Space Stations II

Another quick space colony doodle

Here's a standard mirror/window illumination scheme that I think would work pretty well. Most designs have the windows on the outside of the shell, which can create many problems, as laid out below. In this design there's a large parabolic or spherical mirror facing the sun, which focuses the light onto a secondary convex mirror, just like a Celestron or Meade telescope. In essence, it's your basic Ritchey-Chretien type arrangement, just like we used on the Hubbell space telescope.

The light from the secondary, which is still slowing coming to a focus (the rays aren't parallel yet) encounters a large portal at the axis of the station. This is made up of lots of small windows, say hexagons roughly 1 meter in diameter, made of Pyrex, fused silica, or BK7 borosilicate optical glass. However, each window is forming part of a large Fresnel lens, to realign the rays to parallel, aiming directly along the axis of the station. One large lens couldn't substitute for the Fresnel lens, because the resulting required lens thickness would absorb all the light. Glass is clear, but a hundred feet of it certainly isn't.

As the light travels down the axis it will at some point encounter one of our 45 degree mirrors. These are actuallly all formed from a single aluminum right-angle cone that is sliced into rings and spread out along the much more accute cone that runs along the central axis. The cone segments are coated in mylar for greater reflectivity, and the construction of this final set of mirrors should be extremely simple. Down below you'd see something like a bright set of overhead lights, spaced several feet apart, but from a thousand feet or more away it would just seem to be a big bright light overhead.

The advantages of this design is that there are no fiber optics, light pipes, and only one small window section which is located on axis, leaving the entire floor available for use. The other end of the axis is compeletly free for use as a construction or docking port. The only thing I'd add to my drawing is a 45 degree mirror inserted into the optical path between the main mirrors and the stations axial window, so the station's axis can be aligned perpendicular to the sun. Since a rotating station is a giant gyroscope, aiming the axis directly at the sun only works for about a day or so. If you point the axis at say, Rigel, that's were it will remain pointing as the station continues to orbit the sun, round and round and round. By inserting an extra 45 degree mirror in the path, the axis need not aim at the sun, and can instead aim at the North Star or something, while the large non-rotating mirrors slowly track the sun throughout the year.

Problems with Windows

If a 1 meter square window shatters you've got a major leak. The flow at the sill will probably go sonic, in which case the air is getting sucked out at very high velocity (mach 1) and you're losing 55,000 pounds of air per minute. It's not a loss that can be cheaply replaced, and the worst part is that when you suck a fluid out of a vessel you create a vortex. An earth tornado has a pressure difference of around 1.4 psi, and this window failure has 10 times that amount of pressure differential. Things could get very exciting, depending on how the resulting vortex responds to the coriolis forces inside a rotating station. It might even bend over and start sucking things up. If it did, some big heavy objects are going come hurtling at high velocity toward the neighboring windows, probably converting the crisis into a catastrophe. This is not a very elegant failure mode.

Some early thoughts on the subject were based on the fact that it would actually take quite a long time for all the air to bleed out of the colony, leaving plenty of time for maintenance to fix the window. Putting myself into the shoes of our heroic maintenance man, I find myself holding a 44 lb piece of Pyrex or other such borosilicate glass, and O-ring, and some . I would stand their staring at an F7 tornado as cows go flying past me, and wonder how exactly I'm supposed to get the new window within 50 feet of the hole without having it ripped from my grip as we both go hurtling into space. Then I would think "Somebody missed this problem in the design review..."

The solution I came up with long ago was to have a heavy steel shutter, kind of like a submarine hatch, that could seal each window. The are hinged beside each window and stand vertically, much like the launch doors on an nuclear ballistic missile submarine. The door is merely balanced vertically, so if the window fails the resulting torrent of air will tip the shutter over and cause it to slam shut across the window. Since these shutters are all standing vertically, perpendicular to the light path, they really wouldn't be very noticable from the other side of the colony which is receiving the light from the windows. However, the window design just became a whole lot heavier and more complicated.