Finally, a question about something I've done. The start up for dye penetrant is just buying the dyes, solvent, and developer. I dont know if its necessary, but suppliers sell all three, so it seems reasonable to use the solvent, developer, and dye all from the same source. Search the web for basic technique and supplies, though T Blades technique is good. The dwell time can vary a little, and the time allowed for the developer to do its work can too. Magnaflux supplies some of the stuff we use at work for magnetic particle and penetrant testing. Penetrant is only for surface flaws. You can get flourescent dyes for penetrant testing, but without a black light they arent worth much. If you have a black light, you can see cracks from across the room, if its dark. Its pretty cheap, compared to other testing methods, to get started and works well if you have some experience. I think Bob Engnath used this method to test his japanese style blades. His old website says he used a crack detector spray. I assume it was penetrant testing. For an information overload, check out the American Society of Non-destructive Testing (ASNT). This seems a good way to test blades for macroscopic cracks.