Some what unsatisfying
Seriously FEA analysis was good, I know a little bit about FEA, but the final conclusion was some what unsatisfying, "good enough?".
We know what we mostly already knew:
-If the fork does not dig into dirt/ground, the gear will not dig-in, thus no flipping. No dig, no flip. (makes sense)
-Pilot, skill, don't land hard, bounce or drive into a ditch. (makes sense)
-Tire, pressure, keep it inflated on the high side. Less air pressure less fork/ground clearance. (obvious thought, may be Van should redesign for a 5x5 tire and take the drag hit?)
-Surface, rough or soft surfaces with hills and valleys reduce gnd/fork clearance. (fair to say they implied more risk on soft fields than hard. So RV-A's are not back country Bush planes, we knew that. My input is TAXI slow on soft fields.)
-Nose, weight, keep it light, prop & engine. (the statistical data they quoted did conclusively show a direct correlation to me. analytically it was shown more nose gear load = less grd/fork clearance. makes sense but light prop/eng combos flipped as much or more, I suppose for other reasons?)
-Wheel fairing drag/tire catching is an insignificant factor, a minor contribution to reducing fork/gnd clearance. (I kind-a-sort-of disagree. I seem to recall a few cases, which although did not result in a full flip, it involved taxi on flat hard surface and clearly the fairing catching the tire causing the start of a full dig-in. Vans previous SB or RVator article warned about having sufficient tire/fairing gap.)
-New gear & folk seems like a goodness thing, providing more folk/ground clearance. We shall see. If or when the first RV-A flips with the new gear/fork, proper tire pressure & light prop and engine, taxing on smooth surface & flips, it will throw a monkey wrench into the works again.
FEA is good, but there are limitations. Who ever did the analysis seems to have a good basic understanding of FEA, but there is some user skill involved. The computer will pump out a solution, but it needs to verified with actual structural test, if only for small displacements. I would rather see a full scale test rig, testing a gear to destruction. You could vary gear load, tire press and surface. That would validate the computer model.