A proper safelight for a given paper is one whose transmitted wavelengths are visible to the human eye, but invisible to the paper when exposed to them for the time it takes to process that paper. Whether that safelight is a purpose-built commercial device or DIY is pretty much irrelevent. The light is either safe for a given paper for a given period, or it isn't.
As mentioned earlier in post #27 (before this thread unfortunately went horribly off track), try performing the simple CD/DVD prism test with any safelight you are using. It's quick, it's easy, and it's cheap. I've found it to be a reliable indicator that you only need to look at to see if your safelight is transmitting any grossly non-safe wavelengths.
In other words, no matter what, you know that blues and greens are bad. Oranges that may fall slightly outside of a b&w paper's sensitized range are more problematic. But at least you can visually eliminate the obvious low-hanging fruit with relative ease before following up with a proper pre-fogged safelight test.