Using a Microfiche reader in reverse to print Computer Generated Holograms

Background:

Computers can be used to calculate the actual exposure pattern of a hologram mathematically, without ever having used a real object or laser exposure.  There are various reasons for wanting to do this.  The patterns would look sort of like fingerprint swirls, if you could see them, but they are usually so tiny you would typically need a special microscope.

The problem (among many others!):

Although you can calculate this pattern of fringes with math, actually "printing" them to film (so you can actually view it) is darn near impossible.  Even a high resolution laser printer is still a thousand times too coarse to get a good exposure.

People DO use laser printers.  The holograms they get are viewed at a great distance, and only barely refract light.  So it's good for a proof of concept, but not for the "cool" holograms people dream of.

Microfiche readers to the rescue?

Normally a microfiche reader (or microfilm reader) takes a tiny square of film and magnifies it, and then projects it onto a screen.  The magnification is impressive (TODO: find my notes)

But my idea had been to use the reader in REVERSE; shoving light through in the other direction, having it greatly reduced, and then exposing a piece of film where the microfiche would normally be.  Then the film is moved to the next tiny "frame", and a new image is exposed.  Repeated thousands of times, the entire fringe pattern would be built up.

Steps:

1. A dark lab, of course
2. The screen of the microfiche reader is removed; it's usually translucent  plastic.
3. A computer CRT (or LCD) is placed against where the plastic screen had been, facing INTO the unit.
4. The Microfiche is replaced by a piece of unexposed film.
5. A small segment of the holographic fringe pattern is calculated and displayed on the CRT/LCD screen.
6. That image ENTERS where the old viewscreen was, travels backwards through the machine, and is focused on a tiny spot of unexposed film; the image is greatly reduced!  (and therefore of higher resolution)
7. The computer display screen is then blanked (set to black).
8. The Microfiche platform is moved to the next frame of unexposed film.  Presumably this would be under computer control.
9. The next tiny segment of the hologram is calculated and displayed onto the screen.
10. These steps are repeated over and over, until the entire film has been exposed.

Some optimizations:

• A single color would be displayed on the CRT.   Since blue has smaller wavelengths, perhaps it would produce a tighter image.
• If the hologram were buffered, it could be scrolled across the screen to match the microfiche platform's continuous motion, so that an entire strip of the film could be printed with the constant starting and stopping.
• Comparing a computer screen to a laser printer, you've got perhaps 72dpi vs. 1200dpi, so you are losing some resolution up font.  If laser printouts (on transparency material) were used, instead of a screen, you would not pay that initial penalty.  However, this would not be as easy to automate.
• Possibly a laser printer could be taken apart and its high res system could feed directly into where the view screen would be - basically "printing" onto some type of refraction screen.

A lamer idea might be to just print out all the squares of the hologram on paper, tape them up on a wall, and photograph them in blue light.  This would allow for substantial photographic reduction, though would incur substantial labor, and I'm not sure camera optics are that good.

I would be curious to hear if anybody has heard of this being actually attempted.