HI Bob

That is what I was referring to.

The deformation limit for carbon designs with thin wings appears to be the point
at which it becomes impossible to maintain control movement.

As an example, there are various apocryphal tales of uncommanded airbrake
openings on open class aircraft with thin flexible wings. The Nimbus 4 appears
to be the most common suspect here.

So the deflection limit is not a "x degrees from rest", or a plastic deformation
(although there is a requirement for this in the regulations) but a deflection
beyond which the control actuators do not work correctly or have unacceptably
high resistance.

My point came from published discussions on the construction of the Eta, and the
DG1000 where both constructors commented that the ultimate strength of the
structure was well in excess of the limit load, and that the limit load was
imposed by the deflection of the wing.

There is an interesting test story at:

http://www.dg-flugzeugbau.de/bruchversuch-e.html

The destructive test requirement is that the wing must withstand 1.725* the
limit load for three seconds at a temperature of 54Celsius. The DG1000 wing
withstood this - and eventually failed at 1.95 times the design load limit. This
is one reason why I believe you would probably be able to get away with a brief
overstress load. I am not sure of the limits on older designs, but would expect
there to be less margin of strength.

As I understand it the modern thin section wings are flexible enough that the
load limit is imposed by control freedom limitation, and the wing must withstand
1.725 times this load in test. Flutter is the subject of speed limitation which
give speeds and margins that the designer/manufacturer must demonstrate flying
to. The regulations imply that the glider must be demonstrated safe at a minimum
of 23% margin above the placarded Vne. So your margins for flutter, versus
ultimate strength are 1.23 vs 1.725 in JAR22 (unless I got the math wrong)