RC Sailplane Aspect Ratio
Wayne Angevine
mailing list, but I've also posted it to rec.models.rc and sci.aeronautics.
The sci.aeronautics people may be interested in some of the tradeoffs
that go into the design of an aircraft with this particular set of
constraints, and some of this could even apply to RPVs or other "real
work." I've taken an exclusively empirical approach, unsullied by
any actual mathematics :-). I've also tried to look at the whole
system and all aspects (pun intended) of the task.
The question was, "What is the maximum practical aspect ratio for a
radio-controlled sailplane?" The motivation for the question is the
observation that induced drag is inversely proportional to aspect
ratio, all other things being equal.
There is a competing school of thought which asks the question,
"What is the *minimum* practical aspect ratio for a radio-controlled
saiplane?" These people have had some notable successes, particularly
in span-limited classes, but I won't address that in this article.
For several reasons, small chords are a problem. The reasons are:
1) Reynold's number - proportional to chord. This can have large
and grossly non-linear effects in the range in which most RC
sailplanes fly.
2) Thickness - proportional to chord for a given section. The
strength of the section is proportional to the square of the
thickness.
3) Accuracy - it seems to me that the performance impact of
inaccuracies in the construction of the wing will increase as
the chord gets smaller. I'm talking about errors like waviness
or simply not having the exact profile on your rib or template.
The absolute magnitude of the errors will be approximately
constant, so the magnitude relative to the thickness will be
greater at smaller chords. This may be offset to some extent
by the lower Reynold's number I mentioned above, since errors
may matter less at lower Reynold's number. Accuracy is critical
to the performance of modern laminar-flow airfoils.
Given all these difficulties, it seems to me that the minimum practical
chord is six to seven inches. If so, the maximum practical aspect
ratio would be 13/1 at two meter span, up to 20/1 at 120" span.
Let's look at another question: If high aspect ratio is so good,
how come you (almost) never see a plane which is outside the "normal"
range? The "normal" range is from about 8/1 for some 2-meters to
13/1 or so for unlimiteds.
RC sailplanes have to do several things besides glide. In particular,
they also have to launch, land, and be controllable from the ground.
Safe launching depends largely on absolute weight, because the launching
person and launch equipment have to be able to accelerate the plane to
safe flying speed before it leaves the launching person's hand. The
practical limits today seem to be around 12 oz/sq.ft. wing loading
for launching on a "normal" winch, 9 oz/sq.ft. for high-start launches,
and quite a bit higher for launches on an (old-style) F3B winch with
highly trained crews.
High launching, at least as currently practiced, depends on the ability
to pull many, many Gs during the zoom (ping) phase of the launch. I
have heard numbers like 30-40 Gs bandied about, but I've never seen a
measurement. It's also handy to be able to handle wind gusts, clumsy
pedal feet, etc. without blowing the wings off (believe me, I know.)
There is a three-way tradeoff between strength, weight, and complexity
in any structure. The higher the aspect ratio of the wing, the stronger
the structure needs to be. In practice, this means that, to achieve
the same ability to pull Gs, the wing must be both heavier and more
complex than a wing of the same area with lower aspect ratio. By
"more complex" I mean that it must be made of more exotic materials
and with more exotic processes.
Strength also is important during the landing phase, particularly if
the landing area is not perfectly smooth. Tearing the wings off by
catching a wingtip is not the way to have a nice day.
To control the plane, you must be able to see it. People I have
talked to in the RC cross-country game say that chord is more important
than span for visibility. The possible loss of performance from poor
ability to control the plane accurately could be quite large. Furthermore,
a wing of higher aspect ratio requires more precise control, since
the derivative of lift coefficient with angle of attack is steeper.
I think another reason we don't see high aspect ratio sailplanes is
that the performance of RC sailplanes today is sufficient for the
tasks we fly. The only subfield which needs ever-greater plane
performance is F3B, and the primary need there is for speed, not L/D.
In fact, I think that an argument could be made that the improvement
in performance of F3B planes has leveled out in the last several years.
The pilots keep getting better, though.
Does anyone have any other thoughts?
Wayne Angevine way...@cadnetix.com
Boulder, Colorado {uunet,boulder,nbires}!cadnetix!waynea
Al Bowers
>The question was, "What is the maximum practical aspect ratio for a
>radio-controlled sailplane?" The motivation for the question is the
>observation that induced drag is inversely proportional to aspect
>ratio, all other things being equal.
Some years ago I was posed a similar question which went along the
lines of 'what is the optimum airfoil design'? While at first this
seems some what unrelated in actual fact knowing the minimum Reynold's
number a particular airfoil can stand will indicate the maximum aspect
ratio. This question was posed by Mr. Christopher Ray of Swarthmore,
PA and I hope he doesn't mind my using his name or my response to his
question.
>For several reasons, small chords are a problem. The reasons are:
>1) Reynold's number - proportional to chord. This can have large
> and grossly non-linear effects in the range in which most RC
> sailplanes fly.
Exactly, once Re drops below 500,000 things can get weird real fast
and once you are below about 100,000 your options are seriously
limited.
>2) Thickness - proportional to chord for a given section. The
> strength of the section is proportional to the square of the
> thickness.
>3) Accuracy - it seems to me that the performance impact of
> inaccuracies in the construction of the wing will increase as
> the chord gets smaller. I'm talking about errors like waviness
Errors of this sort tend not to be as important as profile shape.
Smoothness is not as critical as you might at first expect, shape is.
If the roughness is buried in the boundary layer it tends not to
affect the results. Also beware of low speed airfoil codes as they
USUALLY require excessive smoothness for analytical performance which
is not required for high performance in flight. See NASA TM 86035.
E-mail me if you can't get a copy at your library... (Please excuse
my blowing my own horn concerning the report... ;-) ;-) ;-))
> may matter less at lower Reynold's number. Accuracy is critical
> to the performance of modern laminar-flow airfoils.
Not quite true. The major problem at low Renyold's numbers is getting
the boundary layer to become and STAY turbulent. This is the reason
for turbulator strips proliferating on hand launched RC sailplanes.
>Given all these difficulties, it seems to me that the minimum practical
>chord is six to seven inches. If so, the maximum practical aspect
>ratio would be 13/1 at two meter span, up to 20/1 at 120" span.
This number is very dependent on the airfoil. I would expect an old
NACA 6-series airfoil (especially a 65- or 66- series) to have major
problems at these chords on a thermal type aircraft. A slope racer
might get away with it. Some of the newer class of airfoils may be
able to go to slightly smaller chords. This can also explain how some
of the old favorite airfoils can be made to work reasonably at these
chord lengths. Witness Mark's Models Windfree, Clark-Y airfoil
(turbulent) and an aspect ratio 18 (chord of 5.6 inches).
>There is a three-way tradeoff between strength, weight, and complexity
>in any structure. The higher the aspect ratio of the wing, the stronger
You obviously understand the challenge...
>I think another reason we don't see high aspect ratio sailplanes is
>that the performance of RC sailplanes today is sufficient for the
>tasks we fly. The only subfield which needs ever-greater plane
>performance is F3B, and the primary need there is for speed, not L/D.
L/D is speed. All you need for speed is a higher wing loading, if
you're at the same lift coefficient L/D will be the same to a first
order of magnitude approximation. Reynold's number will play a part
but will be secondary as long as you aren't operating at the Reynold's
number of an under priviledged butterfly. ;-)
Newer airfoils that may assist in this area are from the Eppler and
the Horstmann-Quast series. Some of the Ritz airfoils may also be of
use. I wish I could give you a better source but there is a gentleman
at Arnold Engineering Development Center (AEDC) by the name of Chuck
Anderson who has advertised the availability of some of this
information. I will see if I can contact Mr. Anderson and perhaps
post a followup.
--
Albion H. Bowers bow...@elxsi.dfrf.nasa.gov ames!elxsi.dfrf.nasa.gov!bowers
NASA Ames-Dryden Flight Research Facility, Edwards, CA
Aerodynamics: The ONLY way to fly!
Live to ski, ski to live...
Paul Raveling
In article <11...@cadnetix.COM>, way...@cadnetix.COM (Wayne Angevine) writes:
>
> To control the plane, you must be able to see it. ...
> I think another reason we don't see high aspect ratio sailplanes is
> that the performance of RC sailplanes today is sufficient for the
> tasks we fly.
That seems like the key. Full-sized sailplanes go to
extreme aspect ratios and high wing loading in order
to optimize cruise performance. Since an RC sailplane
has to stay local to be seen, it's greatest need is
to work lift rather than to cruise or penetrate sink.
For working lift, sink rate is more important than
glide slope.
For low sink rate, you want low wing loading and low
induced drag. Low wing loading encourages large chord;
low induced drag encourages narrow chord. Somewhere
these opposing factors have to balance, and the natural
balance for RC need not go to extremes for the sake of
cruise performance.
A parallel in full-sized sailplanes would be the Blanik,
which has Fowler flaps that extend the chord but don't
add a lot of camber. The intended mode of operaton is
to extend them for thermalling, retract for cruising.
Ease of construction and durability are probably as
important for RC. Most RC sailplanes I've seen don't
have much taper in their planform; I presume this is
because the wings, and especially the tips, take an
occasional beating on landing & need good strength
to keep from being trashed.
----------------
Paul Raveling
Rave...@isi.edu
Steven Philipson
> > I think another reason we don't see high aspect ratio sailplanes is
> > that the performance of RC sailplanes today is sufficient for the
> > tasks we fly.
Not true. Competition fliers go to great lengths to maximize the
performance of their aircraft, and technological innovation is a major
part of the game. If aspect ratio by itself were that important, we'd
be seeing higher aspect ratio aircraft.
> That seems like the key. Full-sized sailplanes go to
> extreme aspect ratios and high wing loading in order
> to optimize cruise performance. Since an RC sailplane
> has to stay local to be seen, it's greatest need is
> to work lift rather than to cruise or penetrate sink.
> For working lift, sink rate is more important than
> glide slope.
I disagree. First, cruise performance IS important even when
one flies from a fixed location. Aircraft that attain high L/D at
a high airspeed have a strong advantage in windy conditions (it enables
them to fly further downwind and still be able to return to the field)
and when there are strong local downdrafts or only isolated spots of
lift. Their ability to efficiently move from one thermal to another
is a big advantage, even when thermals are less than a mile apart.
"Floaters", i.e. low sink rate, low L/D aircraft, can stay up in very
light lift, but if they do not launch into lift, they have trouble
reaching better conditions (and coming back from there).
Cross-country competition is now the rage in California. We tend
to see gliders with slightly higher aspect ratios than for non-X/C
planes, but they aren't all that much higher.
There are two factors that I believe are responsible for the
relatively low aspect ratios we see on model sailplanes. One is
span limitation by contest class. If a glider is limited to a
relatively small span, it is often more effective to go with a
lower aspect ratio planform than a higher aspect planform of the
same span but with much less wing area, lower Reynolds number, etc.
A higher aspect ratio with the same wing area might be more efficient,
but you can't go to the higher span due to the span limit for the
class.
Second, structural considerations are very important. The aircraft
has to withstand very heavy launch loads (20 g loads are common), be
resistant to "ground contact", and avoid flutter during both launch and
high speed cruise. These goals are more difficult to achieve with
high aspect ratio wings.
Steve
(the certified flying fanatic)
ste...@decwrl.dec.com
Al Bowers
>> > I think another reason we don't see high aspect ratio sailplanes is
>> > that the performance of RC sailplanes today is sufficient for the
>> > tasks we fly.
> Not true. Competition fliers go to great lengths to maximize the
>performance of their aircraft, and technological innovation is a major
>part of the game. If aspect ratio by itself were that important, we'd
>be seeing higher aspect ratio aircraft.
I'm afraid I have to disagree. Aspect ratio is only one part of the
total equation. If you are flying a slow flying airplane then you
have to balance area (or lift coefficient) against aspect ratio
(induced drag). This is also true of faster designs. The trade offs
are well known, and the limits of Reynold's number prevent higher
aspect ratios from being flown in today's 'typical' RC sailplane.
Once Reynold's numbers drop below about 60,000 to 100,000 there just
isn't anything you can do to sustain attached flow over wings.
>> That seems like the key. Full-sized sailplanes go to
>> extreme aspect ratios and high wing loading in order
>> to optimize cruise performance. Since an RC sailplane
... deleted ...
> Cross-country competition is now the rage in California. We tend
>to see gliders with slightly higher aspect ratios than for non-X/C
>planes, but they aren't all that much higher.
> There are two factors that I believe are responsible for the
>relatively low aspect ratios we see on model sailplanes. One is
>span limitation by contest class. If a glider is limited to a
Span is relatively insignificant as a factor. Construction techniques
can keep wing loadings relatively constant. The major problem is
Reynold's number.
... deleted ...
> Second, structural considerations are very important. The aircraft
>has to withstand very heavy launch loads (20 g loads are common), be
I must be misunderstanding what you guys are refering to as 'launch
loads'. I think of loads due to g as acceleration loads, I think what
you are refering to is wing root bending load due to the downward pull
of the high start or winch. In this case it is an equivalent load.
Static load tests of aricraft are normally done in this manner, but is
not a 20 g load. A 20 g load would load up the servo and radio pallet
by 20 g's also which this kind of load does not.
Steven Philipson
> > [...] If aspect ratio by itself were that important, we'd
> >be seeing higher aspect ratio aircraft.
> total equation.
Sounds like you're agreeing to me. I was saying that aspect ratio
isn't the whole story, and you're saying it's only part of the story.
Am I missing something here?
> > There are two factors that I believe are responsible for the
> >relatively low aspect ratios we see on model sailplanes. One is
> >span limitation by contest class. If a glider is limited to a
> Span is relatively insignificant as a factor. Construction techniques
> can keep wing loadings relatively constant. The major problem is
> Reynold's number.
Let's try that again. When span is limited to a maximum by the rules
for each contest class, it by definition is a limiting factor. If you
can't compete in standard class with a span of greater than 100 inches,
increasing span beyond that will elliminate you from competition in that
class.
> > Second, structural considerations are very important. The aircraft
> >has to withstand very heavy launch loads (20 g loads are common), be
> I must be misunderstanding what you guys are refering to as 'launch
> loads'. I think of loads due to g as acceleration loads, I think what
> you are refering to is wing root bending load due to the downward pull
> of the high start or winch. In this case it is an equivalent load.
> Static load tests of aricraft are normally done in this manner, but is
> not a 20 g load. A 20 g load would load up the servo and radio pallet
> by 20 g's also which this kind of load does not.
The wing and fuselage towhook-attach structure must sustain the full
load that the aircraft experiences whether or not all parts of the
aircraft experience that load. Sure, the servo mounts aren't seeing
that load, but that isn't the critical part of the aircraft. In
full scale gliders, auto and winch tow speeds are limited to insure
that the wings and tow hook structures don't have their limits
exceed. The pilot wouldn't feel those loads, but the structures
have to be strong enough to handle them.
Now that I've said all that, there is some evidence that the
flight loads immediately after tow release can exceed the tow loads.
Many structural failures occur after towline release as the pilot
applies up elevator in an attempt to convert speed to altitude.
Towline load measurements have shown tensions of 20 times the aircraft
weight in a few of these cases, and the wings failed after towline
release. Yes, it's a bending load failure. It's also hard to
determine the load at any given point without an on-board instrument.
Most of such failures are a result of errors in pilot technique.
Some pilots make the mistake of applying up elevator before releasing
the towline. Some also dive excessively in an attempt to build
speed. Ah well, it's good for the kit industry.
Here's a data point on structural integrity. Joe Wurtz (famed
model aircraft X/C competitor) has a composite construction aircraft
with about 11 foot span that weighs 9 lbs. One of Joe's favorite
demos is to have two people support the ends of the center span,
and have another person *sit* on the center of the panel. Yeah,
Joe knows how to build 'em strong.