Interesting indeed.
If you look the device on DG:s video you notice how flat it is and it looks like the laser module flat inside and mirrot is used to skew the angle.
I found the patent on that finder....link somewhere and it looks a little different. Interestin g reading though.
Another titbit is the published drawin/writeup that is constructed nicely, looks like for a public domain design or small scale fabrication. You need a google account to download it. Laser module is on this design mounted on a sheet metal cradle that can be pivoted a little. This is hight in height, battery and laser are upward. I think that the laser linked by DG himself relates to this design.
I notices the same than you: Laser beam is almost a line as non collimated laser is, but this has "over colliamation" only with the lense, no pinhole on my model, therefore the "line" or ovality is turned 90 decree after the focal point, no biggie needs just more optics, which I have a coarse feel, but not really that good indepth knowledge. But monkey-see-monkey-do is probably here enough.
I like your pictures, it clearly shows a very nice small dot on the center and then big blob of difraction. Try matte black object and observe how amazingly smaller the dot comes. The difraction/interference is very easy to notice with eyes, but harder to capture on camera.
Now the interesting part: The "dot" has it worst on shiny flat surface. Very sharp edge is ok, but ragged edge perpenticular to beam is bad. Piece that has a side, close to beam shows fairly good resolution - just when most of the beam is blocked by the face of the object and a smal portion of the beam is cast to side.
One factor is a work piece, this laser method is very particular about the part surface, edge and angles, but for some parts this looks very good solution. This is something I want to verify before building the all of it.
Regarding the laser, the quality of it might or might not have major impact on accuracy:
* Shop light and more rpm might be eneough to reduce glare
* DG maintains that ultra small and precise dot is not required and pretty sure he is right on this one
* there are some fun details to find out
* laser he bought might be different, we are just buying here on very vague description
I'm very interested on filter you used on your experiment, it really seems to reduce glare and perceived dot size. Can you give some more information about that filter material?
Somehing like that looks nice, but the price....Variable Reflective Neutral Density Filters:
http://www.optosigma.com/products/optical-components/filters-apertures/variable-reflective-neutral-density-filtersI have some other things to finish today, but might have time to do few more experiments on tonight or very late tomorow.
Are you going to use the unit as it is or are you going to use some laser dot conditioning?
Do you have opinion about the slant angle? I found out that my mill has limited vertical space/travel, I probably need aproximately 10/20 degree two laser modules or adjustable angle. DG makes several remarks about the need for adjustable angle, even on latest design he suggests tilting the whole unit. It must be more than only for convenience.
When I powered up this module first time, my intial reaction was that no way this can be used as described, but after trying it out for real objects (pipes, bosses, bearings), some of my faith was restored.
Pekka