Bruce A. Frank wrote:
Bayard DuPond flys a Defiant with twin Fords. From the very beginning the
front engine cooled just fine and the rear engine cooled in the air using a
P51 type scoop but heated quickly with long taxis and holds. Bayards
original use of the P51 scoop was because he felt that the front prop might
provide enough wash on the ground to keep the rear engine cool. It did not
work. Bayard has always thought that the scoop sticking down likely produced
excess drag. He wanted to eliminated anything hanging out in the slipstream.
Bayard had also talked to Burt Rutan who mentioned that the nice streamline
tapered aft cowlings on the Defiants actually increased drag due to the fact
that the air could not follow the curve and stay attached. When the airflow
detached from the aft cowling it also created an extremely turbulant
enviornment for the rear prop. Bayard had broken several starters and
alternator brackets(even an engine mount) on the rear engine during about
200 hours of flying. The front engine just kept on pluging with none of
these problems.
Bayard decided, with Burt's approval, to remake the rear cowl to have a
squared off rear end -- straight sided with no taper, that ends abruptly
just ahead of the prop. At the same time he decided to design a exhaust
augmented cooling system for the radiator on that rear engine. Bayard has
solved the overheating problem on the rear engine with a electric radiator
fan from a salvage yard. It had worked well and had given no trouble but
Bayard wished to eliminate it.
He tore into it about 1.5 years ago. First move was to design a NACA scoop
for the bottom of the rear cowl and belly of the aft fuselage. Bayard's lead
in ramp started in the belly of the plane ahead of the cowl. He also decided
to install a different style of radiator. He had been using a Taurus (Ford
sedan) 2-row aluminum radiator which he was told by Griffin Rad could handle
more than 200 hp. Bayard wanted a more straight through design than the
large format automobile radiator would allow. Bayard got one of the racing
radiator builders to make one with all the cooling capacity of the one he
had but about 6" high and 6" deep while about 30" long. The technical
information on ducting expansion rate was was that the expanding taper
needed to be about 7 degrees, but as steep as 15 degrees could be acceptable
if length of expansion area was limited.
Bayard created a duct that was virtually straight through from the NACA
scoop out through the squared off aft end of the cowling. To get the ground
run cooling he inserted his exhaust stacks into the dowmstream duct area aft
of the radiators. The exhaust pipes continue aft as the duct shape assumes a
round pipe shape. The exhaust pipe stops a few inches short of the exit end
of the duct. The high velocity of the exhaust gas acts as an eductor and
sucks the cooling air through the radiator. The engine has dual exhaust so
things terminate in two large round exhaust holes in the flat backend of the
cowling giving it a very distinct "Space Shuttle" look from back there.
Bayard stays so busy that he has not had time to fly this plane since the
mod (he just picked up his commercial helicopter and helicopter instructors
license, is AI and manager of the repair facility at New Garden Airport, and
an EAA technical advisor) but has done extended ground running and found
that the exhaust augmentation has eliminated the ground overheating by
creating a front to back minimum pressure differential through the radiator
core of about 1/2" of water in a manometer which is apparently adequate.
Bruce A. Frank
The Aircraft Rotary Engine Newsletter. Powered by Linux.
http://home.earthlink.net/~rotaryeng/ http://www.linux.org