Wednesday, April 26, 2017

Another day, another post about spaceships.

Another day, another post about spaceships.

Imagine a civilization that values modular design, simplicity, and things working together.

I've got this notion of cylinders inside of cylinders. Standard systems core as a 10m long, 10m in circumference cylinder. Standard hab module going around that, so having 20m circumference at the outside. The maybe storage and cargo in the 20m to 30m section, meaning an outside surface that's 30m by 10m with a couple of hard points.

Constant acceleration, but low (like 0.01G), and trading partners -- either planets or habitation stations -- close to each other, like a half an AU. That gives a travel time of ~3 weeks, without the need for FTL.

Make it modular so you can attach another ring segment outward, or extend outward. You could get some funky looking ships.

This making any sense to anybody?

37 comments:

  1. It does, but wrapping things further and further out is, in fact, the opposite of modularization, right? A 50m circumference cylinder has no parts in common with a 40m circumference cylinder.

    Stacking them end to end seems a lot more modular.

    ReplyDelete
  2. Sure, but a 40m cylinder can be swapped out for another. So those are interchangeable.

    ReplyDelete
  3. I can't quite picture it. Can you sketch it?

    ReplyDelete
  4. Which part, Brian Ashford?

    Systems core is just a 10m long cylinder. Hab module around it. Storage module around it.

    What doesn't make sense?

    ReplyDelete
  5. William Nichols​ how much space there is in the mid and outer sections which is not filled with the inner sections. Is it just three concentric walls?

    ReplyDelete
  6. William Nichols: Are you envisioning that in a craft where people have been living in (say) 20, 30 and 40 meter cylinders wrapped around a 10-m spine, they would then (for an upgrade) pull out the 30-meter cylinder and replace it with a different one?

    That seems ... involved. Like, markedly more involved than swapping out both the 30 and 40 meter cylinders, which is (in turn) more involved than just swapping out the whole section.

    ReplyDelete
  7. Ohhh, I see.

    Consider the 2D field, so a single circle for each cylinder.
    Inner: r = 5
    mid: r = 10
    outer: r = 15

    In the mid, the useable area is Pi * 100m^2 - Pi * 25m^2, or 75 * Pi m^2, call it ~225 square meters.

    Looked at another way, the cylinder is a rectangle where the top and bottom are the same, right? And for the mid, the nearer the core version is 10m by 10m, or 100 square meters. If you consider this the "floor", then you've got a reasonable apartment.

    This gives a living square of 22,500 cubic meters. Which is not something I can intuitively understand,as I don't live in zero G.

    ReplyDelete
  8. Tony Lower-Basch Sounds easier just to attach a new 40m section onto the front! And so much room.

    ReplyDelete
  9. Brian, are you asking what the connection method is/what surfaces are shared between the modules? Hmmmm...

    Like, is the inner cylinder 10m in diameter, or is that the inner surface, and then there would need to be 1-2 meters of structure around that... then the second space would actually start at the 12-14m mark and reach out to the 32-34m mark? Or would it be 10m inclusive of the structural walls on the outside, so the inner space would really only be about 8-9m?

    ReplyDelete
  10. Ten meters inclusive. That is, if you roll out the "floor" of the middle one, it'd be 10 meters. If you roll out of "floor" of the third one, it'd be 20 meters.

    Because that's remarkably easier.

    ReplyDelete
  11. But you are right, I need to adjust the useable area a bit to have space between. Doing the math on a conic section sounds obnoxious.

    ReplyDelete
  12. OK, I think I've got it now.

    So this is intended for zero or low-g living? So there isn't a floor really.

    ReplyDelete
  13. Right! 0.01 G is (I think) barely noticeable, but constant acceleration gets the job done.

    ReplyDelete
  14. Oh, blarg. I don't have to do the math.

    With a change in size of 10 meters per section, then the point is we have arbitrary stuff between the levels that means the level size can be exactly 3 meters high. That's where the math goes, and Jesse Cox solved it yesterday.

    So the mid level has exactly 3 m x 10m x 10m interior volume, or 300 m^3. Again, I don't live on a spaceship so I can't intuit that. Consider the 10 x 10 the "floor" means you've got 100 square meters, which is a good sized apartment.

    The next layer out is 3 x 10 x 20, so 600 m^3 if I'm mathing correctly. Each increases linarly, which is cool.

    ReplyDelete
  15. I ... question whether you're mathing correctly. Can you walk me through how taking the cylindrical shell is getting you this "3 x 10 x 20" equation? I feel like there should be a tau in there somewhere.

    ReplyDelete
  16. Note: I grasp that one of the tau's is included in your taking a 10m circumference rather than any sort of diameter ... but a cylindrical shell really isn't the same space as a rectangular parellelopiped. That will be particularly noticeable in the 10-m to 20-m interspace, but noticeable everywhere.

    I mean, technically it's really noticeable in the 0-m to 10m interspace, which is somewhat larger than 3 x 0 x 10 m3.

    ReplyDelete
  17. 3.18 meters is conventionally roughly a story in a multi-story building.

    So if you're thinking living space, it's probably better to consider it as "square meters of floor plan" the ceilings will be smaller than the floors, but once you're a few layers out it won't matter much.

    More interesting may be how much spin you need for gravity, and how gravity carries between layers.

    ReplyDelete
  18. Jesse Cox Who needs spin gravity? Heck, who needs gravity?

    Not spacers!

    ReplyDelete
  19. Actually, maybe not... the concentric tube thing is non-simple.

    ReplyDelete
  20. If you're well-adapted to microgravity, you're unlikely to be able to handle yourself at 10m/s^2.

    Human stock is pretty well adapted for 10m/s^2, and all sorts of things get pretty wonky if left in microgravity for too long...particularly if you have any intention of getting back into gravity. Hypercalicemia is no laughing matter!


    Rough numbers -- if people want to live somewhere between 12 and 8 m/s^2, and they're living near the outside, that means a little more than half the space cylinder volume is living space. The half the volume around the center is low-gravity equipment.

    So if you have 6 layers, the innermost 4 are equipment, and the outer two are living space. Expanding -- add three layers outside and you lose the inner two when you slow down the spin. But you've gone from (for a 10m ring) 1100 m^2 living space to 2400m^2 living space. Not shabby! Then again, you could have just doubled the length of your ship instead.



    This also means the curve of your floor isn't super intense.

    ReplyDelete
  21. For square meterage for living area -- a folksy article, but decent data

    http://shrinkthatfootprint.com/how-big-is-a-house


    The bit about minimum living space in British flats at the end is good -- notably, communal kitchens and bathrooms reduce living space needs considerably, but between 20 and 60 square meters per person covers a wide variety of tastes.

    Oh, and I'd you're doing spin gravity, the outermost layer -- the one with a 40m circumference -- makes a great exercise track.

    ReplyDelete
  22. Right. But, again, when did I say anything about spin gravity? I'm imagining a culture who largely live in space ships and habitats, so they don't need to be able to be cool in a 1 G field.

    ReplyDelete
  23. If you have no interest in spin gravity, cylinders are a dreadful choice for apportioning living space. Why not just cubes?

    ReplyDelete
  24. Ah! Then this gets much simpler. And it makes sense -- pulling ships out of gravity wells is probably the most energy-expensive operation you can imagine.

    ::shuffles growing stack of algebra under his desk, nothing to see here::

    ReplyDelete
  25. Jesse Cox Exactly, going from high G field to high G field is ridiculous expensive.

    If you want to do the math on cylinders and spin gravity, then by all means!

    ReplyDelete
  26. I feel that a civilization with those values would build the shipping container, but for space.

    And there would a standard variant, for "shipping" live goods, that had life support. And the life support would be designed to link together for scalability.

    And, of course, the module would be rectangular, probably roughly 1:2:4. Which probably means the default size is about 3x6x12 meters. Eight modules, turned sideways, can link up into a larger 6x12x24 super-module, ad nauseam. Alternatively, some ratio that involves roots. See A4 paper.

    For communities larger than a few families, you'd want to have communal modules. I'm trying to think up a fractal structure, where you link modules into extended families, and family module clusters into neighborhoods, and neighborhood-module-clusters into communities, and so on.

    ReplyDelete
  27. Nobody likes my cylinders and everybody wants boxes.

    Granted, boxes are super super useful and easy.

    ReplyDelete
  28. On my way home, I remembered: there are two major concerns with building space habitats. Pressure and Radiation.

    Hulls designed for pressure look like spheres, ellipsoids, or rounded cylinders. All of which are less practical than boxes, but might save enough weight to be worth it.

    You can't design the whole hull for radiation. Too much mass is required. Budget a couple tons of solid rock per square meter. Probably in the form of giant bricks between as much colony as possible and the sun.

    ReplyDelete
  29. William Nichols, because it seemed like a good sized living space, however, the more I math it, the less fond I am of it. Check out how A4 paper cuts up into an A5 and two A6. That's a really convenient property, and a three dimensional version would be very useful for space modules.

    ReplyDelete
  30. Jonathan Beverley So an A(n) = 2 A(n+1), which looks to extend down for essentially all n, with A(0) having a specific size such that the ratio of sides stay the same. A(0) is 1 m^2, with an aspect ratio of 1: rt(2).

    This is relatively cool.

    To extend to 3D spaceships, I conjecture the following:
    1. It is preferable to extend forever upwards than downwards.
    2. Instead of 2, break into 4.
    3. Smallest size can be 1 m^3.

    Then let's say, that for Volumes V:
    V(0): 1 m^3
    V(n) = 4 * V(n-1)

    This allows us to conjecture the existence of any V >= 0.

    Two things:
    1. This goes 1-4-16-64-256-2048 m^3, which makes the math geek in me happy but those numbers aren't exactly round. Instead talking about useable space and going: 1-2-10-50-200-2000 could be interesting -- as then everything is nice and round, which makes the non-binary calculations a lot easier.

    2. I do not know what ratios are required to do this, and cannot currently do that math.

    ReplyDelete
  31. Also, if it winds up being 1:4:9, imma gonna punch Arthur C Clark. Well, not actually punch. Just kinda mock.

    ReplyDelete
  32. I'm positive the above math is all crap. You can almost assuredly do a 3D object with doubling from one level to the next. I'm still interested in the ratio.

    ReplyDelete
  33. Heck, if we're free from the constraints of gravity, and just need to worry about shielding, pressurizing, and not breaking when we run the engines, I propose the best design is two nested geodesic domes as the base module.

    The edges of the outer dome have structural airlocks -- twelve of them, I think -- for joining to do sphere packing. Not awesome for space efficiency but, as has been noted, Space is Big. We can afford a larger footprint for the saving of materials.

    Multiple globes also means that if one is breached, the others are ok -- and if the whole structure is breached, the inner globes are still pressurized.

    The question then becomes, can you make a bigger geodesic globe, say one that takes up the space of, say, 13 normal globes and has the airlocks line up to fit with the basic structure.


    Either way, you're not going to get an easy battle-mat style map.

    ReplyDelete
  34. The easy thing is to do 8 cubes making up a larger cube, which is pretty straightforward.

    But, that's not nearly as need as the A3-A4-A5 setup, which is pretty cool.

    ReplyDelete