass to total mass ratio is 3. So wet mass is 3 * 40,000 = 120,000 tons
Starship Volume
1 ton of total ship mass = 3 m3 of volume. 120,000 * 3 = 360,000 cubic meters.

Volume of a sphere is 4/3πr3, so the radius of a sphere is 3√(v/(4/3π)) or

radius = CubeRoot( v / 4.189)

diameter = (CubeRoot( v / 4.189)) * 2

Assumption: a "cigar-shape" for a spacecraft is a six times as long as it is wide, with the proportions indicated in the diagram above. The center body is a cylinder 1 unit in diameter (0.5 units radius) and two units high. The two end caps are cones of 0.5 units radius and 2 units high.

If the monstrous cargo starship is spherical, it would have a diameter of 88 meters. If it is cigar shaped then length = 300 meters and diameter of 50 meters.

A 1500 ton cargo capacity tramp freighter would have a wet mass of 4500 tons and a volume of 13,500 m3. Spherical shape would have a diameter of 30 meters, cigar shaped length = 100 meters long and diameter of 17 meters.

Modular ships dimension would be similar but a bit larger due to being assembled out of component parts.


When a spacecraft built for humans ventures into deep space, it requires an array of features to keep it and a crew inside safe. Both distance and duration demand that spacecraft must have systems that can reliably operate far from home, be capable of keeping astronauts alive in case of emergencies and still be light enough that a rocket can launch it.

Missions near the Moon will start when NASA’s Orion spacecraft leaves Earth atop the world’s most powerful rocket, NASA’s Space Launch System. After launch from the agency’s Kennedy Space Center in Florida, Orion will travel beyond the Moon to a distance more than 1,000 times farther than where the International Space Station flies in low-Earth orbit, and farther than any spacecraft built for humans has ever ventured. To accomplish this feat, Orion has built-in technologies that enable the crew and spacecraft to explore far into the solar system.

Systems to Live and Breathe

As humans travel farther from Earth for longer missions, the systems that keep them alive must be highly reliable while taking up minimal mass and volume. Orion will be equipped with advanced environmental control and life support systems designed for the demands of a deep space mission. A high-tech system already being tested aboard the space station will remove carbon dioxide (CO2) and humidity from inside Orion. Removal of CO2 and humidity is important to ensure air remains safe for the crew breathing. And water condensation on the vehicle hardware is controlled to prevent water intrusion into sensitive equipment or corrosion on the primary pressure structure.

The system also saves volume inside the spacecraft. Without such technology, Orion would have to carry many chemical canisters that would otherwise take up the space of 127 basketballs (or 32 cubic feet) inside the spacecraft—about 10 percent of crew livable area. Orion will also have a new compact toilet, smaller than the one on the space station. Long duration missions far from Earth drive engineers to design compact systems not only to maximize available space for crew comfort, but also to accommodate the volume needed to carry consumables like enough food and water for the entirety of a mission lasting days or weeks.

Highly reliable systems are critically important when distant crew will not have the benefit of frequent resupply shipments to bring spare parts from Earth, like those to the space station. Even small systems have to function reliably to support life in space, from a working toilet to an automated fire suppression system or exercise equipment that helps astronauts stay in shape to counteract the zero-gravity environment in space that can cause muscle and bone atrophy. Distance from home also demands that Orion have spacesuits capable of keeping astronaut alive for six days in the event of cabin depressurization to support a long trip home.

Proper Propulsion

The farther into space a vehicle ventures, the more capable its propulsion systems need to be to maintain its course on the journey with precision and ensure its crew can get home.

Orion has a highly capable service module that serves as the powerhouse for the spacecraft, providing propulsion capabilities that enable Orion to go around the Moon and back on its exploration missions. The service module has 33 engines of various sizes. The main engine will provide major in-space maneuvering capabilities throughout the mission, including inserting Orion into lunar orbit and also firing powerfully enough to get out of the Moon’s orbit to return to Earth. The other 32 engines are used to steer and control Orion on orbit.

In part due to its propulsion capabilities, including tanks that can hold nearly 2,000 gallons of propellant and a back up for the main engine in the event of a failure, Orion’s service module is equipped to handle the rigors of travel for missions that are both far and long, and has the ability to bring the crew home in a variety of emergency situations.

The Ability to Hold Off the Heat

Going to the Moon is no easy task, and it’s only half the journey. The farther a spacecraft travels in space, the more heat it will generate as it returns to Earth. Getting back safely requires technologies that can help a spacecraft endure speeds 30 times the speed of sound and heat twice as hot as molten lava or half as hot as the sun.

When Orion returns from the Moon, it will be traveling nearly 25,000 mph, a speed that could cover the distance from Los Angeles to New York City in six minutes. Its advanced heat shield, made with a material called AVCOAT, is designed to wear away as it heats up. Orion’s heat shield is the largest of its kind ever built and will help the spacecraft withstand temperatures around 5,000 degrees Fahrenheit during reentry though Earth’s atmosphere.

Before reentry, Orion also will endure a 700-degree temperature range from about minus 150 to 550 degrees Fahrenheit. Orion’s highly capable thermal protection system, paired with thermal controls, will protect Orion during periods of direct sunlight and pitch black darkness while its crews will comfortably enjoy a safe and stable interior temperature of about 77 degrees Fahrenheit.

Radiation Protection

As a spacecraft travels on missions beyond the protection of Earth’s magnetic field, it will be exposed to a harsher radiation environment than in low-Earth orbit with greater amounts of radiation from charged particles and solar storms that can cause disruptions to critical computers, avionics and other equipment. Humans exposed to large amounts of radiation can experience both acute and chronic health problems ranging from near-term radiation sickness to the potential of developing cancer in the long-term.

Orion was designed from the start with built in system-level features to ensure reliability of essential elements of the spacecraft during potential radiation events. For example, Orion is equipped with four identical computers that each are self-checking, plus an entirely different backup computer, to ensure Orion can still send commands in the event of a disruption. Engineers have tested parts and systems to a high standard to ensure that all critical systems remain operable even under extreme circumstances.

Orion also has a makeshift storm shelter below the main deck of the crew module. In the event of a solar radiation event, NASA has developed plans for crew on board to create a temporary shelter inside using materials on board. A variety of radiation sensors will also be on the spacecraft to help scientists better understand the radiation environment far away from Earth. One investigation called AstroRad, will fly on Exploration Mission-1 and test an experimental vest that has the potential to help shield vital organs and decrease exposure from solar particle events.

Constant Communication and Navigation

Spacecraft venturing far from home go beyond the Global Positioning System (GPS) in space and above communication satellites in Earth orbit. To talk with mission control in Houston, Orion’s Orion will use all three of NASA’s space communications networks. As it rises from the launch pad and into cislunar space, Orion will switch from the Near Earth Network to the Space Network, made possible by the Tracking and Data Relay Satellites, and finally to the Deep Space Network that provides communications for some of NASA’s most distant spacecraft.

Orion is also equipped with backup communication and navigation systems to help the spacecraft stay in contact with the ground and orient itself if it’s primary systems fail. The backup navigation system, a relatively new technology called optical navigation, uses a camera to take pictures of the Earth, Moon and stars and autonomously triangulate Orion’s position from the photos. Its backup emergency communications system doesn’t use the primary system or antennae for high-rate data transfer.

Spacecraft Parameters

For an given type of automobile, there are parameters that tell you what kind of performance you can expect. Things like miles per gallon, acceleration, weight, and so on.

Spacecraft have parameters too, it is just that they are odd measures that you have not encountered before. I am going to list the more important ones here, but they will be fully explained on other pages. Refer back to this list if you run across an unfamiliar term.

Habitat Module
The pressurized part of the spaceraft where people live. Included in Payload Section. Remember that Rockets Are Not Hotels. Unlike the Starship Enterprise a real spacecraft is a huge expanse of airless machinery with a tiny pressurized habitat module tucked away in a corner where people can walk around without spacesuits.
The part of the spacecraft that is its reason for existance. For a satellite booster, the payload is the satellite it is lifting into orbit. For a transport ship: habitat module, passengers, ship controls. For a warship: habitat module, crew, weapons, defenses, ship controls. For a robot freighter: robot controls and cargo. Some payload like cargo and crew are removable from the spacecraft. Some payload like weapons and habitat modules are fixed parts of the spacecraft. Included in Payload Section.
Engine or Thruster
The rocket engine that moves the spacecraft, and the empty propellant tanks. Included in Propulsion Bus.
Power Plant
Part that generates electricity. Included in Propulsion Bus.
Struture is the skeleton and skin of the spacecraft. Included in both Propulsion Bus and Payload Section.
Propellant and Fuel
Propellant or Reaction mass (remass) is what the thruster fires out the exhaust nozzle to create thrust. Fuel is the source of energy used to propel the propellant. Remember that Fuel Is Not Propellant. In chemical rockets, the chemicals are both propellant and fuel. In nuclear rockets the liquid hydrogen is the propellant and the uranium is the fuel. Included in Propulsion Bus.

Payload Mass (Mpl)
Mass of all the payload. For NASA vessels this is typically 26.7% of Dry Mass. Note that for many spacecraft there is no specific set maximum payload mass. You can strap as much payload to the ship's nose as you want. Understanding that the price will be reducing the spacecraft's delta-V due to the increase in payload degrading the spacecraft's mass ratio.
Payload Fraction (λ)
Payload mass as percentage of wet mass. Mpl / M
Structural Mass (Mst)
Mass of all the struture. For NASA vessels this is typically 21.7% of Dry Mass.
Propellant Mass (Mpt)
The mass of all the propellant in the spacecraft's propellant tanks. Does not include fuel that is retained after it is burnt, e.g., uranium fissioned inside a solid core reactor. For some calculations, you will use instead the mass of propellant that will be expended in a given maneuver.
Power Plant Mass (Mpp)
The mass of the electrical generation system. Includes any heat radiators. For NASA vessels this is typically 28% + 3.4% of Dry Mass
Thruster System Mass (Mts)
The mass of the rocket engines, including the empty propellant/fuel tanks. For NASA vessels this is typically 3.7% of Dry Mass
Propulsion System Mass (Mps)
Thruster System Mass + Power Plant Mass.
Inert Mass (Mi)
Mass of spacecraft with no propellant and no payload. Propulsion System Mass + Structural Mass.
Inert Mass Fraction (δ)
Inert mass as percentage of wet mass. Mi / M
Dry (Empty, Burnout) Mass (Me)
Mass of spacecraft with no propellant but with payload. Propulsion System Mass + Structural Mass + Payload Mass.
Wet (Total, Ignition) Mass (M)
Total mass of spacecraft. Propellant Mass + Propulsion System Mass + Structural Mass + Payload Mass.
Mass Ratio (R)
Ratio of wet mass to dry mass. Wet Mass / Dry Mass.
Parametric Mass Ratio (r)
λ + δ.
Propellant Fraction (Pf, PMF, or ζ)
Percentage of wet mass that is propellant. 1 - ( 1 / MassRatio )

Propellant Mass Flow (mDot or )

How quickly does the Thruster System drain the propellant tanks? Rated in kilograms per second.

mDot constrains the amount of thrust the propulsion system can produce. Changing the propellant mass flow is a way to make a spacecraft engine shift gears.

Exhaust-Velocity (Ve)

How fast does the propellant shoot out the exhaust nozzle of the Thruster System? Rated in meters per second. Exhaust velocity (and delta V) is of primary importance for space travel. For liftoff, landing, and dodging hostile weapons fire, thrust is more important.

Broadly exhaust velocity is a measure of the spacecraft's "fuel" efficiency (actually propellant efficiency). The higher the Ve, the better the "fuel economy".

Generally if a propulsion system has a high Ve it has a low thrust and vice versa. The only systems where both are high are torch drives. Some spacecraft engines can shift gears by trading exhaust velocity for thrust.

For a more in-depth look at exhaust velocity look here

Specific Impulse (Isp)
Another way of stating exhaust velocity. Exhaust Velocity / 9.81 where 9.81 = acceleration due to gravity on Terra in meters per second. Specific Impulse is rated in seconds. It is also a broad measure of the spacecraft's "fuel" efficiency.
Delta V or Δv

Spacecraft's total change in velocity capability. This determines which missions the spacecraft can perform. Arguably this is the most important of all the spacecraft parameters. Rated in meters per second.

This can be thought of as how much "fuel" is in the tanks of the spacecraft (though it is actually a bit more complicated than that).

Velocity Ratio
Δv / Ve
Thrust (F)

Thrust produced by Thruster System. Rated in Newtons. Thrust is constrained by Propellant Mass Flow. Thrust (and acceleration) is of primary importance in liftoff, landing, and dodging hostile weapons fire. For space travel exhaust velocity (and delta V) is more important.

Generally if a propulsion system has a high Ve it has a low thrust and vice versa. The only systems where both are high are torch drives. Some spacecraft engines can shift gears by trading exhaust velocity for thrust.

Acceleration (A)

Spacecraft's current acceleration. Current total mass / Thrust. Rated in meters per second per second. Divide by 9.81 to get g's of acceleration.

In space, a spacecraft with higher acceleration will generally not travel to a destination any faster than a low acceleration ship. But a high acceleration ship will have wider launch windows for a given trajectory.

Note that as propellant is expended, current total mass goes down and acceleration goes up. If you want a constant level of acceleration you have to constantly throttle back the thrust.

5 milligee (0.05 m/s2) : General rule practical minimum for ion drive, laser sail or other low thrust / long duration drive. Otherwise the poor spacecraft will take years to change orbits. Unfortunately pure solar sails are lucky to do 3 milligees.

0.6 gee (5.88 m/s2) : General rule average for high thrust / short duration drive. Useful for Hohmann transfer orbits, or crossing the Van Allen radiation belts before they fry the astronauts.

3.0 gee (29.43 m/s2) : General rule minimum to lift off from Terra's surface into LEO.

For a more in-depth look at minimum accelerations look here.

Thrust Power (Fp)
Power produced by Thruster System. ( Thrust × Exhaust Velocity ) / 2. Rated in watts.
Specific Power (Fsp)
Power density of spacecraft. Thrust Power / Dry Mass. Rated in watts per kilograms.
Specific Mass
Alpha of Thruster System. Thruster System Mass / Thrust Power. Rated in kilograms per watt.


Typically the percentage of spacecraft dry mass that is structure is 21.7% for NASA vessels.

What is the structure of the ship going to be composed of? The strongest yet least massive of elements. This means Titanium, Magnesium, Aluminum, and those fancy composite materials. And all the interior girders are going to have a series of circular holes in them to reduce mass (the technical term is "lightening holes").


A geodetic airframe is a type of construction for the airframes of aircraft developed by British aeronautical engineer Barnes Wallis in the 1930s (who sometimes spelled it "geodesic"). Earlier, it was used by Prof. Schütte for the Schütte Lanz Airship LS 1 in 1909. It makes use of a space frame formed from a spirally crossing basket-weave of load-bearing members. The principle is that two geodesic arcs can be drawn to intersect on a curving surface (the fuselage) in a manner that the torsional load on each cancels out that on the other.


The earliest-known use of a geodesic airframe design for any aircraft was for the pre-World War I Schütte-Lanz SL1 rigid airship's envelope structure of 1911, with the airship capable of up to a 38.3 km/h (23.8 mph) top airspeed.

The Latécoère 6 was a French four-engined biplane bomber of the early 1920s. It was of advanced all-metal construction and probably the first aircraft to use geodetic construction. Only one was built.

Barnes Wallis, inspired by his earlier experience with light alloy structures and the use of geodesically-arranged wiring to distribute the lifting loads of the gasbags in the design of the R100 airship, evolved the geodetic construction method (although it is commonly stated, there was no geodetic structure in R100). Wallis used the term "geodetic" to apply to the airframe and distinguish it from "geodesic" which is the proper term for a line on a curved surface, arising from geodesy.

The system was later used by Wallis's employer, Vickers-Armstrongs in a series of bomber aircraft, the Wellesley, Wellington, Warwick and Windsor. In these aircraft, the fuselage was built up from a number of duralumin alloy channel-beams that were formed into a large framework. Wooden battens were screwed onto the metal, to which the doped linen skin of the aircraft was fixed.

The metal lattice-work gave a light structure with tremendous strength; any one of the stringers could support some of the load from the opposite side of the aircraft. Blowing out the structure from one side would still leave the load-bearing structure as a whole intact. As a result, Wellingtons with huge areas of framework missing continued to return home when other types would not have survived; the dramatic effect enhanced by the doped fabric skin burning off, leaving the naked frames exposed (see photo). The benefits of the geodesic construction were partly offset by the difficulty of modifying the physical structure of the aircraft to allow for a change in length, profile, wingspan etc.

Geodetic wing and fin structures—taken from the Wellington—were used on the post-war Vickers VC.1 Viking, though with a new fuselage and metal-skinned.

From the Wikipedia entry for GEODETIC AIRFRAME

Spacecraft Spine

Many (but not all) spacecraft designs have the propulsion system at the "bottom", exerting thrust into a strong structural member called the ship's spine. The other components of the spacecraft are attached to the spine. The spine is also called a keel or a thrust frame. In all spacecraft the thrust frame is the network of girders on top of the engines that the thrust is applied to. But only in some spacecraft is the thrust frame elongated into a spine, in others the ship components are attached to a shell, generally cylindrical.

If you leave out the spine or thrust frame, engine ignition will send the propulsion system careening through the core of the ship, gutting it. Spacecraft engineers treat tiny cracks in the thrust frame with deep concern.


However, I'm debating if the structures you cite as "keels" make sense when cross-referenced with "thrust frame".

For instance, ISS' truss isn't really a thrust frame—the station is very rarely under thrust, and when it is, it's usually from spacecraft or its own thrusters on the end of the Russian segment, which would actually make the whole main line of modules (Zarya, Zvezda, Pressurized mating adapter-1, Unity, Destiny, Harmony) the main "keel". The job of the truss in such a case is just to stop itself from flexing and hold the solar "wings" in place.

Similarly, there's other space vehicles which lack such a "keel" entirely, such as the DTAL concept or the Altair ascent stage design. In both cases, an engine is basically mounted to a pressure vessel (a prop tank for DTAL and a crew cabin for Altair) and then the rest of the structure "hangs" off of that pressure hull.

I might also note that these kinds of {keel-less spacecraft} will, in a rocketpunk setting, likely be confined to special-purpose craft—landers or scooters, spaceplanes, dedicated fuel tankers, and such. Most "typical" ships will probably have a bit more of spiney spaceframe.

The distinction I might make is "primary structure" and "thrust structure". The thrust structure is just the structural system to distribute the force of the engine, such as the F9 Octaweb. On the other hand, the primary structure is anything that serves a major structural role in the ship, analyzed as a system. In DTAL, it'd be the engine, the thrust structure that mounts that engine to the tanks, the tanks, and then the landing gear and such. For ISS, it's the outer hulls of the core "line" of modules, plus the truss. This primary structure might also be called an airframe or spaceframe.

Engineer Rob Davidoff (2014)

OK, forget what I just said. On top of the engine will be the thrust frame or thrust structure. On top will be the primary structure or spaceframe. The thrust frame transmits the thrust into the spaceframe, and prevents the propulsion system careening through the core of the ship.

The spaceframe can be:

  • A long spine/keel with the propellant tanks and payload section bits attached in various places.
  • A large pressurized vessel, either propellant tank or habitat module. Other propellant tanks and payload section bits are attached to main tank or perched on top.
  • Something else.

The engineers are using a pressurized tank in lieu of a spine in a desperate attempt to reduce the spacecraft's mass. But this can be risky if you use the propellant tank. The original 1957 Convair Atlas rocket used "balloon tanks" for the propellant instead of conventional isogrid tanks. This means that the structural rigidity comes from the pressurization of the propellant. This also means if the pressure is lost in the tank the entire rocket collapses under its own weight. Blasted thing needed 35 kPa of nitrogen even when the rocket was not fueled.

As Rob Davidoff points out, keel-less ship designs using a pressurized tank for a spine is more for marginal ships that cannot afford any excess mass whatsoever. Such as ships that have to lift off and land in delta-V gobbling planetary gravity wells while using one-lung propulsion systems (*cough* chemical rockets *cough*).

This classification means that parts of the propulsion bus and payload section are intertwined with each other, but nobody said rocket science was going to be easy.

Ship Spine Example

This is the von Braun Round the Moon Ship

  1. At the bottom are the four rocket motors. During a burn, the rocket thrust they create pushes upward.

  2. The rocket thrust pushes upward on the thrust frame (dark blue), which is right above the motors.

  3. Built on top of the thrust frame is the spaceframe (light blue), the same way that the skeleton of a skyscraper is built atop the ground. The spaceframe is pushed upward by the thrust frame.

  4. All the other spacecraft components: personnel sphere, hydrazine tank, nitric acid tank, solar mirrors, radar antennae, and everything else is hung from the spaceframe. As the spaceframe is pushed upwards, it drags along all the spacecraft components.

Lighten That Spaceframe!

Getting back to the spine/spaceframe. Remember that every gram counts. Spacecraft designers want a spine that is the strongest yet lowest mass structural member possible. The genius R. Buckminster Fuller and his science of "Synergetics" had the answer in his "octet truss" (which he called an "isotrophic vector matrix", and which had been independently discovered about 50 years earlier by Alexander Graham Bell). You remember Fuller, right? The fellow who invented the geodesic dome?

Each of the struts composing the octet truss are the same length. Geometrically it is an array of tetrahedrons and octahedrons (in terms of Dungeons and Dragons polyhedral dice it uses d4's and d8's).

Sometimes instead of an octet truss designers will opt for a weaker but easier to construct space frame. The truss of the International Space Station apparently falls into this category.


(ed note: In the singleship, the ship is basically build around the long fusion drive tube. The torus propellant tanks are hung on the tube, with attachments for cargo nets. Atop the tube is a thrust frame, with the life support system on top and more attachments for cargo nets around the rim.)

The artifact was the shell of a solid fuel rocket motor. Part of the Mariner XX, from the lettering.

The Mariner XX, the ancient Pluto fly-by. Ages ago the ancient empty shell must have drifted back toward the distant sun, drifted into the thin Trojan-point dust and coasted to a stop. The hull was pitted with dust holes and was still rotating with the stabilizing impulse imparted three generations back.

As a collector's item the thing was nearly beyond price. Brennan took phototapes of it in situ before he moved in to attach himself to the flat nose and used his jet backpac to stop the rotation. He strapped it to the fusion tube of his ship, below the lifesystem cabin. The gyros could compensate for the imbalance.

In another sense the bulk presented a problem.

He stood next to it on the slender metal shell of the fusion tube. The antique motor was half as big as his mining singleship, but very light, little more than a metal skin for its original shaped-core charge. If Brennan had found pitchblende the singleship would have been hung with cargo nets under the fuel ring, carrying its own weight in radioactive ore. He would have returned to the Belt at half a gee. But with the Mariner relic as his cargo he could accelerate at the one gee which was standard for empty singleships.

There are few big cargo ships in the Belt. Most miners prefer to haul their own ore. The ships that haul large cargoes from asteroid to asteroid are not large; rather, they are furnished with a great many attachments. The crew string their payload out on spars and rigging, in nets or on lightweight grids. They spray foam plastic to protect fragile items. spread reflective foil underneath to ward off hot backlighting from the drive flame, and take off on low power.

The Blue Ox was a special case. She carried fluids and fine dusts; refined quicksilver and mined water, grain, seeds, impure tin scooped molten from lakes on dayside Mercury, mixed and dangerous chemicals from Jupiter's atmosphere. Such loads were not always available for hauling. So the Ox was a huge tank with a small threeman lifesystem and a fusion tube running through her long axis; but, since her tank must sometimes become a cargo hold for bulky objects, it had been designed with mooring gear and a big lid.

Nilsson's own small, ancient mining ship had become the Ox's lifeboat. The slender length of its fusion tube, flared at the end, stretched almost the length of the hold. There was an Adzhubei 4-4 computer, almost new; there were machines intended to serve as the computer's senses and speakers, radar and radio and sonics and monochromatic lights and hi-fi equipment. Each item was tethered separately, half a dozen ways, to hooks on the inner wall.

Nilsson nodded, satisfied, his graying blond Belter crest brushing the crown of his helmet. "Go ahead, Nate."

Nathan La Pan began spraying fluid into the tank. In thirty seconds the tank was filled with foam which was already hardening.

"Close 'er up."

Perhaps the foam crunched as the great lid swung down. The sound did not carry. Patroclus Port was in vacuum, open beneath the black sky.

The captive ship was small. Phssthpok found little more than a cramped life support system, a long drive tube, a ring-shaped liquid hydrogen tank with a cooling motor. The toroidal fuel tank was detachable, with room for several more along the slender length of the drive tube. Around the rim of the cylindrical life support system were attachments for cargo, booms and folded fine-mesh nets and retractable hooks.

He did find inspection panels in the drive tube. Within an hour he could have built his own crystal-zinc fusion tube, had he the materials. He was impressed. The natives might be more intelligent than he had guessed, or luckier. He moved up to the lifesystem and through the oval door.

The cabin included an acceleration couch, banks of controls surrounding it in a horseshoe, a space behind the couch big enough to move around in, an automatic kitchen that was part of the horseshoe, and attachments to mechanical senses of types frequently used in Pak warfare. But this was no warship. The natives' senses must be less acute than Pak senses. Behind the cabin were machinery and tanks of fluid, which Phssthpok examined with great interest.

One thing he understood immediately.

He was being very careful with the instrument panel. He didn't want to wreck anything before he found out how to pull astronomical data from the ship's computer. When he opened the solar storm warning to ascertain its purpose, he found it surprisingly small. Curious, he investigated further. The thing was made with magnetic monopoles.

From PROTECTOR by Larry Niven (1973)

A bit more simplistic is a simple stack of octahedrons (Dungeons and Dragons d8 polyhedral dice). This was used for the spine of the Valley Forge from the movie Silent Running (1972), later reused as the agro ship from original Battlestar Galactica.

Saddle Truss

Spacecraft spines are generally down the center of the spacecraft following the ship's thrust axis (the line the engine's thrust is applied along, usually from the center of the engine's exhaust through the ship's center of gravity).

This can be a pain to spacecraft designers if they have anything that needs to be jettisoned. Such items will have to be in pairs on opposite sides of the spine, and jettisoned in pairs as well. Otherwise the spacecraft's center of gravity will shift off the thrust axis, and the next time the engines are fired up it's pinwheel time.

In a NASA study TM-1998-208834-REV1 they invent a clever way to avoid this: the Saddle Truss.

The truss is a hollow framework cylinder with a big enough diameter to accommodate standard propellant tanks, consumables storage pods, and auxiliary spacecraft. One side of the cylinder frame is missing. The thrust axis is cocked a fraction of a degree off-center to allow for the uneven mass distribution of the framework.

The point is that tanks and other jettison-able items no longer have to be in pairs if you use a saddle truss. When it is empty you just kick it out through the missing side of the saddle truss. No muss, no fuss, and no having to have double the amount of propellant plumbing and related items.

Examples of the saddle truss can be found in the Bimodal NTR and the SNRE Spacecraft.

Waterskiing Spacecraft

This is a quite radical method to drastically reduce the structural mass of a spacecraft, allowing a handsome increase in valuable payload mass. It also dramatically increase the separation between a dangerously radioactive propulsion system and the crew, allowing a drastic decrease in the radiation shadow shield mass. This allows yet more handsome increases in valuable payload mass. As the cherry on top of the cake, it allows using the tumbling pigeon method of spin gravity without the direction of gravity inverting.

Please note this has never actually been used in a serious nuclear spacecraft design due to its unorthodox nature.

And warships with such a design would have their manoeuvring critically handicapped (or it's "crack-the-whip" time and the cable breaks).

The concept comes from the observation that for a given amount of structural strength, a compression member (such as a girder) generally has a higher mass that a corresponding tension member (such as a cable). And we know that every gram counts.

Charles Pellegrino and Dr. Jim Powell put it this way: current spacecraft designs using compression members are guilty of "putting the cart before the horse". At the bottom is the engines, on top of that is the thrust frame, and on top of that is rest of the spacecraft held together with girders (compression members) like a skyscraper. But what if you put the engine at the top and have it drag the rest of the spacecraft on a long cable (tension member). You'll instantly cut the structural mass by an order of magnitude or more!

And if the engines are radioactive, remember that crew radiation exposure can be cut by time, shielding, or distance. The advantage of distance is it takes far less mass than a shield composed of lead or something else massive. The break-even point is where the mass of the boom or cable is equal to the mass of the shadow shield. But the mass of a shadow shield is equal to the mass of a incredibly long cable. The HELIOS cable was about 300 to 1000 meters, the Valkyrie was ten kilometers.

But keep in mind that this design has no maneuverability at all. Agile it ain't. If you turn the ship too fast it will try to "crack the whip" and probably snap the cable. This probably makes the design unsuitable for warships, who have to jink a lot or be hit by enemy weapons fire.

Examples include HELIOS, the Valkyrie Antimatter Starship, the ramrobots from Larry Niven's A Gift From Earth and the ISV Venture Star from the movie Avatar.

Certain propulsion systems incorporate the waterskiing concept in spacecraft that use the propulsion. The main one is the Medusa, which sets off nuclear explosions inside a huge parachute-shaped sail. The sail accelerates, and drags along the payload on a long cable. Long because the payload does not want to be any closer to a series of nuclear explosions than it has to be.

The various types of sail propulsion drag the payload with a long cable as well. But for them, the long cable is not because the sail is radioactive, just that it is typically several kilometers in radius.

Cosine Thrust Loss

If the exhaust is radioactive or otherwise dangerous to hose the rest of the spacecraft with, you can have two or more engines angled so the plumes miss the ship.

Angled engines do reduce the effective thrust by an amount proportional to the cosine of the angle but for small angles it is acceptable. The delta V of the spacecraft is also reduced by the same proportion.

Note in the HELIOS design Krafft Ehricke figured that the 300 meter separation was enough to render the exhaust harmless so it does not angle the engine at all. Krafft has a single engine blasting straight at the habitat module. The only concession to the exhaust is mounting the cables on outriggers, so the cables do not pass through the zillion degree nuclear fireball exhaust plume. It would be most embarassing if the cables melted.


The HELIOS has a thrust of 981,000 newtons. Say that Dr. Ehricke figured the exhaust would be dangerous to the habitat module, so the single engine would have to be replaced by two engines with 490,500 newtons each angled off-center by 5°. What would that do to the thrust?

CosineFactor = cos(OffAngle)
CosineFactor = cos(5°)
CosineFactor = 0.9962…

So both the thrust and delta V would be at 99.62%

Each engine has a thrust of 490,500 newtons, this would be reduced by the cosine thrust loss to:

EffectiveThrust = ActualThrust * CosineFactor
EffectiveThrust = 490,500 * 0.9962
EffectiveThrust = 488,636 newtons

488,636 newtons * 2 engines = 977,272 newtons. This means that angling the engines lowers the total thrust by 3,728 newtons, which is an ouch but not a show stopper. If the thrust absolutely has to be 981,000 newtons total, each engine would have to have its thrust increased from 490,500 to 492,371 newtons in order to compensate for the cosine loss.

If the HELIOS had a delta-V of 21,000 m/s, the cosine loss would reduce it to 21,000 * 0.9962 = 20,920 m/s. This is a loss of 80 m/s which is not negligible but not a show-stopper either.

Here's how we can shave off many tons of shielding.

Put the engine up front and carry the crew compartment ten kilometers behind the engine, on the end of a tether. Let the engine pull the ship along, much like a motorboat pulling a water skier, and let the distance between the gamma ray source and the crew compartment, as the rays stream out in every direction, provide part of the gamma ray protection - with almost no weight penalty at all. (ed. note: this should remind you of "Helios") We can easily direct the pion/muon thrust around the tether and its supporting structures, and we can strap a tiny block of (let's say) tungsten to the tether, about one hundred meters behind the engine. Gamma rays are attenuated by a factor of ten for every two centimeters of tungsten they pass through. Therefore, a block of tungsten twenty centimeters deep will reduce the gamma dose to anything behind it by a factor of ten to the tenth power (1010). An important shielding advantage provided by a ten-kilometer-long tether is that, by locating the tungsten shield one hundred times closer to the engine than the crew, the diameter of the shield need be only one-hundredth the diameter of the gamma ray shadow you want to cast over and around the crew compartment. The weight of the shielding system then becomes trivial.

The tether system requires that the elements of the ship must be designed to climb "up" and "down" the lines, somewhat like elevators on tracks.

We can even locate the hydrogen between the tungsten shadow shield and the antihydrogen, to provide even more shielding for both the crew and the antihydrogen.

There is an irony involved in this configuration. Our "inside-out" rocket, the most highly evolved rocket yet conceived, is nothing new. We have simply come full circle and rediscovered Robert Goddard's original rocket configuration: with engines ahead of the fuel tanks and the fuel tanks ahead of the payload.

From FLYING TO VALHALLA by Charles Pellegrino (1993)

Interstellar Ramscoop Robot #143 left Juno at the end of a linear accelerator. Coasting toward interstellar space, she looked like a huge metal insect, makeshift and hastily built. Yet, except for the contents of her cargo pod, she was identical to the last forty of her predecessors. Her nose was the ramscoop generator, a massive, heavily armored cylinder with a large orifice in the center. Along the sides were two big fusion motors, aimed ten degrees outward, mounted on oddly jointed metal structures like the folded legs of a praying mantis. The hull was small, containing only a computer and an insystem fuel tank.

(ed note: cosine of 10° is about 0.9848, so thrust and delta V would be reduced to 98.5%.)

Juno was invisible behind her when the fusion motors fired. Immediately the cable at her tail began to unroll. The cable was thirty miles long and was made of braided Sinclair molecule chain. Trailing at the end was a lead capsule as heavy as the ramrobot itself.

(ed note: "Sinclair molecule chain" is an unobtainium wire that is only one molecule thick and absurdly strong. The theoretical ultimate of low mass cable.)

On twin spears of actinic light the ramrobot approached Pluto's orbit. Pluto and Neptune were both on the far side of the sun, and there were no ships nearby to be harmed by magnetic effects.

The ramscoop generator came on.

The conical field formed rather slowly, but when it had stopped oscillating, it was two hundred miles across. The ship began to drag a little, a very little, as the cone scooped up interstellar dust and hydrogen. She was still accelerating. Her insystem tank was idle now, and would be for the next twelve years. Her food would be the thin stuff she scooped out between the stars.

From A GIFT FROM EARTH by Larry Niven (1968)
Pendulum Fallacy

Now those waterskiiing spacecraft designs look much like Robert Goddard's first liquid-fueled rocket "Nell " (i.e, engine on top). Don't be fooled, it is just a coincidence.

The waterskiing spacecraft designs are trying to subsitute almost mass-less engine standoff distance for ultra-penalty-mass radiation shield. Because on the one hand: Every Gram Counts, but on the other: nuclear radiation kills crewmembers dead.

Goddard's rocket was designed for totally different reasons (which you should have been tipped off by the fact it had no nuclear engine).

Goddard reasoned that if you put the rocket engine at the bottom and build the rest of the rocket on top of that, it would be as unstable as a waiter carrying a tall tippy bottle of wine on a tray held overhead by one hand. One minor shake of the hand and the wine goes crashing to the ground.

But if you put the rocket engine at the top and had it dragging the rest of the rocket, it would be as inherently stable as holding a pendulum by its string. If the engine tips over from its upward flight, the weight of the rest of the rocket will un-tip the engine. Right?

Nowadays rocket designers call this the Pendulum Rocket Fallacy. Meaning it looks good on paper, but it just doesn't work. Having the engines at the top is no more stable than at the bottom. A top-engine design superficially resembles a pendulum, but the system of forces acting on it are totally different.

Goddard discovered this the hard way with Nell's test flight. It rose barely 41 feet, tipped over, flew 184 feet horizontally, then augered into a cabbage field. All of his subsequent designs had the now-standard engine on the bottom arrangement.

Real spacecraft achieve stability by controlling some type of attitude actuator via some type of inertial measurement unit.

Space Trains and Truckers

This section has been moved here

Thermal Protection


Multi-layer insulation, or MLI, is thermal insulation composed of multiple layers of thin sheets and is often used on spacecraft. It is one of the main items of the spacecraft thermal design, primarily intended to reduce heat loss by thermal radiation. In its basic form, it does not appreciably insulate against other thermal losses such as heat conduction or convection. It is therefore commonly used on satellites and other applications in vacuum where conduction and convection are much less significant and radiation dominates. MLI gives many satellites and other space probes the appearance of being covered with gold foil.

Function and design

The principle behind MLI is radiation balance. To see why it works, start with a concrete example - imagine a square meter of a surface in outer space, held at a fixed temperature of 300 K, with an emissivity of 1, facing away from the sun or other heat sources. From the Stefan–Boltzmann law, this surface will radiate 460 W. Now imagine placing a thin (but opaque) layer 1 cm away from the plate, thermally insulated from it, and also with an emissivity of 1. This new layer will cool until it is radiating 230 W from each side, at which point everything is in balance. The new layer receives 460 W from the original plate. 230 W is radiated back to the original plate, and 230 W to space. The original surface still radiates 460 W, but gets 230 W back from the new layers, for a net loss of 230 W. So overall, the radiation losses from the surface have been reduced by half by adding the additional layer.

More layers can be added to reduce the loss further. The blanket can be further improved by making the outside surfaces highly reflective to thermal radiation, which reduces both absorption and emission.

The layers of MLI can be arbitrarily close to each other, as long as they are not in thermal contact. The separation space only needs to be minute, which is the function of the extremely thin scrim or polyester 'bridal veil' as shown in the photo. To reduce weight and blanket thickness, the internal layers are made very thin, but they must be opaque to thermal radiation. Since they don't need much structural strength, these internal layers are usually made of very thin plastic, about 6 μm (1/4 mil) thick, such as Mylar or Kapton, coated on one side with a thin layer of metal on both sides, typically silver or aluminium. For compactness, the layers are spaced as close to each other as possible, though without touching, since there should be little or no thermal conduction between the layers. A typical insulation blanket has 40 or more layers. The layers may be embossed or crinkled, so they only touch at a few points, or held apart by a thin cloth mesh, or scrim, which can be seen in the picture above. The outer layers must be stronger, and are often thicker and stronger plastic, reinforced with a stronger scrim material such as fiberglass.

In satellite applications, the MLI will be full of air at launch time. As the rocket ascends, this air must be able to escape without damaging the blanket. This may require holes or perforations in the layers, even though this reduces their effectiveness.

MLI blankets are constructed with sewing technology. The layers are cut, stacked on top of each other, and sewn together at the edges. Seams and gaps in the insulation are responsible for most of the heat leakage through MLI blankets. A new method is being developed to use polyetheretherketone (PEEK) tag pins (similar to plastic hooks used to attach price tags to garments) to fix the film layers in place instead of sewing to improve the thermal performance.

Additional properties

Spacecraft also may use MLI as a first line of defense against dust impacts. This normally means spacing it a cm or so away from the surface it is insulating. Also, one or more of the layers may be replaced by a mechanically strong material, such as beta cloth.

In some applications the insulating layers must be grounded, so they cannot build up a charge and arc, causing radio interference. Since the normal construction results in electrical as well as thermal insulation, these applications may include aluminum spacers as opposed to cloth scrim at the points where the blankets are sewn together.

From the Wikipedia entry for MULTI-LAYER INSULATION


There are some hazards to worry about with these space-age materials. Titanium and magnesium are extremely flammable (in an atmosphere containing oxygen). And when I say "extremely" I am not kidding.

Do not try to put out a magnesium fire by throwing water on it. Blasted burning magnesium will suck the oxygen atoms right out of the water molecules, leaving hydrogen gas (aka what the Hindenburg was full of). A carbon-dioxide fire extinguisher won't work either, same result as water except you get a cloud of carbon instead of hydrogen. Instead use a Class D dry chemical fire extinguisher or a lot of sand to cut off the oxygen supply. Oh, did I mention that burning magnesium emits enough ultraviolet light to permanently damage the retinas of the eyes?

The same goes for burning titanium. Except there is no ultraviolet light, but there is a chance of ignition if titanium is in contact with liquid oxygen and the titanium is struck by a hard object. It seems that the strike might create a fresh non-oxidized stretch of titanium surface, which ignites the fire even though the liquid oxygen is at something like minus 200° centigrade. This may mean that using titanium tanks for your rocket's liquid oxygen storage is a very bad idea.

An emergency crew at a spaceport, who has to deal with a crashed rocket, will need the equipment to deal with this.

And if the titanium, magnesium, or aluminum becomes powdered, you have to stop talking in terms of "fire" and start talking in terms of "explosion."


As an interesting side note, rockets constructed of aluminum are extremely vulnerable to splashes of metallic mercury or dustings of mercury salts. On aluminum, mercury is an "oxidizing catalyst", which means the blasted stuff can corrode through an aluminum beam in a matter of hours (in an atmosphere containing oxygen, of course). This is why mercury thermometers are forbidden on commercial aircraft.

Why? Ordinarily aluminum would corrode much faster than iron. However, iron oxide, i.e., "rust", flakes off, exposing more iron to be attacked. But aluminum oxide, i.e., "sapphire", sticks tight, protecting the remaining aluminum with a gem-hard barrier. Except mercury washes the protective layer away, allowing the aluminum to be consumed by galloping rust.

Alkalis will have a similar effect on aluminum, and acids have a similar effect on magnesium (you can dissolve magnesium with vinegar). As far as I know nothing really touches titanium, its corrosion-resistance is second only to platinum.


Corrosion in space is the corrosion of materials occurring in outer space. Instead of moisture and oxygen acting as the primary corrosion causes, the materials exposed to outer space are subjected to vacuum, bombardment by ultraviolet and X-rays, and high-energy charged particles (mostly electrons and protons from solar wind). In the upper layers of the atmosphere (between 90–800 km), the atmospheric atoms, ions, and free radicals, most notably atomic oxygen, play a major role. The concentration of atomic oxygen depends on altitude and solar activity, as the bursts of ultraviolet radiation cause photodissociation of molecular oxygen. Between 160 and 560 km, the atmosphere consists of about 90% atomic oxygen.


Corrosion in space has the highest impact on spacecraft with moving parts. Early satellites tended to develop problems with seizing bearings. Now the bearings are coated with a thin layer of gold.

Different materials resist corrosion in space differently. For example, aluminium is slowly eroded by atomic oxygen, while gold and platinum are highly corrosion-resistant. Gold-coated foils and thin layers of gold on exposed surfaces are therefore used to protect the spacecraft from the harsh environment. Thin layers of silicon dioxide deposited on the surfaces can also protect metals from the effects of atomic oxygen; e.g., the Starshine 3 satellite aluminium front mirrors were protected that way. However, the protective layers are subject to erosion by micrometeorites.

Silver builds up a layer of silver oxide, which tends to flake off and has no protective function; such gradual erosion of silver interconnects of solar cells was found to be the cause of some observed in-orbit failures.

Many plastics are considerably sensitive to atomic oxygen and ionizing radiation. Coatings resistant to atomic oxygen are a common protection method, especially for plastics. Silicone-based paints and coatings are frequently employed, due to their excellent resistance to radiation and atomic oxygen. However, the silicone durability is somewhat limited, as the surface exposed to atomic oxygen is converted to silica which is brittle and tends to crack.

Solving corrosion

The process of space corrosion is being actively investigated. One of the efforts aims to design a sensor based on zinc oxide, able to measure the amount of atomic oxygen in the vicinity of the spacecraft; the sensor relies on drop of electrical conductivity of zinc oxide as it absorbs further oxygen.

Other problems

The outgassing of volatile silicones on low Earth orbit devices leads to presence of a cloud of contaminants around the spacecraft. Together with atomic oxygen bombardment, this may lead to gradual deposition of thin layers of carbon-containing silicon dioxide. Their poor transparency is a concern in case of optical systems and solar panels. Deposits of up to several micrometers were observed after 10 years of service on the solar panels of the Mir space station.

Other sources of problems for structures subjected to outer space are erosion and redeposition of the materials by sputtering caused by fast atoms and micrometeoroids. Another major concern, though of non-corrosive kind, is material fatigue caused by cyclical heating and cooling and associated thermal expansion mechanical stresses.

From the Wikipedia entry for CORROSION IN SPACE

Protective Paint

If you want a World War II flavor for your rocket, any interior spaces that are exposed to rain and other corrosive planetary weather should be painted with a zinc chromate primer. Depending on what is mixed into the paint, this will give a paint color ranging from yellowish-green to greenish-yellow. In WWII aircraft it is found in wheel-wells and the interior of bomb bays. In your rocket it might be found on landing jacks and inside airlock doors.

Naturally this does not apply to strict orbit-to-orbit rockets, or rockets that only land on airless moons and planets. Well, now that I think about it, some of the lunar dust is like clouds of microscopic razor blades so they are dangerously abrasive.


Zinc chromate, ZnCrO4, is a chemical compound containing the chromate anion, appearing as odorless yellow powder or yellow-green crystals, but, when used for coatings, pigments are often added. It is used industrially in chromate conversion coatings, having been developed by the Ford Motor Company in the 1920s.


Zinc chromate’s main use is in industrial painting as a coating over iron or aluminum materials. It was used extensively on aircraft by the U.S. military, especially during the 1930s and 1940s, but is also used in a variety of paint coatings for the aerospace and automotive industries. Its use as a corrosion-resistant agent was applied to aluminium alloy parts first in commercial aircraft, and then in military ones. During the 1940 and 1950s it was typically found as the "paint" in the wheel wells of retractable landing gear on U.S. military aircraft to protect the aluminium from corrosion. This compound was a useful coating because it is an anti-corrosive and anti-rust primer. Since it is highly toxic it also destroys any organic growth on the surface. Zinc chromate is also used in spray paints, artists’ paints, pigments in varnishes, and in making linoleum.


Recent studies have shown that not only is zinc chromate highly toxic, it is also a carcinogen. Exposure to zinc chromate can cause tissue ulceration and cancer. A study published in the British Journal of Industrial Medicine showed a significant correlation between the use of zinc chromate and lead chromate in factories and the number of cases in lung cancer experienced by the workers. Because of its toxicity the use of zinc chromate has greatly diminished in recent years.

From the Wikipedia entry for ZINC CHROMATE

Rocket Tumble

The basic idea is that the Axis of Thrust from the engines had better pass through the the spacecraft's center of gravity (CG) or everybody is going to die. In addition, if the spacecraft is currently passing through a planet's atmosphere the axis of thrust had better be parallel to the aerodynamic axis or the same thing will happen.

Specifically, "everybody is going to die" means the spacecraft is going to loop-the-loop or tumble like a cheap Fourth-of-July skyrocket (Heinlein calls this a rocket "falling off its tail"). If this happens during lift-off the ship will auger into the ground like a nuclear-powered Dinosaur-Killer asteroid and make a titanic crater. If it happens in deep space, the rocket will spin like a pinwheel firework spraying atomic flame everywhere. This will waste precious propellant, give the spacecraft a random vector, and severely injure the crew with unexpected spin gravity. If they are lucky the crew's broken bones will heal about the same time that they run out of oxygen.

The axis of thrust is a line starting at the center of the exhaust nozzle's throat, and traveling in the exact opposite direction of the hot propellant. It is the direction that the thrust is pushing the rocket. As long as the axis of thrust passes through the CG, the spacecraft will be accelerated in that direction. If the axis of thrust is not passing through the CG, the spacecraft will start to spin around the CG. When done on purpose this is called a yaw or pitch maneuver. When this is done by accident, it is called OMG WE'RE ALL GOING TO DIE!

Some engines can be gimbaled, rotating their axis of thrust off-center by a few degrees. This is intended for yaw and pitch, but it can be used in emergencies to cope with accidental changes in the center of gravity (e.g., the cargo shifts).

When laying out the floor plan, you want the spacecraft to balance. This boils down to ensuring that the ship's center of gravity is on the central axis, which generally is the same as the axis of thrust. There are exceptions. The Grumman Space Tug has its center of gravity shift wildly when it jettisons a drop tank. To compensate, the engine can gimbal by a whopping ±20°.

Balancing also means that each deck should be "radially symmetric". That's a fancy way of saying that if you have something massive in the north-west corner of "D" deck, you'd better have something equally massive in the south-east corner. Otherwise the center of gravity won't be centered.

For a cargo ship, the Loadmaster has to ensure that the cargo is stored in a radially symmetric balance.

This is another reason to strap down the crew during a burn. Walking around could upset the ship's balance, resulting in the dreaded rocket tumble. This will be more of a problem with tiny ships than with huge cruisers, of course. The same goes for the cargo. The load-master better be blasted sure all the tons of cargo are nailed down so they don't shift. And be sure the cargo is evenly balanced around the ship's axis to keep the center of gravity in the center.

Small ships might have "trim tanks", small tanks into which water can be pumped in order to adjust the balance. The ship will also have heavy gyroscopes that will help prevent the ship from falling off its tail, but there is a limit to how much imbalance that they can compensate for.

Propellent Tankage

A cursory look at the rocket's mass ratio will reveal that most of the rocket's mass is going to be propellant tanks.

For anything but a torchship, the spacecraft's mass ratio is going to be greater than 2 (i.e., 50% or more of the total mass is going to be propellant). Presumably the propellant is inside a propellant tank (unless you are pulling a Martian Way gag and freezing the fuel into a solid block). Remember, RockCat said all rockets are giant propellant tanks with an engine on the bottom and the pilot's chair at the top.

If you have huge structure budget, you have a classic looking rocket-style rocket with propellant tanks inside. If you have a medium structure budget, you have a spine with propellant tanks attached. If you have a small structure budget, you'll have an isogrid propellant tank for a spine, with the rest of the rocket parts attached.

And if you are stuck with a microscopic structure budget, you'll have a foil-thin propellant tank stiffened by the pressure of the propellant, with the rest of the rocket parts attached. But the latter tends to collapse when the propellant is expended and the pressure is gone. This was used in the old 1957 Convair Atlas rocket, but not so much nowadays. You cannot really reuse them.


It had all seemed perfectly logical back on Mars, but that was Mars. He had worked it out carefully in his mind in perfectly reasonable steps. He could still remember exactly how it went. It didn't take a ton of water to move a ton of ship. It was not mass equals mass, but mass times velocity equals mass times velocity. It didn't matter, in other words, whether you shot out a ton of water at a mile a second or a hundred pounds of water at twenty miles a second. You got the same velocity out of the ship.

That meant the jet nozzles had to be made narrower and the steam hotter. But then drawbacks appeared. The narrower the nozzle, the more energy was lost in friction and turbulence. The hotter the steam, the more refractory the nozzle had to be and the shorter its life. The limit In that direction was quickly reached.

Then, since a given weight of water could move considerably more than its own weight under the narrow-nozzle conditions, it paid to be big. The bigger the water-storage space, the larger the size of the actual travel-head, even in proportion. So they started to make liners heavier and bigger. But then the larger the shell, the heavier the bracings, the more difficult the weldings, the more exacting the engineering requirements. At the moment, the limit in that direction had been reached also.

And then he had put his finger on what had seemed to him to be the basic flaw—the original unswervable conception that the fuel had to be placed inside the ship; the metal had to be built to encircle a million tons of water.

Why? Water did not have to be water. It could be ice, and ice could be shaped. Holes could be melted into it. Travel-heads and jets could be fitted into it. Cables could hold travel-heads and jets stiffly together under the influence of magnetic field-force grips.

From THE MARTIAN WAY by Isaac Asimov (1952)

Our running example Polaris spacecraft has a gas core nuclear thermal rocket engine.

The fuel is uranium 235. It will probably be less than 1% of the total propellant load so we will focus on just the propellant for now.

Nuclear thermal rockets generally use hydrogen since you want propellant with the lowest molecular mass. Liquid hydrogen has a density of 0.07 grams per cubic centimeter.

The Polaris has 792.6 metric tons of hydrogen propellant. 792.6 tons of propellant = 792,600,000 grams / 0.07 = 11,323,000,000 cubic centimeters = 11,323 cubic meters . The volume of a sphere is 4/3πr3 so you can fit 11,323 cubic meters in a sphere about 14 meters in radius . Almost 92 feet in diameter, egad! It is a pity hydrogen isn't a bit denser.

If this offends your aesthetic sense, you'll have to go back and change a few parameters. Maybe a 2nd generation GC rocket, and a mission from Terra to Mars but not back. Maybe use methane instead of hydrogen. It only has an exhaust velocity of 6318 m/s instead of hydrogen's superior 8800 m/s, but it has a density of 0.42 g/cm3, which would only require a 1.7 meter radius tank. (Methane has a higher exhaust velocity than one would expect from its molecular weight, due to the fact that the GC engine is hot enough to turn methane into carbon and hydrogen. Note that in a NERVA style engine the reactor might become clogged with carbon deposits.)

Propellant Tank Mass

Robert Zubrin says that as a general rule, the mass of a fuel tank loaded with liquid hydrogen will be about 87% hydrogen and 13% tank. In other words, multiply the mass of the liquid hydrogen by 0.15 to get the mass of the empty tank (0.13 / 0.87 = 0.15).

So the Polaris' 792.6 tons of hydrogen will need a tank that masses 792.6 * 0.15 = 119 tons.

87% propellant and 13% tank is for a rocket designed to land on a planet or that is capable of high acceleration. An orbit-to-orbit rocket could get by with more hydrogen and less tank. This is because the tanks can be more flimsy since they will not have to endure the stress of landing (A landing-capable rocket that uses a propellant denser than hydrogen can also get away with a smaller tank percentage). Zubrin gives the following ballpark estimates of the tank percentage:

PropellentEngineTank %
ArgonIon rocket4
WaterNuclear salt water rocket4
HydrogenNTR / GCR10

But if you want to do this the hard way, you'd better warm up your slide rule.

The total tank volume (Vtot) of a tank is the sum of four components:

  1. Usable Propellant Volume (Vpu): the volume holding the propellant that can actually be used.
  2. Ullage Volume (Vull): the volume left unfilled to accomodate expansion of the propellant or contraction of the tank structure. Typically 1% to 3% of total tank volume.
  3. Boil-off Volume (Vbo): For cryogenic propellants only. The volume left unfilled to allow for the propellant that boils from liquid to gas due to external heat.
  4. Trapped Volume (Vtrap): the volume of unusable propellant left in all the feed lines, valves, and other components after the tank is drained. Typically the volume of the feed system.

Vtot = Vpu + Vull + Vbo + Vtrap

No, I do not know how to estimate the Boil-off Volume. A recent study estimated that in space cryogenic tanks suffered an absolutely unacceptable 0.1% boiloff/day, and suggested this had to be reduced by an order of magnitude or more. When the boil-off volume is full, a pressure relief valve lets the gaseous propellant vent into space, instead of exploding the tank.

Tanks come in two shapes: spherical and cylindrical. Spherical are better, they have the most volume for the least surface area, so are the lightest. But many spacecraft have a limit to their maximum diameter, especially launch vehicles. In this case cylindrical has a lower mass than a series of spherical tanks.

The internal pressure of the propellant has the greatest effect on the tank's structural requirements. Not as important but still significant are acceleration, vibration, and handling loads. Unfortunately I can only find equations for the effects of internal pressure. Acceleration means that tanks which are in high-acceleration spacecraft or in spacecraft that take-off and land from planets will have a higher mass than tanks for low-acceleration orbit-to-orbit ships. My source did say that figuring in acceleration, vibration, and handling would make the tank mass 2.0 to 2.5 times as large as what is calculated with the simplified equations below.

In the Space Shuttle external tank, the LOX tank was pressurized to 150,000 Pa and the LH2 tank was pressurized to 230,000 Pa.

The design burst pressure of a tank is:

Pb = fs * MEOP


Pb = design burst pressure (Pa)
fs = safety factor (typically 2.0)
MEOP = Maximum Expected Operating Pressure of the tank (Pa)

Tank Materials

Allowable Strength
Mass Factor
2219 - Aluminum2,8000.413
0.214 welded
4130 - Steel7,8300.86211.232,500
Graphite Fiber

Spherical Tanks

You have to make Vs so it is equal to Vtot, or at least equal to Vtot - Vtrap.

Vs = 4/3 * π * rs3

As = 4 * π * rs2

ts = (Pb * rs) / (2 * Ftu)

Ms = As * ts * ρ


rs = radius of sphere (m)
As = surface area of sphere (m2)
Vs = volume of sphere (m3)
ts = wall thickness of sphere (m)
Pb = design burst pressure (Pa)
Ftu = allowable material strength (Pa) from tank materials table
Ms = mass of spherical tank (kg)
ρ = density of tank structure material (kg/m3 from tank materials table

Cylindrical Tanks

Cylindrical tanks are cylinders where each end is capped with either hemispheres (where radius and height are equal) or hemiellipses (where radius and height are not equal). As it turns out cylindrical tanks with hemiellipses on the ends are always more massive than hemispherical cylindrical tanks. So we won't bother with the equations for hemielliptical tanks. In the real world rocket designers sometimes use hemielliptical tanks in order to reduce tank length.

What you do is calculate the mass of the cylindrical section of the tank Mc using the equations below. Then you calculate the mass of the two hemispherical endcaps (that is, the mass of a single sphere) Ms using the value of the cylindrical section's radius for the radius of the sphere in the spherical tank equations above. The mass of the cylindrical tank is Mc + Ms.

Vc = π * rc2 * lc

Ac = 2 * π * rc2 * lc

tc = (Pb * rc) / Ftu

Mc = Ac * tc * ρ


rc = radius of cylindrical section (m)
lc = length of cylindrical section (m)
Ac = surface area of cylindrical section (m2)
Vc = volume of cylindrical section (m3)
Pb = design burst pressure (Pa)
Ftu = allowable material strength (Pa) from tank materials table
ρ = density of tank structure material (kg/m3 from tank materials table
tc = wall thickness of cylindrical section (m)
Mc = mass of cylindrical tank section (kg)


When the rocket is sitting on the launch pad, the planet's gravity pulls the propellant down so that the pumps at the aft end of the tank can move it to the engine. When the rocket is under acceleration, the thrust pulls the propellant down to the pumps. Once the engines cut off and the rocket is in free fall, well, the remaining pooled at the bottom turns into zillions of blobs and starts floating everywhere. See the video:

This isn't a problem, up until the point where you want to start the engine up again. Trouble is, the propellant isn't at the aft pump, it is flying all over the place. What's worse, some of the liquid propellant might have turned into bubbles of gas, which could wreck the engine if they are sucked into the pump. Vapor lock in a rocket engine is an ugly thing.

In 1960 Soviet engineers invented the solution: Ullage Motors. These are tiny rocket engines that only have to accelerate the rocket by about 0.001g (0.01 m/s). That's enough to pull the propellant down to the pump, and to form a boundary between the liquid and gas portions. In some cases, the spacecraft's reaction control system (attitude jets) can operate as ullage motors.

In the Apollo service module, they use a "retention reservoir" instead of an ullage burn (but they have to burn anyway if the amount of fuel and oxidizer drops below 56.4%).

Liquid oxygen in the oxidizer storage tank flows into the oxidizer sump tank. During an engine burn, oxygen flows to the bottom of the sump tank, through an umbrella shaped screen, into the retention reservoir, then into a pipe at the bottom leading to the engine. The same system is used in the fuel tanks.

When the burn is terminated and the oxygen breaks up into a zillion blobs and starts floating everywhere, the oxygen under the screen umbrella cannot escape. Surface tension prevents it from escaping through the screen holes. The oxygen is trapped under the umbrella, inside the retention reservoir.

When the engines are restarted there is oxygen right at the pipe to feed into the engine, instead of a void with random floating blobs. The engine thrust then settles the oxygen in the sump tank for normal operation.

As near as I can figure, the 56.4% ullage limit happens when the storage tank is empty, so the sump tank is only partially full. But I'm not sure.

Heat Shield

Aerobraking is used to get rid of a portion of a spacecraft's velocity without using a rocket engine and reaction mass. Or as NASA thinks of it: "For Free!" This can be used for landing, for planetary capture, for circulating spacecraft's orbit, or other purposes.

Robert Zubrin says mass of the heat shield and thermal structure will be about 15% of the total mass being braked.

The general rule is that aerobraking can kill a velocity approximately equal to the escape velocity of the planet where the aerobraking is performed (10 km/s for Venus, 11 km/s for Terra, 5 km/s for Mars, 60 km/s for Jupiter).

This will mostly be used for our purposes designing a emergency re-entry life pod, not a Solar Guard patrol ship. With a sufficiently advanced engine it is more effective just to carry more fuel, so our atomic cruiser will not need to waste mass on such a primitive device.

NASA on the other hand uses aerobraking every chance it gets, since they do not have the luxury of using atomic engines. Many of the Mars probes use aerobraking for Mars capture and to circularize their orbit. Some use their solar panels as aerobraking drage chutes in order to make a given piece of payload mass do double duty. Some of the Space Tug designs listed in the Realistic Design section economize on reaction mass by using a ballute when returning to Terra orbit.

In the movie 2010, the good ship Leonov had a one-lung propulsion system, so they needed an aerobraking ballute to slow them into Jovian orbit. If you are thinking about aerobraking, keep in mind that many worlds in the Solar System do not have atmospheres.

You can find a more in-depth look at heat shields here.


Typically the percentage of spacecraft dry mass that is propulsion is 3.7% for NASA vessels.

For a list of various spacecraft propulsion systems, go to the engine list.

Conserving Payload Mass

Penalty Weight

As you are beginning to discover, mass is limited on a spacecraft. Many Heinlein novels have passengers given strict limits on their combined body+luggage mass. Officials would look disapprovingly at the passenger's waistlines and wonder out loud how they can stand to carry around all that "penalty weight". There are quite a few scenes in various Heinlein novels of the agony of packing for a rocket flight, throwing away stuff left and right in a desperate attempt to get the mass of your luggage below your mass allowance.

Keep in mind that every gram of equipment or supplies takes several grams of propellant. Try to make every gram do double duty.

Tex hauled out his luggage and hefted it. "It's a problem. I've got about fifty pounds here. Do you suppose if I rolled it up real small I could get it down to twenty pounds?"

"An interesting theory," Matt said. "Let's have a look at it -- you've got to eliminate thirty pounds of penalty-weight."

Jarman spread his stuff out on the floor. "Well," Matt said at once, "you don't need all those photographs." He pointed to a dozen large stereos, each weighing a pound or more.

Tex looked horrified. "Leave my harem behind?" He picked up one. "There is the sweetest redhead in the entire Rio Grande Valley." He picked up another. "And Smitty -- I couldn't get along without Smitty. She thinks I'm wonderful."...

...Matt studied the pile. "You know what I'd suggest? Keep that harmonica -- I like harmonica music. Have those photos copied in micro. Feed the rest to the cat."

"That's easy for you to say."

"I've got the same problem." He went to his room. The class had the day free, for the purpose of getting ready to leave Earth. Matt spread his possessions out to look them over. His civilian clothes he would ship home, of course, and his telephone as well, since it was limited by its short range to the neighborhood of an earth-side relay office...

..He called home, spoke with his parents and kid brother, and then put the telephone with things to be shipped. He was scratching his head over what remained when Burke came in. He grinned. "Trying to swallow your penalty-weight?"

"I'll figure it out."

"You don't have to leave that junk behind, you know."


"Ship it up to Terra Station, rent a locker, and store it. Then, when you go on liberty to the Station, you can bring back what you want. Sneak it aboard, if it's that sort of thing." Matt made no comment; Burke went on, "What's the matter, Galahad? Shocked at the notion of running contraband?"

"No. But I don't have a locker at Terra Station."

"Well, if you're too cheap to rent one, you can ship the stuff to mine. You scratch me and I'll scratch you."

"No, thanks." He thought about expressing some things to the Terra Station post office, then discarded the idea -- the rates were too high. He went on sorting. He would keep his camera, but his micro kit would have to go, and his chessmen. Presently he had cut the list to what he hoped was twenty pounds; he took the stuff away to weigh it.

From SPACE CADET by Robert Heinlein (1948)

Long as he had been earthbound he approached packing with a true spaceman's spirit. He knew that his passage would entitle him to only fifty pounds of free lift; he started discarding right and left. Shortly he had two piles, a very small one on his own bed -- indispensable clothing, a few capsules of microfilm, his slide rule, a stylus, and a vreetha, a flutelike Martian instrument which he had not played in a long time as his schoolmates had objected. On his roommate's bed was a much larger pile of discards.

He picked up the vreetha, tried a couple of runs, and put it on the larger pile. Taking a Martian product to Mars was coal to Newcastle.

From BETWEEN PLANETS by Robert Heinlein (1951)

“Is all your cargo aboard? How much did they let you take?”

“A hundred kilos. It’s in the airlock.”

“A hundred kilos?” Norden managed to repress his amazement. The fellow must be emigrating—taking all his family heirlooms with him. Norden had the true astronaut’s horror of surplus mass, and did not doubt that Gibson was carrying a lot of unnecessary rubbish. However, if the Corporation had O.K.‘d it, and the authorised load wasn’t exceeded, he had nothing to complain about.

From THE SANDS OF MARS by Sir Arthur C. Clarke (1951)


In Frank Herbert's DUNE, spacemen had books the size of a thumb-tip, with a tiny magnifying glass.

"If it's economically feasible," Yueh said. "Arrakis has many costly perils." He smoothed his drooping mustache. "Your father will be here soon. Before I go, I've a gift for you, something I came across in packing." He put an object on the table between them-black, oblong, no larger than the end of Paul's thumb.

Paul looked at it. Yueh noted how the boy did not reach for it, and thought: How cautious he is.

"It's a very old Orange Catholic Bible made for space travelers. Not a filmbook, but actually printed on filament paper. It has its own magnifier and electrostatic charge system." He picked it up, demonstrated. "The book is held closed by the charge, which forces against spring-locked covers. You press the edge-thus, and the pages you've selected repel each other and the book opens."

"It's so small."

"But it has eighteen hundred pages. You press the edge-thus, and so . . . and the charge moves ahead one page at a time as you read. Never touch the actual pages with your fingers. The filament tissue is too delicate." He closed the book, handed it to Paul. "Try it."

From DUNE by Frank Herbert

Ruthless Optimization

Other innovations are possible. Perhaps boxes of food where the boxes are edible as well. The corridor floors will probably be metal gratings to save mass (This is the second reason why cadet shipboard uniforms will not have skirts or kilts. Looking up at the ceiling grating will give you a peekaboo up-skirt glimpse of whoever is in the next deck up. No panchira allowed. The first reason is the impossibility of keeping a skirt or kilt in a modest position while in free-fall.) In Lester Del Rey's Step to the Stars all documents, blueprints, and mail are printed on stuff about as thick as tissue paper (have you ever tried to lift a box full of books?).

With regards to low mass floors, the lady known as Akima had an interesting idea:

Unless the deck is also a pressure bulkhead, how about omitting deck plates and beams entirely, and making the floor a metal-mesh version of the trampoline decks used on sailing catamarans? That way, "weights" bearing on the decks would be transmitted into the tubular structure of the hull as an inward tension.

David Chiasson expands upon Akima's idea. There is an outfit called Metal Textiles which produces knitted wire mesh.

The meshes are knitted, as opposed to woven like a screen door. They are manufactured in densities (% metal by volume) from 10% to 70%. There are a wide variety of materials that the mesh can be made from, including aluminum, steels, Teflon, Nylon, even tungsten. Unfortunately, titanium is not on that list, I can only suspect that it must be difficult to get into a wire form suitable for making a knitted mesh.

Direct quote from site's main page: "In compressed form, knitted metal can handle shock loadings up to the yield strength of the material itself. The load may be applied from any direction-up, down or in from all sides."

I can speculate that with some kind of structural forming breakthrough, the mesh could be heated over a (ceramic?) mold to a near-melting point and simply pressed into place, compressing the mesh into a solid.

David Chiasson

Michael Garrels begs to differ:

I need to point out some issues with the idea of mesh floors.

First off there's the idea that bulkheads have to be bulky. In nautical settings, bulkheads have to be bulky to withstand the large pressure of water, to mount things like hatches on, and to provide overall rigidity to the ship during turning and impact. Most partitions in a spaceship would be a thin pressure membrane sandwiched between a mesh to avoid punctures. The skin on the Apollo lander module was thinner than common aluminum foil. If all you're trying to do is partition, pull up pictures of Skylab - you'll see curtains and isogrid all over the place.

Next is your distinction between floors and walls. Unless there is spin or thrust, there will be no such distinction.

Which brings us to the most important point - the floor that you're currently standing on isn't made out of mesh for a reason. Remember that classic description of a gravity well with a weight on a rubber sheet? Many building codes don't limit the weight allowed on floors but instead the amount of deflection allowed. Floors have to be bulky with occasional beams - otherwise you'll never be able to wheel a torpedo or a gurney, and debris will roll toward where you're standing. It might work in a hallway, or as on your boat for stowage of light items, but not for spans more than a couple meters at 1 g using real materials - especially if you want to mount something like a chair and a console in the center of the cabin.

Michael Garrels


If you are dealing with a conventional spacecraft ruled by the iron law of Every Gram Counts, a stowaway is a disaster. If they had not jettisoned a payload mass equal to their mass, there will not be enough propellant to perform the vital maneuvers. The ship will run out, and go sailing off into the Big Dark and a lonely death for everybody on board.

Even if the stowaway jettisions enough mass, there probably won't be enough breathing mix and food aboard for the additional person. Everybody will suffocate and/or starve.

As Dr. Feynman observed about the Challenger disaster, "Nature cannot be fooled." If the equations say that your spacecraft does not have enough fuel, they don't mean "maybe."

For survival's sake, the crew will have little choice but to immediately throw the stowaway out the nearest airlock.

This was highlighted in a famous story called The Cold Equations by Tom Godwin. The story is chilling abet scientifically accurate, but it still caused an uproar when it first came out.

But if the ship is a torchship or uber-powerful faster-than-light starship, things are a little less tense. Since they are not actually threatening the lives of the crew, stowaways will be treated more like their terrestrial counterparts if discovered on a sea-going vessel.


(ed note: Homir Munn has set out in his one-man sports-cruiser hyperspace starship on a desperate mission to fight the Second Foundation. His 14 year old "niece" Arcadia Darrel stows away on board, determined to help.)

     In the luggage compartment, Arcadia found herself, in the first place, aided by experience, and in the second, hampered by the reverse.
     Thus, she met the initial acceleration with equanimity and the more subtle nausea that accompanied the inside-outness of the first jump through hyperspace with stoicism. Both had been experienced on space hops before, and she was tensed for them. She knew also that luggage compartments were included in the ship’s ventilation-system and that they could even be bathed in wall-light. This last, however, she excluded as being too unconscionably unromantic. She remained in the dark, as a conspirator should, breathing very softly, and listening to the little miscellany of noises that surrounded Homir Munn…
     …Yet, eventually, it was the lack of experience that caught up with Arcadia. In the book films and on the videos, the stowaway seemed to have such an infinite capacity for obscurity. Of course, there was always the danger of dislodging something which would fall with a crash, or of sneezing — in videos you were almost sure to sneeze; it was an accepted matter. She knew all this, and was careful. There was also the realization that thirst and hunger might be encountered. For this, she was prepared with ration cans out of the pantry. But yet things remained that the films never mentioned, and it dawned upon Arcadia with a shock that, despite the best intentions in the world, she could stay hidden in the closet for only a limited time.
     And on a one-man sports-cruiser, such as the Unimara, living space consisted, essentially, of a single room, so that there wasn’t even the risky possibility of sneaking out of the compartment while Munn was engaged elsewhere…
     …She tried to poke her eyes outside the door without moving her head and failed. The head followed the eyes.
     Homir Munn was awake, of course — reading in bed, bathed in the soft, unspreading bed light, staring into the darkness with wide eyes, and groping one hand stealthily under the pillow.
     Arcadia’s head moved sharply back of itself. Then, the light went out entirely and Munn’s voice said with shaky sharpness, “I’ve got a blaster, and I’m shooting, by the Galaxy—”
     And Arcadia wailed, “It’s only me. Don’t shoot.”…
     …After a wild moment in which he almost jumped out of bed, but remembered, and instead yanked the sheet up to his shoulders, Munn gargled, “W … wha … what—”
     Arcadia said meekly, “Would you excuse me for a minute? I’ve got to wash my hands.” She knew the geography of the vessel, and slipped away quickly.

From SECOND FOUNDATION by Isaac Asimov (1953)

Stowaways in Space

     When I sat down to write this it was with misgivings. The idea of anyone hiding on board a ship in space seems a little farfetched to me. When Traveller was released in the ’70’s and people began SF roleplaying there was not an inkling that one day everything would have processing capacity and our wallpaper would have micro sensors in it.
     If you wanted to find a stowaway you had to, you know, look for the little sneak yourself.

     Perusing the rules for the anti-hijack program only indicates it will lock people out of control systems. 
     The classic Starship Operators’ Handbook from Digest Group went into great detail about how a ship’s computer could in fact scan a person for weapons, emotional state and then turn snitch on them long before they’d get to anyplace sensitive. However, we must also realize your computer is already pretty heavily tasked and internal sensors and AI cost credits. Not every ship will have them, certainly not many heavily mortgaged tramp fusers plodding along and plying their trade. So let’s go over some ways to stow away.

     Stowing away is defined for my purposes as obtaining passage onboard a ship through nonviolent but illegal means. If you’re holding the captain at gunpoint till he breaks orbit congratulations — you’re a hijacker.
     The easiest way of getting free passage is if you are a powerful telepath (“Aha! Over … wait I was wrong. Just a potted plant in here.”) Let the computer tag you for a drifter all it wants. Who cares if the men sent to grab you don’t seem to see or remember you? The Traveller race, Dronyne, with their cloaking ability would excel at this. S&S Andromedans used to have a molting period where they were invisible and surely an exceptional psi could be found in any race desired. The Dralasites from the Star Frontiers game weren’t invisible but darned near boneless and could easily hide in nooks and spaces no human could reach.
     The problem with all those species is when people begin noticing the missing oxygen, water, and food or see your little footprints on the deck using a thermograph or hear the head flushing. After that it’s a simple matter to order the crew into spacesuits and start evacuating sections of the hull. Note that to get the really satisfying WHOOSH and suck people off their feet we’re talking opening the hangar bay doors.

     Another use for psionics that doesn’t involve making like a Jedi (I am not the stowaway you are looking for …” “Doesn’t matter, you’ll do!”) is to use a power that lets you slow your body’s metabolism down to nothing. Mail yourself wherever in an airtight crate with an alarm clock.
     Medical Fast Drug can serve the same purpose. Make sure you have enough. It also sucks when people do find you as you can’t put up much of a fight (the ship may land before your first punch does.)

     Some stowaways will attempt to ride out their trip in a cargo container keeping a very low profile indeed. These crates are often elaborate affairs with thermal and sonic shielding, cryogenic devices and heat sinks to store and eliminate any thermal traces. This is not too odd when you realize money is not the only reason people stowaway. There are many reasons to let people think you are still on planet. Just be sure you take everything you need. Remember the stowaway who hid in a crated ATV swaddled in chill cans and thermal insulating blankets with water and food bars to spare. They caught him when he emerged after three days to use the bathroom. Yes zero sediment food bars are a thing. Buy some. For added safety bring your own breathing equipment in case they decide to evacuate the cargo bay for whatever reason.
     Of course there are people who convert cargo containers into stowaway modules. Some of them are better than the accommodations you pay for legally.

     There is another way to stow away that few people outside the well travelled circles. We’ve seen how institutions like CT’s Travellers’ Aid Society that gives high passage tickets as dividends.
     Now again back in the ’70’s we figured they were printed tickets (on very nice cardstock) and the recipients lugged them around in wheelbarrows or some such. Actually this works fairly well if you think about it. Give the person a ticket on physical media at each aid station or whatever dated and with a ledger (electronic or physical) that shows the receipt of said tickets. He presents himself, gets a ticket and gets his book stamped all very legal. Note that duplicating a high passage may mean presenting yourself as the recipient or forging a paper trail to prove its sale to you.
     I think those tickets would be very hard to duplicate. Hard but not impossible. counterfeit a high passage and suddenly you are indistinguishable from a paying customer. You can sit around your individual stateroom wondering what the poor stowaways are doing this time of year.
     For my part if I could forge high passage tickets I’d trade them in for the credits then spend them on a middle passage and have a thousand credits to spend on my trip. But then you blur the lines between scam artists and stowaways.

     Another way to blur the lines is to introduce squatters. When you’re hiding aboard a ship you’re a stowaway. When you hide aboard a station you’re a squatter. Being a squatter is an order easier (of course you do need to get to the station first.) Stations are bigger than most ships providing more hiding options. A station with a lot of personnel passing through might not notice the discrepancy in life support and since a station doesn’t go anywhere your mass will not throw it off course. Large space colonies are the ultimate in squatters and may have dozens or hundreds of people living off the grid … in spaaaaace.
     Sadly the results of being found remain the same: possible spacing and most likely arrest and deportation. In the event of deportation they will probably give you a space suit and re-entry kit (used).
     Squatters may even be tolerated on some stations. They provide a steady supply of dayworkers/scabs (or even thugs) if necessary. They could be engaged in all manner of illegal sales and services. If their uses outweighs the oxygen and water loss the station authorities may turn a blind eye to them for years or even decades.

     Of course they can cobble together a stowaway pod. Why do you ask?

From STOWAWAY THE EASY WAY by Rob Garitta (2017)

(ed note: the spacecraft uses some sort of technobabble antigravity device called the "Field Compensation Drive Generator")

The main field went on, and weight ebbed from the Centaurus. There were protesting groans from the ship's hull and structure as the strains redistributed themselves. The curved arms of the landing cradle were carrying no load now; the slightest breath of wind would carry the freighter away into the sky.

Control called from the tower: ‘Your weight now zero: check calibration.’

Saunders looked at his meters. The upthrust of the field would now exactly equal the weight of the ship, and the meter readings should agree with the totals on the loading schedules. In at least one instance this check had revealed the presence of a stowaway on board a spaceship — the gauges were as sensitive as that.

’One million, five hundred and sixty thousand, four hundred and twenty kilograms,’ Saunders read off from the thrust indicators. ‘Pretty good — it checks to within fifteen kilos. The first time I’ve been underweight, though. You could have taken on some more candy for that plump girl friend of yours in Port Lowell, Mitch.’…

…There was no sense of motion, but the Centaurus was now falling up into the summer sky as her weight was not only neutralised but reversed. To the watchers below, she would be a swiftly mounting star, a silver globule climbing through and beyond the clouds. Around her, the blue of the atmosphere was deepening into the eternal darkness of space. Like a bead moving along an invisible wire, the freighter was following the pattern of radio waves that would lead her from world to world.…

…With the silence of limitless power, the ship shook itself free from the last bonds of Earth. To an outside observer, the only sign of the energies it was expending would have been the dull red glow from the radiation fins around the vessel’s equator, as the heat loss from the mass-converters was dissipated into space.…

…An hour after take-off, according to the hallowed ritual, Chambers left the course computer to its own devices and produced the three glasses that lived beneath the chart table. As he drank the traditional toast to Newton, Oberth, and Einstein, Saunders wondered how this little ceremony had originated. Space crews had certainly been doing it for at least sixty years: perhaps it could be traced back to the legendary rocket engineer who made the remark, ’I’ve burned more alcohol in sixty seconds than you've ever sold across this lousy bar.’

Two hours later, the last course correction that the tracking stations on Earth could give them had been fed into the computer. From now on, until Mars came sweeping up ahead, they were on their own. It was a lonely thought, yet a curiously exhilarating one. Saunders savoured it in his mind. There were just the three of them here — and no one else within a million miles.

In the circumstances, the detonation of an atomic bomb could hardly have been more shattering than the modest knock on the cabin door…

…A stowaway was simply impossible. The danger had been so obvious, right from the beginning of commercial space flight, that the most stringent precautions had been taken against it. One of his officers, Saunders knew, would always have been on duty during loading; no one could possibly have crept in unobserved. Then there had been the detailed preflight inspection, carried out by both Mitchell and Chambers. Finally, there was the weight check at the moment before take-off; that was conclusive. No, a stowaway was totally…

(ed note: turns out the stowaway is Henry IX, crown prince of England. He had been trying to travel into space for years but the stodgy prime minister wouldn't hear of it. Henry was smuggled aboard with the aid of the two British crewmen Mitchell and Chambers.)

Saunders swallowed hard. Then, as the pieces of the jigsaw fell into place, he looked first at Mitchell, then at Chambers. Both of his officers stared guilelessly back at him with expressions of ineffable innocence. ‘So that's it,’ he said bitterly. There was no need for any explanations: everything was perfectly clear. It was easy to picture the complicated negotiations, the midnight meetings, the falsification of records, the off-loading of nonessential cargoes that his trusted colleagues had been conducting behind his back. He was sure it was a most interesting story, but he didn't want to hear about it now. He was too busy wondering what the Manual of Space Law would have to say about a situation like this, though he was already gloomily certain that it would be of no use to him at all.

It was too late to turn back, of course: the conspirators wouldn't have made an elementary miscalculation like that. He would just have to make the best of what looked to be the trickiest voyage in his career.

From THIS EARTH OF MAJESTY by Arthur C. Clarke (1955)

Power Generation

This section has been moved here

Energy Harvesting

This section has been moved here

Fuel Cells

This section has been moved here

Solar Thermal Power

This section has been moved here

Solar Photovoltaic Power

This section has been moved here

Radioisotope Thermoelectric Generators

This section has been moved here

Nuclear Fission Reactors

This section has been moved here

Bimodal NTR

This section has been moved here

Dusty Plasma Fission Reactors

This section has been moved here

Nuclear MHD

This section has been moved here

Fusion Reactors

This section has been moved here

Lattice Confinement Fusion

This section has been moved here

Exotic power sources

This section has been moved here

Beamed Power

This section has been moved here

Antimatter Power

This section has been moved here

Vacuum energy

This section has been moved here

Ladderdown Reactors

This section has been moved here

Mass Converters

This section has been moved here


This section has been moved here

Ship to Shore

This section has been moved here

Power Storage

This section has been moved here

Energy Transport Mechanism

This section has been moved here


This section has been moved here


This section has been moved here

Regenerative Fuel Cells

This section has been moved here

Superconducting magnetic energy storage

This section has been moved here

Kerr-Newman black hole

This section has been moved here

Heat Radiators

This section has been moved here

Habitat Module

This section has been moved here


This section has been moved here

Atomic Rockets notices

This week's featured addition is SPIN POLARIZATION FOR FUSION PROPULSION

This week's featured addition is INsTAR

This week's featured addition is NTR ALTERNATIVES TO LIQUID HYDROGEN

Atomic Rockets

Support Atomic Rockets

Support Atomic Rockets on Patreon