Detecting a surface crack with dye penetrant
I bought some dye penetrant to test for cracks in my axles and landing gear legs during the May '04 annual (Update: It is now Mid-August and the plane is still "in annual"!).    Since I hadn't used it before, (and didn't have any idea what I was doing), I thought I'd practice on one of the doorpost cracks that I'd found during the annual.  It's a three-part system; cleaner, penetrant (dye), and developer.   Here are some pictures of the stuff in action, in case anyone is curious about what it looks like.  The techniques shown below may not be optimum, since it was the first time I'd done this, so if you're testing anything critical, find someone who knows what they're doing :-)
First, the suspected area is cleaned with a solvent to remove grease and dirt.
After the solvent dries, a dye is sprayed on the surface and allowed to soak for a few minutes. The dye penetrates cracks, and stays wet (doesn't dry like paint).
After the dye has soaked in for a few minutes, all visible dye is wiped off with a clean cloth, slightly moistened with the same solvent used for the initial cleaning. Not soaked; you don't want the solvent to penetrate the cracks where the dye is.
Finally, a light coating of developer is sprayed on. It reminds me of spray-on foot powder; it goes on as a liquid, but dries in a minute or so to a light powder coating, which absorbs the red dye, drawing it from the cracks so you can see it.
Here the developer is just starting to dry, and the crack is barely visible.
Here you can see the red dye coming out of the crack and spreading over the surface.
If the crack's location wasn't obvious before, it sure is now.
It's hard to believe all that dye came from that small, almost invisible crack.

Web page and images generated using
5/22/2004 23:45