Religious experience

[James Lehman]

There are many cheap laser pointers that come with exchangeable end caps containing holographic diffraction grating beam-forming optics.

This one, for example:

http://www.ubid.com/Mr._Laser_Inc._5...-c1409-s6.html

Do you think this could be done with RGB to make a full color image? That is what I was asking in the first place. This link is a good example of a single color.

James.

[heroic]

Do you think this could be done with RGB to make a full color image? That is what I was asking in the first place. This link is a good example of a single color.

James.

I suspect so, but it would be a computationally difficult problem designing the stamper (these holographic optics are pressed like CDs are ) and I don't think anyone's really tried. Those little beam forming endcaps work because they're close enough to the diode that the light they receive is still coherent (actually, the DI lasers you get in red pointers make pretty good holograms- they have a reasonably long coherence length! You just need to clean the beam up a bit.)

Of course the problem does kinda reduce to a full colour white light hologram of a two-dimensional object... so of course that's possible. I wonder if anyone's actually tried? Anyone know?

[James Lehman]

I suspect so, but it would be a computationally difficult problem designing the stamper (these holographic optics are pressed like CDs are ) and I don't think anyone's really tried. Those little beam forming endcaps work because they're close enough to the diode that the light they receive is still coherent (actually, the DI lasers you get in red pointers make pretty good holograms- they have a reasonably long coherence length! You just need to clean the beam up a bit.)

Of course the problem does kinda reduce to a full colour white light hologram of a two-dimensional object... so of course that's possible. I wonder if anyone's actually tried? Anyone know?

Do you have any idea how these things work? It seems like pure magic to me! If you shoot the beam through any part of the film, you get the same image. I bet for some reason of wavelengths bending at different angles, it would not work for more than one color at a time. I'd be just as happy to be proved wrong though.

James.

[heroic]

Do you have any idea how these things work? It seems like pure magic to me! If you shoot the beam through any part of the film, you get the same image.

That's why it's called holography :-)

In a hologram, every part of the film is exposed to waves from every part of the object, so in principle the entire image can be reproduced even from just one tiny piece. Of course, this doesn't actually work because real materials have resolution limits (film grain, minimum featuresize on etching processes, etc., you know the kind of thing).

The way it actually works is, a hologram is not really a record of the image, but rather of an interference pattern any part of which can be used to reconstruct the image. True holograms are almost impossibly cool; the white light reflection holograms you see everywhere are not quite as awesome (they use a slit optic to reduce the information, so that they can be easily reconstructed by white light).

I'm guessing that, given known wavelengths you could write a program that would generate an interference pattern that would reconstruct to anything you liked... but I'm not sure it'd work for unknown phase relationships. Maybe with pulsed lasers?

[Jem]

True holograms are almost impossibly cool...

Ahhhh... Words that are music to my ears

Jem

[tocket]

Ah, now I understand what James was thinking about.

I'm guessing that, given known wavelengths you could write a program that would generate an interference pattern that would reconstruct to anything you liked... but I'm not sure it'd work for unknown phase relationships. Maybe with pulsed lasers?

Generating these patterns is probably not going to be a (computationally) easy task. This is a bit too much of physics for me to be comfortable with, but us chemists make use of diffraction for one thing; x-ray crystallography. Using this technique you obtain (x-ray) diffraction patterns from crystals and use these to determine its structure. It typically takes a few hours worth of calculations to determine a structure from a simple diffraction pattern. Yet, people do this with proteins, having diffraction patterns like this:
http://www.einstein.cclrc.ac.uk/Acti...Pea.Lectin.jpg

I feel the method should be similar to what you want to do, as you are going from a known diffraction pattern to an unknown physical structure. However, the structures in x-ray crystallography are 3D and I guess if you made your own interference patterns you would want them to be 2D for practical purposes. That might make the calculations easier.

Something cool for the future of laser projectors might be a digital diffraction cube to replace our mechanical scanners. Would take quite a beast of an imaging processor though.

[Alsone]

James there's a googd example of projection holograms in the Ministry Video here (1/2 way through):

http://www.youtube.com/watch?v=uSmpGxIQcPk

So far as I know, its a waterscreen, with projector and laser scanning to add additional effect. Arguably, cheating perhaps but very effective.

(Ministry are 1 of my favourite laser show companies - amazing displays).

Before someone asks, I can't comment on MPE in relation to what they do but I've never heard of anyone being hurt and they are a very respectable company so I'd expect everything to be done safely.

[mixedgas]

james, I have the source code to do what you want, ie takes a two dimensional image, and you then laser print a difrraction grating on transparency. Its called fourier transform imaging. Are you interested? It needs compiled, and I dont have a compiler.

thats what I used to do for a year for UA, creat custom gratings using a big honking UV argon with a etalon.
Steve Roberts

[mixedgas]

here goes

laser printer gratings

google holography without phtography:

http://www.icecube.wisc.edu/~shiu/PH...s/holoexpt.pdf

www-atoms.physics.wisc.edu/papers/holo.pdf

medcosm:

http://corticalcafe.com/prog_CGHmaker.htm

printable gratings:


steve

[heroic]

Something cool for the future of laser projectors might be a digital diffraction cube to replace our mechanical scanners. Would take quite a beast of an imaging processor though.

You'd need a micromirror (or similar) device with pixels with a size on the order of the wavelength of the light you were working with. This is possible (and, indeed, done) with IR, but TI hasn't quite shrunk the DMDs down small enough to work at visible wavelengths.

I used to work one of the guys who worked on the original TI DMD, and digital holography was one of their goals, but I daresay fabrication gets harder as you get below 1 um in size. Maybe it'd work better with silicon mirrors or something- I know that TI continues to develop smaller and smaller DMDs for use in their DLP range of products.