Simple.  Cooling (or heating) can only take place where one molecule
speeds up or slows down an adjacent molecule.  In other words, it takes
both mass and interface area.  If you decrease one, you must increase
the other, in order to keep the same rate of exchange.

Heat exchangers have a place where the heat (or cold) enters the
exchanger, known as the BASE.  The farthest point from the base is the
APEX.  The average rate of heat exchange across the entire exchanger
must meet or exceed the maximum requirement of the device.

Liquid cooling has an advantage in that the entire heat exchanger has a
"wide base" -- heat is being put in all across the face of the exchanger
-- so it will pretty much be the same temperature at any point.  That
means that any cubic millimeter of mass or square millimeter of area
will exchange heat at the maximum rate.  This means that many narrow
fins can be used.  In addition, each fin may have multiple bases and
apexes, as tubes may pass through the fin at multiple points.  Each
point where a tube enters the fin is a base, and the center point
between two tubes is an apex.

Air cooling, however, has a "narrow base" -- all of the heat is put into
the exchanger at one point, whether that be the center or the edge.
That means that thicker fins must be used, to withstand the greater heat
at the base than is found in a liquid-cooled multi-base exchanger, and
it means that the area must be greater, due to the uneven input of heat
into the exchanger.

These uneven heating zones are sinking into the coolant in a liquid
system, so by the time the coolant reaches the exchanger, the
temperatures have averaged out in the fluid.

An example of this last is to compare a single faucet, which mixes hot
and cold water into a single stream (simulating heat entry into a liquid
exchanger), with an older sink which has a separate faucet for hot and
cold water (which simulates heat entry into air-cooling fins).

In other words, the air molecule that rubs up against a liquid-cooled
heat exchanger will be excited at about the same rate, wherever it hits,
while the air molecule that rubs across an air-cooling fin will take
less of the load at the apex than it will at the base.  That means you
will need more area to get the same exchange.

Getting away from physics, look at the Real World(tm).  Compare the
built-in headwind of the Mustang, with a liquid-cooled 1800HP Merlin
engine, with the Corsair and its air-cooled 1800HP P&W.  Look at the
U-Bird fuselage from the front, and 60% of what you see is heat
exchanger cooling fins.  Look at the front of the Mustang fuselage, and
90% of what you see is aerodynamic curves, with only a small scoop at
the bottom to feed the heat exchanger.

Chance Vought was a genius -- you KNOW he wouldn't make a flat plate
nose like that if he had any other option (he was required to use an
air-cooled engine by the Navy).

And you can bet that the guys at Pratt & Whitney didn't pick the "ring
of jugs" design because they liked the way they look!

You won't find a single air-cooled-engine single-seater of WWII which is
as aerodynamically efficient as liquid-cooled-engine single-seaters of
comparable power.

Again, sure, you could make a Mazda rotary into an air-cooled rotary,
but I just don't see why it would be worth doing.

Keith Wood

Keith,
I don't think you could make a Mazda rotary engine air cooled. There are
just so many things against it. Just the difference in temps at the
combustion side compared to the inlet area, would bend the engine into a
banana shape.
All engines benefit from liquid cooling, even Motorbike engine are going to
liquid cooling - especially the higher powered engines. Tighter tolerances
gives better sealing and higher compression, liquid cooling reduces the
problems of detonation or preignition. I would hate to think what the hot
spots in a Mazda air cooled engine would do.

The smaller air cooled Rotary engines in Motorbikes in the bygone era seemed
to be OK, but I don't know what HP they run at - well under 100 hp I would
suggest.

To me - it's just a silly idea!
There is just so much more to be accomplished then dwelling on this, better
to put resources to other aspects of the rotary.
George (down under)


The Rotary Engine NewsLetter. Powered by Linux.
ACRE NL web site. http://www.rotaryeng.net
Copyright 1998-2007 All world wide rights reserved.