Artificial lighting, huh?

On Feb 11, 2008, at 1:57 PM, sailor.barsoom wrote:

> LED's can be produced by a modified ink jet printer. I'm not
> surprised at all; those things seem to be able to do anything.
>
I agree, but it's not usually the most efficient way to mass-produce
anything.
> At this point I'd like to say that I used to post here as Xenophile
> and that I am prepared to utterly reject my earlier advocacy of
> natural sunlight piped into habitats via fiber optics. I'm sure that
> FO will have many wonderful uses, but I no longer think that they are
> the best option for lighting up the habitat.
>
Super, welcome aboard! :)
> I'd like to propose LED lighting, and not just any LED lighting.
> Remember my earlier message in which I linked to the Fremont Street
> Experience in Las Vegas?
>
No, but thank you for the link. I've long pondered similar displays
to produce an artificial sky for a space habitat, but I didn't know
that anything of this scale had ever been made. Should've known to
look to Las Vegas for it!
> I suggested that, while three LED's, each with its own wavelength and
> each capable of 256 levels of brightness (total 768 levels for the
> pixel), is enough to display 16.7 million colors to the human eye, we
> would be using our version for more than seeing and pictures. So I
> suggested twenty distinct wavelengths instead of three. It was
> suggested that even that might not be enough, though most felt that it
> was plenty.
>
Hah! 20 wavelengths sounds like plenty to me, but even better might
be to make use of LEDs with a broader spectrum. As long as the total
combined spectrum when the display is showing "white" is fairly
continuous, I'd think that would be good enough even for nitpicky
artists and photographers.
> So I'm going to call that good enough: each pixel is a half millimetre
> by a half millimetre, contains 100 LED's, each at a different
> wavelength and each capable of 256 different brightnesses, and is
> covered with frosted glass or plastic. It takes four pixels to cover
> a square millimetre, four hundred to cover a square centimetre, and
> four MILLION pixels (and thus 400 million LED's) to cover a square
> metre.
>
Seems excessive to me. One LED per mm^2 would be fine, I would
think, when viewed from 100 m away.
> About now, you are probably thinking, "Are you mad?" No, I'm feeling
> quite peaceful. Remember, these LED's as well as the ungodly amount
> of circuitry needed to support them can be printed, even today (yeah,
> it would be an unworkably huge project today, but the basic tech is
> here). Have a robot with the printer in it's belly crawl along the
> roof and simply deposit it all in place.
>
That is really highly unlikely to be cheaper than having them mass-
produced in a display-panel factory, and then stuck in place on the
roof.
> Now remember, the pixels for the FSE are 400 times the size of these
> (and round), only use three actual separate wavelengths, and are
> typically viewed from 27 metres away. That is enough to create the
> fantastic light shows that leave millions of people enthralled. Our
> pixels are tiny by comparison, square (and thus can be packed together
> without so much blank space between), and will typically be viewed
> from as little as 20 metres and as much as 75 metres away. This
> should be perfectly adequate for all our lighting, display,
> photosynthetic, and other interior light needs.
>
I tend to agree. During the day, have it display blue sky and (if
the inhabitants so wish) fluffy white clouds. At night, show
whatever's actually outside the habitat, in a live view -- but
despun, so that users don't get sick watching the sky swing by every
minute. :)

Best,
- Joe