I'm working on getting a glass grating for more accurate results. There will eventually be a
custom stand for the grating and DUT as well. I got a few Al bars and misc. pieces to put
something together with the other day.
The grating is positioned a set distance from a white ruler. (Usually 100mm) The pattern
is then photographed.
Here is the pattern from a 940nm diode:
660nm:
Then the distance from the center to the first order line is plugged into this:
The results are accurate within ±5nm or so. No lens or slit is needed for a collimated laser
beam source since it is already narrow and has its own lens. I'm still trying to figure out
how to get a good spectrum through a pair of LSGs. bloompyle's PCM would work great
for that.
That's a grating and a formula - not a monochrometer. I did a write-up on that method a couple years back. That's quite a complex formula as well. You can simplify it to:
λ = dsin(arctan(S/L))
Where d is the distance between the slits (10^-6 meters for that grating), S is the distance between the diffraction orders (in meters), L is the length to the screen (in meters), and λ is the wavelength (in meters)
I'm not 100% sure this is safe (please confirm before trying), but have you tried shining the laser at a white paper in front of the entrance slit and looking directly into the exit slit?