GA Airfoils
Terrben
have been digesting it since. He makes a compelling argument for avoiding many
of the popular NACA and NASA airfoils and using his "evolutionary" airfoils,
instead. His airfoils, he claims, stall softer, lift better, and are within in
a drag count or two of the slickest NASA airfoils without the detrimental
characteristics they apparently exhibit. His reasoning and evidence seem
solid.
My question is: If what he says is correct, why is there not a strong move by
designers and homebuilders to incorporate his airfoils into their projects? Am
I missing something?
On another tack but still in reference to Mr. Riblett's work, rotor blades
usually, if not excusively, use very low pitching moment airfoils; not
necessarily symmetrical but with Cm's of essentially zero. Mr. Riblett's
airfoils are derived from actual or modified NACA thickness distributions which
are, in fact, symmetrical and are, apparently, quite good airfoils in their own
right for uses requiring low pitching moment (tail surfaces, aerobatic wings,
and, I would think, rotors). Would anyone be willing to comment on the use of
Mr. Riblett's symmetrical thickness distributions as the planform of choice for
a gyrocopter rotor blade?
Regards, Terry Bendickson
Al Mills
impression that lift and drag were a compromise of the airfoil and washout
(twist) was the function which controlled stall characteristics.
What does he mean by "lift better"? More lift without more drag? I'd
like to see that. What's "evolutionary" about them? What's he doing
different from the many hundreds of airfoils already available?
He may have something new on rotary wings, there's not that many.
>His airfoils, he claims, stall softer, lift >better, and are within in
> a drag count or two of the slickest >NASA airfoils without the detrimental
>characteristics they apparently exhibit.
--
AL Mills
...but we are all as
an unclean thing,
and all our righteousness'
are as filthy rags, and
we all do fade as a leaf,
and our iniquities like the wind
have taken us away.
Is. 64:6
Mike Sieweke
> I purchased Harry Riblett's "GA Airfoils" booklet at Osh Kosh this year and
> have been digesting it since. He makes a compelling argument for avoiding
...
> My question is: If what he says is correct, why is there not a strong move by
> designers and homebuilders to incorporate his airfoils into their projects? Am
> I missing something?
I can think of several reasons, none very complimentary to aerodynamicists.
Engineers and scientists in general are very conservative and not likely to
trust data coming from outside their fraternity - NASA and university labs.
Look at how surprised everyone was by the Facetmobile. And universities do
not emphasize the importance of stall characteristics, or else no one would
ever use the 230xx airfoils.
And then there's the lack of wind tunnel tests. Few people really understand
how to interpret Riblett's data, so they go with well-tested airfoils,
assuming NASA and NACA endorsement is a guarantee of success.
Some newer homebuilts ARE using Riblett's airfoils: Skystar Kitfox, Ultravia
Pelican, the new Rans S-7. As more designs successfully use the GA airfoils,
more designers should jump on the bandwagon.
I find it interesting that a KR-2 group went to great expense to have an
airfoil designed for them and wind-tunnel tested. The design parameters
were very close to those of the Riblett airfoils.
--
Remove the "nospam" from my address if you reply.
Mike Sieweke - msie...@nospam.ix.netcom.com
Yamhill, OR - http://www.netcom.com/~msieweke
Mike Sieweke
> Well, what detrimental characteristics could those be? I've always had the
> impression that lift and drag were a compromise of the airfoil and washout
> (twist) was the function which controlled stall characteristics.
The detrimental characteristics would be poor stall characteristics and high
moment coefficient. The sharp stall and por low-Reynolds-number performance
of many of the NACA and NASA airfoils produce dangerous stall characteristics,
which requires washout to maintain any control. The high moment coefficient
of the NASA GA(W) and NLF airfoils cause excessive trim drag.
Washout is a kluge that compensates for poor choice of airfoil.
> What does he mean by "lift better"? More lift without more drag? I'd
> like to see that. What's "evolutionary" about them? What's he doing
> different from the many hundreds of airfoils already available?
The key is improving the Clmax/Cdcruise ratio. Wing size is determined
by stall speed and Clmax. So to minimize cruise drag, that ratio should
be maximized. At the same time we want to have a benign stall, for safety.
These are the benefits of the Riblett airfoils.
They are evolutionary because they are based on NACA base airfoils and
mean lines, with slight changes to improve the stall and lighten aileron
pressure.
Walter Lounsbery
information, and it seems that you are already on that path by posting this
message. I highly recommend looking at some of the more accepted works by
Eppler, several NASA GA airfoil investigators, some of the older Wichita
State University stuff (thanks Gene!), and even Barnaby Wainfan's material.
I'm afraid I only have dim recollections of some of Riblett's articles from
years ago, and I haven't read his book.
There is a big difference between your question "Have I missed something?",
and my question, "What did Mr. Riblett leave out of his book?" Airplanes
seem to be flying quite OK.
But maybe I can add my two cent productively in some other areas than
disputing Riblett (which is surely an easy debate). I designed a few
airfoils for some Boeing projects about 14 years ago, using cutting-edge CFD
analysis and empirical data. Some airfoils were based on rotor airfoil
design technology obtained from Boeing Helicopters. Since these were
exercises in tailoring, proof-of-concept, and building expertise on
underfunded military studies, the airfoils never made it to actual testing.
The non-symetric airfoils are intended to either obtain a higher maximum
lift coefficient or to shift the "drag bucket" to higher lift coefficients.
I would not discourage use of a non-symmetric airfoil for a rotor as long as
it is already proven in use at a similar size (chord) and speed (actual
speed at the rotor, if the chord is similar it is likely the aircraft speed
will also be similar). In any case you have to be careful of shock
formation on the airfoil. Some airfoils will perform nasty tricks in the
transonic regime. This is why it is best to stick to proven airfoils in
this case. If the Riblett profile has been tested in a transonic wind
tunnel at the proper Reynold's number, as well as on a vehicle similar to
what you are thinking about, have at it. Considering that you will not
likely see significant performance differences from tweaking the rotor
profile on an experimental, GA aircraft, you might want to employ safety as
your guiding factor.
Good luck with your gyrocopter!
Walt Lounsbery
Terrben <ter...@aol.com> wrote in message
news:19991201113935...@ng-co1.aol.com...
> I purchased Harry Riblett's "GA Airfoils" booklet at Osh Kosh this year
and
> have been digesting it since. He makes a compelling argument for avoiding
many
> of the popular NACA and NASA airfoils and using his "evolutionary"
airfoils,
> instead. His airfoils, he claims, stall softer, lift better, and are
within in
> a drag count or two of the slickest NASA airfoils without the detrimental
> characteristics they apparently exhibit. His reasoning and evidence seem
> solid.
>
> My question is: If what he says is correct, why is there not a strong
move by
> designers and homebuilders to incorporate his airfoils into their
projects? Am
> I missing something?
>
Terrben
Second, part of Mr. Lounsbery's post regarded shock formation on an airfoil.
">I would not discourage use of a non-symmetric airfoil for a rotor as long as
>it is already proven in use at a similar size (chord) and speed (actual
>speed at the rotor, if the chord is similar it is likely the aircraft speed
>will also be similar). In any case you have to be careful of shock
>formation on the airfoil. Some airfoils will perform nasty tricks in the
>transonic regime."
Just to recap what I think I know, in the transsonic regime, portions of the
airfoil are subsonic but as the airfoil acts on the air, pushing it around to
get by, so to say, the air must speed up and is forced into the supersonic
range, thereby creating a shock wave in that section of the airfoil. Said
shock waves disrupt the airflow and, hence, lift and drag characteristics of
the airfoil as well as the center of pressure. The net result is performance
unreliability.
I would suspect that for any given airfoil planform, chord, Reynold's number,
and angle of attack there is a speed below which the "transsonic nasties"
become an non-factor. Is there a "rule of thumb" that can be applied; say,
Mach 0.75 or 0.80, to never exceed at any Reynold's number or chord to remain
in the "safety zone"?
I've heard that 750 fps tip speed is a good upper limit. Any thoughts?
Again, thanks for the thoughtful and informative replies.
Terry Bendickson
David Lednicer
1) He does his airfoil design work with the Eppler code, a program with
some serious shortcomings. I am not aware of his airfoils ever being
wind tunnel tested. This is necessary to properly quantify their
behavior.
2) He has claimed that the proper choice of airfoils can reduce induced
drag. This is incorrect and shows a lack of knowledge of the subject.
3) I think he probably has improved the NACA 60-series airfoils a bit.
His modifications probably do produce better stall characteristics.
4) The claim that his airfoils "lift better" is meaningless.
5) His claims about the harsh stall characteristics of the NACA 5-digit
series are correct, but nothing new. His extrapolation of these claims,
calling for the grounding of the ATR-42, show his ignorance of 3-D
aerodynamics. There are a lot of aircraft with NACA 5-digit airfoils
that stall just fine, because the stall behavior has been properly
tailored. For more information, see the Incomplete Guide to Airfoil
Usage at: http://amber.aae.uiuc.edu/~m-selig/ads/aircraft.html
6) His knowledge of US copyright and patent law is lacking. I wanted to
include his airfoils in the UIUC airfoil database, but he claimed they
were copyrighted, which is impossible. Airfoils can be patented (one of
mine is) but not copyrighted. Because of his agitation, his airfoils
are not in the database, but I have the coordinates in my own personal
database. On the other hand, his "book" contains unapproved reprints of
copyrighted material, which is illegal.
7) I am not aware of any rotorcraft airfoils he has designed. However,
the UIUC database contains rotorcraft airfoils designed at Boeing and
Sikorsky.
-------------------------------------------------------------------
David Lednicer | "Applied Computational Fluid Dynamics"
Analytical Methods, Inc. | email: da...@amiwest.com
2133 152nd Ave NE | tel: (206) 643-9090
Redmond, WA 98052 USA | fax: (206) 746-1299
highflyer
I have argued with David Lednicer in the past about aerodynamics and
some
aerodynamic details. David is one of the leading people in the country
in
this field. He earns my respect far above Mr. Riblett. Mr. Riblett
does
do an excellent job of marketing his book and his airfoil designs.
The fact is, that any of a number of airfoil designs will do an
excellent
job for a GA aircraft. Each airfoil design represents a compromise
between
features, and that compromise should be selected in the light of ALL of
the
many compromises that are balanced to design an airplane.
Wing area is normally determined by the low speed characteristics
required
to make the airplane safe and easy to land. Blunting the nose and
increasing the camber and thickening the airfoil will generally increase
the maximum lift coefficient for an airfoil of a given general shape.
Other factors are the Reynold's Number where the airfoil will operate.
This varies widely in GA aircraft. That is why the airfoils of
sailplanes
tend to be thicker and more "tadpole" shaped than those for powered
planes.
Washout is generally used to tailor the lift distribution along the wing
to optimise it. Before they figured that out, there was a tendency for
designers to design beautiful elliptical wing planforms to ensure the
desired elliptical lift distribution without washout! Unfortunately,
cutting into the trailing edge to insert things like ailerons and flaps
destroyed the lift distribution anyway and required compensation. When
designers found that they could take a simple straight tapered wing and
twist it just a little and still get the optimal elliptical lift
distribution
and have a wing that was MUCH easier and cheaper to build, those lovely
"Spitfire" wing planforms went south.
A "soft and gentle" stall depends on many factors, of which the airfoil
lift characteristic near the stall is only one relatively minor one.
Even
the 23012, which has a discontinuous lift curve at stall, can give a
gentle and progressive stall when the wing is designed for that purpose.
Any rectangular planform, with NO twist, will tend to stall
progressively
and give a relatively gentle stall characteristic with good burble for
warning over the tail surfaces. This can be degraded if the wing is
allowed to deform when it picks up a load, and render the stall less
predictable, as in the Tomahawk.
Actually, a "symetrical" airfoil does not often have a zero moment
coefficient. In fact, the symetrical airfoil may have a moment that
is opposite that of the cambered airfoil. The cambered airfoil tends
to have a moment coefficient that is "destabilizing" because the actual
center of lift of the airfoil tends to move forward as the angle of
attack
increases. With most symetrical airfoils, the center of lift tends to
move aft with increasing angle of attack. This movement is
"stabilizing"
since it tends to reduce the angle of attack.
A cambered airfoil can have the moment coefficient reduced to
approximate
zero by merely reflexing the trailing edge. This is how the 23012
airfoil
obtains its near zero moment coefficient. I have heard it said that
this
moment coefficient of the 230xx series airfoils is what made the
monospar
wing of the DC-3 possible. I cannot verify the truth of that, however.
In any event, the moment coefficient can be caused to vary clear off the
scale by the simple expedient of lowering a flap or moving an aileron!
That generally becomes the design constraint in wing torsional rigidity.
Dynamic behavior of the wing/control surface pair also becomes extremely
important. Bad dynamic behavior is a common cause of flutter that will
cause rapid disassembly in flight. :-)
Laminar flow airfoils can give a "drag bucket" for the angle of attack
range where significant laminar flow can be achieved. No airfoil can
sustain laminar flow after about the thickest part of the airfoil. That
is why most laminar flow airfoil designs have their maximum thickness
quite far back from the leading edge. Typically, a "laminar" airfoil
will have MORE drag than a more conventional airfoil outside of the
drag bucket, or with a "standard roughness" surface. Any spanwise
waviness in the wing will totally destroy laminar flow, as well as any
protuberances from the surface.
In summary ... airfoil selection is only one of the many compromises
that
make an aircraft design. It is not one of the more important items, and
there are many others that will have more effect on handling qualities
and
flight performance, although airfoil design is important.
--
HighFlyer
Highflight Aviation Services
highflyer
> I've heard that 750 fps tip speed is a good upper limit. Any thoughts?
>
For a proven airfoil of standard thickness 750 fps tip speed is probably
safe. Remember that the actual tipspeed is the speed determined by the
RPM and diamter of the rotor plus and minus the horizontal translation
speed of the vehicle. For example, most small rotary wing aircraft will
have a horizontal translation speed of about 150 feet per second. That
would imply that you want to keep the tip speed derived from rotor RPM
and diameter below 600 feet per second. :-)
Mike Sieweke
> >
> > My experience with Riblett and his airfoils is this:
> > 2) He has claimed that the proper choice of airfoils can reduce induced
> > drag. This is incorrect and shows a lack of knowledge of the subject.
I'm not sure what you are saying here. It is pretty obvious that an airfoil
with a high moment coefficient will cause a relatively large tail down load
at cruise. This tail load increases the lift required of the wing, and thus
increases the induced drag. Choice of an airfoil with low moment coefficient
will reduce induced drag over an airfoil with high moment coefficient, all
else being equal.
There is a common misconception that a tail down load can be compensated
by "thrust" from the horizontal tail operating in the wing's downwash.
NASA tests indicate that this is false, and that a negative tail load
always increases drag.
> > 5) His claims about the harsh stall characteristics of the NACA 5-digit
> > series are correct, but nothing new. His extrapolation of these claims,
> > calling for the grounding of the ATR-42, show his ignorance of 3-D
> > aerodynamics. There are a lot of aircraft with NACA 5-digit airfoils
> > that stall just fine, because the stall behavior has been properly
> > tailored. For more information, see the Incomplete Guide to Airfoil
> > Usage at: http://amber.aae.uiuc.edu/~m-selig/ads/aircraft.html
Where did you hear that Riblett called for the grounding of the ATR-42?
I haven't heard that elsewhere. Riblett WAS concerned that the FAA was
grounding a whole class of airplanes due to a problem with icing on the
ATR-72.
His concern with this class of airplane has nothing to do with stall
characteristics - it has everything to do with icing characteristics.
Riblett merely explained why the ice appeared where it did, and why not
all airplanes in this class should be limited in operation. Riblett's
claim was that the FAA was being TOO severe in grounding airplanes with
airfoils that should not display this icing characteristic.
Highflyer wrote:
> I have argued with David Lednicer in the past about aerodynamics and some
> aerodynamic details. David is one of the leading people in the country in
> this field. He earns my respect far above Mr. Riblett. Mr. Riblett does
> do an excellent job of marketing his book and his airfoil designs.
Unfortunately, Lednicer shows a limited knowledge of Riblett's work.
> Washout is generally used to tailor the lift distribution along the wing
> to optimise it. Before they figured that out, there was a tendency for
I have never heard of this from anywhere else. My readings indicate that
washout is used to tailor stall characteristics. When NACA tested tapered
wings, they discovered that several tapered wings with zero washout performed
as efficiently as the elliptical wing (i.e. high e factor). Some tapered wings
had higher e factors than the elliptical wing NACA tested.
> A cambered airfoil can have the moment coefficient reduced to approximate
> zero by merely reflexing the trailing edge. This is how the 23012 airfoil
> obtains its near zero moment coefficient. I have heard it said that this
> moment coefficient of the 230xx series airfoils is what made the monospar
> wing of the DC-3 possible. I cannot verify the truth of that, however.
The 230xx series airfoils have straight camber lines over the aft section
of the airfoils - they are not reflexed. They have their max camber located
extremely far forward - 15% in the 230xx - which causes a rather high
negative pressure peak on the forward part of the airfoil. This pressure
peak causes a flow separation bubble, but the flow reattaches in a short
distance at low angles of attack. At the stall, flow fails to reattach
after the separation bubble, which causes lift to drop from 1/3 to 1/2
with only one degree of increased angle of attack. This is the sharp stall
the 230xx series airfoils are famous for.
The DC-3 used NACA 4-digit airfoils. Perhaps you were thinking of the
DC-4?
> In summary ... airfoil selection is only one of the many compromises that
> make an aircraft design. It is not one of the more important items, and
> there are many others that will have more effect on handling qualities and
> flight performance, although airfoil design is important.
Airfoil selection can have a marked effect on the stall characteristics
of an airplane. Witness the Piper PA-24 Comanche, which had an untwisted,
tapered wing with a good choice of airfoil and a docile stall. Also witness
the BD-5A, which had unusually high stall speed due to bad airfoil choice, and
the Questair Venture, which had to graft on several mods to tame the dangerous
stall characteristics of the 230xx airfoil series.
I agree that airfoil choice is just one of the many compromises that make
an aircraft design. But it IS an important one and often misunderstood.
Steven Eberhart
> I find it interesting that a KR-2 group went to great expense to have an
> airfoil designed for them and wind-tunnel tested. The design parameters
> were very close to those of the Riblett airfoils.
I was envolved in the KR-2 airfoil project. Anyone interrested in the airfoils that
resulted from the project can get the details from http://www.newtech.com/nlf
The airfoils were designed by Dr. Ashok Gopalarathnam and Dr. Michael Selig at the
University of Illinois and were wind tunnel tested in the UofI wind tunnel. The
first plane to fly with one of the new airfoils was Troy Petteway's KR-2, N100TP.
So far all flight tests have been consistant with the CFD and wind tunnel analysis.
Low drag, gentle stall, good performance in the rain, etc. There are approximatly
15 KR-2 and KR-2S aircraft under construction with the new airfoils.
Steve Eberhart
new...@newtech.com
David Lednicer
your post. Just a few comments:
> Washout is generally used to tailor the lift distribution along the
> wing to optimise it. Before they figured that out, there was a
> tendency for designers to design beautiful elliptical wing planforms
> to ensure the desired elliptical lift distribution without washout!
I am not certain which aircraft you are referring to here. If you are
referring to the Spitfire, it did have washout. My analysis shows that
it didn't have an ellptical loading due to this. R.J. Mitchell used the
elliptical planform to get greater airfoil depth outboard, without
increasing airfoil t/c. He needed the depth for the outward retracting
landing gear and the ammunition trays.
> This can be degraded if the wing is allowed to deform when it picks up > a load, and render the stall less predictable, as in the Tomahawk.
An even better example of this is the Fw 190. Under load, as much as
50-60% of the wing stalls all at once.
> Actually, a "symetrical" airfoil does not often have a zero moment
> coefficient. In fact, the symetrical airfoil may have a moment that
> is opposite that of the cambered airfoil.
Here I disagree. Most symmetrical airfoils have near zero moments until
separation begins to develop on the airfoil. This is why they were long
used on helicopter rotor blades. Modern helicopter rotor blades now
have cambered airfoils, but this is because the means have been found to
design low moment cambered airfoils.
> I have heard it said that this moment coefficient of the 230xx series > airfoils is what made the monospar wing of the DC-3 possible.
Interesting idea, but not true. The root airfoil on the DC-3 is the
NACA 2215 and the tip is either the 2209 or 4412, depending on who you
believe (I am still working on tracking this one down).
Charles K. Scott
David Lednicer <da...@amiwest.com> writes:
> An even better example of this is the Fw 190. Under load, as much as
> 50-60% of the wing stalls all at once.
This can be verified by the numerous comments made by German pilots who
lauded the 190 A series for it's snappy rate of roll but often
mentioned that like the Mustang, it would suddenly blow out of a high G
turn into tumbling flight without warning.
Corky Scott
David Lednicer
msie...@nospam.ix.netcom.com (Mike Sieweke) wrote:
> I'm not sure what you are saying here. It is pretty obvious that
> an airfoil
> with a high moment coefficient will cause a relatively large tail
> down load
> at cruise. This tail load increases the lift required of the
> wing, and thus
> increases the induced drag. Choice of an airfoil with low moment
> coefficient
> will reduce induced drag over an airfoil with high moment
> coefficient, all
> else being equal.
> There is a common misconception that a tail down load can be
> compensated
> by "thrust" from the horizontal tail operating in the wing's
> downwash.
> NASA tests indicate that this is false, and that a negative tail
> load
> always increases drag.
Mr. Riblett's arguments were presented in a forum at Oshkosh. I
think the forum was in 1988. He claimed that the pressure distribution
of a 2D airfoil could in itself reduce induce drag. At the time I was
working for John Roncz, and I remember spending some time discussing
Mr. Riblett's assertions.
> Where did you hear that Riblett called for the grounding of the
> ATR-42?
You obviously don't read the letters column in Aviation Week and
Space Technology! He had a letter published there after the Roselawn
ATR-72 crash.
> Unfortunately, Lednicer shows a limited knowledge of Riblett's
> work.
No, I think you do. I have read his book, heard his forums, read
his letters to me and argued with him on the phone.
> I have never heard of this from anywhere else. My readings
> indicate that
> washout is used to tailor stall characteristics. When NACA tested
> tapered
> wings, they discovered that several tapered wings with zero
> washout performed
> as efficiently as the elliptical wing (i.e. high e factor). Some
> tapered wings
> had higher e factors than the elliptical wing NACA tested.
Washout is usually used to tailor stall characteristics, not span
loading.
> Also witness
> the BD-5A, which had unusually high stall speed due to bad airfoil
> choice,
Most BD-5s that have nasty stalls, have the characteristic due to the
method of wing construction. They make the leading edge bend in such a
way that a very tight radius develops in the metal, giving a full span
stall strip. BD-5 wings constructed by creating the leading edge bend
in a better fashion have better stall characteristics. This info comes
from a BD-5 builder I know.
> the Questair Venture, which had to graft on several mods to tame
> the dangerous
> stall characteristics of the 230xx airfoil series.
The 230XX airfoil series has a sharp, not dangerous stall. There
are many airplanes flying with the 230XX airfoils that stall just fine.
-David Lednicer
* Sent from RemarQ http://www.remarq.com The Internet's Discussion Network *
The fastest and easiest way to search and participate in Usenet - Free!
Mike Sieweke
> Mr. Riblett's arguments were presented in a forum at Oshkosh. I
> think the forum was in 1988. He claimed that the pressure distribution
> of a 2D airfoil could in itself reduce induce drag. At the time I was
> working for John Roncz, and I remember spending some time discussing
> Mr. Riblett's assertions.
Interesting. What is induced drag anyways? Of course it comes from
the energy that goes into the wingtip vortices, but how is it passed
on to the wing as drag? I have yet to hear a good explanation.
I'm not saying Riblett is right. He may be completely wrong. But I'm
not going to form an opinion until I understand the topic better.
> > Where did you hear that Riblett called for the grounding of the
> > ATR-42?
>
> You obviously don't read the letters column in Aviation Week and
> Space Technology! He had a letter published there after the Roselawn
> ATR-72 crash.
Do you know what issue this was in, or perhaps the year? I would like
to look it up. So far I have only your assertion that Riblett's stance
was not reasonable.
In Riblett's book he published a letter he sent to the NTSB specifically
outlining why the FAA should only limit the operation of airplanes that
use the 230xx or 430xx airfoils. This came after the FAA forbid all
turboprop-powered regional aircraft from flying in freezing rain and
drizzle.
> > Unfortunately, Lednicer shows a limited knowledge of Riblett's
> > work.
>
> No, I think you do. I have read his book, heard his forums, read
> his letters to me and argued with him on the phone.
Perhaps that is so. I have only his book, and everything in there
makes perfect sense to me. I'm willing to ignore some of Riblett's
eccentricities if he has a few valuable things to contribute.
> > Also witness
> > the BD-5A, which had unusually high stall speed due to bad airfoil
> > choice,
>
> Most BD-5s that have nasty stalls, have the characteristic due to the
> method of wing construction. They make the leading edge bend in such a
> way that a very tight radius develops in the metal, giving a full span
> stall strip. BD-5 wings constructed by creating the leading edge bend
> in a better fashion have better stall characteristics. This info comes
> from a BD-5 builder I know.
The factory demonstrator BD-5 also had unusually high stall speed. This
came from the use of an airfoil that does not perform well at the low
Reynolds numbers seen at the BD-5's stall.
> > the Questair Venture, which had to graft on several mods to tame
> > the dangerous
> > stall characteristics of the 230xx airfoil series.
>
> The 230XX airfoil series has a sharp, not dangerous stall. There
> are many airplanes flying with the 230XX airfoils that stall just fine.
They stall just fine straight ahead. Washout does a good job of taming
the straight-ahead stall, because the wingtips stall after the root -
keeping aileron control during the stall.
But what happens if the plane stalls while yawing left or right? While
the plane is yawing, the retreating wingtip is at a higher angle of
attack than the advancing wingtip. With sufficient yaw rate, the effect
of washout is eliminated and the wing will stall everywhere at once or
first at the tip.
You may say "Who would enter a stall while yawing?" Think of the turn
from base to final during landing. The pilot keeps the ball centered
during the turn, but his feet are a bit slow exiting the turn. He applies
a little too much right stick to level the wings, causing adverse yaw.
The plane yaws left at the same time the left aileron goes down, raising
the angle of attack of the left wingtip. The entire left wing stalls.
The pilot is smart. He knows what he did wrong, so he centers the
stick, pushes forward, and kicks in hard right rudder. But the wing
isn't going to unstall right away because the 230xx airfoils have
hysteresis in the stall. That means you have to lower the angle of
attack several degrees before the wing unstalls. By the time the
pilot can get the wing unstalled, it is too late. He's a statistic.
Or look at a light twin on losing an engine. Lose the left engine and
suddenly the plane starts yawing left and decelerating. The pilot had
better act quickly or the whole left wing will stall and the plane will
go into an sharp roll to the left. Another statistic.
The same thing happens if the plane slows below Vmc. The rudder can
no longer counter the yawing force of the engines and the deflected
aileron on the dead engine's wing. The plane is already flying slowly,
and it starts yawing into the dead engine. With a sharp-stalling
airfoil, you get a nice snap roll. Yet another statistic.
These situations are not unique to sharp-stall airfoils. But with a
sharp-stall airoil, everything happens more quickly. Things progress
much more slowly with a soft-stall airfoil. But that's not all.
The problem with the NACA 5-digit airfoils is not just the sharp stall;
the larger problem is what happens after the stall. The larger problem
is the hysteresis loop in the lift curve. Once the wing stalls, it
takes so much time and altitude to get it unstalled that you become
a statistic if you don't have a lot of altitude beneath you.
It is your opinion that this is not a dangerous stall; it's not a fact.
It is my opinion that this is a VERY dangerous stall characteristic.
David Lednicer
Induced drag is a 3D effect that comes from the creation of lift.
Lift is created by deflecting a volume of air downwards. This means
the deflected air has a small vertical velocity, called downwash. This
downwash has an influence back at the wing, raising the angle of attack
slightly. This then tilts the force vector, which is normal to the
oncoming flow, back slightly, resulting in small component of force in
the drag direction. This small component is called induced drag.
There is no way to escape induced drag, but there is a way to minimize
it, by tailoring the 3D lift distribution.
> Do you know what issue this was in, or perhaps the year? I would
> like to look it up. So far I have only your assertion that Riblett's
> stance was not reasonable.
I'll have to dig for this. It appeared quite a while after the
Roselawn crash. We all got a chuckle, but I didn't save a copy.
> The factory demonstrator BD-5 also had unusually high stall speed.
From what I have been told, the factory demonstrator had a wing
constructed the "wrong" way. I have been told that BD-5s constructed
the "good" way have much better stall characteristics. However, I
think you have a point - some of this sensitivity is probably due to
the low Reynolds number at stall.
Yes, I can picture where yawing can change a OK straight-ahead stall
into a bad stall. With one exception, I have only stalled Cessnas
which strike me as quite benign. My exception is the Tomahawk. There,
I was more aware of the tail buffet than the sharpness of the stall.
Come to think of it, I have also "stalled" a 777 fixed base simulator,
but there the stick pusher was really what I was experiencing.
However, look at the list of aircraft that have NACA 5-digit
airfoils. (http://amber.aae.uiuc.edu/~m-selig/ads/aircraft.html).
There are a lot of them. Do they all have nasty stalls characteristics?
-Dave Lednicer
Mike Sieweke
> Induced drag is a 3D effect that comes from the creation of lift.
> Lift is created by deflecting a volume of air downwards. This means
> the deflected air has a small vertical velocity, called downwash. This
> downwash has an influence back at the wing, raising the angle of attack
> slightly***. This then tilts the force vector, which is normal to the
> oncoming flow, back slightly, resulting in small component of force in
> the drag direction. This small component is called induced drag.
> There is no way to escape induced drag, but there is a way to minimize
> it, by tailoring the 3D lift distribution.
I've been trying to gain an intuitive understanding of induced drag, and
this explanation sounds very good. But at ***, doesn't this reduce the
effective angle of attack, forcing the wing to fly at a higher angle of
attack to get the same lift?
What intrigues me is that something as simple as sheared wingtips can
reduce induced drag. It doesn't quite fit in with what I think I know.
Perhaps the sheared tips do not reduce the induced drag by changing the
flow field around the wing - perhaps they just increase the e factor of
the wing.
> From what I have been told, the factory demonstrator had a wing
> constructed the "wrong" way. I have been told that BD-5s constructed
> the "good" way have much better stall characteristics. However, I
> think you have a point - some of this sensitivity is probably due to
> the low Reynolds number at stall.
I think so. In "Theory of Wing Sections" it shows that even at a RN
of 3 million, the 64-212 tends toward a very sharp stall - indicating
that the airfoil might not perform well at even lower RNs. Since the
BD-5 stalls at a RN of around 1.5 million, this seems a poor choice
of airfoil for the wing. The 64-218 used on the wingtip would be a
much better choice for the whole wing, since the "stall" is almost a
flat line. The 64-218 also performs better with flaps.
> Yes, I can picture where yawing can change a OK straight-ahead stall
> into a bad stall. With one exception, I have only stalled Cessnas
> which strike me as quite benign. My exception is the Tomahawk. There,
> I was more aware of the tail buffet than the sharpness of the stall.
> Come to think of it, I have also "stalled" a 777 fixed base simulator,
> but there the stick pusher was really what I was experiencing.
Most (if not all) single-engine Cessnas use the 2412 airfoil. That's
not the softest-stalling airfoil around, but it's not bad. I don't
think it has hysteresis in the stall characteristic.
I'm not an expert on Cessnas, but most Cessna twins I've read about
use 230xx airfoils. At least Cessna had the good sense not to use the
23009 at the wingtip. The 23009 stalls about 3 degrees before the
23012 or 23015, which eliminates the benefit of the commonly used
3 degrees of washout.
> However, look at the list of aircraft that have NACA 5-digit
> airfoils. (http://amber.aae.uiuc.edu/~m-selig/ads/aircraft.html).
> There are a lot of them. Do they all have nasty stalls characteristics?
I admit to having precious little experience flying, so I can not
say. I read everything I can get my hands on, and I see far too
many pilot reports that state "stalls fairly well straight ahead,
but drops a wing very quickly if you slip it." You seldom see
that kind of report on a plane that uses the Clark Y, for example.
Just as an example, the Questair Venture had such a bad stall that
they went to North Carolina State University to get help. NCSU
grafted on wingtip droops and vertical slots to try to tame the
stall. (The Venture used the 23017 at the root and 23010 at the
tips.) It worked, mostly, but this is another of those planes that
could go into a snap roll if slipped too close to the stall.
Beech seems to have used the 230xx airfoils quite extensively. How
do Beeches and Cessna light twins stall? Light twins in general have
a bad reputation, but I have no experience there.
How does the Beech Bonanza (230xx) compare to the Piper PA-24 Comanche
(64A215)?
Mark Hickey
>In the limit of infinitely long wings, you move an infinite
>mass of air at zero velocity and induced drag drops to zero
>(of course the parasitic drag may be a problem :-)
I'd be more concerned about finding hangar space!
Mark Hickey
Bob Chilcoat
Or even just a tiedown slot.
Bob (better be a high wing so you can walk under it)