This is the first part of a series of questions I'll ask about a self-sustaining colonization ship. I'll keep editing this post as things unfold.

Part II - Landing

Edit 1: Added two movie references to try to make the ship's idea clearer and removed the points that were making this question too broad, as pointed by JBH. Also, life support is not being taken into consideration for physical space anymore. It will be discussed in a further question.

Some Context

The setting is kind of a low sci-fi. In this universe there are no FTL drives - in fact, the maximum level of technology is not even close to that. Humans rely on "old fashioned drives" with a little twist. They're more efficient than the ones we have today, but bear in mind that this is a very-near-future-tech scenario (50 to 100 years in the future).

Humans are in the early stages of the space colonization era. "The Colonists", as they're known, are huge ships equiped with the tools to ensure the crew's safety and to help populate the new worlds.

Since these trips might take decades, the ideia behind the ships is basically the same as seen in Pandorum and Passengers - most of the tripulation is in a suspended animation state while AI's or a minimal crew keep the thing on track.

For the sake of simplicity, assume that the colonized planets are exactly like Earth (but untouched by men). For now, this series focuses on the ships and their trips to the new worlds. In order to estabilish boundaries, these trips might take from 20 to 200 years.

This first question focuses on the building process of a Colonist. The original idea is that the ships are built completely outside of Earth in Space Shipyards located in orbit (kind of like the ISS).

The Question

Is it possible to build such an enormous ship? If yes, is the Space Shipyard a feasible idea?

Here's a list of things to consider for this question:

  • The Moon is a fully terraformed colony in this setting. This colony is able to provide 1/5 of a given raw resource, like iron or wood;
  • This is a collective effort of humanity, which means that, as long as the Earth and the Moon have the resources to do it, money is of no consequence;
  • Each ship has to have enough space for: 100 tripulants plus at least a dozen members of the crew, heavy machinery and fuel;
  • I decided to not take life support in consideration just yet because most of the crew will be in a suspended animation state. There will be another question to deal exclusively with this;
  • The Shipyard is already there and it has all the manpower and the tools to assemble the ship - but not the materials. They must come from either the Earth or the Moon;
  • Since we can't know what kinds of fuel, alloys or any other new tech might arise in the coming years, consider the costs for all transportation of materials as half what they are today;
  • The shipyard idea came from the notion that a ship this big would never be able to leave Earth because of its weight. I don't know if that's correct (but I think it is). In any case, if the shipyard is not necessary, feel free to kill it.
  • 1
    $\begingroup$ First, you need to define 1. approximately how many people the colonists carry, and 2. if any new alloys or materials have been created in this future. I can try answering once those two are cleared up. $\endgroup$ Dec 28, 2018 at 18:11
  • 1
    $\begingroup$ Also, it seems like it might be a given, but you have not explicitly stated how easily you can send workers and supplies to/ from the shipyard. $\endgroup$
    – cobaltduck
    Dec 28, 2018 at 18:13
  • 2
    $\begingroup$ 20 to 100 years for sublight speeds is incredibly short a time, I think you should increase it a bunch. Just getting to Alpha centauri at lightspeed means you are busy 4.3 years, so unless theres a bunch of earthlike planets at similar distances and you can get up to 0.2C or more its going to be a lot more than 20 to 100 years. This has an important Impact on the ship design. $\endgroup$
    – Demigan
    Dec 28, 2018 at 18:22
  • 1
    $\begingroup$ @Magus for materials, if it pans out Graphene will most likely be your best bet. So airtight it can hold helium, very light, extremely strong and depending on how you build it can be transparent, ridgid or extremely flexible while also great for conduction of heat or electricity. Future materials could also include 3D printed limbs and organs... Or entire people. This would mean that your "crew" is a biological paste most of your trip and printed into people at the last stages of the trip where they get enough time to get educated and trained in their duties. $\endgroup$
    – Demigan
    Dec 28, 2018 at 18:29
  • 1
    $\begingroup$ Should we assume that Earth has space elevators or other means to lift materials to the orbit at low cost? $\endgroup$
    – Alexander
    Dec 28, 2018 at 18:46

3 Answers 3


First off: passenger space.

(TL;DR: The passenger space would have to be pretty big, but we can cut corners and bring it down to Star Wars size.)

If you want to prevent the 100 or so passengers on the ship from getting claustrophobic and eventually insane, you'd have to provide a good amount of space for each passenger, especially if some of them live together. 10 x 10m might be a nice start, since you could most likely fit every basic house thing in that area. However, this already means that the ship would have to be at least 1000 x 1000 m in size for passenger space alone.

So, you could instead offer the passengers some form of 'class' system, where the ones who pay more get bigger rooms. Let's say those who opt for smaller rooms get a 6.5 x 6.5 meter space. We'll also say that out of the 100 rooms, 35 are the large ones. So, this takes the ship's total size for passenger space down to at least 772.5 x 772.5 m.

When you consider that such a huge area is just passenger space, it might not be super attractive, but at least all the passengers will have a nice time.

(For reference, that's about the same length as an Acclamator Class Landing Ship in Star Wars.)


(TL;DR: The oxygen storage, if concentrated in one room, would have to be almost as large as the equatorial radius of Ceres. Instead, we'll opt for a life support system spread across the entire ship's roof, which saves lots of space and only adds a few meters to the ship's height.)

Oxygen may or may not be a bit of a problem. Since we, on average, breathe in around 11,000 litres of air a day, over just 100 years all 100 people on board would need about 401,766,420 litres of air. That's about 401,766.4m³. Let's say everyone has personal oxygen systems that don't take up much space and thus we can subtract around 40-50% of the oxygen space. We'd still need 220,972m³ for oxygen space.

As a final caveat to reduce this ridiculous required space, let's say we've invented an alloy from tungsten and iron or something. The alloy is ultra-resistant to leaks and therefore we can remove the 10% of extra oxygen we took (just in case). That's down to 198,875m³. Which is still WAAY too much space.

So, we'll instead have life support systems in each individual room instead of having one big oxygen storage room, and since space stations like the ISS have it installed along the roof, it takes up far less space since it's spread out. We'd probably only need to add 5 or 6 meters to the ship's roof. We're at 772.5 x 777.5 meters now.

Leaving Earth

(TL;DR: This section is pretty short, so there is no TL;DR.)

While the space shipyard idea is presumably more efficient, it adds extra danger since the colonists must take a transport up to the shipyard. The transports could crash and explode. So, let's try to employ some methods to get it to leave Earth from the surface.

-A giant magnetic 'sling' on a huge runway, similar to the system that aircraft carriers use but for a giant spaceship.

-Simply tilting the ship up and attaching some incredibly strong thrusters to it. This might not be very efficient and could have a high chance of failure.

-A runway that leads up to a super-strong launching ramp. With the addition of some medium-strength thrusters, this could get it to escape velocity.

Any of those three are doable, as well as the space shipyard. I say you should pick which one to employ. The problem is, the shipyard might be a bit too futuristic, since you'd somehow have to get the shipyard into a stable orbit, and I can't even start there.

Final Dimensions

So since we've sacrificed around 700 square meters for passenger space, we'll have to keep everything else low. The cockpit will probably be the largest of all the extra rooms, since a large crew would be required to control a spaceship of this magnitude. We'll say 10 people man it. A cockpit of around 15 x 15 meters could comfortably house the people, their positions, and all the crazy machinery.

There should be a few kitchens and living-room type places to cook up simple snacks and relax with friends. These could be around the sides of the ship and would probably take up about 16 x 16 meters, or maybe a bit more.

Since those specific rooms are all we really need, plus maybe an engine room (we'll give all that stuff 50 x 50 meters and put it in the back), we're now at a size of 853.5 x 858.5 meters. Approximately.

That's about the same as a Devore Imperium Warship from Star Trek.

As a few final notes, I'd estimate cost to be around 5,000,000,000,000 dollars, which seems like a lot but would actually be feasible if the entire world came together.

Hope this ridiculously long post helped!

  • $\begingroup$ Seems like you're using area in your calculations instead of volume. You're also thinking of layouts like aircraft instead of spacecraft. think more "sky scraper", and less "single-story house". $\endgroup$
    – Stephan
    Jan 11, 2019 at 19:03
  • $\begingroup$ Ah, that was my mistake. $\endgroup$ Mar 21, 2019 at 4:34

The main issue with long term colony ships of this sort is the mass of supplies and requirement for virtually 100% recycling of everything. This isn't possible with today's technology, but we do know that humans need a certain amount of food, oxygen and water per day, and then try to discover the machinery or mechanism to make up the CLSS (Closed Life Support System). Your ship will then be scaled to carry that amount of mass, plus whatever "backup" materials you think you need (raw materials to "top up" the system, spare parts to keep the system in repair etc.

This is a NASA document which should give you an idea of the order of magnitude masses needed: https://ntrs.nasa.gov/archive/nasa/casi.ntrs.nasa.gov/19670025254.pdf

And this is a more modern iteration: https://ttu-ir.tdl.org/bitstream/handle/2346/73083/ICES_2017_311.pdf?sequence=1&isAllowed=y

The next issue is radiation shielding. Any spacecraft going into deep space needs to protect the crew from cosmic radiation and other events, so will require a massive amount of shielding. The "Millenial Project" (my go to book on these maters) suggests that a shield of water 5m in depth is needed. An equivalent amount of rock or metal is likely needed if you don't want to use water. The mass of the spacecraft can then be calculated by using the shape (i.e. a cone or sphere) to calculate the area and working out the mass of a 5m thick shield of water or rock surrounding it.

Since we now have a vessel which is lily the size an mass of an aircraft carrier, conventional rocket technology isn't going to work. The problem is you have to accelerate the mass of the ship, plus the mass of the fuel (including the fuel you are going to use to decelerate at the other end), which means that you are looking at a geometric expansion of mass as you try to increase the ships acceleration or speed.

The best way to get around this is to use some form of external power, like a massive solar sail (at this scale perhaps a series of solar sails tehtered together like a bunch of kites). You could either dive close to the Sun and unfurl the light sail, or use massive banks of lasers to drive the sail.

Paradoxically the way to get out of the Solar System fast is to first arrange that your probes dive towards the Sun with the probes facing edge on to minimize radiation pressure. Then turn almost face on at perihelion(closest to the Sun) and blast away.

If closest approach is 1/10 of an Au, the final velocity is:

420 km/sec

And if the materials(some sort of unobtanium:)) could withstand an approach to 1/100 Au, only 1.5 million km from the center of the Sun, and therefore only 800,000km over the seething surface(!), the final velocity of our interstellar probe would be:

1330 km/sec

Astute readers will notice that just as the escape velocity of a massive body varies inversely as the square root of the starting distance R from the center, so does the final velocity of our Solar Probe.


enter image description here

Actually, at this scale, you would not really be able to see the ship. Solar sails will have to be unimaginably large

So you will be looking at a very large, very massive ship. Propelling it by diving into the Sun to provide the maximum amount of Solar energy provides the velocity necessary to reach Alpha Centauri in about 1000 years. Presumably the crew at the time of arrival will arrange to "dive" into Alpha Centauri to provide the energy to brake into stellar orbit and then explore the system.

  • $\begingroup$ You can't substitute an equivalent amount of rock or metal for water when it comes to shielding. Effectiveness is measured by Half-Value Layer. A good write-up summary is available at Hanford.gov $\endgroup$
    – Stephan
    Jan 11, 2019 at 19:23

Location, Location

The shipyard is located in the Asteriod Belt, or in solar orbit just past Neptune.


  • Located in the Asteriod belt, resources are easy to acquire and resources are required in large amounts.
  • Located in High solar orbit, the launched ship will have the perfect launching position to use the large planetary bodies and the sun itself to perform an initial gravity assisted slingshot. Additionally moving large amounts of mass to the shipyard can be done "safer" as any rogue masses are more likely to be caught by the large planetary gravity wells, or be ejected into cold space.
  • In either orbit, delta-v requirements are reduced for solar-system escape in comparison to earth orbit.


Providing continuous power to such a large scale construction project, not to mention any living requirements for humans, is going to take more than can be provided by solar cells at those orbits. Nuclear Fission reactors might be able to do the trick, but these require specific isotopic elements, and produce nasty by-products. Alternately Nuclear Fusion reactors (while still requiring specific isotopic elements) don't produce nasty by-products, but also currently don't produce any net-positive power. Perhaps this would be possible in a few decades, perhaps not.

Also we have not considered how the ship itself will be powered once launched. Doubtlessly the design will contain multiple redundant systems, but operating its own power plants will require consumable resources, and what do you do with the rubbish?

Thinking large scale what about a Dyson Array? Even a very modest array will provide power in copious amounts for not only construction, but also for transmitting to the vessel for 200+ years.


People moving to and from the shipyards can be provided by a smaller 500+ person ship. Again powered by a mix of solar, reactor, and dyson array. The engines can be ion based, with a solar sail carefully pushed by the dyson array.


Perhaps the most complex operation, while the asteriod belt contains resources this would present some difficulty for humans to extract. Robots could do the job, but we are still fumbling the robotics in our own factories on Earth as Tesla can show. A near term source would be a mining colony on the moon, or mercury.

  • The moon has hassles in terms of atmosphere, and cosmic radiation.
  • Mercury has hassles in terms of gravity

Life Support

This is the most difficult aspect. We aren't sure what is needed for long-term health in space, even within the solar system. The best bet is to engineer the space station to provide something convincingly earth like: Atmospheric pressure, actual terran biomes, gravity, and sun-like light.

The best design for this is actually a ring (or rings) with a central counter rotating shaft for docking, and engine mounting. The ring should have a circular cross-section to maximise both internal space, and strength of the overall structure. If there are multiple rings, perhaps some cross ring bracing for extract structural stability.

So as to not induce too much discomfort the rings should have an inner diameter of at least 3km. The cross-sectional diameter will depend in large part on the resources available, assembly technique, etc..

Bootstrapping the First Shipyard

The first such shipyard would be constructed modularly from Earth, not unakin to the ISS. The modules would be connected at angles to each other forming a small ring. It could probably only rotated to emulate <1/6 earth gravity. Enough to allow work crews to begin fabrication of the actual ring. This inner ring might be kept on the final station as a maintenance yard, or it might be detached to start construction on the next station for deployment elsewhere in the solar system.

Resourcing this proto-shipyard will be difficult, as resources from earth are expensive. Moving this proto-shipyard to lunar orbit would be one option. We currently posses technology sufficient to establish a space-elevator on the moon, providing a cheap way to move resources to the proto-shipyard. Alternately as the moon is very important to the Earth-Moon system, pilfering resources from the moons of mars would be easier as they have less gravity. Though this will make the human logistics harder.

Once the proto-shipyard has constructed the first shipyard, humanity can really start building at scale the generation ships.

  • $\begingroup$ Fusion power currently produces NO net power. We currently use Fission generators like Radioisotopic Thermoelectric Generators. Since OP hasnt specified that net-positive generation fusion has been achieved, this may need to be addressed. Also, another benefit to being in the asteroid belt is low Delta-V to solar escape. $\endgroup$
    – Stephan
    Jan 11, 2019 at 19:07
  • 1
    $\begingroup$ @Stephan, fair point i've updated my answer. $\endgroup$
    – Kain0_0
    Jan 13, 2019 at 23:07

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .