J. Nieuwenhuize wrote:
Papa3 schreef:
Not quite along the quantitative lines we've been following, but it is
pretty amazing to see how little the average pilot pays attention to cg
and its effect on performance in their common flight attitudes. Ask
some of the stockier pilots in your club to "trim for 55kts" (or some
similar, reasonable speed) and leave the trim there. On landing,
take a look at the elevator. I would wager at least one in two is at
or near full up deflection. Assuming this puts them at the outside
edge of the Cl ranges discussed, that's an awful lot of downforce
being produced (1/2RhoV2ClS IIRC). Aside from the "negative lift",
what's the typical induced drag that goes along with this? I'm
assuming it's pretty high given the relatively low aspect ratio,
especially of older models...

P3

Not quite correct actually; induced drag is proportional to the squared
lift coefficient and inversely proportional to the aspect ratio.
Knowing that the lift coefficient of your stabilizer is always lower
(main wing stalls first) induced drag is fairly low and certainly lower
than the weight penalty of a heavier tail. Also bear in mind that while
thermalling a glider you're flying at a relatively moderate angle of
attack, not at stall speed. (At the Discus for example you're flying
about 30% above stall speed in a thermal) This is different in
landing...

Nevertheless I usually fly at the back end of the cg-range; mainly
because of the difficulty to achieve "natural" ballast ;-)

Aha! Now things make much more sense!

Being one of those "stockier" types I find a fairly different
experience in the 304C that some of us fly. For me, full back trim
results in about 50 kts (nominal landing speed), and thermalling beyond
about 30 degrees of bank seems to massively increase the sink rate.
However, in an L33 full back trim flies about 5 kts slower and it loves
steep banks with me. Other pilots in the 304 (at the rearward end of
the CG range) report performance much more like I get with the L33. It
must be that the elevator design is different...