A problem we continue to see at high altitude airports is low altitude
visiting pilots not accounting for the higher true airspeed. As they get
into the flare they feel they're flying too fast (TAS) and slow down too
much (IAS). We see some hard landings.


wrote in message
...
On Friday, January 31, 2014 7:52:24 AM UTC-8, JJ Sinclair wrote:
On Friday, January 31, 2014 6:57:01 AM UTC-8, wrote:

Additional compounding illusion that most of us know from theory and
instruction but may forget when flying instinctively under stress:

Above a certain altitude when you make a coordinated turn with a given bank
angle the inside wingtip traces a circle in the opposite direction vs the
ground. In a low turn to final the wingtip goes the same way around versus
the ground leaving a subtle sensation that the turn lacks enough rudder.

So you are too slow from a high horizon and pulling the nose up to the back
side of the polar in a misguided effort to stretch your glide, while at the
same time you have a tendency to under-bank to keep the wing away from the
ground and over-rudder due to the wingtip motion illusion.

All these illusions nudge you in the wrong direction at the wrong time.

On a related topic - I installed a new flight computer/vario last year (an
LX 9000). Like a lot of current generation instruments it has accelerometers
built into it. I believe the primary purpose for these is to add some
additional capability to separate horizontal gusts from vertical lift (that
surge in the seat of your pants that we all know and love). They seem to
still be working out exactly how to do this in the software.

The other think they appear to be doing is using some combination of
airspeed, g-load and pilot input wing loading to estimate angle of attack
and provide a stall warning. I've only done a little testing with it but it
appears to be pretty reliable.

Given that stall/spin has been the cause of 39% of all fatal glider
accidents and 36% of fatalities over the past 20 years a decent warning
system might help. Has anyone done any serious testing of the accuracy of
these systems under the kind of scenarios in this thread?

9B