I have wanted to give my 120mm f/7.5 more of a chance with afocal NV. Diagonal goes along with that. I have three 2" dielectric diagonals, and one 2" 45 degree prism with a smaller clear aperture of 36mm. I'd prefer the full 46mm. I was concerned the dielectrics would do poorly unfiltered. It seems to be fine with my set of diagonals so far.
I compared the views with an without diagonal. I didn't notice any difference whatsoever, except scattering from the dusty diagonal. I was still getting the faintest stars and spillting the same doubles. I just carried on with the diagonal. Focusing without a diagonal is not always possible though.
To try and exclude dumb luck on targets, I just used half view through diagonal, comparing sky brightness. Same brightness. However, I was still looking at a different patch of sky through the diagonal and past the diagonal.
So I used the best broadband IR source I have - a turned OFF hot incandescent bulb's filament. For added safety, I also used the 642. I could certainly watch the filament dim as it cooled. These peak above 1 micron for ordinary color temperature bulbs, so the spectrum flattens in IR as it cools. The curly filament also has some structure. No difference in brightness really or any effect on structure.
It seems promising so far. Omega Centauri will be the final test I try before I'm sated. For good measure, I'll do tests 2 and 3 with my dob's diagonal. I have only been looking at brightness and resolution for visual use. Is there anything else I should be looking out for visually? What have been your experiences with dielectric diagonals? I'm trying to avoid poisoning the well with brands, but how do your unknown named dielectric diagonals do with these fairly simple tests?