[ExI] Fwd: Village in space, was marmalade agenda

Adrian Tymes atymes at gmail.com
Thu Nov 3 08:04:00 UTC 2016

On Wed, Nov 2, 2016 at 8:05 PM, Keith Henson <hkeithhenson at gmail.com> wrote:

> > Or it may be through the use of zero-gee research and manufacturing that
> > yield high value knowledge and/or products.
> So far this has failed.  One of the more promising projects was
> growing large hard-to-crystallize proteins in zero g so the structure
> could be determined by X-ray crystallography.  Alas, the people
> involved delayed getting around to doing it for so many years that an
> alternative method, femtosecond lasers, was developed that gets
> entirely around the need for large crystals.

I've spoken to people making money from zero-G manufacturing.  They are
able to turn quite a profit from "Vomit Comet" brief microgravity in
climbing and falling airplanes, but they do not have nearly enough of a
market to justify even small orbital manufacturing plants.  That said,
there has not been much room to affordably experiment with this yet, so
perhaps non-trivial products might justify orbital manufacturing better (at
least in the current stage, where there is little enough demand for almost
anything on-orbit that manufactured products must economically justify
themselves being sold back to Earth markets, if they are to economically
justify themselves at all).

> "It is generally believed that at 2 rpm or less, no adverse
> effects from the Coriolis forces will occur, although humans have been
> shown to adapt to rates as high as 23 rpm.[6]"
> https://en.wikipedia.org/wiki/Artificial_gravity
> For the same thickness of shielding, the Stanford Torus used a lot
> more mass than the Bernal Sphere designed the same year.  (Roughly the
> area of a torus to a sphere.)

I have favored the O'Neill Cylinder when I have tinkered with space
colony.  Assuming spin gravity, only a certain band of the torus or sphere
is habitable, but most of the surface area of a cylinder (everything but
the caps) could be subject to 1G.  1G and 2 RPM can be achieved with a
radius of just under 250 meters.
http://www.artificial-gravity.com/sw/SpinCalc/ is a cool little utility to
let you play around with G vs. RPM vs. radius vs. tangential velocity.

(Further, a cylinder can - in theory - be extended as the population grows:
build more cylinder past the enclosed section, put a new end cap at the new
end, then disassemble the old end cap that is now in the middle of the
cylinder.  This is critical when starting out with a few hundred people -
and thus, only the resources to build a habitat for a few hundred people -
but then growing to a medium-sized city equivalent - 10,000 to 100,000.
After that you'd want to just build new cylinders, but by that point you're
probably generating enough resources to do so.)

> We have to pick a number for the size of the habitat and the
> population--just to get some numbers for the artists.  For a first
> try, how about 1/10th of the diameter of O'Neill's Island One?  That
> was 500 meters diameter and held about 10,000 people.  For this one, I
> am going to assume 50 meters and 400-500 people.  That makes the
> radius 25 m, the spin rate 5.98 RPM, and the volume 65450 cubic
> meters.  If the population is 400 people, they have163.6 cubic meters
> per person.  For a population of 500 it would be a bit over 130 cubic
> meters per person.

Except that much of that would be at different RPMs, and quite a bit in
near zero G near the axis of spin.  That is not area I would consider
habitable.  (Also, 5.98 > 2; if humanity in general can live in 2 but no
higher, then 5.98 as the lowest RPM is right out.)

Estimates of livable population density vary wildly, but I would take my
home town, Mountain View, as a good area: generally suburban density, along
with substantial commercial and industrial space.  (My preferred design has
the outermost layers of the cylinder dedicated to agriculture, since that
is where much of the colony's water would be - thus, taking advantage of
that placement to shield the inhabited area.)
https://en.wikipedia.org/wiki/Mountain_View,_California gives a population
density of 2,300 per square kilometer.  Going beyond just workers and going
for a viable colony, https://en.wikipedia.org/wiki/Minimum_viable_population
suggests we want at least a bit over 4,000; to simplify the math, let's go
for 4,600 people and thus 2 square kilometers.

and other sources suggest that with good modern techniques, you could feed
at least 1,000 people per square kilometer of arable land, and use of
hydroponics et al - which would be a given in this case - can more than
double this amount.  Also note that, if the agricultural layer is stacked
underneath/outside the habitable layer, a given square kilometer of
habitable area means more than a square kilometer of agricultural area.
Multiple agricultural layers are possible if this estimate proves high,
meaning 1 km^2 of habitable area could have 10 or more km^2 of agricultural
area beneath it.  All of which means that 2,300 people per km^2 can be fed

2 square kilometers of livable area, and for simplicity assume 250 meter
radius, comes to just under 1.3 km length.  If we go for 500 meter radius
(which, using the utility linked above, drops the RPM to a bit over 1.33 -
which means we have safety margin for habitability), we're at just over 630
meter length.  While the former is more efficient use of shielding (when
calculating the total surface area, only the size of the end caps differ),
the latter is probably more stable (lower length/diameter ratio).

Further, for the "from 100 to 100,000" problem above, the length scales
linearly.  100 people at 500 meter radius is a 7ish meter wide petri dish -
livable, especially if it's only for a few years until hundreds more can be
attracted (bringing funds to pay for lengthening the habitat), while
100,000 is slightly under 7 km (which is where you'd want to stop and build
another cylinder, if you hadn't already).

> The habitat also needs power to run fans, pumps, a few lights and a
> lot of computer workstations and low latency teleoperator stations.
> Detailed design will modify this, but I am going to initially figure a
> kW/person load or ~400 kW.

http://www.indexmundi.com/g/r.aspx?v=81000 (which sources from the CIA
World Factbook) gives, for the US, an annual consumption - including
industrial consumption - of just over 13 MWh, or on average just over 1.4
kW/person.  Even in the environment you envision, adding in all the
electronic conveniences people are used to and adding industrial space
station needs on top of that, you're probably correct to within an order of
magnitude, if slightly low.

PS.  When we cut a new power satellite loose, do we smash a bottle of
> Champagne?

 Following up on Spike's suggestion that this is in a sense the opposite of
a battleship, maybe use the satellite's energy to 3D print a bottle?
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.extropy.org/pipermail/extropy-chat/attachments/20161103/5a007f8a/attachment.html>

More information about the extropy-chat mailing list