|
I think it boils down to this.
You need a certain pressure difference across
a given heat exchanger to create the right convection coefficient at the
right mass flow to move enough heat to keep things cool.
What is the most efficient way to achieve
this.
Everybody wants a black and white answer.
There is not one.
The NACA works by making two vortices that curl
down into the inlet. The high speed air enters under the lip due to this. To
create these vortices there must be something "sucking" on the
inlet.
Once flow is established, you get about 80% of
dynamic pressure recovery from a properly designed NACA and downstream
diffuser.
A conventional diffuser slows and expands the air
thus increasing the pressure. If the air stacks up it can also create external
diffusion, something that the NACA cannot do. It just slips past with no
vortices being produced.
A canard typically flies with some positive angle
of attack. It also has a bluff body at the back of the cylinders. I know this
will offend the canard guys, but the aft end of a canard is bad...really bad.
And the bluff body creates a negative pressure. Which is bad for drag, but good
for cooling. The suction here due to this and the prop disk can suck the air
past the cylinders and allow the vortex to form. So it works.
Would a NACA in front of a thick 4 core dense
radiator work...probably not.
Would a NACA in front of a thin radiator with exits
in the right place work....probably
Would a P-51 scoop be better? maybe if it didn't
add a lot of frontal area, surface area, interference drag, and such that cancel
out the 10% increase in dynamic pressure recovery that is
available.
I have never understood the statement that the
original paper made about how the flush inlet was not suitable for a carb air
inlet. Seems like it would work fine if you only want 80% vs 90% pressure
recovery. The original report just says this without really saying why. The
argument about radiator inlets makes sense-most of the time. But induction air
is not as clear cut an issue.
They also admonish you not to place it in a thick
boundary layer.
I would point out that the NACA inlet has not
even been used for jet engine inlets very often (other than APU's). The only
non-research example that comes to mind is the Caproni jet that used one over
the cockpit. Most jet engines are very sensitive to turbulence in the inlet, and
the NACA is not very good in this regard. The Caproni used two small centrifugal
type jets which can tolerate a lot of inlet turbulence. Pressure recovery is
also very important to a jet.
In short....it depends.
If it works....it works.
Monty
|