|
|
I think the keyword here is drag.
Yes, if you could hang a thin radiator in the air, with enough
area to provide adequate cooling, that would not slow down the air
as it passes through it, you would not need any ducting, diffusers
or plenums.
Unfortunately in the real world we run into things like a wide
range of airspeed, angle of attack of the radiator and turbulence
(=drag) caused by air passing over cooling fins.
Not that I know what I'm doing. Still trying to get small
radiators (2 cu in/HP) to work. One thing I have learned is the
difficulty in making two rads and an oil cooler share limited air
exit area. Point being that if you want to limit cooling drag,
both entry and exit air flow and paths become critical. And for
those starting down this road, start with bigger than optimum rads
if you have room for them (and can tolerate the weight increase).
It's a lot easier to throttle air flow (e.g. exit flap) than tying
to optimize everything to get small rads to work. And yes, one
radiator would make life much, much simpler, if you have room for
it!
Finn
On 7/21/2021 11:54 AM, Ernest Christley echristley@att.net wrote:
I'm just a computer
programmer, not an aerodynamicist, but I think the whole
converting speed to pressure thing is a red-herring that
causes more confusion than it solves.
What you're after is flow
through the entire heat exchanger. Building big plenums so
that the you can stop the air and then accelerate it is one
way to get there, but where else in aviation do we attempt to
stop the air so that we can start it again?
The first radiators just hung
out in the airstream and had terrible performance. The
problem was that they created so much turbulence that air went
around them instead of through, and some parts of the radiator
didn't get used at all. As airplanes got faster, that
turbulence would occur even between fins, obstructing the
flow, and making some parts of the radiator unused. The point
that K&W was making, at least by my reading, was that the
air needed to be slowed to be smooth through the WHOLE heat
exchanger.
The ducting I have on top of
my Corvair engine doesn't look like any others I've seen.
Bascially, all I have is a sheet of aluminum that taper down
as it goes back. I do not have a "plenum" of any sort, and I
do not try to slow the air down. I just make sure that it
smoothly spreads out and moves through all the cylinders.
When first flying, I was concerned about the CHTs being too
low. . . that the cooling might be TOO good, because they stay
below 250*F.
I guess my message is that
you're not looking for "pressure recovery". That is just one
means to reach the end, which is to make sure that ALL of the
radiator is radiating.
I could very well be
completely off base.
As I recall the K&N curves, the duct/plenum is
supposed to decrease the speed approaching the radiator
(or fins), thusly increasing the pressure, which does
the cooling. Not speed, pressure differential. After the
fins, the deal is to regain the speed and have an
adequate size smooth hot air exit.
So a couple of inches of H2O pressure drop across
the fins?
The P51 arrangement seems to be about the best?
Or wing root LE like the Hurricane/Mosquito?
M
Sent from my iPhone
Seems
what we're missing is a curve that's the
product of these curves.
In other words some kind of bell or parabolic
curve with top where you have max
cooling/drag.
Obviously you can push fluid (and air) through
a radiator at a furious rate, but the drag
will go up.
So for both fluid and air rates there must be
an optimum spot.
Finn
On 7/21/2021 8:42 AM, Stephen Izett stephen.izett@gmail.com
wrote:
This graph from Mocal
might be helpful. It's for their oil coolers
but the trends may be transferable to water
exchangers.
The solid line is
Pressure Drop.
The two dotted
lines tell the story of two different oil
flow rates/tube.
<Screen Shot 2021-07-21 at 7.58.40
pm.png>
| m/sec |
5 |
10 |
15 |
20 |
25 |
| kmh |
18 |
36 |
54 |
72 |
90 |
| mph |
11.2 |
22.4 |
33.6 |
44.8 |
55.9 |
Increasing the air
flow 5 fold from 11 to 56 mph only increases
the heat transfer:
2 fold with an oil
flow of 0.02 L/sec/ tube and 2.3 fold by
doubling the oil flow rate per tube to 0.04
L/sec/tube
While pressure drop
increased 13 fold.
So, diminishing
returns from increasing airflow or fluid
flow.
Steve Izett
Charlie,
No, no reference, just what I
have read and also talking to
Rad manufacturers such as BWR in
Brisbane. You can check it out
by passing your hand through a
naked flame. Quickly and there
is no heat transfer. Pass
slowly and you will see what the
argument is. As I said the
truth is there somewhere and as
Lyn so aptly puts it “I could
well be wrong”..
Neil.
Hi Neil,
Do you have a reference for
that? Slowing a medium down so
it has time to absorb the heat
seems to conflict with physics
as I've been led to understand
it.
Charlie
On 7/20/2021 5:01 PM, 12348ung@gmail.com wrote:
Charlie,
Much wisdom out there, you
just have to find the truth!
Max cooling is apparently 30
MPH, so Any faster and it does
not pick up heat before going
past. Look at big trucks,
that grill is not only for
looks, they slow the air to
get max cooling. If too slow
they have a quite large fan
that kicks in to drag air
through at 30 MPH not 100!
As you say, what do I know – I
have seen too many that do not
work – without any degree.
Neil.
Subject: Inlet
cooling article
I remember
the Laboda article about
enlarging their cooling inlets,
but not many of the details.
This:
The
plenum receives air through
two circular air intake ducts
behind the propeller and
squeezes it, Bernoulli-style,
so that the air accelerates
across the cylinders and
between their fins, carrying
the heat back, down and out an
outflow "gate" at the back and
bottom of the engine area,
forward of the firewall.
Is contrary to everything I've
ever read about cooling
efficiently. Faster relative
flow will always have higher
drag, all else being equal.
Accelerating the air even faster
than freestream just sounds
crazy. My understanding is that
there's a balancing act between
having the room in an a/c to
'recover' (increase)
differential pressure across the
heat exchanger (engine fins, in
this case), and causing too much
drag from the air going through
the fins too fast (there's
aerodynamic drag in the heat
exchanger, just like over the
a/c itself). It's surprising to
me that James made the plenum
the way he did. The rest sounds
like putting bandaids on stuff.
The next-to-last image, of the
final inlet, shows what appears
to be a *much* smaller plenum
inlet than the cowl ring in
front of it, and a rather sharp
edged lip where the plenum
starts. It looks like the air
would accelerate until it hits
that sharp lip, and immediately
go turbulent, which will kill
any pressure recovery and
actually slow flow into the
cylinder fins.
Most Lyc plenums I've seen (even
the ones James made for the 4
cyl engines) have significant
volume above the cylinders with
smoothly expanding ducts feeding
the plenum. That allows the air
to slow in an organized fashion,
which increases *pressure*,
which is what actually makes the
air move through the fins.
But what do I know; I have an
Economics degree....
Charlie
|
|