My guess would 15-20 seconds. Any B737 pilots here? You'd think you would at least have a little fun and loop the thing before checking out, like that guy that stole the Q400 from Seatac. Not sure he did loop it, but he certainly did some semi-aerobatic stuff with it.
IIRC, TWA actually used that as a possible excuse when one of their planes went down back in the 90’s
With trim rolled full down it could be held in a vertical dive, recently demonstrated by the 737 MAXes. I got a hunch the rumor is based on data from a flight systems recorder that shows the stick was pushed forward and THEN the trim adjusted, which precludes the possible suspect of electric trim-runaway condition. That does happen every once in awhile, the fix is pull the breaker and use the manual trim wheel to re-adjust. This is however one of the most-beloved gremlins for the guys conducting simulator training to spring on their students this condition has killed (AFAIK) nobody yet though.
I did the math with SOHCAHTOA and the published 30,000 fpm descent yesterday when @Raz'rtreated the vertical speed as the composite velocity, I came up with about 380 knots composite minimum, potentially faster than 400 knots, from the published velocity profile that I looked at online.
You'd be toast within seconds. If the 737 was flying at normal cruise speed (0.78 mach) and you initiated a roll, pulled hard towards the ground, you'd be at the 737's +2.5g limit pretty quick. In a normal aerobatic loop you pull about 3.5g - 4.0g, and that's with the plane decelerating as it comes over the top of the maneuver. If the 737 was pointed straight at the ground doing 0.78m you would most certainly exceed the Vne. However, exceeding Vne doesn't necessarily mean the plane instantly disintegrates. Being at Vne and then deflecting a control suddenly is an entirely different story. Don't ask me how I know this.
It's a pretty good story, but I'll wait for the Stupid Shit That Should Have Killed Me Anarchy thread for that one. It had Darwin Award potential.
