S D oaring igest

S D oaring igest
SoaringDigest
Radi
June 2014
C ntr lled
Vol. 31, No. 6
CONTENTS
June 2014
4
Vol. 31, No. 6
Front cover: Morgan Hill's ASH 31Mi at the 2014 Jerilderie Aerotow. Full
coverage of this event starts on page 63. The story behind the transmitter
in the cockpit is that the size of your sailplane has become a little bit of a
status symbol, and the new benchmark is that if your transmitter doesn't fit
in the cockpit – it’s not big enough... Bit of a local joke.
Photo by Henryk Kobylanski.
Nikon D5200, ISO 100, 1/400 sec., f10, 78mm
Bell-Shaped Versus Elliptical Lift Distribution
Wings, Proverse Yaw-Roll Coupling
Philip Randolph
41
62
63
Sixteen Foot Flying Wing for Slope
and Aerotow
Steve Pasierb, John Appling, and Erich Schlitzkus
Graeme Phipps photo
Colin Taylor's Discus 2
Jerilderie Scale Aerotow 2014
Review: ST-Model DG-1000 RR
81
Renato De Cecco photo
90
CEWAMS Saddle Mountain Slopener
91
Tom's Tips: Large Scale CG Balancer
100
Dan Ouellet
Las Aguilas Sailplane Club in Caracas, Venezuela
Coverage by Philip Randolph
Tom Broeski
Event coverage by Henryk Kobylanski
Back cover: All-round slope glider Swing 88, span 2.2m, flying
at Le Col du Glandon at 2000m of altitude in the French Alps.
Photo by Pierre Rondel
Canon EOS 650D, ISO 250, 1/2000 sec., f5.6, 300mm
2
R/C Soaring Digest
R/C Soaring Digest
June 2014
Volume 31 Number 6
Managing Editors, Publishers
B2 Kuhlman
Contact
[email protected]
[email protected]
http://www.rcsoaringdigest.com
Yahoo! group: RCSoaringDigest
FaceBook: https://www.facebook.com/RCSoaringDigest
R/C Soaring Digest (RCSD) is a reader-written monthly
publication for the R/C sailplane enthusiast and has been
published since January 1984. It is dedicated to sharing
technical and educational information. All material contributed
must be original and not infringe upon the copyrights of others.
It is the policy of RCSD to provide accurate information. Please
let us know of any error that significantly affects the meaning
of a story. Because we encourage new ideas, the content of
each article is the opinion of the author and may not necessarily
reflect those of RCSD. We encourage anyone who wishes to
obtain additional information to contact the author.
———
Copyright © 2014 R/C Soaring Digest
Published by B2Streamlines <http://www.b2streamlines.com>
P.O. Box 975, Olalla WA 98359
All rights reserved
———
RC Soaring Digest is published using Adobe InDesign CS6
June 2014
In the Air
Various types of lithium batteries have been available to the
aeromodeling community for a number of years - LiPoly, LiIon,
LiFe, etc.. Originally used for supplying the power to electric flight
motor systems, with the advent of servos capable of handling
higher voltages there is a growing trend to use lithium cells to
directly power on-board electronic equipment. Two battery-related
articles have appeared within the last few days...
First is an article which appeared in NASA's magazine "Tech
Briefs" which explored power supplies for extreme environments
and focused on Tadiran Batteries LiSOCl2 cells. The Tadiran
cells come in AA and AAA sizes, with the AA version capable of
output pulses of 5A and recharging capability of more than 5,000
cycles. These cells are currently available from Digi-Key, House
of Batteries, and Mouser Electronics. See the Tadiran website
<http://tadiranbat.com> for further information, cell terminal
options, and a complete list of distributors.
The second article appeared in GizMag.com. The Japanese
company Power Japan and Kyushu University have announced
the development and planned mass-production of a disruptive
dual carbon battery that can be charged twenty times faster than
an ordinary lithium-ion cell. Power Japan is planning to produce
the battery using an organic carbon complex, developed inhouse from organic cotton, to obtain a greater control over the
size of the carbon crystals in its electrodes. Originally destined
for use in electric automobiles, photos on the GizMag page
portrayed what appears to be a AA cell along with a couple of flat
cells. See <http://www.gizmag.com/dual-carbon-fast-chargingbattery/32121/>.
Time to build another sailplane!
3
Bell-shaped versus elliptical lift distribution wings
A few translations of basic mathematical aerodynamic truths into physical explanations
Proverse roll-yaw coupling
and flying the undersides of shifted lift/drag curves.
Philip Randolph, [email protected]
Profili plots by Adam Weston
Where to: BSLD wings. Proverse
roll-yaw coupling.
I’ve split this article into two parts. First
is a discussion of the structural and
aerodynamic efficiency benefits of bellshaped lift distribution (BSLD) wings
as pioneered by Ludwig Prandtl in a
1932 paper, as compared to elliptical
lift distribution (ELD) wings. Much of the
article covers basics, especially trailing
vortex pressure drag, necessary for our
next step.
Second is a look at wings that will
perform coordinated turns without a
rudder – proverse roll-yaw coupling. The
pioneers here were Walter and Reimar
Horten. The Horten brothers, from
1933 – 1950, working in pre-war and
WWII Germany and later in Argentina,
designed and built tailless flying
wings that achieved proverse roll-yaw
coupling via BSLDs. Because of their
work, discussions of proverse yaw are
4
usually linked with BSLD wings and their
structural and aerodynamic benefits of
BSLD and with ‘tailless’ aircraft, ‘flying
wings.’ For achieving a combination of
efficiency, light weight per structural
strength, and proverse roll-yaw coupling
BSLD is one ideal, but in the aero world
of compromises there are other options.
The basics important here are wingtip
vortex upwash and the use of lift/drag
curves to predict proverse yaw.
In the two halves of the article we’ll
separate the influence of trailing vortices
and wingtip vortices.
• Trailing vortex pressure drag: Drag
is mainly from trailing vortices. If we’re
talking about comparative drags of BSLD
and ELD wings, we look at trailing vortex
pressure drag. That’s a major tool for the
first part of the article.
• Wingtip vortex upwash: Roll-yaw
coupling of BSLD wings is related mainly
to wingtips flying within the upwash
of wingtip vortices, plus adverse drag
influences of trailing vortex pressures,
and ultimately to adjusted lift/drag
curves. These are the tools for our
second half, proverse roll-yaw coupling.
Right up front I have to say I am not
completely sold on ‘pure’ BSLD wings.
That means I’m somewhat objective.
BSLD wings may have applications.
The concepts innate to comparisons of
BSLD and ELD wings are important.
The compromises and approximations of
BSLD do have applications. Most airliner
wings are not purely ELD. They narrow
the lift distribution near wingtips for
structural reasons,1 a slight shift toward
BSLD.2 Then commercial designers add
1
According to a 4/27/2014
conversation with S. Allmaras, Ph.D.
2
Ludwig Prandtl, in the 1932 article
and diagram discussed later in this article,
diagrammed a compromise between ELD
and BSLD. And Robert T. Jones, in his
Wing Theory, elucidated the benefits of
R/C Soaring Digest
winglets, which if horizontal would be
another tilt toward BSLD. Please see the
section on ELD-BSLD compromises.
To some extent the timing of BSLD
implementation is backwards. BSLD
wings optimize lightweight structure,
usually measured by root bending
moment, for a given lift. And that helps
with fuel efficiency. But structural weight
considerations are most important with
weak materials – bird muscles, tendons,
and bones; the willow sticks of Otto
Lilienthal’s gliders; or wood and fabric. To
some extent aluminum and then carbon
fiber allow great strength with little weight
penalty from varying from optimal lift
distributions.
On a strength continuum from 98-pound
weakling getting sand kicked in his
face at the beach to Spiderman strong,
modern materials are partway to
fictional-but-ideal materials of infinite
strength, where varying shape would
make no weight penalty. On the other
hand, in a non-stop flight of 8,000 miles,
every bit of weight savings makes fuel
savings.
that compromise. (Robert T. Jones, Wing
Theory, Princeton University Press, 1990,
113–114.) Also see the diagram on page
16 of: (Bill Kuhlman, “Twist Distributions
for Swept Wings, Part 4,” Radio Controlled
Soaring Digest, June 2003, 16, http://
www.rcsoaringdigest.com/OTW/on-thewing4/164-HCP4.pdf.)
June 2014
A tertiary benefit: Writing the article
made me clean up more of my basic
aerodynamic understandings. So it has
been educational, and should be for
readers.
In the history of aerodynamics there
have been many neglected approaches.
To bring a historic idea into acceptance
requires an advocate. Albion Bowers,
Deputy Director of Research at NASA,
Dryden, is that advocate for the Horten’s
approach and the structural, efficiency,
and proverse yaw benefits of BSLD
wings. You can find his TED talk at:
http://www.youtube.com/
watch?v=223OmaQ9uLY
There is also excellent, ongoing
discussion at the Nurflugel (‘flying wing’)
Yahoo discussion group, in which Albion
Bowers frequently comments. It’s at:
https://groups.yahoo.com/neo/groups/
nurflugel/info
This article has thoroughly benefited
from the discussions on Nurflugel.
Extensive information about BSLD
wings and how the upwash outboard of
wingtip vortices affects proverse roll-yaw
coupling has been presented there. If I’ve
unintentionally scooped anyone’s serious
research I’m happy to yield precedence.
I have been privy to information I
conceptually disagreed with and did not
repeat. What I’ve repeated has been
part of the very general discussion. For
example, it is well understood that flying
aileron-equipped wingtips in vortex
upwash can produce proverse yaw
forces.
There are a few areas within which I have
independently derived contributions
(which doesn’t at all mean that I thought
of them first):
• First and most basic is an explanation
of wingtip/trailing vortex formation
and ‘trailing vortex pressure drag’ as
a foundation for how the wingtip and
trailing vortices affect roll-yaw coupling
via thrusterons and dragerons.
• Second is clearing up basic
misunderstandings about the forces
wings put on air, the forces that create
wingtip/trailing vortices: I show that
by ‘downwash’ Prandtl meant ‘netdownwash,’ and show that three
concepts are equivalent: Prandtl’s (net)
downwash, Prandtl’s induced angle of
attack, and Lanchester’s sinking vortex.
• Third (and the second part of the
article) is the use of lift/drag polar curves
to predict roll-yaw coupling. I posted
the beginnings of this approach on
the Nurflugel site (8/22/2013).i Here I
add how the curves shift upon aileron
deflection and location in vortex upwash.
Preview: Proverse roll-yaw coupling
forces happen when roll-control surfaces
are located in wing areas flying in
negatively sloped areas of the lift/drag
polar curves. Even in vortex upwash that
only occurs in wing segments that are
lightly or negatively loaded.
5
There is still a great lack of information.
For bell-shaped lift distribution wings
(BSLDs) there have not been published
wind-tunnel smoke-stream photos or
sophisticated CFD analyses that show
trailing vortex pressure profiles, nor
‘crossover points’ (different from trailing
vortex centers!). The true experts will
probably change that soon. (Outboard of
the crossover point flows slope upward.
A bit of winglet in that upward flow can
gain thrust in the same way that a glider
gains thrust from thermal updrafts.)
Prandtl diagrammed the crossover point
at about 70% of half-span. (See his
diagram.) My very questionable modeling
with XFLR5 indicated the crossover
point at about 80% of half-span. An
aerodynamics Ph.D. friend informally
estimated the vortex center at 91% of
half-span. (That may also be suggested
by the BSLD elliptical net-downwash
pattern in figure 2!) In the BSLD
illustrations I’ve stuck with Prandtl’s 70%
for the crossover, and about 2/3 halfspan for the vortex center. For reality
we’ll have to wait for the quantitative
guys.
Fishing for clarifications, corrections,
and good information.
Disclaimer: Artwork herein is for
conceptual purposes. It is generally not
to scale. If anyone out there can correct
it or put it into correct scale, please
contact me.
6
Comments, corrections, quantifications,
or supplemental graphical or CFD
analyses: Please. If you’ve got expertise
I’m happy to give credit.
BTW, I’m writing a book, mostly on
great historical aerodynamic theories
that were bypassed either for no good
reason or because they predated the
supercomputers required to turn them
into engineering.
BSLD wings
Elliptical versus Bell-shaped lift
distribution wings: design optimization
follows design parameters
In aerodynamic engineering the question
determines the answer. An early question
was, “What wing planform is the most
efficient?” Answers to such questions
are determined by choice of constraints.
The simplest constraint was picked
first – wingspan. It’s not the constraint
that gives the best answer, unless, of
course, wingspan really is a constraint
-- for example, Standard Class gliders
maximum span is 15 metres. Discus
launch model glider competitions (DLG)
are limited to 60". The prevalence
of competition classes limited by
span biases design efforts toward a
commonplace focus on elliptically loaded
wings.
In 1908, published in 1918, Ludwig
Prandtl developed the elliptical load
distribution.ii Wingspan, airfoil, and load
were held constant. Prandtl determined
that for a given wingspan an elliptical
lift distribution (ELD) yielded the most
efficient flight. Which is true. However:
A decade later Prandtl questioned
whether he had picked optimal design
parameters. In a 1932 paper he
attempted to approximate the answer
to a more sophisticated question, “For
the same lift, spar weight, and wingroot bending moment (strength) as an
elliptical wing, what lift distribution and
span offers the greatest efficiency?” His
answer was a bell-shaped lift distribution
(BSLD). With a 22% increase in span the
BSLD wing was just as light and strong
as the ELD wing and carried the same
load with about 11% less induced drag.
See Figures 1, 2 and 3
The Horten brothers, from 1933 – 1950,
working in pre-war and WWII Germany
and later in Argentina, designed and
built tailless flying wings that used ‘bellshaped lift distributions,’ or BSLD. In
addition to structural and efficiency
benefits, their aircraft achieved
coordinated turns without the use of a
rudder, or ‘proverse roll-yaw coupling.’
That’s our second article. In this first
article we’ll cover basics.
The Hortens achieved BSLD with
a combination of planform, twist
(washout) and airfoil changes along span
(aerodynamic twist). They weren’t the
first to do use all three design elements
– a May 2014 article in Air & Space
describes a swept-wing, tailless biplane
R/C Soaring Digest
Ludwig Prandtl’s 1932 diagram,
with his elliptical lift distribution
highlighted in red.
Ludwig Prandtl’s 1932 diagram, with his
bell-shaped lift distribution highlighted.
Vortex
upflow
0
Net downwash
Red: Airfoil pressure forces
Dark red: Wingtip vortex
pressures
Blue: Flows
Light blue & gray: Vortex flows
2D airfoil drag (from D/L)
Figure 1: Prandtl’s 1932 diagram of elliptical
and bell-shaped lift distributions
designed by Starling Burgess in 1912.
Google images show significant twist near
wingtips, where he used thickened airfoils.
It did rely on large ‘end curtains’ between
its biplane wingtips for yaw stability but
was rudder free, a step toward tailless
coordinated flight.
And BSLD? Any wing with twist will, at
some wing loading and angle of attack,
loosely approximate BSLD. It may require
very light Gs, such as when pushing over
at the top of a high-speed arc. When a
combination of speed and low angle of
attack put the twisted wingtip at zero-lift
AoA BSLD will at least be approximated.
June 2014
Trailing vortex drag
(strong near wingtips),
from centrifuged
low pressures
2D airfoil drag (blue) is very low. It’s lift
times the inverse of the 2D airfoil
sectional L∕D ratio, or D∕L.
Net downwash
0.0
0.2
0.4
0.6
0.8
1.0
Figure 3: Prandtl’s 1932 diagram with bellshaped lift distribution highlighted
Figure 2: Prandtl’s 1932 diagram with elliptical lift distribution highlighted in blue.
Actual net-downwash velocities near wingtips are curved because of the pressure
gradient around the end of the wing, but perhaps ‘rectangular downwash’ refers to
the vertical component of net downwash. Wingtip vortices are asymmetrical and only
partially formed. The wingtip partial vortex is formed by pressure gradients and by
sheer of downwash flows with upflows. Downwash flows and the pressure gradient
around the wingtip help form the wingtip vortex, not the other way round. As the
wingtip vortex becomes the trailing vortex it centrifuges a low-pressure center which
‘pulls’ back on the wing and helps defeat pressure energy recovery near the wingtip.
It’s important to distinguish between: (1st) Pressure forces exerted by the wing (red)
which are always normal to its surface; (2nd) Pressure forces around the wingtip (dark
red); (3rd) Resulting flows (blue, light blue & gray).
7
And any wing designed for BSLD will
stray from that optimum at different
speeds. The Horten designs used a lot
of twist. If designed to have wingtips at
zero-lift AoA in takeoff, when AoAs are
highest, at the lower incidence of cruise
speeds wing twist would make tips lift
negatively. If designed with twist for BSLD
at cruise speeds, during takeoff and
landing higher AoAs would put wingtips
in positive lift, which could challenge
the wing’s proverse roll-yaw coupling
capabilities. A rudder might be necessary.
But the lift distribution would be between
elliptical and BSLD. That’s like having
more wing for takeoff and landing.
And then there is inverted flight. Wings
that achieve BSLD via twist don’t like it.
Upside down, the twisted BSLD wingtips
find themselves at high positive angles of
attack, making tip stalls. That will either
lead to a roll to upright or to a spin. For
aerobatics tiperons off an untwisted or
lightly twisted wing could work.
BSLD may be attempted by planform
alone, making a wing with a bell-shaped
profile as viewed from above. But then
the very narrow tips operate at low
Reynolds numbers, potentially making
other problems.
Terminology, and how the trailing
vortex makes high wingtip drag:
First, basics: we now look at the
mechanisms of what should properly be
called ‘trailing vortex drag.’
8
“Deflected ailerons deform the
load distribution away from the
ideal near-elliptical shape, and
hence increase induced drag.iii” –
Mark Drela, quoted by kcaldwel on
RC Groups.
The drag on wings is from pressure or
friction. Most of the drag on fractional
subsonic wings is from pressures.
And most of that is from the lowered
pressures in the trailing vortex. Terms will
get us to how that works:
• ‘High wingtip drag’ is a correct term
that merely indicates that drag is usually
highest near wingtips, at least for elliptical
lift distribution wings. BSLD wings
have the highest vortex drag somewhat
inboard of wingtips.
• ‘Wingtip vortex drag’ is an incorrect
term; the wing puts forces on air to
make the wingtip vortex, but the wingtip
vortex mostly doesn’t put drag forces
on the wing. (Well, actually all wingcaused forces have ‘interference’
affects on all other parts of a wing. Lift
forces form vortex forces which leak
spanwise. Wingtip vortex forces on
vortex upflows are critical to BSLD thrust
and thrusterons. Still, the actual drag
forces on a wing are not from the wingtip
vortices, but from pressures within the
trailing vortices.) We’ll examine this in
detail.
• ‘Trailing vortex pressure drag’ is a
correct term. The low-pressure center
of the trailing vortex ‘pulls’ back on the
wing (mainly near the wingtip) and the air
surrounding it, reducing pressure energy
recovery. It also ‘pulls’ forward on trailing
air, with the equal and opposite force.
The low-pressure center of the trailing
vortex is created in two ways. First,
pressure energy is used up creating the
circular velocities of the wingtip vortex.
Where the velocities are highest, near the
center of the vortex, pressures are lowest.
Second, as the wingtip vortex becomes
the trailing vortex its rotational velocities
centrifuge its core pressures even lower.
The energy input per second required to
keep a plane moving forward is equal to
the energy-per-second lost to wake. In
the wake that energy is a mix of trailing
vortex pressure gradients pulling forward
and inward on air, and resulting forward
and rotational wake air velocities, plus
turbulence and heat.
Lanchester pictured his wingtip/trailing
vortices centered near wingtips, accurate
for most wings. For an ELD wing the
crossover from downflow to upflow
happens close to the center of a ‘wingtip’
(sic) vortex and roughly at or a little in
from the wingtip. The ‘crossover points’ of
Prandtl’s 1932 BSLD wings are centered
well in from wingtips. (For a BSLD wing
the crossover happens just outboard of
the vortex center.) However, in his 1932
article Prandtl didn’t mention vortices.
He approached the problem in a more
mathematical manner.
See Figures 4 and 5
R/C Soaring Digest
Competeting wingtip forces on BSLD flows
Elliptical Lift Distribution
Downward
pressures
on air
Elliptically loaded wing
Downwash velocities
at trailing edge are
approximately vertical
within the curve of flows.
Crossover
Vortex center
BSLD wing
Competeting wingtip forces on BSLD flows
determine the crossover point from downward
to upward ‘induced angles of attack.’
Figure 4: The forces that determine the crossover
point. Outboard of the BSLD wing crossover
point, wingtip vortex upward pressure forces
exceed downward lift forces on air, for a net
upward force on air. That makes the rising flows
within which a BSLD wingtip may gain thrust.
The vortex center doesn’t make an upward
or downward force, so at the vortex center
downward lift forces on air are unopposed.
Therefore a BSLD wing’s crossover point to
upflows is always outboard of the vortex center.
June 2014
Strong pressure gradents
around wingtips make intense
wingtip vortices
Rectangular downwash pattern
makes strong sheer at wingtip,
adding to rotational velocities.
Centrifuged low pressures in
the trailing vortex ‘pull’ back on
the wingtip for very high wingtip drag.
Bell-Shaped Lift Distribution
Downward
pressures
on air
Vertical velocities
at trailing edge
Weaker pressure gradients near
wingtips make gentler wingtip
vortices.
Weaker downwash near wingtips makes gentler sheer, lower
rotational velocities, and more
gently lowered trailing vortex
core pressures for lower drag.
Figure 5: Lift forces on air and net-downwash. Bell shaped
lift distribution wings create gentler trailing vortices for lower
trailing vortex pressure drag. Their wingtips ride in vortex
upwash. Red arrows show lift pressure forces. Blue arrows
show net-downwash momentums.
9
To make thrust, BSLD wingtips must overcome trailingvortex pressure drag. A primer on how trailing vortices
make pressure drag.
The accompanying XLFR5 plot shows negative induced
drag, or thrust, outboard of about 80% of a BSLD wing halfspan. I used XLRF5’s VLM, or Vortex Lattice Method feature.
Unfortunately when running the more sophisticated XLFR5
features, viscous analysis and 3D analysis, it announced
errors. And when I had it build a graph of ‘induced drag’
(vortex drag) for an ELD wing it didn’t show high wingtip
drag. So whatever it was doing was suspect. XLFR5 is
phenomenal wing analysis freeware, but please take that
80% ‘crossover’ from drag to thrust and the pattern of
‘induced’ drag (trailing vortex pressure drag) with a grain of
salt. Supercomputer CFD results would be more trustable.
Still, it’s illustrative. And leaves a mystery.
See Figure 6
When one designs a wing, whether in XFLR5 or some
industrial CFD program, one can ask for a graphic of drag
by span. That’s great. XFLR5 will even animate, so you can
watch how drag changes with angle of attack. For a BSLD
wing, near the tips, you can see how drag is negative,
meaning the wingtips can produce thrust. And it’s possible
to do all that without having an idea of what the various
influences are around a wing, what causes what, and how
they add up to the total effect you are watching on your
computer screen.
To start to get a grasp on how various wings work
it’s necessary to understand the various forces and
momentums at play. One needs to investigate just how the
wingtip/trailing vortex system forms, where it is located on
the wing, and the pattern of its pressures and velocities.
And to do that we have to chase some old ideas about what
‘downwash’ means or should mean.
10
Trailing vortices and a more
probable vortex pressure-drag
profile are superimposed on
XFLR5 induced drag output
Figure 6: XFLR5 BSLD 7° Vortex Drag, with trailing vortices and
a more probable vortex drag superimposed. XFLR5 inviscid VLM
very approximate but illustrative plot of ‘induced’ drag of a BSLD
wing. ‘Induced’ drag should be a map of trailing vortex pressure
drag, and thus an approximate map of trailing vortex pressures.
Meaningful accuracy would require an industrial CFD program or
wind tunnel results. Note the areas of negative drag or thrust near
wingtips. Also note the drag spikes. Vorticity and vortex pressure
drag increase wherever there is a sudden change in lift, as at the
junctures between trapezoidal wing sections. Anyone who can
supply a more accurate plot, please contact me.
R/C Soaring Digest
Basics: What did Prandtl mean by ‘downwash?’ (Net
downwash.) Prandtl’s (net) downwash and ‘induced angle
of attack’ as equivalent to Lanchester’s assertion that
airplanes always fly in sinking air (with the exception of
Prandtl BSLD wingtips!)
α
Another case of good math making good results even when
applied to questionable physical understandings.
First we’ll look at truths that mainly date to Frederick William
Lanchester’s work from 1894, 1897 and 1907: Lift is from the reversal
of upwash momentums ahead of a wing to slightly greater downwash
momentums aft (per second). The difference is net downwash. Net
downwash contributes to lift, but also carries energy into a wing’s
wake, part of the energy that must be replaced by thrust to keep a
plane moving forward.
See Figure 7
That a wing loses energy to its wake is equivalent to three nearly
synonymous but seemingly disparate descriptions.
• First: All airplanes fly in sinking air, in a ‘sinking-vortex’ pattern. That
makes flight like walking up a sand dune, with sink at every step.
Energy lost to wake implies sink. Lanchester diagrammed a wing
flying in air that sinks inboard from its wingtips and rises outboard of
its wingtips.
Figure 7: Frederick William Lanchester’s 1907
diagram showing greater angle and velocity
of downwash than upwash. Lanchester
correctly asserted that lift came from upwash
momentums ahead of a wing being reversed
to greater downwash aft. Although he was first
to visualize an idealized ‘circulation’ around a
wing he was too realistic to be a true believer
in ‘bound vortex’ symmetry. He translated
wing-flow waveform into ‘circulation,’ but
didn’t believe literally in the useful but idealized
symmetry of either.
See Figure 8a
An airplane’s weight, exercised through the action of the wing, makes
air inboard from the centers of its wingtip vortices sink. Viewed
from ahead the wing flies in a sinking vortex (though wingtips may
stick into rising air). Lanchester, a physical intuitionist, generally had
causally correct analyses.
See Figure 8b and 8c
June 2014
11
Figure 8a: Lanchester’s 1907 diagram showing
sink (f f f f) inboard of wingtips and rising air (o o o)
outboard of wingtips.
Figure 8b: Lanchester’s 1907 diagram with equivalent downwash
patterns superimposed:
Upper: Elliptical load distribution wing has rectangular netdownwash
Lower: Bell-shaped load distribution wing has net-downwash
inboard of ‘crossover points’ and ‘net-upwash’ outboard of
crossover points.
12
R/C Soaring Digest
Trailing vortices form
behind the wing, and
mostly don’t affect net
downwash. They do
precess downwards.
Figure 8c: Prandtl’s causally backwards explanation
of (net) downwash.
• Second: ‘Net-downwash.’ For the entire span of traditional
wing sections, downwash momentums aft are always
somewhat greater than upwash momentums ahead. The
difference between vertical momentums of upwash at the
leading edge and downwash at the trailing edge is ‘netdownwash.’ BSLD wingtips have net upwash.
Downwash
velocities
Net-downwash
Rectangular net downwash velocities of elliptical lift distribution wings
Upwash
velocities
Prandtl’s idealized symmetrical
‘circulation’ around a ‘lifting-line’
engineering-substitute for a wing.
Since upwash and downwash
appeared equal, he attributed
(net) downwash to the inner
downflows of trailing vortices. But
trailing vortices are a result of a
wing’s net downwash, not a cause.
Net-downwash momentums
are constant for elliptical
lift distribution wings
The difference between upwash ahead and downwash aft is ‘net
downwash.’ The largest wing chords are far from wingtip losses that
sap upwash. Thus far from the wingtips upwash is nearly equal to
downwash.
If there were no losses near wingtips, smaller chords would make
smaller downwash. The loss of upwash near wingtips makes greater
downwash there. Upwash velocities lessen near wingtips because
‘leaks’ of pressures up around wingtips make failure of pressure energy
recovery into upwash. And downwash velocities also increase near
wingtips because the low pressure centers of wingtip vortices ‘pull’
back on flows, the opposite of pressure energy recovery.
Figure 9a: Net downwash. Local net downwash is from differences
between upwash ahead and greater downwash aft.
See Figures 9a and 9b
The idea that elliptical wings have a rectangular pattern of
net downwash momentums was developed mathematically
by Prandtl in the second decade of the last century. Figure
9c attempts a physical, causal explanation.
See Figure 9c
June 2014
13
How small wingtips can make big net-downwash.
Figure 9b: Rectangular net-downwash of elliptical lift
distribution wings. Many aerodynamics texts consolidate
upwash and downwash into net-downwash at the
quarter-chord, Prandtl’s ‘lifting line.’ Unfortunately this
net-downwash is often simply called ‘downwash,’ which
leads to confusion of ‘net downwash’ with ‘trailing-edge
downwash’ and to forgetting that lift comes from reversing
upwash momentums ahead to downwash momentums aft.
• Third: ‘Induced angle of attack’ and
‘sink’: The ‘induced angle of attack’ is
the local downward angle of flow an ELD
wing encounters because the air it rides
in is sinking. The induced angle of attack
is theoretically constant along the span
of an elliptical wing. The induced angle
of attack varies along BSLD wingspan,
and becomes positive outboard of the
‘crossover point,’ where wingtip vortex air
is rising.
• Also third: ‘Induced angle of attack’
and ‘net-downwash.’ Net-downwash
velocities can be translated into induced
angles of attack: We make vector sums
14
Away from wingtips, upwash and downwash momentums are nearly
equal. In relation to its stillness before the wing passed, air at the
trailing edge is dragged slightly forward (blue arrows).
Near wingtips, upwash momentum is sapped by the pressure losses
that form the wingtip vortex.
Near wingtips, the low pressures of the trailing vortex accelerate flows
down and back along the airfoil surfaces. A component of that velocity
adds to downwash momentums.
The difference between lessened upwash and trailing vortex
enhanced downwash makes high wingtip net-downwash, and high
energy lost to wakes at wingtips. Wingtip energy loss is equivalent to
high trailing vortex pressure drag at wingtips.
Figure 9c: How narrow ELD wingtips can make ‘rectangular’ net-downwash. If these
were 2D wing sections in a wind tunnel, and one had twice the chord of the other,
part of the answer would be that the air at the surface of the smaller section drops
half the distance in half the time, for the same vertical velocity at the trailing edge.
That’s deceptive. The larger wing section affects a larger volume of air, and so would
create greater net-downwash momentum. A real wingtip has less upwash because
of wingtip losses of pressure differences between upper and lower flows. And its
downwash is increased as air is not only accelerated downward but is also accelerated
at a downward angle backwards toward the low-pressure center of the trailing vortex.
Theoretical net-downwash remains constant along the span of elliptical wings.
R/C Soaring Digest
Freestream velocity
‘Induced’ re αi, ε
lative airfl
ow
Sink or net-downwash
Freestream velocity
αi, ε
‘Induce
Sink or net-downwash
d’ angle
d local
flow
Sectional li
ft
90° to ‘indu
ced’
local ‘effecti
ve
relative air
flow’
ε
ELD induced AoA
Weight
Chord
line
αo Loca
α
l relative airfl
ow
αi, ε
Freestream velocity
‘Effective lift’
90° to flight path
induced drag
BSLD induced AoAs
Figure 10a: For each wing section, the vector sum
of the freestream velocity and net-downwash (or its
equivalent, sink) yields the ‘induced’ angle of local
‘effective relative airflow’ or ‘relative wind,’ and the
‘downwash angle,’ ε, equivalent to the ‘induced angle
of attack, αi. For ELD wings the induced angle of
attack is downward for all sections.
Figure 10b: Induced angle of attack is positive for
BSLD wingtips.
Sectional lift forces are perpendicular to the
‘induced’ local flow or ‘effective relative airflow.’
A component of this sectional lift force is in the
direction of drag, and a component is opposite to
weight, making effective lift. To stay up, wings have
to angle up so that their zero-lift line is steeper than
the average (negative) angle of attack.iv
Prandtl’s causally backwards idea of ‘downwash.’
June 2014
of the wing’s forward speed and local ‘net-downwash’ velocities (or,
outboard of the ‘crossover point,’ ‘net-upwash’ velocities.
See Figures 10a and 10b
Prandtl’s idea that a wing encounters down-flowing air is equivalent to
Lanchester’s more physically accurate diagrams showing that an airplane
always flies in sinking air. Prandtl, as a mathematical engineer, built
methodologies that gave engineering results with correlations confused
as causalities.
Within aerodynamics misinterpretations abound. In this section we’ll
see a common misinterpretation of Prandtl’s 1932 diagram. It (probably)
15
accurately shows elliptical lift distributions
making constant (rectangular)
‘downwash’ (momentums), and bellshaped lift distributions making elliptical
downwash momentums. Unfortunately,
at least in that paper, Prandlt wasn’t clear
about what he meant by ‘downwash,’
though it can be parsed that he meant
‘net downwash.’
Prandtl substituted his highly artificial
(symmetrical) ‘bound vortex’ engineering
idea for a wing. His idea was that his
symmetrical bound vortex couldn’t make
vertical velocities, so his rectangular
‘downwash’ must come from the inner,
downward velocities of the trailing
vortices.v That is technically equivalent to
saying that the upwash and downwash
of the bound vortex are equal (false),
while net downwash is caused by the
trailing vortices (false again, and causally
backwards).
Rather, the flows, forces, losses, and
pressure-energy recoveries around a
wing are asymmetrical and cause net
downwash and create the trailing vortex
system, not the other way around.
To assert that the trailing vortices cause
net downwash is a bit like pulling a
bucket out of a well, and then claiming
that since there is a net force upward the
bucket must be pushing the rope up. It’s
as if Prandlt was self-hoisting on his own
petard and claiming that he, rather than
his backward causality, was the lifting
force. In the mythical ‘lifting oneself by
16
one’s bootstraps’ the equivalent notion
would be that the boot puts the upward
force on the strap.
We could split Prandtl’s ‘circulation’
approach into two parts. If ‘bound
vortex’ ‘circulation’ were symmetrical
(it isn’t), it would have upwash equal to
downwash, and by Kutta-Joukowsky
would make lift without losses. Second,
if the (net) downwash were from trailing
vortex action (false) rather than from the
action of the wing, then net downwash
wouldn’t contribute to lift but would be
part of losses of energy to wake. Actually,
the ‘net’ downwash thrown down by the
wing does contribute to lift, but also is
part of the energy losses to wake. Net
downwash is the expensive part of lift
creation.
And yes, trailing vortices do precess
downwards, in the sinking or traveling
vortex pattern typical of smoke rings. But
that’s another result of the wing forces
that set up the vortex motions, and not a
cause of net downwash.
Prandtl’s idea that trailing vortices create
net downwash has another flaw. Trailing
vortex velocities roughly follow the rule
that V = k/r, except near the center of
each vortex, where velocities are more
proportional to radius. Such vortices
would not make a rectangular netdownwash velocity pattern. Again, it is
the wing that makes the wingtip/trailing
vortices, not the other way around.
See Figure 11
Even though Prandtl’s idea that trailing
vortices cause net downwash was
causally backwards it was mathematically
passable. Engineering requires only
quantitative knowledge of ‘what happens’
rather than ‘why’ or ‘how.’
The wingtip and trailing vortices and
trailing vortex pressure drag
Note: I use the term ‘trailing vortex
pressure drag’ because it’s accurate and
explanative. The usual terms are ‘induced
drag,’ ‘vortex drag,’ or ‘high wingtip drag.’
These terms are often expressed vaguely
in terms of ‘energy’ going into wingtip
vortex formation. While that can be made
to add up, drag is a force and the force is
pressure difference on wing area. ‘Trailing
vortex pressure drag’ makes this explicit.
We can divide wingtip/trailing vortex
formation into the forces that create the
wingtip vortices, wingtip vortices, and
trailing vortices.
The components of pressure forces that
make the swirl of wingtip vortices are in
the y-z plane (the vertical plane crosswise
to a plane’s travel, up through the quarterchord of an unswept wing). This is the
pressure gradient around wingtips, from
slightly raised below to lowered above.
These forces are like a spade bit in a drill
used to stir your coffee, or a single beater
in your eggbeater. Unlike a propeller,
they impart a rotary force without adding
thrust.
R/C Soaring Digest
Okay, true, ahead of the wingtip air is
accelerated both in a swirl and up toward the
low pressure atop the wing. So it speeds up.
With poor pressure recovery near wingtips
some of that speed remains in the trailing
vortex flows, rather than transforming back
into pressure. So that’s one reason there
is low pressure in the center of the trailing
vortex that drags back on a wing. But we’re
going to focus mainly on the y-z plane forces.
The pressures that form the wingtip vortex
extend ahead and outward from the wingtip.
They are a mix of the x-z pressure gradients
that form upwash ahead of a wing and the
y-z pressure gradient around the tip of a
wing. These pressures make the wingtip
vortex swirl up and around the end of the
wing. The vortex (y-z) component of motion
around the end of the wing lowers pressures
below the wingtip and raises pressures
above the wingtip. Pressure gradient energy
is exchanged for rotational velocity energy.
This is another way in which pressures are
lowered in what becomes the center of the
trailing vortex. Again, the lowered pressures
Figure 11: Prandtl falsely visualized the ‘bound
vortex’ as symmetrical, and therefore imparting
no downwash. He then (falsely) concluded
that rotational velocities of the trailing vortices
must create (net) downwash. He apparently
ignored that this would lead to unrealistic
wake downwash patterns, sticking with his
‘rectangular’ downwash pattern for elliptical
spanloaded wings, probably moderately
accurate.
June 2014
A causally backwards notion
Prandtl substituted a symmetrical ‘bound vortex’ around a lifting
line for a wing. This was just for engineering purposes, but it
appears Prandtl believed in vortex symmetry. In this idealized
bound vortex, upwash ahead would equal downwash aft, so he
apparently figured the source of [net] downwash must be from
somewhere else -- the inner, downward flows of the trailing vortices.
Trailing vortices don’t cause net-downwash, though they do ‘suck’
backwards on tip downwash velocities. Trailing vortices are caused
Falsely idealized ‘bound
vortex’ has equal upwash by wing net-downwash and pressure gradients around wingtips.
and downwash.
Trailing vortices
An incorrect pattern of net-downwash
follows from Prandtl’s idea that trailing
vortices cause downwash.
If Prandtl’s idea that wingtip/trailing vortices cause ‘net-downwash’ were true, then netdownwash would never be in his rectangular or elliptical patterns. The overlapping downward
vortex velocities would sum to a pattern of ‘net-downwash’ strongest near vortex centers (black
line).
The opposite causal sequence is true -- a wing’s net-downwash and pressure differences
around wingtips cause wingtip/trailing vortex formation.
17
in the trailing vortex drag back on the wingtip.
Equivalently, along each streamline spiraling up around the wingtip air is
accelerated up, centripetally (in toward the center of the spiral), and back.
The strongest accelerations are near the center of the forming wingtip
vortex. For elliptical wingtips this strongest acceleration is at or just inboard
of the wingtip. For BSLD wingtips the vortex center is further in. Pressure
is used up accelerating air along streamlines. The poor pressure energy
recovery near wingtips means this process is not completely reversed.
These lowered pressures persist as the low-pressure center of the wingtip
vortex.
Rotational velocities of the trailing vortex centrifuge its
low-pressure center. Centrifugal forces are in red.
The centripetal (x-z) acceleration around the wingtips makes the rotational
velocities that are the trailing vortex and that further centrifuge the low
pressures at the center of the trailing vortex.
Vortex rotational velocities are also reinforced by the net-downwash pattern
of the wing. An elliptical lift distribution wing’s approximately rectangular
pattern of net-downwash makes a powerful addition to the rotational
momentum of the trailing vortex, as does the strong pressure gradient
around its wingtip.
A BSLD wing’s transition from central downwash to wingtip vortex upwash
mixed downwash momentums make a more complex and softer influence
on trailing vortex formation. Inboard of the crossover point the downwash
momentums are strong and reinforce trailing vortex rotation. Outboard of
the crossover point downwash momentums are weaker but fight vortex
rotation. So the downwash momentums of the BSLD wing help to make its
trailing vortex more diffuse. In combination with a more diffuse pressure
gradient around wingtips, the resulting BSLD trailing vortices are broader,
more diffuse, and have lower rotational velocities and weaker centrifuging
near their centers. That means higher-pressure centers for less conflict with
pressure energy recovery and less vortex drag than for elliptical wings of
similar lift and root bending moment.
The low-pressure center of the trailing vortex pulls
back on the wingtip and the air passing over and
under the wingtip. Accelerating these flows destroys
pressure energy recovery, a second reason the air
behind wingtips is of low pressure. The difference
between higher pressures ahead and lowered pressures aft is trailing vortex pressure drag. The lowpressure center of the trailing vortex also pulls forward
on wake air.
Figure 12: Trailing vortex pressure drag.
Again it should be emphasized that wingtip vortices are being formed by
asymmetrical forces and momentums and only approach symmetry well
behind the wing, as trailing vortices, at about the time they break up into the
unevenness one observes in the aft part of contrails.
See Figure 12
18
R/C Soaring Digest
The persistent false notion that wings
gain lift only from downwash, rather
than from Lanchester’s reversal of
upwash to downwash momentums
Aerodynamic misinterpretations persist.
Similarly to how the false notion of
‘longer path/equal transit times’ was
perpetuated by the misBernouligans
through most of a century, there are
others. One was carried forward by
almost everyone. It is re-perpetuated in
Robert T. Jones otherwise excellent Wing
Theory (1990). It’s the idea that lift comes
only from downwash.
In 1894, 1897, and 1907, with his wave
theory of lift, Frederick W. Lanchester
had correctly asserted that lift comes
from the reversal of upwash momentums
ahead to somewhat greater downwash
momentums aft.
Oddly, Jones, who champions
Lanchester, had a partial grasp on
Lanchester’s correct explanation of
lift. He even includes an extremely
rare mention of Lanchester’s theory of
wave lift! (Which predated Lanchester’s
conceptual development of ‘circulation
lift.’) In Wing Theory Jones writes,
Recall that the lifting wing in twodimensional flow does not require
a continuous supply of energy to
maintain its course if its speed is
subsonic. The wing rides on a kind of
wave having fore and aft symmetry,
June 2014
with upwash ahead and downwash
behind. – Robert T. Jones, Wing
Theory
That’s close, if only true in a universe
without turbulence. Even inviscid 2D
wing polars show drag as the result of
turbulence, ‘bubble’ formation (partial
flow detachment), and stall. Thus even
infinite wings require energy input to
keep going.
But then Jones slips back into a
conventional misunderstanding.
The fact that the wing derives its lift by
imparting downward momentum to the
air… We can then think of the wing as
encountering a circular jet of air with
diameter equal to the span of the wing
and as deflecting this jet downward.
That’s a scoop notion combined with
the false idea that the momentum
of horizontally flowing air ‘deflected’
downward is all that makes lift, rather
than Lanchester’s sum of the reversal
of upwash to downwash momentums.
It’s the notion that wings stay up only
by throwing air down. And in some
interpretations this false notion of
‘downwash’ is plopped right into the
center of Prandtl’s analysis of BSLD
wings. Referring to Prandtl’s 1932 paper,
Jones writes,
This problem was considered
many years ago by Prandtl. Prandtl
suggested that the integrated or
averaged bending moments along
the span be used as a constraint…
Thus for minimum drag with limited
bending moment and given lift, the
downwash should have a parabolic
distribution. The span load distribution
corresponding to this downwash can
be obtained…
So: Whenever you see the diagrams of
rectangular downwash for elliptical wings
or parabolic downwash for BSLD wings,
interpret the vaguely labeled ‘downwash’
as ‘net downwash.’ Also, note that
‘parabolic downwash’ is what lift forces
do to air before the wingtip vortex bends
that air upwards.
A more detailed diagrammatical
summary. Adding the forces.
The following two diagrams sequentially
trace how ELD wings and BSLD wings
create the wingtip vortices, trailing
vortices, trailing vortex pressure drag
profiles, and the effect on drag reduction
or thrust at BSLD wingtips. Not to scale.
19
Rectangular net downwash pattern
Bell-shaped load distribution
Elliptical net downwash pattern
Rectangular 2D sectional drag pattern
Elliptical 2D sectional drag pattern
Asymmetrical wingtip
vortex velocities
Actual location of BSLD wingtip vortex
needs more research! Estimates range
from 2/3 to 91% of half-span.
Net downwash
Net downflows
The bottom line is that portions of
wingtips flying in vortex upwash may gain
a bit of thrust if their Cl/Cd (L/D) glide
angle is greater than the angle of vortex
upwash. That thrust will at least make
drag reduction by fighting trailing vortex
pressure drag. Whether one can get an
actual push out there isn’t so important,
but if so, that happens only when thrust
is greater than vortex drag.
See Figures 13a and 13b
BSLD span/2
Net downwash
Trailing vortex velocities
20
Pressure gradient
around wingtip
BSLD span/2
Pressure gradient
around wingtip
Elliptical load distribution
Ellipse span/2
Prandtl bell-shaped lift distribution
Ellipse span/2
Prandtl elliptical lift distribution
For a BSLD wing, the vortex center is
inboard of the crossover point. If there is
a region of actual wingtip thrust it starts
outboard of the crossover point. Even for
drag reduction from a winglet riding in
vortex upwash, drag reduction will start
where wing airfoil section’s L/D ratio
makes an angle shallower that vortex
upwash. That also will happen outboard
of the crossover point, since near the
crossover point upflow angles approach
zero.
Trailing vortex velocities
Upflows
Net downflows
Upflows
Figure 13a: From load distributions to trailing
vortex velocities
R/C Soaring Digest
Trailing vortex pressure profile
Trailing vortex pressure profile
Trailing vortex pressure drag is less severe than
pressure profile because of narrowing wingtips
Vortex pressure drag is low on skinny wingtip.
(D = P x A)
Sum sectional and trailing vortex pressure drag
Sum sectional and trailing vortex pressure drag
Thrust
High
wingtip
drag
Figure 13b: From trailing vortex pressure
profiles to BSLD wingtip drag reduction.
June 2014
BSLD span/2
Ellipse span/2
Prandtl bell-shaped lift distribution
BSLD span/2
Ellipse span/2
Prandtl elliptical lift distribution
Upflows
Vortex center
Crossover
to up flow
Net downflows
Drag
Wingtip
L/D
Net downflows
Upflows
21
ELD-BSLD compromises
A page from Bill Kuhlman’s five-part
“Twist Distributions for Swept Wings”
summarizes the concept that at least
part of the benefits of a BSLD wing can
be achieved with a more tapered lift
distribution than elliptical.
See Figure 14
Figure 14: Jones low induced drag wing planform is a
compromise between ELD and BSLD wings.vi vii viii
22
R/C Soaring Digest
Proverse roll-yaw coupling.
See Figures 15 and 16
Piloting is easiest when airplanes are well behaved and
do what a pilot wants with minimal correction. To make
coordinated turns pilots generally correct adverse rollyaw coupling with rudder deflection. Full-scale pilots
use the rudder pedals. Model pilots use the left stick. Or
they program in aileron-rudder mix, generally with aileron
‘differential.’ Model flying wing pilots trust to fins. That’s
technically sloppy, but Zagis get by just fine. It’s possible to
do better. There is a long history of designing airplanes with
neutral or even ‘proverse’ roll-yaw coupling. However, there
are always tradeoffs.
Lift
adverse yaw
0
0
drag increase
Adverse roll-yaw coupling
Drag
Drag
Left wing: In the adverse
part of the L/D curve,
drag increases with lift
Right wing: In the adverse
part of the L/D curve, drag
decreases as lift decreases
Proverse roll-yaw coupling. Plane rolls right, yaws right.
Lift
The problem: In most airplanes and model airplanes when
aileron deflections roll the plane to the right the nose usually
(with exceptions) yaws left. Or visa versa. That’s adverse rollyaw coupling. It’s generally present even at cruise speeds,
but it’s strongest at times of high coefficient of lift, at a high
angle of attack, approaching stall. That happens most often
in high G maneuvers or when a plane is near its slowest
speed. For example, when a plane loses power shortly after
takeoff pilots sometimes try to turn back to the runway while
gliding at near-stall angles of attack (AoA). That combination
of maneuvering when adverse yaw is strongest threatens
spin without sufficient altitude for recovery.
roll
roll
proverse yaw
0
Lift
Dragerons and thrusterons.
Adverse roll-yaw coupling. Plane rolls right, yaws left.
Coordinated turn requires rudder correction
Lift
Flying the undersides of shifting lift/drag curves with
winglets in vortex upwash.
The effects of roll-control surface deflections on unequal
trailing vortex drag.
0
drag decrease
Drag
Left wing: In the proverse
part of the L/D curve,
drag decreases as lift
increases
Drag
Right wing: In the proverse
part of the L/D curve, drag
increases as lift decreases
Note: The above simplied lift/drag curves work for AoA
roll-control devices, e.g. ‘tiperons.’ Cl/Cd curves for aileron
deflections are more complex. Vortex influences add a
third level of complexity.
Figure 15: Roll-yaw coupling
And for many, the main benefit of looking at BSLD and
proverse roll-yaw forces will be greater understanding of
what happens around wings.
June 2014
23
Even though engineering is beyond this
article, it contains hints for building a
proverse roll-yaw coupling engineering
methodology via adjusted Cl/Cd curves.
And we’ll get to practical examples
of what should work and what won’t,
and problems with a couple historical
approaches. We’ll look at the Horten
brothers’ proverse approach, the high
adverse roll-yaw coupling of the Wright
Flyers, implications for elliptical and
bell-shaped lift distributions, and an
imaginary Piper Cub equipped with
tiperons – rotating wingtips.
Please be aware that structural and
aerodynamic benefits of BSLD, proverse
roll-yaw coupling, induced wingtip thrust,
and flying wings are separate subjects,
though usually interrelated. It’s true that
most solutions for proverse yaw will have
a lift distribution closer to bell-shaped
than elliptical, but in some optimums lift
at wingtips may even be negative. And
there are proverse roll-yaw solutions for
standard-planform aircraft and canards
as well as for flying wings.
Inverted cambered airfoil wingtip
Cl
For most purposes this is the most adverse solution.
Proverse yaw section is only at strongly negative lift.
Symmetrical airfoils are also poor for proverse yaw.
Normally-cambered airfoil wingtips are most likely
to produce proverse roll-yaw.
L
0
R
Cd
Strongly reflexed ailerons of some swept flying
wing wingtips use moment arm to fight pitching
moment. The result may approximate negative
camber in negative lift. That may work, but normally
cambered tiperons would be more proverse.
Figure 16: Inverted cambered wingtips generally make adverse
roll-yaw coupling forces.
24
R/C Soaring Digest
Pre-summary:
The easy but partial argument: Within the
rising air outboard of the vortex center
and even a bit outboard of the ‘crossover
point,’ by changing angle of attack or by
aileron deflection, a lightly loaded wingtip
may gain thrust in the way a glider gains
thrust within rising air. A component of
the lift force may be in the direction of
flight -- thrust. For a bit of wing section
to contribute thrust the air must be rising
at an angle steeper than the section’s
actual glide ratio, which will be worse
than its 2D sectional Cl/Cd because of
wingtip pressure difference losses. And
not all such thrust will be greater than
trailing vortex pressure drag, but that
doesn’t matter for proverse roll-yaw
coupling. What matters is that on left and
right aileron or tiperon deflections, left
and right changes in thrust and (adverse)
changes in trailing vortex pressure
drag add up such that a right roll is
accompanied by right yaw.
With wingtips in significant vortex
upwash: For the left wingtip, a roll to the
right is accompanied by the left wing
pushing forward, for proverse roll-yaw
coupling. The right wingtip, with aileron
deflected upwards, may decrease thrust
as it drops, again for proverse roll-yaw
coupling.
The more complete argument: All
that is required for proverse roll-yaw
coupling is that sectional airfoils operate
in negatively sloped portions of their
June 2014
adjusted lift/drag curves, where lift and
drag move in opposite directions. That
usually means proverse roll-yaw forces
are generated from airfoil sections in
fairly low or negative lift, in relation to
local flows.
Proverse roll-yaw coupling analysis
requires adding a number of effects.
Flying winglets in vortex upwash shifts
their Cl/Cd (lift/drag) curves up (added
lift) and to the left (reduced drag). Wingtip
vortex formation lessens pressure
differences between upper and lower
wingtip surfaces, shifting Cl/Cd curves
down (lessening lift). Trailing vortex
pressure drag and airfoil drag shift the
curve back to the right (increased drag).
The summed result is that tiperons
or even lightly loaded wingtips with
ailerons can often operate in the area of
their adjusted Cl/Cd curves, below the
‘drag bucket,’ where an increase in lift
makes a decrease in drag. That makes
a coordinated turn, a proverse roll-yaw
coupling, without the use of a rudder.
But even a standard cambered elliptical
wing, exerting no lift while briefly in a
ballistic (zero gravity) parabolic trajectory,
will generally be operating in a negatively
sloped area of its lift/drag curve, and will
respond proversely to aileron deflections.
In contrast, at a wing’s highest Cls,
adverse roll-yaw forces are strongest.
As we’ll see, tiperons will generally be
more proverse than ailerons. That’s partly
because tiperons can maintain proverse
angles of attack regardless of the
incidence of the main wing. A morphing
tiperon would be able to set variable
angles of attack along its span, to
optimize drag reduction or thrust within
the different slopes of vortex upwash.
The morphing could be via aileron.
Summary of the wingtip vortex
upwash in which a lightly loaded
wingtip may fly
As said, a winglet, whether vertical or
horizontal, can catch a bit of thrust
if it extends into the upward swirl of
flows around a wingtip. That seems
simple enough. It isn’t. The location and
profile of the upward flows is a result of
several forces. For a BSLD wingtip the
downward force on air from lift is small,
and too weak to overcome the upward
forces from pressure differences that
make the upward wingtip vortex swirl.
The resulting flows are from a balance
of lift forces and wingtip vortex forces.
Since the wingtip vortex center exerts
no downward or upward force, it is only
further out that wingtip vortex upward
forces exceed wing lift downward forces
on air. Thus the vortex center is inboard
of the ‘crossover point.’
All wings have to fight the downward
‘induced angle of attack’ (caused by lift
forces on air). But outboard of the vortex
center this downward angle of flows
is lessened even before the crossover
point. So outboard of the vortex center
the upward wingtip vortex forces lower
25
drag, perhaps transitioning further out to
actual thrust.
An additional paradox is that ‘netdownwash’ for a BSLD wing is generally
pictured as elliptical, while we know that
outboard of the ‘crossover point’ flows
go up. What gives? Probably the elliptical
pattern of net-downwash ignores vortex
upwash.
Complexity
Back when I posted the beginnings of
this article to Nurflugel I had an idea,
not necessarily new but new to me, that
toward lower drag or even thrust. And
every deflection of roll control surfaces
changes lift distribution unequally, left
and right, which changes the strength,
spread, and spanwise location of trailing
vortices and associated drag, again,
unequal left to right, making adverse
roll-yaw coupled forces only quantifiable
with a 3D analysis. Since a BSLD wing
can have proverse roll-yaw coupling,
its proverse airfoil forces exceed these
adverse trailing vortex pressure drag
forces.
of lift/drag polars, and the discussion of
how the conflict of BSLD downward lift
forces on air with upward wingtip vortex
forces on air determines the ‘crossover
point’ and the ‘induced’ angles of flows
outboard of the wingtip vortex center.
Proversely flying the lift/drag polar
‘drag buckets’ and their shifts with
AoA and camber changes and with
their span location in vortex up-ordown flows
The definitive tools for conceptually
analyzing roll-yaw coupling are vortex-
In aerodynamics a polar diagram graphs two
interdependent variables.
proverse roll-yaw forces are from areas
of an airfoil’s (2D) lift/drag curve where as
lift increased drag decreased, making a
coordinated turn. (I think the formatting
of my rather crude graphs worked when
Yahoo sent them to members as emails,
but not on the site. Oh well.) Comments,
corrections, and time have helped.
But things are not as simple as that
partially scrambled starting point. Each
truth roused additional complexity.
Ailerons make more complex shifts of
lift/drag curves than do simple angle
of attack changes. Wingtips fly in
vortex upwash, usually shifting curves
26
This is a conceptual article. I expect the
true experts to publish a comprehensive
and quantitative article in the not-toodistant future. Do I have something that
will help? Perhaps, or perhaps what I
write here will be old hat. Still, it’s my
observation that, within aerodynamics,
computational fluid dynamics (CFD)
and wind tunnel data produce excellent
engineering even while concept lags. So
perhaps I can make a contribution, or at
least stimulate the discussion.
Two areas I haven’t heard used by others
for roll-yaw coupling analysis are the use
shifted lift/drag polar diagrams. These
are potentially also good tools for design,
though that would require building a
reliable engineering methodology. There
are of course other approaches, from trial
and error to CFD, excellent for results but
usually poor for comprehension. The goal
is to leave readers with understanding
and concept sufficient for gut-level
guidance.
It’s inadequate to use standard lift/
drag polar diagrams to analyze rollyaw coupling. That’s because standard
‘polars’ show the lift and drag of a wing
R/C Soaring Digest
section at various angles of attack or
aileron deflection in relation to local
flows, and generally before wingtip
pressure-difference losses. But subsonic
flows are always either tilted down
(induced angle of attack) or up (wingtip
vortex upwash) within the 3D forces of
lift. Thus L/D polars don’t initially show
lift and drag in relation to flight path,
which is what counts. We’ll adjust L/D
polars for induced angle of attack and
wingtip vortex upwash at different parts
of various wings’ spans.
These polars adjusted for vortex upflows
or downflows then allow a consistent
rule: Roll control deflections of wing
sections flying in positively sloped
areas of the vortex-shifted L/D curve
make adverse yaw. That’s where an
increase in lift makes an increase in drag.
Control deflections of wing sections
flying in negatively sloped areas of the
vortex-shifted L/D curve make proverse
yaw―where an increase in lift makes a
decrease in drag. The curve is king.
Portions of ELD wings with roll-control
surfaces fly the in local down-flows
(induced AoA, net downwash, or vortex
sink), which makes for adverse roll-yaw
coupling. Ailerons or tiperons on BSLD
wingtips or other lightly loaded wingtip
extensions fly in wingtip upwash. There
they are more likely to exert proverse rollyaw coupling forces.
Polar diagrams: In aerodynamics a polar
diagram graphs two interdependent
June 2014
variables. That’s in contrast to simpler
mathematical functions with an
independent variable unaffected by
a dependent variable. For example,
when Galileo dropped a weight from
the leaning tower of Pisa the changing
velocity of the weight over time didn’t
affect time. Disambiguation: In sailing,
‘polar diagram’ just means ‘circular,’ a
graph of headings-in-relation-to-the-wind
(the independent variable) versus speed
(the dependent variable.) Which is a poor
use of the data and bad third-grade
arithmetic, but that’s another story.
Angle of attack devices: A tiperon is a
wingtip that pivots around its quarter
chord line. Wingeron wings rotate in
opposite directions for roll control, with
pitch controlled by an elevator. Pitcheron
wings also rotate in opposite directions
for roll control, but collectively change
angle of incidence (decolage) in relation
to the chord line of a fixed horizontal
stabilizer. They control pitch with variable
longitudinal dihedral or decalage.
Wing-warping changes AoA without
changing camber, supposedly. (The
upper fabric of the 1910 Wright Model
B wing was secured only at leading and
trailing edges. In flight it would belly up,
increasing camber!ix) Wing-warping was
used on the Wright flyers, the Bleriot XI
(1909), the Fokker Eindecker monoplane
(1915), and others till about 1915. After
1915 ailerons predominated, mainly
because they allowed stronger wing
structure and thus better roll control. AoA
changes shift L/D along the lift/drag polar
curves.
Pure camber changing devices are
commercially rare or nonexistent.
Perhaps someone working with wing
morphing has built one. Camber changes
move the L/D curve nearly vertically.
Camber/AoA changing devices include
ailerons, flaperons, and elevons. As an
aileron is deflected down it increases
camber and AoA. If we hold the
incidence of the airplane constant we
can graph the L/D changes with aileron
deflections. Deflections of ailerons move
L/D by a combination of the near-vertical
camber-change shifts of the L/D curve
and the AoA shifts along the L/D curve.
Drag buckets. Flying wingtips with rollcontrol surfaces beneath the wingtipvortex-adjusted drag bucket for proverse
roll-yaw coupling: The nearly vertical left
portion of a lift/drag polar is the drag
bucket. It’s where drag is lowest. Most
airplanes are designed to cruise near the
top of their drag bucket, ideally where the
slope of Cl/Cd is steepest, for best L/D
and fuel efficiency at that cruise speed.
The drag bucket for an entire airplane
is in relation to flight path and is not the
same as drag buckets of wing sections in
relation to local flows (given by standard
polars), which need to be adjusted in
relation to flight path, which is what
counts.
To make proverse roll-yaw coupling
forces a wing section with roll-control
27
surfaces must fly beneath the drag
bucket of the appropriate, local-flowangle-adjusted Cl/Cd curve.
CFD for design. Three conceptual
approaches to achieving proverse yaw.
There is one design approach and three
conceptual approaches to looking at
roll-yaw coupling. The definitive design
method is via CFD (computational
fluid dynamics) 3D analysis. That’s
beyond the scope of this article. It’s
superb for design, great for verification
of conceptual analyses, and doesn’t
necessarily offer good explanations of
what is going on.
Conceptually there are three required
approaches to understanding roll-yaw
coupling:
• First is gaining thrust or reducing drag
by flying a lightly-loaded wingtip with
control surfaces in vortex upwash. When
equipped with ailerons the differential
effects of left and right deflections on
thrust or drag can make proverse yaw
forces. BSLD wings have such lightly
loaded wingtips as well as structural and
drag advantages; these topics are the
focus of most discussions.
• Second is conceptually adjusting 2D
airfoil Cl/Cd polar diagrams, to see where
lift and drag move in opposite directions
upon roll-control surface deflection, for
coordinated turns.
• Third is adverse yaw forces from
changes in trailing vortex pressure
28
drag caused by aileron deflection. Rollcontrol surface deflections unequally
affect lift distribution; lift distribution
changes affect the location and strength
of wingtip vortices. Wingtip vortices
affect the location and pressure profiles
of trailing vortices and thus the induced
or vortex drag profile of the wing. For
positively lifting wingtips with ailerons,
trailing vortex pressure-drag is always an
adverse yaw force! But for BSLD wings
the vortex center is probably sufficiently
inboard that this third effect is not the
dominant yaw producing force.
So trailing vortex formation is a function
of lift distribution. At one end of a
continuum of lift distributions, if Cl is zero
across a wing’s span it doesn’t produce
a vortex. In contrast, if aileron deflection
increases winglet lift till lift distribution
approximates elliptical the vortex center
will move outward to near the tip,
eliminating the possibility of the tip riding
in vortex upwash. Aileron deflection
then exerts the usual adverse roll-yaw
coupling typical of ELD wings.
In various planforms and lift distributions
these three effects can reinforce or fight
each other. We’ll make sense of it all.
Thrusterons and dragerons
There are two interrelated approaches
to design for proverse roll-yaw coupling.
Each is limited by the adverse vortexdrag yaw effects of aileron deflection.
The first is to extend a carefully designed
lightly-loaded roll-controlling bit of
wing into the rising flows of the wingtip
vortex. It will either be a tiperon or will be
equipped with ailerons. Such a wingtip
can act like a glider in rising air. On
deflection to slight positive lift (in relation
to local vortex upflows) it can provide
thrust that can help yaw the airplane
in the direction of roll. The thrust may
not be absolute – such thrust may not
exceed vortex pressure-drag (which
adversely increases with increasing lift).
But that thrust will lower drag. We could
call that the ‘thrusteron’ approach.
As the opposite wingtip lessens its
lift it may increase drag, also creating
proverse roll-yaw forces. We’ll see the
specific conditions where this works.
Even within wingtip vortex upwash a
thrusteron must be lightly or negatively
loaded to make proverse roll-yaw forces.
If it is too heavily loaded it will operate
in an adverse yaw area of its (shifted) lift
drag curve, it will chase the crossover
point further out toward the wingtip,
and will create adverse trailing-vortex
pressure-drag yaw forces.
Second, in regions without sufficient
vortex upwash to make proverse yaw
on aileron or AoA deflection: to create
proverse yaw forces, portions of wings
with control surfaces must operate in
regions of their lift/drag polar curves
where control input moves lift and drag
in opposite directions. These proverse
regions of the lift/drag curve are generally
at low or negative coefficients of lift, in
R/C Soaring Digest
turn found at local low or negative angles of attack, for example,
during low-or-negative G pushovers. High speeds may allow some
airfoils to operate in a low-but-positive-lift proverse region of their Cl
curves. Roll-control surfaces that change wing drags such that roll
and yaw are coordinated could be called dragerons. (However, the
term is taken. It refers to devices that increase drag to control yaw.
Split ailerons are an example that also work as air brakes.)
The two effects can be combined in tiperons or in wingtips with
ailerons operating in vortex upwash. There is a continuum from
lowering drag to increasing thrust while increasing lift.
As mentioned, the effects of aileron deflection on vortex pressure
drag are roll-yaw-adverse. They either make adverse roll-yaw
coupling worse or partially fight proverse yaw forces. Because of
vortex location, adverse roll-yaw vortex pressure-drag forces will be
stronger for elliptical wings than for BSLD wings if each has ailerons
near their tips.
See Figure 17
For tiperons the bottom line is the vortex-upwash-shifted lift/
drag polar. For wingtips with ailerons the bottom line is the vortexupwash-shifted aileron deflection Cl/Cd curves, that we’ll look at
later.
Thrusterons on a Cub
Imagine a short-takeoff-and-land (STOL) airplane, perhaps a Super
Cub, flying slowly with enough altitude for safe recovery. Its pilot
isn’t using its ailerons or rudder. Instead it has ‘tiperons’ protruding
outward from its wingtips. They have a three-foot chord, are five
feet long, and sport a symmetrical airfoil to avoid pitching moment
forces. They are mounted on shafts just ahead of their quarter-chord
lines. The mechanical linkage is such that the pilot can rotate them
a few degrees in opposite directions with left-right movements of
the stick, but collectively they align with the local airflow. They aren’t
an optimum, but they work. Because the plane is mushing along
there is steep vortex upwash outboard of its normal wingtips. The
pilot tips their control stick to the right. The left tiperon increases
June 2014
Untwisted wing at zero-lift angle
of attack makes no vortex!
BSLD wing with mild upward
wingtip aileron deflection. Vortex
moves inwards and weakens. Lift
and vortex drag are both lessened
making adverse roll-yaw forces.
But negative lift likely puts the
wingtip in negative Cl/Cd slope,
overcoming the adverse vortex
pressures, for net proverse yaw.
BSLD wing in lift forms vortex
inboard of wingtip
BSLD wing with mild downward
wingtip aileron deflection. Vortex
moves partway out toward wingtip and strengthens. Increasing
vortex drag and moment arm
lessens BSLD proverse roll-yaw
coupling.
BSLD wing with strong downward
wingtip aileron deflection. Vortex
moves to wingtip and strengthens, so wingtip doesn’t fly in
vortex upwash. Roll-yaw coupling
force is adverse.
Figure 17: Adverse roll-yaw coupling effects of trailing
vortex pressure drag.
29
Re = 500000
Mach = 0.0000
NCrit = 9.00
NACA 4412 =
NACA 0012 =
NACA 4412_Inverted =
Cl(Cd):
1.5
1.0
0.5
Cl
0.0
-0.5
-1.0
0.00
0.01
0.02
0.03
0.04
Cd
Page 1 of 5 - Drawn by Profili 2.30a Pro on data processed by XFoil - Copyright (C) 1995-2011 - All rights reserved.
Figure 18: Camber shifts the Cl/Cd curves vertically. Profili plots by Adam Weston
30
R/C Soaring Digest
its AoA relative to local upwash from zero to three degrees. Just
like a glider in a thermal updraft it gains lift and thrust, rolling and
yawing the plane to the right. The right tiperon hits a negative
angle of attack relative to local up flows, for negative lift and
increasing drag, also rolling and yawing the plane to the right.
Our pilot-designer has not achieved the energy or structural
efficiency possible with BSLD wings. But with a small sacrifice
of efficiency he has used the stronger upwash outboard of a
standard elliptical or Hershey-bar wing to achieve stronger
proverse roll-yaw coupling than would be expected from a true
BSLD wing. Trade-offs.
The pilot improves on his design. He uses cambered airfoils for
his tiperons, and adds trim tabs set so that in slow straight flight
his tiperons maintain low lifting angles of attack relative to the
vortex upwash. The tiperon winglets provide thrust, salvaging
a bit of energy from wintip vortices. For cruise he gets fancy.
At cruise speeds upwash outboard of wingtips is weaker and
a bit too flat to gain much thrust. But he knows that at low
positive angles of attack his cambered airfoils will still move roll
and yaw in coordination. He builds trim tabs adjustable in flight
and searches for optimums of efficiency and proverse roll-yaw
coupling.
If the pilot operates his tiperons at a coefficient of lift similar to
that of the rest of the wing they become merely a strongly lifting
extension of the wing (good for short landings, but adverse).
That moves the wingtip vortex outward so that the tiperons are
no longer flying in vortex upwash. Then as he moves his tiperon
stick left and right the lift and drag of each tiperon move in the
same direction for standard adverse roll-yaw coupling. Thus the
pilot has to use his rudder. Wingtips with control surfaces only
produce proverse roll-yaw coupling when operated at fairly low
coefficients of lift, and even that depends on the Cl/Cd curve of
the airfoil.
Figure 19: Aileron deflection changes both AoA
and camber.
The factors
Whether roll-yaw coupling from control surface deflection is
adverse, neutral, or proverse is determined by several factors
plus their interactions.
These are:
1: The effects of aileron or tiperon deflection on wingtip/
trailing vortex drag. This is always an adverse roll-yaw force,
which must be overcome by proverse forces.
2: The effects of AoA (angle-of-attack) roll-control devices
(tiperons, wingerons, and pitcherons, and AoA wing-warping).
AoA devices slide L/D along the lift/drag curve. This is a twodimensional, airfoil sectional analysis in which freestream
velocities are assumed parallel to flight path. See figure 15
3: Camber change shifts the entire lift/drag curve nearly
vertically. That’s critical information, though pure camberchanging devices aren’t in use.
See Figure 18.
4: The effects of camber-and-AoA-changing roll-control
devices (ailerons, flaperons, elevons, and the Wright’s wing
warping).
See Figures 19 and 20
June 2014
31
5: Angles of attack – what part of the appropriate Cl/Cd curve the
wingtips are at.
Cl
AoA 5
1.5
Aileron deflection
Aileron Cl/Cd curves are more adverse than their AoA
curves. At each pre-deflection AoA (black dots) the
aileron Cl/Cd curve is steeper than the AoA curve.
+8°
0°
-8°
AoA 4
1.0
.5
AoA 3
AoA 2
See Figure 21 and 22.
AoA 1
-.5
Aileron deflection
0
+8°
-1.0
6: The location of roll-control devices within wingtip vortex upwash.
Here our analysis becomes 3D. Generally for proverse yaw, wing
segments with control surfaces are designed to fly in wingtip vortex
upwash. Like little glider wings they gain thrust (or at least reduce
drag) by riding in up-currents. For the L/D of the airplane in relation
to freestream velocities (as opposed to local upwash velocities), this
shifts the lift/drag curve up and to the left, toward lower drag or even
negative drag, thrust.
0°
-8°
Cd
Along each blue-black line the angle of incidence of the
airplane remains constant but aileron deflection varies.
Blue-black lines are aileron Cl/Cd curves by deflections
from five AoAs. Proverse areas of the aileron Cl/Cd
curves are in blue; adverse in black. Not to scale.
Figure 20: Aileron Cl/Cd curves are more adverse than
tiperon (AoA) Cl/Cd curves. Cambered tiperons are
probably the best choice for achieving proverse rollyaw coupling.
7: The effects of lift distribution on location of wingtip vortex location,
strength, and vortex upwash or downwash. Generally segments of
wings with roll control surfaces must be lightly or negatively loaded
to fly in wingtip vortex upwash; a strongly loaded segment will move
wingtip vortex upwash outboard of itself. Hence lift distributions
designed for proverse yaw usually (but not strictly or always)
approximate a ‘bell shaped lift distribution’ (BSLD) rather than the
more common elliptical lift distribution.
8: The effects of trailing vortex pressure drag on wingtip thrust or
drag.
See Figure 23
9: The effects of roll-control deflections on lift distribution; wingtip
and trailing vortex location and the strength; location, and pressure
profile of trailing vortex drag. E.g. when an aileron is deflected down
it increases lift, shifting the wingtip vortex outward.
10: Different airfoils have differently shaped lift/drag curves and will
be better or worse for proverse roll-yaw control wingtips.
See Figures 24, 25 and 26
11: Other: Spanwise flows and flow attachment, etc., mostly ignored
here.
32
R/C Soaring Digest
Figure 21: Aileron Cl/Cd curves (black) are only proverse at lower Cls than AoA Cl/Cd curves. Profili plot by Adam Weston.
June 2014
33
2D tiperons (unrealistic)
Lift
Cl
2D sectional drag
Total aero force
Stall
3D tiperons. Wingtip vortex upwash shifts the Cl/Cd curve up and to
the left, potentially making thrust, but vortex drag shifts it back up
to the right, probably eliminating thrust. But tiperons in vortex
upwash will at least lower drag.
Cl
2D sectional drag
plus vortex drag
Adverse
Thrust
after
vortex
drag!
Cd
Proverse
Thrust
Stall
Cd
Stall
Proverse
3D tiperons in wingtip vortex upwash
Cl
Thrust before
vortex drag!
2D sectional drag
plus vortex drag
Thrust
Lift in relation to local flow
Sectional airfoil drag plus vortex drag
Stall
Total aero force
Flight path drag
Note that cambered tiperons would
raise the Cl/Cd curve further, and would
be more proverse than symmetrical
tiperons.
Cd
Proverse
Lift in relation to local flow
Sectional airfoil drag plus vortex drag
Total aero force
Flight path drag
Figure 23: Upflow shifts with trailing vortex pressure drag.
Even where drag is lowered, shifting the Cl/Cd curve to the
left, there is also trailing vortex pressure drag, which shifts
the curve back to the right. Whether thrust is achieved or
not, tiperons in vortex upwash can usually lower drag.
Figure 22: Upflows shift tiperon Cl/Cd curves up and to the left, for higher lift and less drag. Locating a winglet within vortex upwash
raises its lift coefficient in terms of flight path. That is, it gains lift from upwash. And its lift vector may angle ahead, attempting
thrust, or at least fighting trailing vortex pressure drag, moving the Cl/Cd curve to the left.
34
R/C Soaring Digest
Figure 24: The AG455ct is fine for discus-launch gliders, but has a narrow range of negatively sloped Cl/Cd curve.
Profili plot by Adam Weston.
June 2014
35
Figure 25: The RG15 is a great airfoil with a broad drag bucket, but flying its negatively sloped lift/drag curves for proverse roll/yaw
could be tough, and might result in sudden oversteer if wingtips were in negative lift. Profili plot by Adam Weston.
36
R/C Soaring Digest
Re = 500000
Mach = 0.0000
NCrit = 9.00
NACA 4412 =
NACA 0012 =
NACA 4412_Inverted =
Cl(Cd):
1.5
1.0
0.5
Cl
0.0
-0.5
-1.0
0.00
0.01
0.02
0.03
0.04
Cd
Page 1 of 5 - Drawn by Profili 2.30a Pro on data processed by XFoil - Copyright (C) 1995-2011 - All rights reserved.
Figure 26: The NACA 4412 has predictable, negatively sloped Cl/Cd curves at low but positive lift, and thus
would be fairly good for proverse roll-yaw tiperons. Contrast the NACA 0012 symmetrical foil and NACA 4412_
Inverted foils. The latter probably could only make proverse roll-yaw coupling as a tip stuck into the strongest of
vortex upwash. Profili plot by Adam Weston.
June 2014
37
Aileron differential
Aileron differential exists when with given input, from
yoke or transmitter stick, one aileron deflects up more
than its opposite deflects down. On some airplanes and
models it’s mechanically built in. RC flyers usually set up
differential from programmable transmitters. Model airplane
manufacturers typically suggest the amount of up and down
throw for ailerons. RC flyers also can easily try what works.
See Figure 27
For most wings downward deflection of an aileron increases
both lift and drag, making an adverse roll-yaw force.
Downward deflection of an aileron generally moves lift up
into the strongly adverse part of the lift/drag curve. Aileron
differential minimizes this force by lessening downward
aileron deflection.
The upward deflection of an aileron may slide its lift down
into the area of the lift/drag curve where lift decreases as
drag increases, for a proverse yaw force.
Cambered airfoil wingtip with ailerons
Left aileron down,
lift and drag increase
Right aileron up, lift and drag decrease
L
Adverse
roll-yaw R
region
Weakly L
proverse
with
R
differential
Left aileron down, lift and drag
increase weakly
Right aileron up, lift decreases,
drag increases
Proverse
roll-yaw
region
L
R
Fixed AoA 4°, L/D by aileron deflection
Figure 27: Aileron, tiperon, or wingeron differential.
And how well does this work? It depends on the lift-drag
curves of the portions of the wings with ailerons.
Proversely stable thermal turns and spins, even with no
aileron deflection
Sometimes after initiating a turn a model glider will tend to
stay in the turn, even when the ailerons are allowed to return
to neutral positions.
In a tight turn in still air, left and right wings sink at the same
rate. The inside wing has lower forward velocity and thus
a higher angle of attack. The result is indeterminate – the
lower velocity makes lower lift and probably lower drag,
but the higher angle of attack makes a higher coefficient of
lift and probably higher drag, at least till it stalls. It’s more
pronounced in a glider in a tight spiral. When its inner wing’s
higher angle of attack moves the lift/drag up into an area
38
of the curve where lift flattens and drag increases, then there may
be a combination of low lift (from the low speed) and high drag.
Proverse roll-yaw coupling now requires minimal aileron deflection.
In a spin the outside wing is flying but the inside wing is shedding
vortices, for low (stuttering) lift and very high drag. Spins are a
generally undesirably stable excess of proverse yaw forces. Even
with no aileron input, spins are proversely stable turns.
Closely related: ubiquitous examples of small fins stuck into
wingtip vortices to reduce drag
The winglets on most commercial jets stick up. And they’re fixed.
They don’t swivel or have control surfaces. But they are the
ubiquitous example of small fins stuck into the wingtip vortices
R/C Soaring Digest
to reduce drag. They are designed to improve efficiency at
cruise speeds and altitudes, where most of the fuel is spent.
Any savings at other speeds is fortuitous. Since they are
generally vertical they don’t directly add lift. They do angle
into the inwash above the wingtip to gain a bit of thrust. It is
your author’s conjecture that a major part of what they do is
move the center of the wingtip vortex up and slightly in, for two
additional effects: It gets the lowest-pressure center of each
wingtip vortex above the wing where it (1) doesn’t ‘pull’ back
directly on the wing and thus reduces drag, and (2) speeds
airflow across the top of the wing for increased centrifuging and
increased lift. They also lengthen the pressure gradient around
wingtips for a broader, softer, lower velocity wingtip vortex
center and lower pressures atop the wing.
Elliptical Lift Distribution tiperon
Possible planform with
tiperon. ELD is achieved
with twist.
Lift distribution of tiperon
on downward deflection
(yellow) conforms to vortex
upwash intensity.
Tiperon deflection uses up
some of the vortex upwash,
and moves vortex upwash
outboard.
Now, especially at lower speeds, if winglets swiveled and were
horizontal, they’d be proverse roll-yaw coupling control surfaces
-- tiperons.
Optimal tiperons, morpherons
Birds have the ultimate morphing wings, plus the millions of
years of bio-flight-computer evolution to do the right thing at the
right time. Birds don’t require rudders, though they can twist
their tails to achieve rudder-like control when they choose.
Some compromise of ELD and BSLD wings, perhaps with
tiperons, may make flight easier on a pilot and even improve
structural, weight, and lift/drag efficiencies. And it is always
possible to sacrifice performance for stability. But chances are
no pilot-controlled rudderless solution will avoid oversteers,
understeers, or adverse roll-yaw coupling perfectly. Hence in
full-scale airplanes without fly-by-wire controls rudders are here
to stay. Mostly. Zagis and many hang gliders get by just fine
with fins, partly due to the dihedral effect of swept wings.
Commercial aircraft winglet designs are optimized for one
speed, cruise. For that speed winglet height, and each local
chord, airfoil, and angle of attack can be optimized. A tiperon or
a BSLD wingtip with ailerons is harder to optimize. But for each
June 2014
Figure 28: On deflection, tiperon lift distribution should
conform to the pattern of wingtip vortex upwash angles.
Where vortex upwash is weak local tiperon sectional lift
should be weaker.
part of the vortex upwash at each speed there will be an ideal
combination of left and right airfoil shapes and angles of attack
that will optimize proverse roll-yaw coupling without creating so
much lift that the vortex is chased outwards. Only a morphing
wingtip could achieve perfection. Worse, the more flexibility
in a wingtip the more chance of destructive flutter. The ideal
provides a target. Reality requires compromise, simplification,
and most often, rudders.
See Figures 28 and 29
39
(Endnotes)
BSLD tiperon aileron combination
Possible planform with
tiperon-aileron combination.
BSLD is achieved with twist.
Lift distribution of tiperon
aileron on downward
deflection (yellow)
conforms to vortex upwash
upwash angle.
Tiperon deflection uses up
some of the vortex upwash,
and moves vortex upwash
outboard.
Simpler solutions: An aileron
and a tiperon with lift distributions conforming to
wingtip vortex upwash
angles. Proversity of the
aileron tip may vary as angle
of attack varies.
Figure 29: To make BSLD wing’s tiperon’s lift distribution
conform to the pattern of wingtip vortex upwash is more
complex. Ideally the tiperon/aileron shouldn’t make sharp
transitions in lift that form local vortices and drag. To
operate the wingtip at low angles of attack while the main
wing operates at varying angles of attack, the wingtip may
need to be a tiperon. To get its lift to conform to varying
angles of wingtip vortex upwash an arc-shaped aileron
may be required, or even one which warps to provide arcshaped lift. A simpler compromise is to just use a wingtip
aileron or a twisteron.
40
i
Philip Randolph, “The Nurflugel Mailing List, a Discussion
of Historic and Modern Flying Wings. - Yahoo Groups,”
accessed March 31, 2014, https://groups.yahoo.com/neo/
groups/nurflugel/conversations/messages/13938.
ii Bill Kuhlman, “Twist Distributions for Swept Wings, Part 4,”
Radio Controlled Soaring Digest, June 2003, 3, http://www.
rcsoaringdigest.com/OTW/on-the-wing4/164-HCP4.pdf.
iii Mark Drela, “Mark Drela Quoted on RC Groups,” accessed
March 31, 2014, http://www.rcgroups.com/forums/
showthread.php?t=1748293&page=2.
iv Bill Crawford, “Flightlab Ground School, 3. ThreeDimensional Aerodynamics,” www.Flightlab.net, 2009, 3–4,
http://www.flightlab.net/Flightlab.net/Download_Course_
Notes_files/3_Three-DimensionalA%232BA154.pdf.
v CW Fan, “Incompressible Flow over Finite Wings, Chap.
5,” December 2003, 10, http://www.google.com/url?sa=
t&rct=j&q=&esrc=s&source=web&cd=1&cad=rja&uact=8
&ved=0CCkQFjAA&url=http%3A%2F%2Fwww1.ytit.edu.
tw%2Fedu%2Fme%2F145%2FA05.ppt&ei=SnBRU7n4CMy
VyASnpIHwAQ&usg=AFQjCNGJxMm2ES9WGgDho25pdUi
GoL0Bdw&sig2=a7Wx_Fyf5nJ1lJvaBNkArA.
vi Kuhlman, “Twist Distributions for Swept Wings, Part 4,” 16.
vii Robert T. Jones, Wing Theory (Princeton University Press,
1990), 113.
viiiRobert T. Jones, “Minimizing Induced Drag,” Soaring,
October 1979, 26–29.
ix Drew Landman et al., “Wind Tunnel Testing of the Wright
Brothers Model B Airfoil,” AIAA 2001, no. 0310 (n.d.): 5,
http://www.wrightexperience.com/pdfs/airfoil.pdf.
R/C Soaring Digest
Sixteen Foot Flying Wing
for Slope and Aerotow
It seemed like a good idea at the time
Steve Pasierb, [email protected], John Appling and Erich Schlitzkus
A 48 inch span flying wing can be good
fun. An 8 foot span flying wing is better. A 16 foot span monstrosity of a wing is
just plain stupid!
Born out of mortal combat at the spring
2013 Cumberland Maryland Soar For
Fun and blended with both a dash of
capability a dose of culpability results
in a bad idea realized. Sure, we could
double the span of the classic Bash
Enterprises Mongo, an 8-foot flying wing.
But instead of smartly making it 16 feet
by adding long tips -- as the few Super
Mongos produced once did -- let’s
make the whole thing 200 percent of the
original?! And, change the airfoil. And,
well, change everything else.
While Steve Pasierb is the instigator of
this project, the majority of the thinking,
design effort and plain old hard work
cutting foam cores for four wings, was
June 2014
1
41
2
that of John Appling. He’ll be picking white foam
beads out of his house and Jeep for the next 10
years!! A third wing was constructed by Erich
Schlitzkus.
What goes into one of these wings? A whole
bunch of four-foot EPP and expanded bead foam
sheets, a pile of pultruded fiberglass tubes and
rods, The better part of a quart of polyurethane
Gorilla Glue, rolls and rolls of two-inch fiberglass
strapping tape, many square feet of Oracal selfadhesive sign vinyl, a roll of servo wire, add eight
235-ounce torque metal gear servos plus a tow
release set-up, throw in a receiver (or two) and
provide some electrons. Easy peasy!
Our spokesmodel poses with raw cores in the
summer of 2013. Main body of the wing is white
expanded bead foam while the leading edges are
expanded polypropylene foam.
42
3
R/C Soaring Digest
4
Spar and joiner stock. Spars are 1 inch OD and ¾ inch OD
pultruded fiberglass tubes to add span width rigidity to the
panels. Joiner tubes between panels are a mix of 1 inch and ¾
inch with ¾ inch and ½ inch fiberglass in both tube and solid
rod formats. This first picture, #4, is all the four foot spars/joiner
segments sufficient for two completed wings.
Construction of the control surfaces was left to the preferences
of each builder. Shown here in #5 is Steve’s balsa elevon stock
for the tips of the yellow wing. The other two wings used the
foam core sections sheeted with carbon and fiberglass in a
traditional vacuum-bag.
Panel cutting was completed in Maryland. Shown in #6 is
John’s set-up to hot wire the lower surface of a starboard inner
panel. The foam blank is positioned on a 40" X 60" drafting
table. The hot wire bow handle that is resting on a brick will be
hung from a traveler suspended by winch line attached to the
ceiling joists.
5
June 2014
43
6
7
Before committing to cutting the spar holes in the foam, a
confirmation of the spar’s strength for the inner panels was in
order. Photo #7 shows the 48" long 1" OD glass tube holding
20 pounds in front of a second tube. Deflection appears to be
about ¾". 3" pieces will be added to each spar to achieve the
required length.
The spar holes were cut with a hot wire simply following a circle
template.
Confirming the ¾" hole in the outer panel is to size. The foam
“rod” removed from the hole is shown in front of the core in #8.
To get to the proper CG (15% - 20% of MAC) we decided to
do some calculations to determine the moment about the
CG. Good thing we double checked because the amount of
ballast planned was close but the location needed to be moved
WAAAAY forward of what was anticipated.
44
8
R/C Soaring Digest
9
10
So no expense was spared to create a gas-fired foundry and
custom mold to produce two lead ingots. See Photo 9. What
the wife doesn’t know won’t hurt.
Photo 10. This should get us within a workable range to fine
tune the CG using slugs of lead in the 1 inch OD X ¾ inch ID
main spar tube. Using 2 inch lengths of EMT, John cast slugs
that were oversize in diameter and turned them down on a lathe
to fit neatly inside the main spar tubes.
Photo 11 is an image showing the pieces that went into the four
panels of John’s wing. Ribs (two on each panel), sub ribs (to
anchor joiner tubes), joiner tubes and spar tubes. If you look at
the rib second from the bottom you will notice John went with
the more complex, heavier and labor intensive releasable tow
mechanism. It just “seemed like a good idea at the time.”
June 2014
11
45
12a
12b
Servos were installed in the wings built by Steve and
Erich using a “top-hat” approach of a plywood plate
base and traps of bent aluminum sheet screwed to the
plate. Details of this installation can be seen in Photos
12a and 12b.
Wing #3 while under construction in Pennsylvania can
be seen in Photo 13. It quickly fills Erich’s workshop!
The plan on this one is to have a simple elevon,
approximately two feet in length in the center of
the main panels. Again, each wing build was done
differently based upon the personal preferences and
dementia of each builder.
OK, this is when things starting getting a little crazy
for Erich. After putting the wing together and trying to
pick it up, it was easy to see that we would need more
carbon to stiffen the main panels. It was like a darn
seagull flapping in the wind.
46
13
R/C Soaring Digest
14
As shown in Photos 14 and 15, on the
inner panels, a router with a 0.125" bit
was used to make slots for 1.0 inch tall
carbon which was then put in place with
epoxy.
This addition made a significant
difference.
The wing was now ready for several rolls
of packing tape.
Meanwhile, up north in Connecticut, wing
#2 was at a similar stage.
Right wing is shown here in Photo 16 all
prepped with trailing edge carbon sheet
cut to size. The carbon was installed with
June 2014
3" fiberglass tape and epoxy overlapping
the surfaces.
Balsa elevons for the tip and main panels
were vacuum bagged with 3 ounce
fiberglass and carbon mat on the bottom
for stiffness. See Photo 17.
In early January, Steve finally got to
unwrap his Christmas present from Erich!
Combination tow hook and tow release
mount. On steroids. Shown in Photo 18.
15
Gorilla Glue makes everything inside
the core as strong as possible. The glue
is the yellow you see on the foam. The
small hole in the middle of each section
is a fill spot Steve used to add extra glue
once the tubes were in place.
Yeah, as can be seen in Photo 20, this
baby is not designed for transportation in
a compact car. Main joiners are shown Photo 19 on
Steve’s wing. Each is a solid fiberglass
rod. Each rod tube is captured at the root
and also at a sub root 9 inches into the
panel.
47
17
16
48
18
R/C Soaring Digest
19
20
June 2014
49
21
22
The lead ingots to achieve CG balance are shown in Photo 21
all tucked in place on the top side of the main panels of Steve’s
wing. The cut-outs were then capped with pink extruded foam.
When it all cured, Steve restored the airfoil shape with a
sanding bar, easily sculpting the foam.
Photos 23 and 24 show the top and bottom tip strapping tape
scheme employed by Steve. The main panels have much more
aggressive use of strapping tape.
Photos 24 and 25 show covering the tip panel bottoms with
orange Oracal vinyl.
Covering tip panel tops in yellow for a bit of contrast in flight is
shown in Photos 26 and 27.
We used 3M 77 under all the wide strapping tape (dust it on as
the new propellant can eat foam) and then a full dusting of 77
under the vinyl even though the vinyl is a self-adhesive product.
The finished panels appeared to be incredibly stout.
50
23
R/C Soaring Digest
June 2014
24
25
26
27
51
28
29
Photo 28 shows servos in place, wiring completed and control
horns installed. Just waiting for the elevons. Steve installed his
elevons with sections of 2" strapping tape perpendicular to the
hinge line, the hinge lines were then fully sealed with vinyl hop
and bottom. This is the final stretch of construction! Contrasting to Steve’s approach, John used aluminum piano
hinge cut into short lengths. Show here in Photo 29 is a core
section with hard points installed to accept the hinges.
Each tip on the yellow wing is made of Coroplast material
covered in black vinyl, shown in Photo 30. They attach with
two fiberglass pegs that insert into the tip root, and are held in
place with sections of Velcro. The yellow wing used two large
tip sheets. The other two wings used slightly smaller vertical
surfaces between each panel. Both approaches appeared
to work equally well. The vertical sections between the wing
sections look much cooler!
52
30
R/C Soaring Digest
31
32
June 2014
33
53
Photo 31 shows the tow release installed
and working perfectly. Another 235
ounce torque metal geared servo.
Making up carbon + tungsten pushrods,
Photo 32.
This sucker is massive in its final state
ready to fly. Steve’s large center island
workbench is dwarfed! Wing sections
are just slid together for Photo 33, not
securely taped.
For ease of assembly, each wing half of
Steve’s model has its own switch, 2.4G
receiver and battery. The receivers are
simply both bound to the same program
in the transmitter. The switches are
recessed. You can see the receivers to
the side of each switch in Photo 34.
34
Oh, and as you can see in Photo 35, it
does have a “bottom.”
Test flights. Technically, wing #1 built
by John flew twice in the November of
2013, just slightly exceeding the Wright
Brothers’ distance record, but with
lesser results. It became quickly obvious
that this beast needs to be aero towed.
Forget the bungee. Remember, this is
NOT a copy of a Super Mongo (standard
Mongo with skinny tips added) this is a
200% Mongo. As they say in Brooklyn,
it’s freakin’ euge!
Bungee technique is a four to five person
ballet operation as evidenced in Photos
36 and 37. Towing it ROG will be a much
better solution. 54
35
R/C Soaring Digest
36
37
On flight #1, the bulky tow release stripped the servo when released under
pressure -- or non-release as the case may be. The resulting impact rendered
the gears of the center panel servos toothless. Metal gear servos with high
torque are essential on a project like this. Wings #2 and #3 received 235 ounce
torque metal gear servos.
Sortie number two was done with a new tow release servo and the center
panels locked in place. The good news is the metal gear tow servo worked
fine. Bad news is that the center panel control surfaces are vital. And, they
were likely locked too cambered instead of reflexed. The tip sections alone
were not sufficient to effectively change pitch -- up elevator flexed the tips
while the main body panels kept moving forward in a straight line.
While an engineering masterpiece, shown in Photo 38 is the original tow
June 2014
38
55
39
40a
release that caused issues on the test
flight. It has since been replaced with
the simpler version.
Wing #1 came away from all this
excitement essentially undamaged other
than needing all the servos replaced with
more substantial offerings. Key to all of
this foolishness and experimentation was
to learn that the wing does indeed fly
and it travels perfectly flat just as a lifting
body should -- no tendency to be bowed
in the middle. See Photos 39, 40a and
40b. At approximately 35 pounds of allup weight, this was a bit of a concern.
It does fly! If the tow release on flight #1
had worked, it had sufficient altitude to
clear the tree line and head out into the
valley.
56
40b
R/C Soaring Digest
41
Fast forward a few months. Two of
the three wings then flew in March at the
2014 Cumberland Maryland Spring Soar
For Fun. Pictured in Photo 41 are the
creations of John Appling (The previous
white test bed now done in red-whiteblue), Steve Pasierb’s yellow monster,
and Erich Schlitzkus’ gray overcast.
Each is a bit different in their construction
methods, but each is certainly a beast.
John’s wing had wiring and radio issues
that kept it grounded for the day.
June 2014
Towing was done by Len Buffinton flying
an Aero Works Carbon Cub powered by
a DA-150 motor. There was no question
this mighty tug could pull against
any issues that might have occurred.
Without yaw control, we were concerned
that the wings might slide awkwardly
on the ground. These concerns were
completely unwarranted.
First off, these wings aero tow
fantastically. See Photos 42, 43, 44 and
45. ROG occurs in about a dozen feet
and both wings that flew this day tracked
perfectly behind and just above the tow
plane. Towing speed needs to be kept
moderate or some oscillation begins
across the entire wing. More to be said
on that subject later.
What they are is floaters. Majestic at
that. The wing tracks well, carves nice
turns and is actually a joy to fly. Steve
Pasierb did the first flight of the day
and everything went smoothly until
he dropped the nose to gain speed to
57
58
42
43
44
45
R/C Soaring Digest
46
attempt a loop. The entire wing began to
oscillate. Not an unknown quality in flying
wings. The left outer elevon fluttered and
he quickly returned to normal, level flight.
The rest of the time in the air was a long
majestic glide.
Landing was a non event as Steve turned
the big wing in off the slope, made a
short base leg (See Photo 46) and rolled
onto final. The wing got into a bit of
ground effect flying low and refusing to
go lower than a foot off the turf. A shot
June 2014
of down control resulted ground contact
and an amazingly uneventful landing.
Erich was up next with the big silver and
gray wing. He also had a great flight.
He had not programmed enough reflex
into the control surfaces, so it had to fly
while holding the right stick back a bit
throughout. His wing also grooved nicely
in turns and performed very well. Another
long flight in the books. Another smooth
landing.
Steve flew again, and it was just like
gliding along with a giant woody sailplane
of the 1970’s. Don’t push it and you’ll be
happy, push it too hard and too fast and
odd stuff happens. A long happy flight.
Erich flew his again and that was where
serious issues were uncovered. Erich
and his friend Russ Bennett were trading
the controls having a fun time. The
wing got a bit too fast, began a shallow
dive, picked up more speed, did not
respond well to up elevator stick and
59
47
simply IMPLODED!! It was found near
the bottom of the mountain. Half in a
tree, half on the ground, as can be seen
in Photo 47. Nothing salvageable. He did
get the receiver back and they hauled
what little litter remained back up the
mountain. A giant oak tree still owns the
rest.
Steve flew two more flights. On the last,
the tug got too fast before turning out
over the valley into the slope lift and
the wing oscillation bit once again. He
came off tow low, stalled and eventually
flopped it onto the ground with no
damage. It was time to not tempt fate
60
anymore and put her away until next
time. See Photo 48.
The verdict? More rigidity in the
airframe is desperately needed! They
fly great, but must be flown at slow to
moderate speeds. Forget aerobatics. A
structure this large and heavy is more
than the typical slope wing construction
of miles of strapping tape and lots of
polyurethane glue can withstand. The
structure needs to be much more robust
and stiff. Erich’s was taped, cross taped
and glued at wild excess, had a carbon
cross member and it still did not survive.
It’s just too damn big!
48
While much thought was put into the
spars to provide strength across the
span, as a lifting body the primary issue
is panel rigidity. The wing does need
to resist bending into a giant “U” shape
under pitch control, but twisting forces
at speed are its worst enemy. Knowing
what we do now, there would still be a
significant spar in the main panels, but
the tip panels appear to be in less need.
Reinforcement, running in a diagonal
orientation (think Disser wing) to prevent
twisting needs significant thought and
improvement if there is ever to be a 2.0
version of these wings.
R/C Soaring Digest
Steve’s plan at the moment, although his
is still in one piece, is to cut the center
panels approximately in half, reducing the
huge chord, total area and total weight
significantly. The center will become one
rigid section with a major spar. The two
tips will receive more cross reinforcement
and plug into that. We’ll see in a couple
months what a more rigid 12 or 14 footer
will do?!
John’s 16 foot wing will likely get a test
tow at an aerotow event this summer.
That should be fun, but these are flat
land events, not the mountain slope
conditions of Cumberland, MD.
The “Mongostrocity” as it has come to
be known, is still a work in progress.
Stay tuned.
As we’ve said all along... It seemed like a
good idea at the time!
Specifications
Airfoil: Span: Root chord: Tip chord Sweep angle: Area:
June 2014
MH-64
192 inches
39 inches
16 inches
22 degrees
5280 in. sq.
Washout inner:
Washout tip:
Elevon root:
Elevon tip:
LE material:
Panel material:
0 degrees to 0 degrees
0 degrees to - 2 degrees
7.8 inches max (20%)
3.5 inches max (22%)
1.3 lb/cu ft EPP
1.3 lb/cu ft expanded bead styro
61
A 4.5m Discus 2 from Top Models CZ owned by Colin Taylor from Kapiti, NewZealand. Gotta love scale saiplanes
with realistic pilots in place. Photo taken at the Burnham Aerotow in March of this year by Graeme Phipps.
FujiFilm FinePix JX370, ISO 100, 1/180 sec., f11
(Complete coverage of the 2014 Burnham Aerotow event can be found in the May 2014 issue of RC Soaring Digest.)
62
R/C Soaring Digest
Jerilderie Scale Aerotow 2014
Henryk Kobylanski, [email protected]
Written by Henryk Kobylanski, [email protected]
rctrader.com
Photos by Henryk Kobylanski
The full gallery from the event is available
on the Oz Scale Soaring Website:
http://www.scalesoaring.com.au/
Jerilderie Scale Aerotow 2014
Jerilderie is a small town in southern
New South Wales where twice a year,
June 2014
RC soaring pilots gather for two very
distinct disciplines; the first being for the
F3J nationals, and the second for the
Jerilderie scale aerotow.
large scale soaring event in Australia.
Situated in the middle of one of the larger
country racecourses in Australia, one
could not ask for any larger an area to
conduct large scale soaring.
In its 10th Year, the Jerilderie scale
aerotow was the brain child of a small
group of scale aerotowing enthusiasts
championed by Gregg Voak. Over the
years, Gregg, with the help of a few
dedicated soarers has turned the event
into what has now become the premier
Over the years, the event has had many
guises, and while Gregg was away
overseas, run by a few with the same
vision. This year Gregg took the formula
back to its original simple format of a fun
weekend’s flying aerotow… which was
achieved in spades.
63
With over 30 pilots coming from all over Australia, this year we
had a special guest, Ross Biggar, coming all the way from New
Zealand with his 6.6m Arcus in tow (pun intended). That’s what
we call dedication.
One of the immediately obvious things that have changed
from the early years of Jerilderie is the number of tow planes
that were available (I counted 12). The tugging “fleet” were
kept well entertained all weekend but at times you would be
excused thinking we were at a 3d competition rather than a
Scale aerotow, given the number of IMAC/3D tow planes. As
much as they are not totally a scale look, the idea of these
airframes is more functional than visual. They make for one
of the most stable towing platforms with very friendly landing
characteristics, which is an important aspect when a pilot can
tow up to 50 sailplanes per day.
From the outset, Jerilderie was always a “big sky” event
allowing for the flying of large scale sailplanes. The thing that
has become very obvious over the past 10 years is that the
average size of the sailplanes has increased. With notable
exceptions, namely Bill Bland’s half scale sailplanes, in the
beginning, third scale was the exception. This year, it really
seemed the norm, with the majority of sailplanes being round
the 1:3 scale and a few pushing the 1:2.5 barrier. With such a
lot of investment in airframes, the other significant difference
was the number of trailers transporting them. In Jerilderie’s
first year there may have been one or two, and again the 10th
anniversary shows how much more the average person in
investing in the sport.
64
Pilot briefing
Some of the tow planes
R/C Soaring Digest
Jim navigating the pits
Morgan Hill, Rod Watkins and David Hobby ponder if the
weather is quite perfect yet...
was wonderful to see that Martin, now a resident of Melbourne,
had made the trip to Jerilderie and brought with him some of
his wonderfully handcrafted models. Two of the outstanding
vintage models were Martin’s beautiful Golden Wren and Leon
Carlos’s meticulous Fafnir.
young Riley spent quite some time hanging round Henryk’s
trailer looking inside, walking round and looking inside again.
Finally came up to me and with a cheeky grin asked me if that
was Dusty in my trailer?
Unlike past years, the weather was very mild and probably the
greenest we have ever seen the racecourse. This did not affect
the fine flying conditions and other than the odd windy morning,
flying went on right through to Tuesday afternoon.
Sunday night saw us have the communal BBQ in the
Racecourse Pavilion. With meat sizzling and wine-a-flowing,
everyone settled in for a viewing of “Planes” thanks to Jim
Houdilakus’s portable cinema. Young Ryley Bishop, who was
the weekend’s mascot and champion “wing-walker” was
transfixed as Dusty prevailed yet again. The next morning
June 2014
One of the new elements that evolved this year was the
evenings on the field. Rather than packing up and heading off
to the pub, a few of us stayed at the field and made good use of
the facilities at the Hacienda JaseBro. ( I am sure Jason’s trailer
is a Tardis). Flying till after sunset and setting in for an evening
of great food, fine company and where long stories were
a-plenty. There is something magical about that big open space
at night as well.
It’s a privilege that we are given these facilities to use each year.
The Jerilderie Shire and all of the people in Jerilderie make us
very welcome giving us full use of such a great set-up. Special
65
Gregg Voak and Ryley
thanks to David Tamlyin from the Shire for making it all available
and making sure that the field was ready for our arrival.
Nobody wanted to leave and head back to reality.
Thanks to all those that came – and I hope we did not miss
anyone:
Gregg Voak Antares, Duo Discus, Extra 260
Martin Simons
Golden Wren, Weihe
Leon Carlos
WoodStock, Birgfalke, Fafnir
Rod Watkins
Mx2, Ventus 2c, Minimoa
Caroline Antares
Jason Sagaidak
Yak 54, Pawnee, ASG 29, ASW 27
Morgan Hill
Grunau, ASH 31
Henryk Kobylanski Yak 55sp, Antares, Cessna (Dusty’s cousin),
DG303, SZD 54
Mark Doyle
Reheir, Ventus 2c
66
Joe Rafenaet
ASW 22, ASH 26
David Millward
Discus 2c, DG1000
Kevin Jolly
Antares
John Copeland
ASH 26, Fox, Wilga
Colin Boothy
K8
Jim Houdelakus
ASG29, Duo Discus
David Hobby
Extra 260, Pawnee, Ventus 2c, DG 600, K8
Simon Bishop
DG600, Extra 330, Ventus
Theo Arvanitakis Extra 260, Pilatus B4
Chris Carpenter
ASH 26, PSW 101, Pawnee
David Raccanello Radian
Chris Graham
Blanik
Ross Biggar
Arcus
Wayne Jones
Fauvette, LS4
Dave Bell
ASW 27, DG808
Gary Whitfield
ASW 28, Hots Special
Darrel Blow
Lunak, ASW 28, Ka8
Ryley Bishop
Mascot & Wing walker
Warren Edwards Photography
Sandra Voak and
Roseanna Millward Moral Support
The following pages are filled with other images from this event.
The full gallery from the event is available on the
Oz Scale Soaring Website:
http://www.scalesoaring.com.au/
R/C Soaring Digest
June 2014
67
Gary McDougall’s scratch built Minimoa going for a walk
68
R/C Soaring Digest
June 2014
69
70
R/C Soaring Digest
June 2014
71
72
R/C Soaring Digest
June 2014
73
74
R/C Soaring Digest
June 2014
75
Rod Watkins’ Minimoa
76
R/C Soaring Digest
Gary McDougall’s Minimoa
June 2014
77
78
R/C Soaring Digest
June 2014
79
80
R/C Soaring Digest
Review
ST-Model DG-1000 RR
Dan Ouellet, [email protected]
The ST-Model DG-1000 is a 2 meter
scale rendition of the DG-1000 sailplane.
It features easy assembly, high aspect
ratio airfoils, and a Retractable Motor
System! It even has enough power to
allow it to rise off the ground from a
grass runway!
June 2014
First Impression
This model looks good, both close-up
and in the air! The molding is of good
quality. The foam is mostly smooth.
There are just a few visible molding
marks on the underside of the wings,
wing tips and horizontal stabilizer.
It flies well under power, with the RMS
fully extended. It becomes a good
sailplane when the RMS is retracted. It
has a broad speed envelope. This allows
it to thermal in light lift, yet speed up for
good penetration in moderate turbulence.
81
The ST-Model, Sheng Teng Electric R/C Model Plane Co.,
Ltd., located in Jiaxing, China, was established in 2000 to
produce R/C airplanes made from EPP foam material, which
are “convenient” for consumers. This means that their models
are affordable, easy to assemble, and have excellent flight
performance.
The DG-1000 is no exception. It meets all these corporate
goals. Overall, it is fun to fly, and I am having a good time with
it!
Unboxing and Assembly
The model arrived packaged in a box that seemed somewhat
light. It is adequate to ensure a safe journey, since after
unpacking, all the parts looked good. There was no visible
damage, warping or dents, on any of the components.
This model requires very little assembly to get it in the air.
The servos, servo covers, linkage rods, and control horns are
already in place, having been assembled at the factory. The
RMS is installed and setup to function out-of the box, using a
sequencing board.
The package includes a basic decal sheet, which can be
applied in just a few minutes. All that remains to be done to
complete the assembly, is to attach the linkage rod to the
elevator and bolt-on the horizontal stabilizer with a single screw.
Then just mount the wings, install and bind the receiver, and
adjust the control surfaces.
The nose compartment is tight. There is just enough room for
a small receiver between the servo and battery compartments.
Plus, to help stow the antenna, ST-Model thoughtfully molded
in a small plastic tube, in the bottom at that location. One of my
Spektrum AR600 receivers fits nicely in this area.
My 1300 mAh 3-cell LiPo battery fits perfectly in the molded
battery space. I just had to replace the ESC connector provided
by the factory with an EC3 connector to match the ones on my
battery.
82
My 1300 mAh 3-cell LiPo battery fits perfectly in the molded
battery space. You can see the clam-shell doors for the RMS at
the wing trailing edge.
R/C Soaring Digest
This setup worked just right. The aircraft
balanced perfectly on the recommended
“Centers” of Gravity: 53mm behind the
leading edge of the wing with the RMS
stowed, and 46mm with the RMS fully
deployed. There is no need to add any
weights.
Since there is so little to do, the aircraft
can be ready to fly in just a few minutes,
unless you would like to customize it.
Custom Paint Job
To make the DG-1000 more visible, I
painted the underside of the wings and
the horizontal stabilizer “flat black.”
Then I painted the wingtips, rudder and
horizontal stabilizer tips “red”.
The black bottom really stands out! It
helps me to see the model much further
out! So far, I have flown it over ½ mile
away, possibly as far as ¾ mile distant,
without losing sight of it.
As to the “red” on the tips and rudder, I
just think it makes the model look cool!
As to the “red” on the tips and rudder, I just think it makes the model look cool!
June 2014
Should you decide to paint yours, note
that ST-Model uses “very efficient” mold
release compound in its manufacturing
process. To give the paint a fighting
chance to hold, I found that it helped to
first sand the areas to be painted with
400 grit sandpaper, then clean them with
denatured alcohol.
83
RMS Sequencing Board
To better understand how the RMS
worked, I read the manual several times.
The principle involved with the RMS is
mechanically similar to how a retractable
landing gear operates. However, there
is a little more going on than that, since
in addition to the folding arm, there is a
motor and a propeller to account for. To
make things simple, ST- Model wires in a
sequencing board, which allows the use
of a simple 4-chanel radio transmitter to
operate the RMS from just the throttle
stick.
Unfortunately, this did not work in my
case. I ended up removing the board and
configuring a custom programming mix.
Custom Programming
The problem I encountered using
the included sequencing board, is
that try as I may, I could not get it to
power up normally. It kept showing
a rapidly blinking light. According to
the instructions, this indicates “not
enough current to power up.” The
manual suggests to use the throttle trim
to increase the current to the board.
However, the throttle must also be below
its 24% travel setting, or the ESC will not
arm. This combination just would not
work with my setup.
The RMS extended and retracted.
84
Therefore, I decided to remove the
sequencing board from the model, and
R/C Soaring Digest
create my own custom mix to control the RMS in the DX8
transmitter, using the gear switch.
The first thing I did, was assign the throttle cut-off function to
the gear switch, so that when I select retract, the throttle will be
completely “off.” This is done to prevent accidentally powering
the motor when the mechanism is stowed. Then I programmed
in a two second delay on the RMS (gear) servo, to slow down
the deployment and retraction of the mechanism.
The way I use the system to deploy the RMS is by selecting
“gear up” on the gear switch to extend it, then count 3 seconds
to make sure that the RMS is fully extended, and then power up
using the throttle normally. To retract the unit, I do the opposite.
I make sure that power is “off” using the throttle, count 3
seconds to allow the motor/propeller to slow down, then select
gear retract with the switch.
The two second RMS servo delay not only provides a smooth
deployment, but also proved helpful one time as I accidently hit
the gear retract switch while the throttle was in the “full open”
100% position. The RMS unit stowed normally without any
incidents. Still, just to make sure, I prefer to manually count 3
seconds between using the switch and the throttle.
For the following two mixes involving the use of the motor, I was
not sure of the required settings and used my best guesses:
Considering that the propeller trust line is “way above” the
centerline of the model, I expected a noticeable nose down
tendency, whenever applying power. Therefore, I configured a
23% up elevator mix, to the throttle movement. This looked to
be just about the right elevator deflection when applying power!
Close-up of the extended RMS.
June 2014
The propeller is otherwise functioning in a conventional manner.
This means that I expected conventional p-factor and torque
– a left yawing tendency with the application of power. To
compensate, I dialed in a 25% right rudder mix, to the throttle
movement. This mix should be most beneficial when climbing
at full power.
85
Since I like the radio to do as much of
the work as it can, I configured some
additional mixes. To help control the
landing approach, I configured the
wing with dual aileron servos and setup
spoilerons on the 3 position flap switch.
The 1st position is programmed for
normal ailerons, the mid-position is set
for 35% up on both ailerons, and the 3rd
position is set at 75% up. To minimize
abrupt changes when deploying the
spoilers, I programed it in a slower 2
second servo speed. To assist with
coordinated turns, I set up a 55% aileron
to rudder mix.
The clam-shell doors open, the RMS rotates upward and forward, and the motor
starts. Reversing the procedure creates a clean rear fuselage and , making it more of a
sailkplane than a glider.
The manual only suggests maximum
control surfaces deflections. It does
not recommend any dual rates, or
exponential settings. To archive a
smother flight, I configured dual rates,
and dialed in a lot of exponential. The
full rates is set to 100% travel on all
servos, with 40% exponential. The low
rates is cut down to 80% on the ailerons,
elevator, and the rudder with the same
40% exponential.
Maiden Flight
The DG-1000’s maiden flight was on
Sunday, April 6, 2014, at the Seminole
Radio Control Club facility in Tallahassee,
Florida. The weather was in the mid-70s,
clear sky, with variable wind at 8~15 KTs
from the South.
Just for fun, I decided to try to have
the DG-1000 rise from the ground in a
86
R/C Soaring Digest
conventional manner. Therefore, instead
of hand tossing it, I placed the model in
the center of the grass runway, facing
into the wind and deployed the RMS. I
slowly added power, and at about half
throttle, the model begin to slide forward.
I had good rudder authority so I
continued adding power, and the model
gently lifted off a few inches above the
runway, all on its own. It accelerated
nicely while in ground effect, as I
continued adding power to full throttle,
and maintained its height and direction,
all without my intervention.
I commanded a slight nose up. The
model transitioned easily to a healthy
climb attitude, which it maintained hands
off after just 1 click of down trim!
So much for best guesses with the mixes
– Sometime when the planets align just
right, you get lucky! ;-)
I decided to fly one full circuit to see how
the DG-1000 felt under power with the
RMS fully deployed.
I turned left crosswind and leveled at a
couple hundred feet above the runway.
The DG-1000 maintained level flight with
about ½ throttle, so I proceeded to turn
downwind while keeping the throttle at
the ½ position.
Once established on downwind, the
model flew hand off without any further
trim.
June 2014
I turned the model on a close base and
reduced power to about ¼ throttle to
begin the decent. Immediately, the DG1000 nosed down and begin to lose
altitude quickly.
Note to self: This RMS system is a
“GOOD SPEED BRAKE”!
I added a little power to stretch the glide
and turned on short final.
This worked well for a short approach!
I flared the model at about a foot off the
runway and proceeded to make a normal
touch-and-go.
Just add power to go back up! Not
something I am used to in a glider.
This time, I climbed the model straight
out to a comfortable altitude to test the
RMS retraction. I leveled it at about 500
FT and cut the power. After counting
3 seconds, I selected the gear retract
switch.
The RMS retracted normally with mild
indication of the process. Visually, this
was a slight altitude loss as the motor
powered off and the “air brake” took
hold. This was quickly followed by
an increase in speed and the model
leveling out on its own as the RMS
finished stowing itself. Overall, the model
responded nicely to the transition.
It was easy to tell that the RMS was
retracted - The model automatically
settled in to a good glide and became
more responsive.
Shortly thereafter, I noticed the right
wing lifting slightly, possibly indicating
lift on that side. I immediately turned
90 degrees to the right, and the model
started rising in light lift. I established
a tighter right turn to get it closer to the
core of the thermal, while pulling a little
on the elevator to keep the fuselage level.
I was immediately rewarded with the
model rising faster.
This thing does glide well!
Some additional minutes of flying to get
comfortable with the DG-1000, showed
me that it can slow down and turn fairly
tight, without any bad tendencies, such
as tip stalling. However, it prefers moving
along at a good clip. This model has a lot
of penetration for its size, and will retain a
decent amount of energy, if you let it.
It is defiantly more of a sailplane than a
glider!
While still high, I tested the spoilerons to
find out what to expect before coming
in for the next landing. They behaved
as expected, without any unwanted
characteristics or tendencies, or require
any trim changes. It is good to know
that the spoilerons work well. However,
should the DG-1000 need to come down
quickly, just deploy the RMS at partial
power. That works better!
87
The ST-Model DG-1000 sailplane is a model that should be in any aficionados’ hangar.
88
R/C Soaring Digest
I elected to make a normal glider landing
approach. I used the first notch of spoilers on
base, and went to full spoilers on final. It worked
well enough. The approach was predictable, and
the model touched down 30 feet or so in front of
me in a good 3-point attitude.
Altogether, my timer showed that I used about 3
minutes of power for the maiden flight. Total fly
time was approximately 20 minutes. According
to my charger, it put 650 milliamp back in to
recharge the battery pack.
Therefore, the 1300 milliamp battery should safely
supply about 4 minutes of “full” power – Enough
for about 4 one minute climbs to 500 FT.
Test Model Specifications
Wingspan:
2010mm (79.1”)
Length:
970mm (38.2”)
Weight:
750g (36.5 oz.) w/ Battery
Battery:
Turnigy 1300 mAh 3 Cell 20~30C LiPo
Receiver: Spektrum AR600 6-chanels Full Range
Transmitter:
Spektrum DX8
Sequencing:
Custom mix on DX8
CG:
53mm (2.1”) behind leading edge of wing
when RMS is completely folded and 46mm
behind the leading edge when the RMS is fully
deployed.
With a little luck finding thermals, it should be very
easy to stay up 20~30 minutes per flight, possibly
much longer.
Additional Flights
I flew the DG-1000 twice more that day. Both flight
lasted over 30 minutes each. I encountered good
lift and was able to perform mild aerobatics using
only momentum. The wings have noticeable flex
under load but seem to hold up well.
Interestingly, neither flight required any additional
trim changes.
Conclusion
The ST-Model DG-1000 sailplane is a model that
should be in any aficionados’ hangar. It is simple
to setup and a joy to fly! The RMS really makes
it stands out at the field! Best of all, it is on sale
today at Tower Hobbies for only $124.00!
June 2014
Sources
Author:
Dan Ouellet, [email protected]
ST-Model DG-1000 PnP:
<http://tinyurl.com/plvnbh2>
Distributor: Tower Hobbies
<http://www.towerhobbies.com/>
Manufacturer:
Sheng Teng Electric R/C Model Plane
Company, Ltd.
<http://www.sheng-teng.com/>
89
This photo was taken at Las Aguilas Sailplane Club in Caracas, Venezuela. On the ground queue, 4m Baudis Salto,
followed by 4m Tangent ASH 26, and last, 3.7m Tangent DG600. In the air, a 5.6m Graupner Discus coming in at a
low pass. Salto - Jesus Esteller, ASH - Juan Ramon Brunet, DG600 - Renato De Cecco, Discus - Jose Cruz.
Photo by Renato De Cecco. Samsung GT-I9300, ISO 50, 1/614 sec., f2.6
90
R/C Soaring Digest
CEWAMS
Chris Erikson’s Wild Arsed Mountain Slopers
Saddle Mountain Slopener, April 2014
Philip Randolph, [email protected]
June 2014
91
Frustrations are the rent one pays to life, worth it, for
those who like being alive. And we who like being
alive are a sample biased by Evolution, which favors
entities that attempt to stay alive, probably imparting
the innate enjoyment of it all as sort of a cheap-shot
bio programming incentive to not playing lemming.
But as I was saying, somewhere back there, and in
case you have already forgotten: Frustrations are the
rent one pays to life. Applied to the arcane hobby of
throwing toys off ridges, frustrations are the minor
costs of dealing with the persons (as corporations
are, according to our Supreme Court) who sell us
our toys, and who sell us the pieces with which
we assemble our toys. It’s part of the hobby. One
takes a little of the questionable with a lot of the
good. And thus it behooves us to gracefully accept
a little frustration in our progress toward the greater
enjoyment. I have not always been graceful in my
acceptance. That’s another story, and embarrassing,
so I won’t tell it. Here I will merely grumble a little, in
what I am hoping to pass off as good humor.
Most of this article will merely be a slope report. But
I’ll start with what mostly doesn’t get written about,
the grumbles within getting ready to go. And how
a goofy HobbyKing switch caused me to test (and
crash) an odd plane that had been hanging from my
ceiling for five years. Some of these outfits could
use a retiree from the Gong Show as BS Czar.
Title page image: Damian at Sentinel, looking north
across lowered Wanapum Lake & sandbars. The
local growers had half their crops covered in plastic.
92
Slope Report Part One:
Preparation Grumbles
So I’m getting ready for the slopener. I put all my old NiMH batteries in a
pile and cut the leads off. One of the guys on the trip, Erik Utter, has great
luck with NiMH. He uses a Triton charger. He says he treats his batteries
“poorly.” Or, “Like dog meat.” Darn. I can’t remember his exact expletive. He
always charges at a high rate. Never has problems. Me, I have lost planes to
false NiMH charges. So I’m sticking with NiCad, LiPo, or LiFe.
I have two foam chevron flying wings I’m getting ready. One is my old 48"
Sonic. And I’ve mostly put together a 60" Scout Bee. I’ve equipped it with
a big NiCad. In each I’ve stuck a $5.75 Lemon 2.4 GHz receiver. These
are great for slopers because they don’t have fancy failsafe programming
that will keep on sending a signal to a servo on loss of signal. That
means standard lost model alarms work with them, unlike with a lot of
2.4 receivers. The weak point is just that they sometimes die if you put
the battery plug onto its pins backwards. At six bucks that’s just cheap
instruction.
I put HobbyKing combo switch/light/charge-jacks into the Sonic and the
Bee. Then I’m ready to cover the 60" Bee. I decide to charge it first. Which
is when I discover the charge jack has male pins. That’s backwards from
normal battery connections, where charge receptacles have male casings
and female pins. Phooey. So I soldered up some backwards charge leads.
And that allowed me to charge the Sonic. But I was not going to bury
a bogus switch/light/charge-jack into a new Scout Bee. So that wasn’t
making the trip. Bong.
The solution: I’d been wanting to test-fly a strange model I picked up for
free with a broken nose. It’s a Jade Shogun, but built with the wings swept
forward rather than back. The Shogun was kitted by Richard Jarel, probably
sometime around 1990, before he became a model builder for Hollywood.
He made the taxis for The Fifth Element. The Shogun was billed as “The
Ultimate Warrior,” and sported USAF markings. I’m not generally into
the military stuff, but I liked its lines. Mostly. It came equipped with huge
horizontal faux air scoops, that just looked draggy. So I filled them with
foaming polyurethane, and patched on a nose out of Kevlar and EPP foam.
R/C Soaring Digest
It looked like Frankenplane. And then I
lost the cockpit cover for a couple years.
Found it again in a box of foam rubber.
Oh, yeah: It’s built of thin plastic, with
balsa-covered white-foam wings. Really
brittle. Dumb to take to a rocky site. So I
decided to haul it along.
The price one pays when one gets
those great deals from HobbyKing is
that occasionally something is not right.
Like one of the LiPo batteries I ordered.
Somewhere in the fine print or the
picture it probably explained it only had a
charge lead, but no power plug. Oh well,
it was cheap. Or the latest frustration
with their order fulfillment: Back in midMarch I ordered from the International
Warehouse, all stuff supposedly in stock.
In the second week of April I asked
where the stuff was and was told it
was held up because one item was on
backorder. I grumbled, “Your mistake.
Delete that item and send the rest fast,
free shipping, by air mail.” A few weeks
later I said, “Where’s the stuff?” They
said, “Like you asked, we sent it airmail.
Via Malasia Air. It should be there within
45 days.” What? Oh, and they did charge
me for the shipping. What? Yep, at the
bottom of the shipping confirmation it
says, “45 days.” I grumbled. Pointless.
“Oh sir, we appreciate how you feel.”
I resigned to waiting another month. It
arrived that afternoon. Then I tried to
order a charger and the site kicked me
into something called PayDollar which
June 2014
wouldn’t talk to my bank. Gawrd.
But every time I say, “I should just deal
with domestic suppliers” I recall Horizon
Hobby. I ordered a micro E-flight ASK21. I was hoping it would be a bit like
the Liftworx Seeker, a great tiny light-air
plane, but it appears Horizon designed
more for scale appearance than
aerodynamics. Oh well. It was impossibly
inexpensive after a reduced price and
crossing a free-shipping threshold. But:
We got it out to a field. Bound it to my
transmitter. The ailerons would go up and
down together, like flaps. An electronics
buddy and I spent a couple hours trying
to figure out the transmitter programming
mistake. Read the little manual. Now,
you’d think a manual written by native
English speakers would explain, “If the
ailerons go up and down together, call
Horizon.” You’d think that they’d mention
that the two aileron servos are linked by
a mixer in their control board, and run off
one channel in the radio. That means that
even when the transmitter says that left
and right ailerons are moving in opposite
directions, since they’re both controlled
by the left aileron channel, trying different
programming won’t fix anything. I called
Horizon. They sent me a little dongle
that reprogrammed the tiny onboard
chip to move the ailerons correctly. I
grumbled about the incompleteness of
their manual, to no avail. Bong. Oh, well.
Whoop-te-do. I brought the
ASK-21 along, in case winds were rising
up the draw at our grassy, Saturday
night/Sunday morning ‘cow corner’
campsite. Nope.
Plus there’s my JR 9303. Made by
techies for techies. That means goofily
Bong complex. I vaguely understand that
the programming of other transmitters of
similar capability is more straightforward.
Slope Report Part Two:
Slope Report
In which we are chased off a ridge
by lightning (though it was almost
time to go anyway)
I had planned to leave Friday, May 25, at
about 3:00 pm. I’d drive east across the
Cascades and get to Saddle Mountain
about 5:30 pm. As it was, I left at 5:30.
Then I stopped to socialize for twenty
minutes near Issaquah, which turned into
two hours, so I arrived at the campsite
after 10:00 pm.
Our favorite campsite is up against a
bunch of basalt columns overlooking
Wanapum Lake, a damned part of the
Columbia River. (Woops, post writing
I caught that. Dammed.) The guys had
a fire going – Chris Erikson and his
sweetheart Melissa and his brother Sean,
Damian Monda, Erik Utter with four-yearold Cole and ten-year-old Riley, and Mike
Zanol. The usual fire chatter.
The next morning Steve Allmaras and
Chris’s father, Erik Erikson, showed
93
CEWAMS founder Chris Erikson and the Deathmobile, a 1970
Datsun 510
up. Chris has set up his sun scope. Someone comments on
a scrape on my head. Damian asks what kind of a hat I’m
wearing. I say “It’s a Filson. I found it in the remains of oour
26’ geodesic dome that blew down this winter, up on Waldron
Island, in the San Juans, when we were doing demo. Geodesic
domes were Buckminster Fuller’s revenge on hippies. I’ve been
leaving it in the sun in my kitchen window because it smelled a
little of mold. But I’m sixty-five, so I can smell a little moldy if I
choose to.” Sean says, “It’s not an option.” Slam dunked. Gud
shot.
Breaking camp, Mike Zanol says, “Has anyone seen my water
bottle? I have the cap.” It’s a gallon water bottle. I’d grabbed
the nearest thing, pretty late, to hydrate, as I headed for my
94
Desert flowers; Arrowleaf balsamroot, flox, sage
R/C Soaring Digest
Melissa, Erik E., Steve, Cole at the “Cow Corner” campsite
Steve, Damian, Erik, and Chris launching his EZ Glider
truck and sleeping bag. I say, “I have
it. I used it for a pee bottle.” Quoth
Mike: “So where’s Sherman?” For you
young persons, that’s a reference to Mr.
Peabody and Sherman, of the “Bullwinkle
Show.” Gud Gawrd, Michael.
should have a board meeting.” I say, “I
met with the board last night. It was a
1x4. When I fell across the wood box.”
Erik U. says, “That explains the cut on
your head.” Uh, let’s just say it was dark
and I tripped.
We drive past boat ramps, their lower
ends fifteen feet above what remains of
Wanapum Lake. By noon we’re at the
Mattawa taco stand, where I ordered
Lengua tacos. Tongue.
And then up to Sentinel. Sentinel is 1400’
above Sentinel Gap, where the Columbia
cuts through the ridge called Saddle
Mountain. It’s a choke point behind
which the ice age floods which scoured
Eastern Washington (back when it wasn’t
Someone is joking about CEWAMS’ lack
of formal organization. Damian says “We
June 2014
called that) backed up to reiteratively
create Lake Lewis.
The winds were variable out of the SW.
Immediate scent of sage. Half clouds,
half sun. We’re looking north at the
lowered levels of Wanapum lake, great
sand bars usually covered by water. It’s
looking a bit more like a river than usual.
The Columbia River. Wanapum is the
dam with the 65’ crack in a spillway. They
dropped the lake 26’ to reduce pressure
on the dam. They plan to lace it back
together with big cables down into the
95
Looking south from Cow Corner past Damian and Erik to the
Columbia river and the Hanford Reach
bedrock. Maybe it will look like Mother Hubbard’s shoe.
Damian is first up with a 60" Bowman Hobbies Javelin. That’s
a 60", standard tail, EPP foam plane. He is on his third set
of wings. He cut the 2nd and 3rd sets with a hot wire and
templates. We watch his plane circling above the rocky
point and its sagebrush. Chris has good flights with his EZ
Glider, equipped with flaps. Erik flies his new Herring. Steve
zips around with his Bowman Hobbies Super-Scooter and a
Boomerang. I get my Sonic up. From across the river to the
southwest we hear the Army blowing things up in its Yakima
Firing Range.
Erik Erikson and the Cow Corner camp spot
Erik E: “Why are clouds flat on the bottoms?” A discussion of
dew points. Philip: “Because dew points are horizontal.” Well,
96
R/C Soaring Digest
Philip with Franken-Shogun about to smash it on the rocks at
Wahatis
Steve with busted-nose Super Scooter
it’s actually isotherms which are horizontal. Nerd stuff.
By the trucks Erik blasts rockets off with his kids. Four-yearold Cole gets to push the button on the igniter. 500’ up, a
parachute, and they send Philip chasing through the sage, sort
of an olden retriever.
Every once in a while everyone drops out of the air. I mean, our
toy gliders drop out of the air. But since we’re into vicarious
flight it seems like we fall. Lift is intermittent. We wander back to
where the guys who only fly beers are.
Sean is stretched out on a fair-sized faux Persian carpet.
Folding chairs all around. Erik E. shows me Glacial Lake
Missoula by David Alt. He’s a geologist. Great maps of how
this area was ripped repeatedly by 500 cubic miles of water
breaking loose every fifty or a hundred years from an ice dam
on the Clark River in Montana. Ice age.
June 2014
Back up at the point Steve and I cheat. He lofts his electric
Radian, and I toss an electric EZ Glider with a brushless.
Around six Mike and Sean leave. The rest of us head back
down toward Mattawa, where it would have been good to get
the smallest-fry into a regular bathroom. Oh well. We drive
twenty highway miles east and then back up on to Saddle
Mountain ridge a few miles East of Wahatis Peak. I’m last there
because I did stop in Mattawa. By the time I get there Steve
97
has shoveled clean the campsite we call ‘Cow Corner.’ It’s in a
little draw looking south across where the Columbia meanders
across the Hanford Reach. Gorgeous view, with the Hanford
Nuclear Plant in the distance. Melissa says, “I generally don’t
want to camp in view of a nucular plant.” She reluctantly admits
it’s kind of pretty. Damian sets up tripods topped with stereo
speakers powered by a big amp and a car battery. Some kind
of hilly-billy tunes. Everyone cooks steaks. Fire. I go to bed by
10:30.
In the night I wake to light rain on my truck canopy. And I hear
coyotes. I get up to put my planes under the truck. Because
of the rain, not the coyotes, though I have heard coyotes take
an interest in planes. That was back in 2000, kayaking on
Wanapum lake. California bighorn sheep herds rumbling by
in the night, beaver, otter. A girlfriend and I were camped at
Whiskey Dick Creek. A few A-6s from the Whidbey Island Naval
Base came winding down the river, between the cliffs, making
big noise. The coyotes across the river howled back at them.
Heh. Tonight’s coyotes turn out to be Cole crying. Something
about the rain on the tent where he sleeps with Riley. Erik takes
him into his truck.
Breakfast. Plenty of time to remove an old 72 MHz Rx from
FrankenShogun and stick in a Lemon 2.4GHz. I fool with the
programming.
Wahatis peak. We generally fly on a south-facing point a few
hundred yards east of the peak and its cell towers. But the wind
is shearing from the southwest. It’s blowing hard, gusty, and
cold. It’s a second jacket, long underwear, and tuck-the-pantsinto-the-socks day.
The planes stay up. We fly to the right side of the point, in a
big bowl. Chris flies his EZ glider. Steve, his Super-Scooter,
Boomerang and a 60" Lumberjack. Damian flies a 48"
homebuilt delta. Good flights.
I’m determined to fly my FrankenShogun. Steve throws it for
98
some tests, back in the flat where the sage is forgiving and
there are few rocks. It wants to nose under. I add weight to
the nose and give it more up. A thirty-foot flight into the wind.
Seems good enough. Steve launches it from the rocks at the
lip of the bowl. Probably turbulence, but it dives straight down,
nose in. It and a couple chips cartwheel thirty feet off to the
right and downwind. Phooey.
Oddly, for something so fragile, it’s hardly damaged. Two
chunks of the right wingtip have broken off. I can glue them
back on with Gorilla glue mixed with a drop of water, which will
foam to fill any voids. I’ll find a grassier spot to trim the thing.
FrankenShogun will fly again. It won’t look much worse than it
already does.
Chris says, “Why don’t you throw something made of foam?”
Yeah, of course. My Sonic, dependable, flies great. So does
Steve’s Super Scooter, till for the second time in it’s history its
nose gets crunched. He gets out the 60" Lumberjack. For some
reason it wants to do the same nose dive as my Shogun. Being
EPP it survives to fly quite elegantly.
We have about five planes in the air. There is a large black
cloud slowly moving toward us along the ridge, from the west.
Hail is bouncing off my sunglasses. Erik says, “Was that the
Yakima firing range?” Steve says, “I think it was thunder.” Chris
says, “Naw. It’s probably the firing range. We were hearing
them all day yesterday.” The debate continues till we hear an
extended peal. The black cloud is getting closer. We start to
see small flashes. Steve says, “I hear lightning can strike from
ten miles away.” Erik says “I hope the lightning will hit the
towers on Wahatis instead of us.” I say, “Maybe it’s time we got
off this ridge.”
Within a few minutes all of us are landing our planes. We all
overmisunderestimate how far back the wind will sweep us.
Steve, Chris, and I all land within twenty-feet of each other,
seventy yards back on the flat.
R/C Soaring Digest
Cole and Reily with Philip’s Sonic and Steve’s Super Scooter
As we leave, the hail cuts loose. Within
minutes the dirt roads are peppered
partly white. A mile east and we are out
of the precipitation.
Now, finishing up. A world without
coincidences would be so statistically
improbable that it would make me
believe in a higher being. Perhaps one
who had flunked statistics.
Final quote: “It was scary when I saw this
guy charging around the corner at me.”
Leaving for a slope trip last year I had
gotten as far as the Cle Elum Safeway.
June 2014
Erik & Damian and a couple of cell towers. Looking ENE at
Wahatis Peak
That’s about 80 miles out of Seattle.
I was off at the end of the milk and
cheese-fude isle when all of a sudden
there is this big bearded guy in a camo
jacket and hat charging his cart straight
at me with a glower on his face. It was
scary but I acted like nothing was out of
the ordinary. Then I recognized Damian.
So we laughed. Well, here it is a year
later about half-way back from Saddle
Mountain, and I’m getting decaf in the
Cle Elum MacDonalds. I’m waiting at the
counter, which is arranged so that I can
see through the drive-in-window. There’s
Damian, in his truck. I try to get his
attention, but that’s not happening. So
I go outside and do my scariest charge
around the corner straight toward the
hood of his truck. After recognition and
the adult guffaws that substitute for littlegirl giggles, insert the above quote here.
I remind him of when he charged me, in
the Safeway. He was stopping for real
coffee. We talk about setting up the next
trip.
99
T
om’s
ips
Large scale CG balancer
Tom Broeski, T&G Innovations LLC, [email protected]
Here’s a simple CG balancer for large
scale planes.
Place under the plane and push down.
Lets you lift plane from ground, rather
than having to lift plane by hand and then
setting it on something.
I used 1/2" aluminum rod, but you can
use a pipe and “T” rig also. Works great.
100
R/C Soaring Digest
June 2014
101
Was this manual useful for you? yes no
Thank you for your participation!

* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project

Download PDF

advertisement