Lower Drag: Rounded or Sharp?

[Jay]

I've been wondering which shape is lower drag, a rounded one or a
sharp one. I can understand why the leading edge of a wing would be
rounded, this allows a larger range of AOA before flow speration and
stall, but what about the vertical stabilizer? Wouldn't this be
better with a sharp leading edge?

When the air hits a blunt leading edge it has to accelerate quickly to
get out of the way. At the very front, the molecules are actually
moving in a direction normal to the direction of travel of the wing.
This has got to cause drag. If the edge were sharp and the air didn't
have to react as quickly to displace, this would seem to cause less
drag.

I know that the tear drop shape is the lowest energy state, but it may
not be the lowest drag shape. I'm thinking that an eye shape may be
better for many profiles.

So I see tear drop shape profiles all over on struts, whell pants,
vertical stabilizer, etc. Looking for explanations other than "Thats
what everybody does, so they couldn't be wrong."

Regards

[Dave Driscoll]

The answer to this question varies depending upon how fast you want to
go. All bets are off once you start looking a compressible flow and its
effects (roughly sonic velocity). For the speeds that most homebuilts
experience, the rounded shape promotes laminar flow around most of the
shape and hence lower drag for a wide variety of angles of attack. The
problem with the "eye shape" is that the point will start developing
turbulent flow at the point at very low angles of attack. This
turbulent flow will cause a lot of drag. Even gear fairings have angle
of attack issues during some flight regimes, steep climb, slips, etc.
As far as a vertical stab, your control surfaces are the last thing you
want to have behind a turbulent flow transition as they need clean air
to be effective.

Play around with dragging some different shapes at various angles of
attack through a tub of water and you'll see the effect very clearly.

Dave

Jay wrote:

>I've been wondering which shape is lower drag, a rounded one or a
>sharp one. I can understand why the leading edge of a wing would be
>rounded, this allows a larger range of AOA before flow speration and
>stall, but what about the vertical stabilizer? Wouldn't this be
>better with a sharp leading edge?
>
>When the air hits a blunt leading edge it has to accelerate quickly to
>get out of the way. At the very front, the molecules are actually
>moving in a direction normal to the direction of travel of the wing.
>This has got to cause drag. If the edge were sharp and the air didn't
>have to react as quickly to displace, this would seem to cause less
>drag.
>
>I know that the tear drop shape is the lowest energy state, but it may
>not be the lowest drag shape. I'm thinking that an eye shape may be
>better for many profiles.
>
>So I see tear drop shape profiles all over on struts, whell pants,
>vertical stabilizer, etc. Looking for explanations other than "Thats
>what everybody does, so they couldn't be wrong."
>
>Regards
>
>

[Kevin Horton]

On Sat, 08 May 2004 18:44:30 -0700, Jay wrote:

> I've been wondering which shape is lower drag, a rounded one or a sharp
> one. I can understand why the leading edge of a wing would be rounded,
> this allows a larger range of AOA before flow speration and stall, but
> what about the vertical stabilizer? Wouldn't this be better with a sharp
> leading edge?
>
> When the air hits a blunt leading edge it has to accelerate quickly to get
> out of the way. At the very front, the molecules are actually moving in a
> direction normal to the direction of travel of the wing. This has got to
> cause drag. If the edge were sharp and the air didn't have to react as
> quickly to displace, this would seem to cause less drag.
>
> I know that the tear drop shape is the lowest energy state, but it may not
> be the lowest drag shape. I'm thinking that an eye shape may be better
> for many profiles.
>
> So I see tear drop shape profiles all over on struts, whell pants,
> vertical stabilizer, etc. Looking for explanations other than "Thats what
> everybody does, so they couldn't be wrong."
>
> Regards

I looked for some wind tunnel data in my reference books, but I couldn't
find anything conclusive. Hoerner's Fluid Dynamic Drag as a bit of stuff
on page 3-17, but it doesn't really properly answer your question.

You need to understand that the air doesn't come straight at the blunt
leading edge, and then suddenly have to make a 90 deg turn. There is a
high pressure area ahead of the leading edge, and it acts to start
deflecting the air well before it reaches the surface. So the air makes a
smooth transition around the object. You can see this effect in action if
you are ever driving at night in heavy snow, with low density fluffy snow
flakes. The snow flakes provide a good visual clue as to what the air is
doing as the car approaches. You can observe the snow flakes rising quite
a few feet ahead of the windshield, so they make a smooth transition over
it.

Also, we can't ignore the problems of airflow separation. That vertical
stabilizer needs to work properly in a sideslip too. So we need that
rounded leading edge. Pretty much every part on an aircraft will see
quite a variety of airflow angles if we consider all the different flight
conditions. So whatever shape is used needs to be quite tolerant of
variations in airflow angle.

--
Kevin Horton RV-8 (finishing kit)
Ottawa, Canada
http://go.phpwebhosting.com/~khorton/rv8/
e-mail: khorton02(_at_)rogers(_dot_)com

[Wright1902Glider]

Wilbur Wright approached this same question in January of 1903. Up until then,
he had not formally tested various shapes for strut profiles. The struts of
the 1900 and 1901 gliders are believed to be roughly a long teardrop in
cross-section. The 1902 glider was built using the 1901's struts, so it would
be the same as well. However, after wind tunnel testing, Wilbur found that he
obtained the lowest "head resistance" or drag, with a shape that was
rectangular in section, but with 1/4-round chamfered corners. This doesn't
seem to make sense considering what we now use, but that's what he found, and
that's what's on all of the later Wright machines.

Just a little historical perspective,
Harry

[George A. Graham]

On 8 May 2004, Jay wrote:

> I've been wondering which shape is lower drag, a rounded one or a
> sharp one.

The most instructive seminar I've attended was one by John Ronce at
Oshkosh a few years ago. He began by drawing a three foot long airfoil
shape, then drew a one-quarter inch dot (the profile of a cable),
he said that each had the same drag. Very hard to believe.

He also said that air will create its own nose profile, against a blunt
or flat shape, that is why wings can have a round front for better angle-
of-attack tolerance.

The drag is caused more by turbulent flow trailing the part.

I left the talk dumbfounded, and still am.

George Graham
RX-7 Powered Graham-EZ, N4449E
Homepage <http://bfn.org/~ca266>

[Jay]

George wrote the following e-mail to me:
-----------------------------------------------------
"The most instructive seminar I've attended was one by John Ronce at
Oshkosh a few years ago. He began by drawing a three foot long
airfoil
shape, then drew a one-quarter inch dot (the profile of a cable),
he said that each had the same drag. Very hard to believe.

He also said that air will create its own nose profile, against a
blunt
or flat shape, that is why wings can have a round front for better
angle-
of-attack tolerance.

The drag is caused more by turbulent flow trailing the part.

I left the talk dumbfounded, and still am."
---------------------------------------------------------------------------

So the vibe I'm getting is that in general, the tear drop shape is
lower drag with the eye shape being better for a very narrow range of
AOA depending on the "sharpness" of the leading edge.

I guess you could optimize for one particular AOA, say cruise, at the
expense of all other regimes. So different parts or the plane would
have different shapes depending on whether it was a flight surface,
horizontal, vertical, in prop blast, etc.


Regards

Dave Driscoll > wrote in message >...
> The answer to this question varies depending upon how fast you want to
> go. All bets are off once you start looking a compressible flow and its
> effects (roughly sonic velocity). For the speeds that most homebuilts
> experience, the rounded shape promotes laminar flow around most of the
> shape and hence lower drag for a wide variety of angles of attack. The
> problem with the "eye shape" is that the point will start developing
> turbulent flow at the point at very low angles of attack. This
> turbulent flow will cause a lot of drag. Even gear fairings have angle
> of attack issues during some flight regimes, steep climb, slips, etc.
> As far as a vertical stab, your control surfaces are the last thing you
> want to have behind a turbulent flow transition as they need clean air
> to be effective.
>
> Play around with dragging some different shapes at various angles of
> attack through a tub of water and you'll see the effect very clearly.
>
> Dave
>
> Jay wrote:
>
> >I've been wondering which shape is lower drag, a rounded one or a
> >sharp one. I can understand why the leading edge of a wing would be
> >rounded, this allows a larger range of AOA before flow speration and
> >stall, but what about the vertical stabilizer? Wouldn't this be
> >better with a sharp leading edge?
> >
> >When the air hits a blunt leading edge it has to accelerate quickly to
> >get out of the way. At the very front, the molecules are actually
> >moving in a direction normal to the direction of travel of the wing.
> >This has got to cause drag. If the edge were sharp and the air didn't
> >have to react as quickly to displace, this would seem to cause less
> >drag.
> >
> >I know that the tear drop shape is the lowest energy state, but it may
> >not be the lowest drag shape. I'm thinking that an eye shape may be
> >better for many profiles.
> >
> >So I see tear drop shape profiles all over on struts, whell pants,
> >vertical stabilizer, etc. Looking for explanations other than "Thats
> >what everybody does, so they couldn't be wrong."
> >
> >Regards
> >
> >

[L. Darte]

George A. Graham wrote:
> On 8 May 2004, Jay wrote:
>
>
>>I've been wondering which shape is lower drag, a rounded one or a
>>sharp one.
>
>
> The most instructive seminar I've attended was one by John Ronce at
> Oshkosh a few years ago. He began by drawing a three foot long airfoil
> shape, then drew a one-quarter inch dot (the profile of a cable),
> he said that each had the same drag. Very hard to believe.
>
> He also said that air will create its own nose profile, against a blunt
> or flat shape, that is why wings can have a round front for better angle-
> of-attack tolerance.
>
> The drag is caused more by turbulent flow trailing the part.
>
> I left the talk dumbfounded, and still am.
>
> George Graham
> RX-7 Powered Graham-EZ, N4449E
> Homepage <http://bfn.org/~ca266>
>

I'd be more concerned about flow seperation at high angle of attacks
being induced by the sharp leading edge. Typically, stabilizer surfaces
have very low aspect ratios to insure that the airflow remains stable
across the stabilizers even when the main wing is fully stalled. Having
a sharp leading edge may cause problems with early flow seperation. This
might cause stability problems or even possibly some flutter. I don't
have any hard data to prove this, but I don't think you're going to gain
enough drag reduction to justify risking the potential problems.

----- quote:
However, after wind tunnel testing, Wilbur found that he obtained the
lowest "head resistance" or drag, with a shape that was rectangular in
section, but with 1/4-round chamfered corners. This doesn't seem to
make sense considering what we now use, but that's what he found, and
that's what's on all of the later Wright machines.
----------

Welcome to the wierd world of low reynold's numbers aerodynamics.

[Kevin Horton]

On Sun, 09 May 2004 12:51:48 -0700, Jay wrote:

> George wrote the following e-mail to me:
> ----------------------------------------------------- "The most
> instructive seminar I've attended was one by John Ronce at Oshkosh a few
> years ago. He began by drawing a three foot long airfoil
> shape, then drew a one-quarter inch dot (the profile of a cable), he said
> that each had the same drag. Very hard to believe.
>
> He also said that air will create its own nose profile, against a blunt
> or flat shape, that is why wings can have a round front for better angle-
> of-attack tolerance.
>
> The drag is caused more by turbulent flow trailing the part.
>
> I left the talk dumbfounded, and still am."
> ---------------------------------------------------------------------------
>
> So the vibe I'm getting is that in general, the tear drop shape is lower
> drag with the eye shape being better for a very narrow range of AOA
> depending on the "sharpness" of the leading edge.
>
> I guess you could optimize for one particular AOA, say cruise, at the
> expense of all other regimes. So different parts or the plane would have
> different shapes depending on whether it was a flight surface, horizontal,
> vertical, in prop blast, etc.
>
>

Don't forget that you may have a handling problem if you have a large
surface were the airflow becomes separated as you change angle of attack.
I.e. control surfaces may lose their effectiveness if they are in the wake
of the separation, or the control forces may become very strange. For
example, it is possible that a vertical stab with a sharp leading edge
could cause the rudder to go all the way to the stop all by itself if you
get some sideslip going. You might need a lot of force to get it back off
the stop again. That could get pretty interesting.

So, unless you are talking about small surfaces (like antennae), then you
should be pretty cautious about going for sharp leading edges. You can
probably get away with it on an antenna, as the separated area would be
small.

[Jay]

"L. Darte" > wrote in message >...

> I'd be more concerned about flow seperation at high angle of attacks
> being induced by the sharp leading edge. Typically, stabilizer surfaces
> have very low aspect ratios to insure that the airflow remains stable
> across the stabilizers even when the main wing is fully stalled. Having
> a sharp leading edge may cause problems with early flow seperation. This
> might cause stability problems or even possibly some flutter. I don't
> have any hard data to prove this, but I don't think you're going to gain
> enough drag reduction to justify risking the potential problems.

I think you're right. Probably better leave control surfaces well
enough along, but go after the other "protruberences": Antennas,
struts, landing gear, etc. Places that you don't care if they produce
drag in a side slip but clean up nicely in cruise.

[alexy]

(Jay) wrote:


>I think you're right. Probably better leave control surfaces well
>enough along, but go after the other "protruberences": Antennas,
>struts, landing gear, etc. Places that you don't care if they produce
>drag in a side slip but clean up nicely in cruise.

You better think that through carefully, and test. E.g., think of the
pitching moment if your wheel pants suddenly got a lot draggier in a
slip.
--
Alex
Make the obvious change in the return address to reply by email.