Going back over my notes and pictures, I did not fully explore H4895 like I should have. I was using the Hodgdon 140 grain data before I had found out that I can use the 139 grain data.
Whats significant about that is the 140 grain data tops out just over the start data for the 139. So I could have gone a lot further in my tests, plus the top load(40.5) did show some promise.
