This looks great. Now all we need to do is find
out where we can buy them. I am working on that.
http://www.csfimports.com/home.asp
--
Paul Lamar ...No rotor no motor.
I've been AFK for a few days, or would have come in on this sooner.
Back in my misspent youth, I worked for a division of Chromalloy. We
made radiators, cores and custom heat exchangers -- for a while,
we were
THE source for new exchangers for your P-51 or P-38, not to mention
radiators for various antique cars.
Radiators are SIMPLE. They are just a way of bringing coolant and
airflow as close together as possible while still keeping them apart.
They are primarily of two types: crossflow and vertical flow. The
advantage of vertical flow is that sediment and other solids
falls down
into the bottom tank. The advantage of crossflow is that you can
make a
radiator that goes all the way across the front of the car.
There is a third type of common automotive heat exchanger, not
generally
considered a radiator but that's what it is anyhow -- or, more
precisely, a radiator is just a subgroup of heat exchangers --
and that
is commonly known as a "heater core."
Yep, that's right, the forgotten little "radiator" under the dash. It
is just as strong as the big radiator that won't fit into your
airplane,
and they are available in a bunch of sizes and configurations.
Go find your friendly, neighborhood radiator shop, the kind that has
been there so long the wooden structure is starting to petrify.
Take a
look at the book they order replacement cores from. They have
long and
short, rectangular and square cores, even some which are
essentially a
cube. The tanks and tubes go every which way, and if you can't find
exactly what you like, they can take a bare core and build it into
whatever you need, and probably supply the header tank as well.
If you
get the guy interested in the project, he'll grab that ball and run,
because it's not just another Chevy!
So have some idea what you are after, but still have an open mind
when
you go see them, because they are the parts experts. The guy at the
auto shop counter only knows how to get the box the computer told
him to
get, the radiator shop guy can tell you why the one from the '65
Mustang
is better than the one from the '67 Mustang (or vice versa).
Keith Wood
Keith,
What's the difference between a radiator core and an
air-conditioning core.
George (down under)
George,
The major difference is around 300 - 400 psi in the a/c core v.s.
around 20 psi for a radiator. A/C cores are thicker walled round
tubes to prevent them from distorting when subjected to the hgh
pressures in the system. Radiator cores can use a rectangular tubes
with thinner walls and will have more surface area per square inch
of radiator surface, hence better heat transfer.
Joel
This may have been discussed before, but being the new kid on the block,
I will bring it up again.
How about, instead of the fin-and-tube type radiator, use a surface
effect-type radiator? They could be mounted on the outside of the
airframe and the slipstream would cool the coolant running underneath.
We are not talking about taking the aircraft into combat, so we don't
have to worry about it being holed by AAA.
The only example of surface effect radiators that springs immediately to
mind is Brabham's F1 car of a few years ago, the BT46:
http://s12.photobucket.com/albums/a234/JacobPotts/Atom/Aerodynamics/?action=v
iew¤t=BrabhamBT46002.jpg
<http://s12.photobucket.com/albums/a234/JacobPotts/Atom/Aerodynamics/?action=
view¤t=BrabhamBT46002.jpg>
Yes, I know the BT46 is a POOR example, but that is because they
undersized the radiators, not because the radiators were, of themselves,
bad.
What do you think, not about the efficacy of the example, but the
efficacy of the idea: surface effect radiators?
Les Nordman :-)
They just don't work well. Most if not all aircraft designers have given
up on them in the late 1930's. The early P38s had an inter cooler built
into
the leading edge of the wings. They eliminated that feature in later
P-38s. The problem is the boundary layer heats up and then the heat
transfer
from the following surface is greatly diminished. The same effect is
noticed
in thick radiators. The first half does 3/4 of the work.
Here are some picture I took in the Super Marine museum in South Hampton.
Schnider cup racers of the early 1930's used surface mount radiators.
Almost every square inch was covered with them.
Paul Lamar ...No rotor no motor.
Heated leading edges can't be all bad for free anti-ice, but will the
runback refreeze? Plumbing the whole wing will be heavy and, as mentioned,
inefficient. Still, any drag free cooling can't hurt. Water cooled
snowmobiles come with foot heaters, aka floorboard radiators. Any thoughts
on getting similar extrusions to match the inside radius of my composite LE
with very poor heat transfere? I know the idea will come back into use, I
just can't think how...yet...
Tom Lee
You will never get out of the pattern if it don't cool let alone
worrying about de icing the wings. If you can figure a way of doing
away with the boundary layer than you have a chance of making it work.
Of course once you do that the entire aviation world will beat
a path to your door.
A typical radiator with ten fins to the inch has a 1/2 inch tube
spacing. A rad 20 inches long by 18 inches wide has 36 rows, 200 fins per
row or 7200 fins. Each fin has an area of about 1 square inch if the rad
is 2 inches thick. This is the size of rad it takes to cool a Mazda 13B.
Now you need to double that as heat is transferred from both sides
of the fin so you have 14,400 square inches of heat transfer surface
not counting the area of the tubes themselves. The boundary layer
on the fins does not get a chance to build up as the fins are only
2 inches long. So if the wing skin cooling is on the first two inches
of the wing leading edge you need 14,400 square inches.
If my arithmetic is right you need a wing 300 feet long. Maybe only
150 feet long if I made a mistake some place.
That is a big airplane to power with one 200 HP Mazda rotary engine.
Paul Lamar ...No rotor no motor.
Paul,
I wonder if your calculation is correct on the above heat exchanger for
water cooling. Total size is 18X20X2" = 720 cubic inches. Tad large?
It was my impression you were recommending about 600 cubic inches.
Doug Fir
That was before the 264 HP we got with the RX8 p-port :)
So it is only a 100 foot wingspan.
Paul Lamar ...No rotor no motor.
The Rotary Engine NewsLetter. Powered by Linux.
ACRE NL web site.
http://www.rotaryeng.net
Copyright 1998-2007 All world wide rights reserved.