Tue, 18 Apr 2000 21:25:58 -0700
Mark Simmons was woofing about the open type colony's mirrors. Here's a
bit more detail on the problem.
Unlike virtually all of the other space colony designs of the last hundred
years, only the O'Neill cylinder has mirrors that rotate along with the
colony and extend out into space from it. The Bernal sphere has an annular
parabolic mirror on the far end, but that's a mere flange and firmly
attached along the circumference. The Stanford torus uses a mirror
floating freely in space, not subject to any rotation whatsoever. But the
O'Neill design has these three rectangular mirrors, 30 kilometers long and
3 kilometers wide, deployed at a 45-degree angle from the rear. And it's
got to be a full 45-degrees, too, in order to reflect sunlight coming in a
straight line parallel to the cylinder in through the sides at a 90-degree
angle. (The angle of incidence has to equal the angle of reflection)
| | | | | | | | | | |
(| +--+--+--+--+--+--+--+--+ |)
As you can see, the sunward end of the mirror is 30 kilometers out -- it's
a right triangle, so both sides have to be of equal length. And that end
is spinning at the same 1/2 RPM as the colony to which it's attached, so
it's pulling upwards of six gees. And the only thing holding it in place
is the attachment base at the outer shadeward end and whatever
reinforcement can be woven into its necessarily thin length.
No, in the original O'Neill design, there was a ring of farming modules and
solar power stations around the forward end, to which the sunward end of
the mirrors could be anchored. O'Neill never explained how the crops could
stand up to the gee, treating the as if the either had the same gee rating
as the cylinder or were stationary and therefore in zero gee, neither of
which is a workable arrangement.
There's a better way, of course. Instead of one big honking mirror for
each skylight, how about eight mirrors, each covering an area 3.2
\ | |
\ | | |
\ | | | |
\ | | | | |
\ | | | | | |
\ | | | | | | |
| | | | | | | | | | |
(| +--+--+--+--+--+--+--+--+ |)
Nice, solid attachment -- but a lot harder to draw, which is probably why
we ended up with the three strip mirrors in the first place. You could
make these complete rings around the cylinder, too, and line the hull with
solar-electric panels. Or you could dispense with mirrors entirely and use
prisms, achieving the same "staircase" effect without extending so far out.
One of the CG outtakes from G-Saviour shows the interior of the
colony. It's noteworthy because it not only shows the usual alternating
ground and skylight panels, but also shows a series of pylons, in triads,
extending inward from the middle of the ground panels to a ring. Internal
bracing? Transportation system? A similar arrangement appears in Babylon
5, where the pylons rise up to meet the axial lighting fixture, which
doubles as a monorail line. The pylons presumably contain elevators and
similar elevators have been shown in Gundam, in both open (O'Neill) and
closed ("Minovsky") type colonies.
Well, whether it's Babylon 5 or Gundam, those elevators shouldn't run
straight from the hull to the axis. Riding them would be murder because of
the same Coriolis effect that produced the artificial "gravity" at the
hull. As the elevator "rises" from the hull to the axis, the passengers
are going to be pushed downspin at the same rate as they are inward, with
the result that the "floor" is going to feel as if it's been upended 45
degrees. The same applies going "down" from the axis to the hull, except
that the push is going to be upspin.
A body dropped from the axis to the hull would fall in a Nautilus-shell
helical spiral, appearing to travel in an upspin arc around the axis until
it finally impacted, not on the ground panel immediately below, but the
ground panel upspin from there. The fall would take about 5 minutes and 20
seconds and make one and one-third revolutions, with a terminal impact of
644 KPH (400 MPH).
Presuming that the elevator accelerates and decelerates at the same rate,
minus the sudden sharp stop going from axis to hull, travel time would be
the same as it is for a free fall, with the Coriolis effect converted into
lateral forces on the vertically restricted passengers.
That being the case, the best design for the elevator would be an upspin
spiral for the cars going from axis to hull and a downspin spiral for the
cars going from hull to axis. The cars would not run "vertically"
(perpendicular to the "ground"), but drive "parallel" to the hull the
Since the first illustrations in The High Frontier, the O'Neill cylinders
have been drawn as a Cartesian grid. That's because a Cartesian grid is
easy to plot and draw and makes for nice, neat wireframes. And it makes
mapping the streets of a city to the interior of the hull a lot simpler, too.
But it wouldn't be so good for the people living on those streets, again
due to the Coriolis effect. Driving in a straight line along the
circumference of the hull, one increases in weight going downspin and
decreases in weight going upspin. The net effect is that downspin seems
like uphill and upspin seems like downhill. Great for skating or skiing,
so long as you don't mind always going the same direction, but not so good
for walking and driving.
A geodesic grid, with straight paths going from end to end and at 60 degree
angles (instead of 90 degree angles) across would be a lot more
comfortable. Regional divisions could be mapped in hexagons instead of
squares. The sky panels would be honeycombs instead of gridirons.
And you game-players would all feel a lot more at home....
Gundam Mailing List Archives are available at http://gundam.aeug.org/
This archive was generated by hypermail 2.0b3 on Wed Apr 19 2000 - 13:27:03 JST