I wonder if this sort of thing is normal.
No, it is not! This “coffin corner” thing described above only applies to certain aeroplanes that are fast enough for transonic airflow (> Mach 0.8 something) and that can fly high enough for their stall speed to come very close to their maximum operating Mach number. Certain types of Learjets only have a speed range of +/- 10KT between stall and overspeed at their maximum cruising level. Most airliners don’t have a coffin corner because they can not fly high enough and most light jets are not fast enough. I would say that no civilian propeller aeroplane currently in production (piston or turbine) is affected.
Since Air France 447, high level stall recovery is included in our syllabus of the simulator recurrent training. We have to do a clean stall at max. operating FL. The aeroplane behaves exactly the same way as in lower levels, but the controls feel very “soft” and control response is slower than usual. Altitude loss during recovery is higher, but well within 1000ft. If you don’t recognise the stalled condition and let high (or low) angles of pitch develop, it can take some time to come back to normal pitch. Maybe this is what happened to the CitationJet described in Neil’s posting above. But it does not really explain what the TB21 pilot experienced.
was just amused at the thought of a TB21 experiencing transonic flow.
When Peter gets his AoA sensor watch out. The first TB20 to have to obey the 250ktspeed restriction below 10,000ft. Windows shattered all over Sussex due to the shock waves. 
The 737, the most widely used airliner, apparently has the capability of operating very close to the corner, to name one. At least, that’s what I’ve been told.
The TB21 not so much. I guess my point was that the TB doesn’t necessarily have to exhibit the exact same behavior at F250 as it would at 2500ft, but thinking about it some more it probably should. There must be additional parameters involved in this stall behavior.
It’s also possible this guy was actually doing something else and messed it up.
It would not be the first time that was done.
I do not think a TB is capable of passing through the stall inverted backwards in pitch, because the first thing that happens in a stall is that the nose drops down. It should not rise up unless perhaps the plane is massively wrongly loaded with an aft CofG.
A poorly executed stall turn – not suggesting this is the case – may result in the aircraft tipping back – but you need around 3 1/2 G to get into this situation in the first place.
An over-enthusiastic pull up to create a power on stall could get you into a tail slide, which could certainly get you inverted (like the hung loop mentioned earlier).
Question: Stalling speed is expressed in relation to IAS, but is control surface loading related to TAS?
When Peter gets his AoA sensor watch out. The first TB20 to have to obey the 250ktspeed restriction below 10,000ft. Windows shattered all over Sussex due to the shock waves
That reminds me. Our aircraft insurance tells us we’re not covered for any damage we do while generating a sonic boom. I think if the Auster is generating a sonic boom, insurance is the least of my worries. (Our insurance also excludes damage caused by nuclear war…)
Just like “what next” I have tried a stall at FL430 in a simulator. My last sim training was just after the CJ2 incident and I wanted to see how it looked in various configurations, so we tried it with power and without. It is strange as the controls are rather ineffective and speed build up on recovery takes a long time to happen, almost slow motion in fact.
Maybe the simulator is inaccurate at that level, although I was assured it was representative, and clearly it’s different when you aren’t expecting it. Might some people might be slow to recognise a stall at that level?
Ted.P
I found this link which has a good explanation why flutter is linked to TAS and not IAS.
http://www.australianflying.com.au/news/vne-and-flutter-explained
