|
On a flight in a DC-9 landing in sleet
at Syracuse, NY, I was seated just ahead of the wing's LE. It was night, and the
landing light was on. I saw the most incredible thing; the sleet in the air
ahead, being illuminated by the landing light, was rising up toward the LE.
"How was that happening?", I mused. When I returned to California, I looked up
an acquaintance who was a Cal Poly Aero grad and told him of my observation and
inquired if he knew why that was happening. With a straight face, in all
seriousness, he said the air at the LE was broadcasting its pressure. I then,
with a straight face, asked him if the medium was AM, FM, or some more
complex modulation scheme. When I asked him to further illuminate me on the
cause of this phenomenon, he told me that that's what his professor said, and he
really had no idea what the mechanism was. After much rumination, I came to the
conclusion that the only way that could take place is that if the molecules of
air that were traveling in the direction of the LE were not replaced from the
air in the vicinity of the LE, a net flow would occur.
Now keep in mind, the mean free
velocity of air is about M 1.4 in all directions. Molecules are constantly
moving from any given parcel of air into all directions toward other parcels.
Those parcels in turn are also radiating molecules out in all directions, so
that at any one instant in time, some parcels have more molecules, and some have
less, but over time, the average density is the same if the air is not being
disturbed. On a micro level, there is a continual flow going on in all
directions, but on a macro scale the air is still. Another tidbit: air does
not have pressure as an intrinsic part of itself. Air has density and velocity,
the latter is what we call heat, which is why the velocity of sound is
proportional to the absolute temperature. Pressure is a force that acts on a
membrane that gets in the way of those fast moving molecules. If we were able to
view the pressure being generated, we would see all these little molecules
banging into the membrane like a ball-peen hammer beating on it. If the
membrane is surrounded by air on both sides, it will absorb blows to it on both
sides, which will average-out, and so it will tend to remain where it
is.
So what was taking place on that wing's
LE is that the passage of the wing through the air divided it into jets of air
over the upper and lower surfaces. That wing, just as in Brent's venturi tube,
organized the air flow in a particular direction. I doing that, more of the
little devils were flowing in a direction parallel to the surface, and fewer
were normal or perpendicular. When Bernoulli says that the pressure goes down
when the velocity goes up, what that really means is that the pressure normal to
the flow direction is lower, but it is higher in the direction of the flow,
so that the total pressure remains the same. I once told someone that he should
place an inlet facing forward at the point of maximum thickness of a wing where
the velocity is highest to get the highest pressure, and he argued that no,
the pressure would be lower there. I said that if he put a probe facing forward,
a pitot, into the airstream, the pressure would be higher. No! he affirmed,
Bernoulli says that when the velocity is highest, the pressure is the
lowest, and that was good enough for him!
But what Henri saw in his attempt in
1911 to make a jet-propelled airplane was that when he dumped fuel into the
cowling behind an engine-driven compressor, and provided curved plates from the
cowling onto the sides of the fuselage for its exit, that instead of the burning
gases being ejected at an angle from the cowling, they instead followed the
curve right around onto the side of the fuselage, which, of course, they set on
fire! He was so fascinated looking at this that he almost flew into a wall in
his path. It wasn't until the 20's that he finally formulated the "Coanda
effect". There's a logic element for a fluidic computer that is a bistable
element, a flip-flop. It is a small jet that emerges between two outward curving
walls. Each wall is provided with a small hole through which controlled air
may enter. When the jet first emerges, it will attach itself to one of the walls
and stay there. There is less pressure acting outward from the wall into the
jet since the air in the jet is organized in one direction, so that the air
ouside of the jet holds it against the wall. If the hole in that wall is
opened so that air may enter on the wall side of the jet, the pressure unbalance
is removed, and the jet immediately detaches from that wall and attaches to
the other one where its hole is closed. That is how it becomes a
bistable element, under control of some other element which controls the
holes.
That jet of air flowing up and around
the LE has centrifugal force since it has mass. It is held against the
curve by the molecules in parcels of air above and forward of the jet which
are flowing toward the LE. That and the fact that there is no air between the
jet and the LE which would push it away and counter the force from the molecules
on the other side of the jet. Remember the hole in the fluidic element's walls?
So what is then obtained is this flow upward toward the LE, which is really what
the prof meant when he said the pressure was broadcast!
You can use this Coanda effect in
several places on your plane to reduce its drag. Right behind the spinner on the
front of the cowling we usually make the edges of the cowling follow immediately
behind the spinner's curvature to provide a seemingly smooth flow. But there's a
fly in the ointment! Pressurized air from the cooling plenum will flow forward
along the crankshaft and enter the space between the back of the spinner and the
cowling. From there it will flow outward and disturb the airflow from the
spinner, making the flow turbulent. If you made the cowling immediately behind
the spinner about 1.5" larger in diameter and shaped it into a quarter-round
shape with a 0.75" radius, the Coanda effect would take that jet emerging
from behind the spinner and turn it around to flow smoothly back. That's also
the drag-lessening effect that can be done by doing the same to the LE of
the ailerons and elevator. By thickening their LEs by 15% on each side with
a smooth radius, gap flow can be turned into the flow
direction.
Enough!
|