Al Gietzen wrote:
Attached are photos of my air flow test setup to measure flow
distributions
and pressure drops. It is shown here with the scoop and custom Ron
Davis
radiator. This is one of two radiators for my 20B, this one goes in the
cowling of my Velocity pusher, and another of different configuration is
in
the wing root. This is an "armpit" scoop (under the wing at the
fuselage),
and has a diffuser ratio a bit over 4 to 1.
I measured flow and flow distribution with a small handheld anemometer
from
Radioshack; a very cool little instrument that gives digital readout of
air
speed (see at their website); pressure measurements with water
manometer.
The output from the centrifugal blower unfortunately turned out to be
about
half my design point for the rad, but it gave me some very useful
information on placing some vanes internal to the scoop/diffuser in
order to
get a good flow distribution through the rad.
The max flow rate from the blower through the rad was 1300 cfm. The
design
point for the installation is 2600 cfm. The radiator is a two row 1"
tubes
with 16 fins per inch and a core thickness of about 2 ½ ". The core
is 16
x 11.5 for an area of 184 sq in. The measured pressure drop across the
core
at this flow rate was only 0.4 in water. I think the pressure drop goes
roughly as the cube of flow rate, so at design point flow, the drop
would
be about 3.2" water. This compares favorably with Ron Davis' computed
value
of 3.7".
Note the length of duct from the blower to the scoop. The output from
a
centrifugal blower is very turbulent, so if you want meaningful
measurement
of flow distribution, be sure to use a length of duct to help straighten
things out.
Now; about that gentle breeze:
As you can see the pressure difference across the right radiator
is 5.11 - 3.61 is only 1.5 inches of water or about .055 psi
which is hardly enough to generate a gentle evening breeze.
In effect this radiator is not functioning. Well hardly at all.
I'd say the 1.5" drop across Tracy's radiator gives a lot more than a
gentle
breeze. That's enough pressure to produce significant flow, especially
through a fairly open core as I believe the air conditioning cores are.
That's enough for about 2000 cfm through my 16 fins/in radiator. That's
a
pretty strong wind - or I guess one could say - it's a lot of hot air.
I've also done some measurements on my custom oil cooler. More on that
later.
Al Gietzen
As I recall Tracy's AC cores are 3.5 inches thick and not 2.5. As I also
recall,
but I could be wrong, the flow per square inch is inversely proportional
to the square of the thickness. 1/6.25 = .16 while 1/12.25 = .081 or
about half the flow per square inch of frontal are. Due to the thicker
core.
Now as to the volume or CFM of air going through you must realize
Tracy's rads are only 10 by 9 inches for a frontal area of 90 square
inches
while your rad is 184 square inches or twice the total flow for the same
pressure
drop if the thicknesses were the same which they are not. They are much
thicker.
Also the fin spacing looks much closer on the AC cores than it does on a
real water
cooling radiator. My guess is Tracy's AC heat exchanger cores are flowing
one
quarter of the air that your rads are at the same pressure drop.
Also the left AC core has twice the pressure drop so it should be really
pumping air.
We also need to measure some real world water and air temperatures going
in and
out of the rads.
Paul Lamar
Al,
You've basically got the experimental setup I was going to use to get some
radiator data. The addition of some water flow through the radiator and four
thermometers would finish the setup. To get a flow of hot water, at a steady
temp, I was going to put a copper coil over my fish cooker (may require some
finessing of the gas and water flows to get it right). To get a flow rate
for the water, measure the flow into a bucket of known volume with a stop
watch (take a series of five runs). I would use dial thermometers that can
be sealed using a compression fitting or pipe fitting for the water side.
The data needed is, water flow, air flow, air temp in/out, water temp
in/out. If you do this make sure to check the water flow to see that it does
not vary, and to calibrate the thermometers with an ice bath and boiling
distilled water. Also measure the temp and barometric pressure, a basic Idea
of relative humidity would also be nice (just get it off the weather
channel, no need to buy an fancy instrumentation) If you had that data for a
series of flow rates of air and water you could optimize your system. I can
crunch the numbers if you like, I just can't build anything right now. All
my toys are in storage until the house is finished :-(
This info is sorely needed to be able to DESIGN a proper cooling system, not
just guess.
Keep up the good work!!
Monty
The AirCraft Rotary Engine NewsLetter. Powered by Linux.
ACRE NL web site.
http://home.earthlink.net/~rotaryeng/ http://www.linux.org