FrothyChimp said:
hydrogenman15 said:
Because of diffraction, a smaller output aperture yields a greater divergence. HeNe tubes had apertures made from concave output-coupler mirrors that were, what, a couple of 16ths of an inch or so across? (Dave?) The optical cavity of a diode laser is basically a microscopic scratch on a semiconductor. This makes the light waves come out much like the waves in the 'double-slit' experiment - they 'wrap around' the edges of the slit. They do this in both directions, but since the scratch is much shorter than it is wide there is greater diffraction in the vertical direction. This contributes to the non-circular shape of a laser diode's spot... generally the diffraction in the vertical direction is about 3X that of the horizontal direction.
In a DPSS laser, the light from the diode is used to excite crystals that then emit the light. The size of this crystal is quite large compared to the aperture of the diode. This contributes to the smaller diffraction and therefore smaller divergence of the output beam. So no, a red diode does not generally have a better divergence than a DPSS laser. A red HeNe, on the other hand, will most likely have better divergence than either the diode or the DPSS laser.
(Side note: I think diffraction is just a facet of the Heisenberg uncertainty principle... it just seems like the harder we try to pin down one variable the more squirrelly the others get, any time we're anywhere near Maxwell's equations.)
Different colors diffract by different amounts. This is one of the reasons why the double-slit experiment needs monochromatic light: otherwise, the diffraction patterns of each wavelength would smear into each other, fuzzing out the results. (Refraction has this property as well... you could even define colors as being wavelengths that refract by a specific angle. This isn't voodoo, it's basically just solving Maxwell's equations for different variables).
The M[sup]2[/sup] value FrothyChimp speaks of is a measure of beam quality that doesn't depend on wavelength, so we can use it to compare lasers of different colors without getting hosed by the different diffraction angles. There's a value called the beam parameter product (BPP) that is just the divergence multiplied by the radius of the narrowest point of the beam (the beam waist). The lowest-possible BPP comes from a beam with a Gaussian profile (which you might have seen called TEM[sub]00[/sub]). M[sup]2[/sup] is just the ratio of your beam to that ideal TEM[sub]00[/sub] beam, i.e. you can focus your laser to a spot whose area is M[sup]2[/sup] times as large as that of a similar TEM[sub]00[/sub] beam. (So smaller numbers are better, and you can't get better than M[sup]2[/sup]=1.)
================
On a semi-related note: look at a prism. (If you don't have a prism, the cover of Pink Floyd's Dark Side of the Moon will do.) (If you don't have Pink Floyd's Dark Side of the Moon, may the gods have mercy on your soul.)
OK... note that red 'bends' less than blue. All the colors in between bend by proportionate amounts, 'smearing' white light into a rainbow. This is why, for instance, you might have seen the discussions here about the shorter focal-length violet diodes have relative to red wavelengths through the same lens. (There's been a lot of discussion about this in the thread about 405nm AR-coated lenses, particularly revolving around the need to screw the lens-holder deeper in to the module - blue bends so much more strongly than red that not only does the lens need to be much farther in, the diode itself needs to be pressed into the module farther than 'flush'.) The early PS3 diodes from the KES-400AAA sled had a single 5-pin diode with red, IR, violet, and a photodiode all in the same can. Since these all refract differently, they all have different focal lengths for any specific lens. To compensate for this, there was a lens on a little motor-controlled gantry that slid back and forth to adjust the focus depending on which wavelength was in use at the time. Later iterations have taken 2 forms: completely separate paths through the sled for red/IR and violet (the head actually has 2 lenses on it), or partially-shared paths (where each color takes a different path, but they all pass though a single lens on the head.)
I almost prefer Wish you were here, The Wall or Relics though, but there is no denying the greatness of the dark side! Nice so see a fellow floyd fan!
At first it looked like a normal dead diode then the interference pattern and then the pattern started to warp out. 
