I applaud the sensible application of physics that has been used here. It's been a few years since my last physics class, but from what I can remember the math looks sound. We have to remember (and TDC20 recognized this himself) that we are making a lot of assumptions when trying to deal with this problem.
That said, how do we explain the extreme velocity spread that many have reported when shooting high powered ammo with the stock spring that is significantly reduced when switching to an aftermarket 22+ pound spring?
If it is true that running an aftermarket spring greatly reduces velocity spreads when shooting high powered ammo, then we could reasonably infer that the barrel is staying locked up for a longer period of time before unlocking.
That would then lead us to ask how this could happen, in spite of the mathematical evidence we have been shown to the contrary? I would postulate that either: (1) there are significant variables that we are not accounting for; or (2) there are variables that have not been correctly represented.
A big problem that I can see is that while we're talking about a 17lb stock spring vs. 22+ lb. aftermarket spring we do not know all of the important specifications of these springs. Are these the true spring weights? Are there overall length differences between these springs? What is the actual amount of force being placed on the slide while at rest (pre-load)?
I don't have an answer these questions, but I do think that in light of some evidence we have that contradicts our current mathematical findings maybe we need to take a deeper look before concluding that different springs do not significantly alter lockup time/characteristics.