Message-ID: <3FBD8F4E.AE9B94F0@mozcom.com>
Date: Fri, 21 Nov 2003 12:06:38 +0800
From: Marc de Piolenc <piolenc@mozcom.com>
Reply-To: piolenc@reporters.net
Organization: Autodidactics
MIME-Version: 1.0
To: Rotary motors in aircraft <flyrotary@lancaironline.net>
Subject: Radiator size
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

Dear Listmates,

I've been a lurker on this list for some time, monitoring it for
information on aircraft rotary engine installations. I've learned a good
deal, but haven't had much to say. I'm chiming in now because the topic
is one that I am studying intensively in order to write a book about it.

I'm co-author of the first in a series of books about ducted propulsors,
called (appropriately enough) Ducted Fan Design, Vol. 1. It presents a
very simplified method of designing ducted fans and propellers in which
the duct or shroud length is at least equal to the diameter of the
rotor. Subsequent volumes will deal with short shrouds and other topics,
but volume 2 is now reserved for cooling systems. This was originally
supposed to be just a chapter or two in a book devoted primarily to
propulsion, but as I did my research I discovered that, just as
aerodynamicist Bruce Carmichael had said in his book on drag reduction,
there is a huge gap between what cooling systems should cost in terms of
drag and shaft power, and what they actually do consume. Even more
surprising, even people who devote their whole lives to reducing drag,
more often than not have only a vague empirical notion of how to achieve
that with cooling systems. Hence in an airplane where external drag has
been carefully optimized, a large increment in performance is often
still available from improving cooling drag. Better still, the cooling
effect can usually be improved as well.

Good information was available in the past on this subject for those
willing to dig for it. Some engine design texts could teach optimization
of the coolant side of things, while a few aerodynamics texts - notably
Kuechemann & Weber's _Aerodynamics of Propulsion_ - could teach radiator
installations. All that is out of print. What is more, putting all the
necessary guidance into modern language and in one consistent notation
should make Volume 2 more accessible in every sense. So much for the
commercial. 

There are a lot of questions on the list about the proper size for a
radiator, but it is far more important to define the cooling duct's
parameters first. The only mention of the duct in the latest digest of
messages concerned the inlet, but the EXIT of the duct is far more
important, because that and the pressure drop through the radiator
control, in the final analysis, the mass flow through the radiator - and
that is what produces the cooling effect. It is possible to spoil the
effect of a good exit by poor inlet design - by making the inlet lip too
sharp or cambering it the wrong way - but it really takes an effort! I
doubt there is anybody on this list whose inlet is wrong. 

The proper duct parameters, in turn, depend on the pressure-drop and
heat-transfer characteristics of the radiator core, as expressed in the
modification of Miley's equation that I used in my article published in
Contact! issue number 62. Two parameters are needed to express a
particular core's presure drop characteristics, and another two for heat
transfer. Each relation requires only two measurements to determine the
two parameters, since the form of the relation is known. What's more,
the pressure drop measurements can be carried out with no heat transfer
taking place, as there are procedures for correcting for the occurrence
of heat transfer in designing the duct. For heat transfer, obviously,
entry and exit temperatures DO need to be taken for both the coolant and
the air, simultaneously with the pressure drop and flow rate. Even so,
these can be taken at the SAME two speeds previously calibrated for the
pressure drop measurements. As a result, both the test rig and the test
procedure can be simplified by comparison with NACA's early radiator
installation test rigs, which may have inspired Rube Goldberg.

The only problem is, I don't know anybody in the amateur-builder
community who is doing this, and I don't know whether the cores that you
folks are using come with the necessary information already provided by
the manufacturer. It is, after all, much easier to pick two points off a
plot than to derive them from tests... I would very much like to know,
because aside from needing to give valid guidance in volume 2, I will
soon be starting on a hovercraft project for a Catholic parish located
upriver from town and needing reliable transportation for farm goods one
way, building materials the other way. Although I'll be buying plans for
a proven h/c design, I will still be on my own where radiator
installation is concerned, as the hover crowd are at the pre-1921 stage
of development where a bare radiator was thought to give the best
cooling effect.

If anybody out there has manufacturer's pressure drop and/or heat
transfer data for a core or finished radiator that he could scan and
send me, I would be most grateful. It doesn't matter what core - I just
need to test my work so far against real hardware currently on the
market.

Regards to all,
Marc de Piolenc
Mass Flow
http://www89.pair.com/techinfo/MassFlow/ductbook.htm
-- 
"The troops returning home are worried. 'We've lost the peace,' men tell
you. 'We can't make it stick.' ... Friend and foe alike look you
accusingly in the face and tell you how bitterly they are disappointed
in you as an American. ... Never has American prestige in Europe been
lower.... Instead of coming in with a bold plan of relief and
reconstruction we came in full of evasions and apologies.... A great
many Europeans feel that the cure has been worse than the disease. The
taste of victory had gone sour in the mouth of every thoughtful American
I met." --Life Magazine, January 7, 1946