i printed some transparency gratings on the laser printer. Looking at them with a magnifier, I found that the 600/1200 dpi (according to specs) printer at work could really only achieve about 4 lines/mm.Lower than this, the lines are made up of circular dots so are too rough.
I also got a few printed on a big machine at the local print shop. A little better resolution.
However the main problem is that the transparencies are not very optically good (at least after being exposed to the heat of a laser printer), and seem a little...cloudy is the best way to describe it. Putting them on the front of my 200mmf6 newtonian, I can see spectra, but the zero order image and spectrum clarity is poor, presumably due to the transparency not being very good.
I did find an old paper by Arthur Vaughan made full aperture gratings (called amplitude gratings) for the Palomar 48 inch schmidt camera to make spectra of the very wide field survey pictures which this instrument could make (back in the days of glass plates bent onto a curved film holder (wince!). They used the thinnest commercially available grades of mylar film, and ruled the grating with a finely lapped fountain pen. Tedious I guess, but it worked apparently. I guess the problem I had with overhead transparencies might have been related to them being much thicker, and then varying in thickness even more after exposure to the heat of a laser printer.
http://adsabs.harvard.edu/full/1970PASP...82.1133V
The paper has curves for the theoretical grating efficiency at different diffraction orders, which are a function of the amount of black / clear space per ruling.