ORIGINAL: Rotaryphile
Wobblewobble mentioned having crashed because of a boost tab. I had my rudder servo linkage disconnect once, and the rudder went hard over - more than 45 degrees, with a very large 3-D type rudder. I had to fly the airplane in knife edge to a gentle upwind crash landing that slightly damaged a wingtip. I had been experimenting with the travel of the passive boost tab, where the servo controlled the rudder directly, with the tab assisting.
It turned out that the tab was too effective, causing the rudder to slam hard over when the resistance of the servo was removed. I simply reduced the travel of the boost tab, and everything was OK. The rudder had been a bit hard to trim, because the tab was forcing the rudder to uncommanded movement within the slight slop in the control linkage.
Interesting experience and good that you saved the model.
But one question. How do you know your fix actually was better, or the change actually did anything at all? Did the rudder servo linkage fail after you reduced the travel? If you think about it, removing our model servos from the system removes the absolutely rigid, constant lock the servo has on rudder position. As long as a servo has a solid grip on model surfaces, we really wouldn't know how much boost the tab gives. OK, unless we put a recorder into the servo wires and checked the amps drawn before and after.
Did your "3-D type rudder" have significant area forward the hinge line?