Can you post your Part file? I rarely use derived sketches, so while I may not be able to help someone else will almost certainly be able to, and having the file instead of a description almost always helps, even if it's a good description (and yours is better than usual).
I use derived sketches often with the parent sketch controlled by design tables defining hundreds of part configurations. I have never had a problem with the derived sketches updating.
- remove all relations from the derived sketch(s)
- open and close the design table to update values
- switch configurations to see if the derived sketches fail
My guess is that they will update correctly which would mean that you need to position and orient them with a different relation technique. The pierce point is probably the culprit.
Also, why not use a helical sweep with one sketch instead of a loft with three sketches?
I didn't think to do a helical sweep first, at the time I was all juiced up to try the derived sketch deal, it sounded like something I can employ with some other projects I have going0--=, and when I checked the geometry with the 3D scanned geometry of the actual part, the modeled geometry is within the tolerance (.01mm) to the scanned geometry. Keep in mind the scanned geometry itself is only +/-.001". I will look into the helical sweep technique to see if I can recreate the same tolerance to the 3D scan as the loft command. I agree that the pierce point is the culprit, so that leads me to believe that in the future I should avoid this constraint or think of it as a almost constraint. It almost works but not quite.