Coanda effect, exhaust extraction, and bumblebees

["Janet Scruggs" <scruggs@mbay.net>]

Paul Anderson mentioned the Coanda effect relative to exhaust devices and
that it was described as, "...not a smooth transition, it is an abrupt
change in the fluid flow."  That is likely the description offered by an
advertisement, but if that is what is happening it isn't Coanda flow, or if
it is Coanda flow then it is not an abrupt change in anything... in fact,
the contrary.

Although not generally known, every time your airliner lowers flaps for an
approach you are the beneficiary of the Coanda effect.  Coanda effect is
the characteristic of laminar airflow to adhere to the contour of a curved
surface.  When you see the demonstration where someone blows across the top
of a drooping sheet of paper and the paper rises and becomes essentially
aligned with the flow stream it is usually attributed to Bernoulli's law. 
Well... Bernoulli's law would not be possible were it not for the Coanda
effect.  

Essentially it is the tendency for moving air molecules to 'attach'
themselves to a boundary curvature in an _unconstrained_ flowstream. 
Coanda effect does not happen inside a tube as the flowstream is
constrained to follow the curvature of the walls.  Looking further into the
effect will uncover the influence of terms like kinematic viscosity, shear,
laminar flow and such.  But we need not go there.  The effect is so strong
there is enormous gain to be had through its employment when designing a
STOL (short takeoff an landing) aircraft.  In trying to gain more lift to
get off the ground earlier and to have as slow an approach as possible, the
Coanda effect comes in very handy.  

Examples of the extremes to which the Coanda effect can be taken can be
found in the experimental the U.S. YC-14 and 15 transports and the Antonov
AN-72/74, NATO name "COALER," which was in production for a number of
years.  All these designs use two turbofan engines mounted slightly above
their wing such that their exhaust blows over the top of the wing and when
needed, over the top of radically deployed slotted Fowler flaps.  By doing
so the flow over the top of the wing can be turned downward in excess of
90deg over a distance of two meters.  By turning the flow two things are
accomplished: the extra flow across the top of the wing sharply increases
aerodynamic lift and, there is now a downward component to the exhaust flow
which results in a vertical thrust vector.  Normal production aircraft have
lift coefficients on the order of 0.85 -1.25, airliners have huge lift
coefficients at 3.0 - 3.5, but this configuration STOL aircraft have
coefficients in excess of 9.0!  (If you have goose bumps at this point, you
_are_ a Nerd. But don't worry... if you rub them with your calculator
they'll go away in a minute or so.)

The most radical aircraft to explore the use of the Coanda effect was the
NASA project called the QSRA.  Short for Quiet Short-haul Research
Aircraft, the "Q" used four engines blowing over the wing of a highly
modified DeHavelin Buffalo and could take off in about twice it's own
length.  I saw the last flight the Q made and it was a very impressive
performance.  As I watched with a friend who was a flight test engineer on
the program, I kept saying... It can't do that... It can't do that
either... No, that's not possible... as it turned, climbed, descended,
landed in a 20 min routine in a space the size of a football field.  

And how, you are no doubt asking, does this effect the performance of my
Audi?  If a muffler manufacturer claims that his device is special because
it uses the Coanda effect you may, with great justification, inquire how
that is possible within a confined flowfield as the effect has definition
only in an unconstrained environment.

Mercifully, I've skipped over the parts involving laminar flow and
turbulent boundary layers... which is why golf balls have dimples.

Chris Miller mentioned using the low pressure area behind the car to
evacuate the exhaust system.  That will work, but here's what you'll have
to do.  Make the exhaust exit about two feet in diameter and in the shape
of a funnel.  Problem is that as the exhaust gasses exit the normal size
pipe their velocity is such that the static pressure around the stream is
lower than the pressure area behind the car and the pressures are 'upside
down' for extraction to take place.  Bernoulli's law applies here... the
higher the speed the lower the pressure and vice versa.  (We are talking
static pressure here not dynamic pressure.  Dynamic pressure is that which
is caused by the impact of the air molecules on a surface, static pressure
is that which is felt as the air molecule rushes past.)  So, in order to
increase the static pressure of the exhaust stream so that it will be
higher than the low pressure area behind the car, we must slow the velocity
of the stream.  This can be done by increasing the area of the exhaust exit
plane.  (Same volume through a larger pipe travels slower.)  And what you
end up with is a huge flared bell mounted directly behind the boot of your
car.  Please post the pictures when your project is completed.

Like most everything else in the world of physical phenomena this exhaust
evacuator works best only at one set of conditions, determined by vehicle
speed, exhaust stream volume (engine rpm), and exit area of your extractor.
 This means that, for instance, the extraction force (vacuum) felt at 70mph
in third gear will be different when you shift to fourth because the
exhaust volume is different being a function of rpm and throttle opening. 
One would have to design a computer controlled exit orifice which monitored
all the factors and picked the optimum diameter for conditions.  Could be
done... and once the hardware is in place I suspect there would be a
dramatic decrease in the number motorists prone to tailgating you.  

Bumblebees.  I would be remiss were I to pass an opportunity to correct a
slanderous myth regarding my profession.    It is said that, "...according
to all known aerodynamic laws, bumblebees can't fly."  A bit of research
reveals the originator of this idea was a 19th century French zoologist
named Antione Magnan.  Professor Magnan, apparently noting that the way
that birds fly differs from that of bees approach the matter, said, "Je ne
sais pas," or words to that effect.  Somehow that statement has morphed
it's way through contemporary mythology and been reattributed to aero
engineers until an unknowing person would rightly resist getting on an
aircraft that was designed, apparently, by ignoramuses.  It is not the case
that this bee-havior is not understood.  I'll happily post the formula for
lift coefficient of ornathroptic flight, should there be even one of you
remotely interested.  There, I feel better.

More car aero that you didn't want to know either: You're driving down the
road on a rainy day.  Ever notice the vortex off the car in front of you? 
Off the left side of the car it rotates clockwise and off the right side it
rotates counter-clockwise, when viewed from behind... and it will always be
this way.  Reason is that as the air passes over  the hood/bonnet and
encounters the windscreen a rolling vortex is established at this junction.
 Sitting in the car the top of the vortex rolls toward you and the bottom
rolls away from you.  There is a law of continuity which states that once
established a vortex will not just end, it will continue until it
dissipates.  In this case it rolls off the edge of the windscreen and along
the side of the car and, once established, continues back on to the car
behind you.  

Regards, Gross

PS. alright!... I won't write any more.