http://ozreport.com/forum/viewtopic.php?t=35640&start=27

Steve Morris writes:

Al Bowers’ discussion on how birds fly without vertical surfaces
has been popping up on a lot of forums I read and the re-kindled interest in the
Horten style flying wing design has been a surprise to me. I’ve never been a big
fan of the “potato chip” wing and have avoided their approach in all of my
tailless designs for what I feel are sound reasons involving the trade-offs
between performance, handling, and safety. I admire the effort the Hortens put
into their work but their expected performance gains over a wing with vertical
surfaces has never been realized and in many cases people have been injured
trying to prove their point. In the modern era of fly-by-wire aircraft all
design concepts should be revisited because computer stabilization may enable a
formerly “bad” idea to have a new life. For example, the B-2 has proven that the
all-wing configuration can been made to fly well enough for use as a bomber when
low radar cross section is a design goal. Unfortunately, the world of hang
gliders and sailplanes has yet to adopt fly-by-wire computer stabilization and
the same design issues that the Hortens faced still remain for these aircraft
types.

Before I get any more negative on the whole subject, I would like to see Al’s
paper which I heard is to be published soon. I have yet to see a time history of
the aircraft’s dynamic motion while performing a roll maneuver into a steady
turn. Without seeing the actual sideslip and roll rate response due to aileron
commands it’s impossible to gauge the level of his success in improving the
handling qualities of a flying wing. I do believe that high aspect ratio
aircraft with “unloaded” wing tips (such as the Horten or Prandtl lift
distributions) will have fewer tendencies for adverse yaw when rolling, but I
don’t believe they eliminate it, at least without using negative load at the
tips which is very bad for drag. Al’s videos mention that his wing twist and
elevon design produces a proverse yaw torque when the ailerons are deflected.
Most stability and control engineers know that this is only a small part of the
adverse yaw problem, most adverse yaw is caused by the rolling motion which
changes the local angle of attack along the wingspan resulting in the downward
wing thrusting forward and the upward wing pulling backward. It is the tilting
of the local lift vectors along the span as the wing rolls that contribute the
major effect in the adverse yaw problem. The drag change due to aileron
deflection is only a minor effect in comparison and this is the only effect that
Al mentions in his discussion. There still remains the question of adequate yaw
damping and Dutch roll stability when there is no vertical surface for yaw
stability. I look forward to seeing the flight test and simulation results of
the complete aircraft to help quantify his results.

The proposed Prandtl lift distribution may work well at a single trim condition
(i.e. max L/D) but may result in drag penalties, particularly at higher speed
where the excessive twist is undesirable. Al will need to offer detailed
analysis of specific design examples to convince guys like me that there is
something to this approach rather than simply using winglets or an aft tail
configuration.

Regarding the mysteries of bird’s handling qualities, I think there’s much more
to it than nonlinear wing twist. Birds have many degrees of freedom to change
their wing shape in flight and do not obey the same stability constraints that a
manned aircraft does. I studied this problem some time ago and one result that
came from it was the use of anhedral to improve roll rate in the presence of
adverse yaw. If you can’t eliminate adverse yaw, a bit of anhedral will prevent
it from slowing down your roll rate. Ever notice that many high AR soaring birds
have anhedral wingtips? Perhaps they do this to maintain agility in the presence
of adverse yaw. I have performed design studies that confirm this effect on
rigid wing aircraft but don’t have enough data on birds to say any more about
it. I’ve attached a few photos I took of Albatross dynamic soaring in New
Zealand. In all photos the birds are gliding and not flapping. I can’t say I see
any “Prandtl style wing twist” but I do see anhedral.

Referring to the
slide show
above, Steve writes:

The slides show that Al is not considering the yaw due to roll
rate effect that I mentioned (Cnp) and he seems to only consider the aileron
deflection effect (Cnda). That explains our differences in opinion on the
matter. The bad news is that the dominant effect, Cnp, is much bigger at low
speeds and large wingspan, but diminishes at high speeds. This correlates with
what most sailplane pilots already know: You need a lot of rudder when flying
slow to coordinate the turn entry. I look forward to seeing their flight test
results to see if they really have accomplished much for adverse yaw during
45-45 degree roll reversals. This is what most of us really care about, but it
doesn’t seem like that’s what his theory addresses. His approach is mostly
concerned with the yaw moment at the instant the ailerons are deflected and not
what happens next. The reduced load at the tips will minimize the Cnp effect
when rolling, so I expect some improvement in adverse yaw reduction from his
lift distribution.

We’ve been able to achieve all of this with winglets and movable rudders on the
SWIFT, plus we get the added stability of the vertical surfaces. The winglets
also produce “thrust” at low speed like Al’s unloaded wing tips do, so there
probably isn’t much performance improvement there. The only possible way to use
Al’s results for increased performance is to increase the span of a Swift with
reduced tip loading so that you’d have a light structure and OK handling
qualities but higher L/D. If this isn’t done properly the L/D at high speed
could suffer.

The other option is to simply increase the span of the Swift and use the
optimized span loading without regard to unloading the tips. This will give an
even higher L/D but will require a few pounds more carbon in the wing (not much)
and the winglets will still provide good handling. So, for a few pounds more
weight it appears you can have even better performance without resorting to any
of this new theoretical approach, at least for an aircraft like the Swift.