"Lemings, Bob" wrote:
Paul,
Has any consideration ever been given to air-cooling a Rotary? I realize this is a fire-breathing heat generator, but one liquid on board is better than two. What if cooling fins were added to the rotor and side housings and oil flow added to the coolant passages instead of coolant-- increase the size of the oil cooler, oil flow, and the quantity of oil available to recirculate. Maybe someone out there is proficient in Thermodynamics that could do some calculations? What are the heat transfer properties of engine oil (Mobil 1, of course) compared to Ethylene Glycol and water. Ethylene Glycol by itself is very poor, I know.
Bob Lemings
Its been done but the big problem is cooling between the rotors.
No problem on a one rotor engine. Come to think of it that might
be an easier way to make an all aluminum one rotor. I should apply
for a patent :) Also while we are on the the subject it is easier
to make an aluminum front housing if you use an external oil pump.
Are you listening Richard? If you mill some parts of the ribs out of the
rotor housing you could feed coolant into and out of just the rotor
housing and cool the end housings with just air. No casting would
be required for these air cooled end housings as it can all be done
with a manual mill if necessary.
Paul Lamar
Further, computer cooling has come a long way through the advent of heat
pipes, passive water cooling, etc. Would it be so hard to use a heat pipe to
pull the heat from the core out to the fins? What about doing something crazy
like filling the water pipes with honeycomb/fin somethings and using fans to
force air THROUGH the block? I realize there are thousands of issues to
answer, but if there's any place to pose a crazy idea, this must be it. =)
It just seems as though we could learn a lesson from computer cooling. CPUs
have enormous heat outputs in very small packages, so heat density is a
problem there, too. You're only talking 150degF or so max there for most
purposes, but the concepts are the same - extract heat from a high density
area to a lower one. And you don't have the benefit of ambient eventually
being 40 degrees or less.
OK, monkey math time. A P4 3.2GHz with hyperthreading CPU dissipates 82 watts
in a ceramic package of about 0.8 cu. in. That's about 0.14hp/cu. in. Some
heat is dissipated by the board below it, but not much so just assume it all
has to go up, through the cooler.
Now take an 80 cu. in. rotary. Oversimplified already because this is just
combustion chamber volume if memory serves, but moving on... According to one
of Paul's excellent articles
(http://home.earthlink.net/~rotaryeng/how-to-cool12.txt) the rotary (12A, so
again oversimplified but at least a starting point) puts out about 24% of its
power to waste cooling heat, so at 180hp that's 0.54hp/cu. in. If I recall
correctly the oil cooler handles about a third of that so the water cooling
system needs to remove 0.36hp/cu. in. That's a bit over twice what a CPU
requires, but the CPU in this case is being cooled via "traditional" means -
no heat pipes or etc.
Somebody ought to check my monkey math but it seems as though SOMETHING should
be possible. I'm not opposed to water cooling, but it seems as though we
haven't done much in the automotive world to capitalize on cooling
improvements made in other industries. Am I just full of hot air? =)
Regards,
Chad
The AirCraft Rotary Engine NewsLetter. Powered by Linux.
ACRE NL web site. http://home.earthlink.net/~rotaryeng/
Copyright 1998-2003 All world wide rights reserved.