Re: Practically achievable beam divergence for white, non-coherent light

From: Ian Stirling (root_at_mauve.demon.co.uk)
Date: 02/26/05


Date: 26 Feb 2005 22:11:09 GMT

Joe D. <joe@nospam.invalid> wrote:
> I have two questions about beam divergence for white, non-coherent light:
>
> (1) What is the lowest practically achievable beam divergence for a
> searchlight. The GE WWII searchlights had about 2 degree divergence. Can you
> build a searchlight with, say, 0.5 degree beam divergence? What are the
> governing principles that bound this?

Fundamental diffraction limit:
1.22 * wavelength/diameter.

This is more-or-less irrelevant, it only means that you cannot have a
1/100th radian searchlight beam that's under about 122* lightwavelength, or
about 60um.
In practice, you'll never get a significant amount of light through a hole
that diameter (barring the use of multicoloured lasers).

> (2) Can you design a Cassegrain optic system where the beam off the
> secondary mirror has near-zero divergence? IOW instead of the secondary
> having a fixed focal length, it has infinite focal length? So that at, say,
> 400 meters the beam width off the secondary is still the diameter of the
> secondary?

There is nothing magic about a Cassegrain (or any other system).
You end up focussing one point on the light source to one angle of output
from the main mirror/optic.
The mirrors are locally flat, and if you reverse the direction of the bounced
ray, you can consider what would happen if you replace a small portion of the
mirror with a pinhole.
You get an image projected of the light source.
Then, it gets into how large a mirror you have, compared to the source.
A mirror 1m away from a .5cm light source will have a divergance of 1/200
radians, or about 2m at 400m.
The bigger the mirror, the more light it catches from the bulb and reflects.
This rapidly leads to wanting bigger and bigger mirrors.

R/G/B lasers are obtainable, that are mixed into one output beam, and
will not diverge much over 400m, with a beam over a few mm.
Expensive though.



Relevant Pages

  • Thrity Years From Now
    ... The 'plug in the wall' could be any number of energy creating tech- ... ped with laser energy outputs of 10**4 joules, ... provide initial damping for the lense, mirror housings, pulse ... The reason for beam shaping is to compensate for spatial gain ...
    (sci.space.policy)
  • Re: Wang / Sagnac Devices
    ... mirror image of the source. ... to the reflection issue, if there is no relative motion between source ... the angle of incidence and the angle of ... Now we'll follow the clockwise beam from start to finish and numerate ...
    (sci.physics.relativity)
  • Re: Beam Expansion Question
    ... The idea is to acheive a collimated beam from a collimating mirror. ... Thus with a paraboloid there are 5 adjustments ...
    (sci.optics)
  • Re: A modification of Fizeaus toothed-wheel experiment
    ... So you did not get an answer to your question from Gisse. ... A light beam ... |> The mirror is a fast-rotating [~1400 rpm, but lab notes are gone so I ... | a laser I've not tested it. ...
    (sci.physics.relativity)
  • Re: A modification of Fizeaus toothed-wheel experiment
    ... A light beam ... > mirror is on the circular flat side. ... a laser I've not tested it. ... and the light speeds differ ...
    (sci.physics.relativity)