On 06/10/2012 06:21 AM, Bud Yerly wrote:
> If I do say so myself, I never have cooling issues in Florida at 95
> degree summer days using 50/50 Anitifreeze. Judging by the calls and
> email success stories, these simple techniques work pretty much world wide.
Although your solution might indeed work world wide it is less suitable
for Europeans.
There are a few reasons for this:
1) Your solution is focussed on more cooling air flow. Unfortunately
more cooling air also means more cooling drag. More cooling drag means
that higher power settings are required to achieve a desired cruise
speed. Although that might not be significant for those living in a part
of the world where fuel prices are only a fraction from what it is here,
for many of us here the fuel price is by far the largest cost involved
with flying our Europa's.
In another posting you wrote:
"135 with the old pants at 5500/34 inches at 1000 MSL,"
"His trigear only goes 130 at max cruise."
I do 130 with 27", about 155 with 34". This is a heavy hi-top tri-gear.
Although I selected a very efficient prop and have Fred Kleins wing root
fairings, I believe that a large portion of the reduction in drag is
achieved by my modified cooling system.
2) Climate. I believe that in your area the weather is somewhat more
constant than it is here. This month I have been flying in cruise mode
in freezing temperature conditions (high altitude in Norway) and in
prolonged full power climbs with 35C OAT (Croatia). Unless you use
thermostats you can't achieve constant temperatures in all conditions.
After numerous experiments this is what I have learned:
0) Cooling drag is the main drag of modern efficient airplanes. Imagine
the air swirling around in the cowling, colliding in all kind of
obstacles, and flowing over tubular objects (worst airodynamic shape
possible) and finally leaving the airplane in the wrong direction and
with the wrong speed...
1) The Rotax engine is for 95% liquid (oil, coolant) cooled. Forget
about airflow over the engine, apart from the cylinder walls no cooling
air is needed at all. Sure you can compensate for a poor liquid cooling
with air flow, but it is extremely inefficient. Once you have the liquid
cooling working correctly, you can close off all holes. That's right,
ALL holes. This includes the two "eye's". the gills, the naca inlets,
the nose wheel opening, everything. There is only one small opening
needed to connect the Rotax shroud for cylinder wall cooling.
2) The main problem in the stock setup is the radiator. The stock
radiator is too thick (not even to mention the tandem design). The
pressure difference required to maintain enough air flow is too large,
and you need very large openings to keep enough air flowing. I have
tried many carefully designed diffusers but I never got the stock
radiator working sufficient enough for prolonged full power climbs in
hot weather.
3) Use an oil to water heat exchanger. This means that you have to focus
on the air flow of only one radiator, and also don't need an oil
thermostat and still have superfast warm-up times.
4) Get away with the entire "dog house". It is ugly and it serves no
purpose.
5) Use an adjustable cowl flap to control engine temperatures.
6) Reroute the exhaust so the cowl opening points to the rear. I tried
to build an exhaust augmentor but I'm not sure if it really works. In
any case, the cowling air escaping around the exhaust opening flows in
the correct direction and doesn't upset the air stream.
So. what I'm using now is a thin radiator, mounted flush with the
underside of the cowling. (Custom made, 300 Euro's). It is slanted so it
follows the shape of the cowling. Because of the angle the horizontal
frontal area is very small, you can say it forms a natural diffuser.
Flow is controlled by an exit cowl flap. If the air flow is restricted
by the cowl flap, the air in front of the radiator even doesn't "see"
the radiator, the air flows over the surface and follows the shape of
the cowling as if the radiator doesn't exist at all.
The air leaving the radiator is recycled because it flows under the
engine, taking the heat of exhaust and turbo with it on its way out. No
separate openings for these items are necessary.
The coolant is also used to cool (or heat!) the oil via a heat
exchanger. Apart from the heat-up time, the oil temperature is always 5C
higher than the coolant temperature, which I consider to be perfect. I
keep the water at 105C and the oil at 110C in all conditions. This is
the best for the engine and the efficiency.
For cylinder cooling you can use the standard Rotax shroud with the
opening under the propeller. I made my own shroud because of the vaccuum
pad alternator which prevents the use of the standard shroud, but the
idea is similar.
After one year of using this, I can state the following results:
1) Adequate cooling in ALL conditions. I have excecuted a full power
climb from 0ft to FL095 at 80 knots in an inversion layer with a OAT of
35C for most of the trajectory, with the water temp not exceeding 110C
and the oil not exceeding 120C. This was with the cowl flap not yet
fully open. Although not yet tested, I'm convinced that the cooling
would work ok in OAT's of 45C as well.
2) In cruise the cowl flap is typically only half an inch open; i.e. it
protrudes only a half inch below the belly of the airplane. Compare that
with the huge tunnelexit of the stock XS design!
3) Unlimited ground operations, even in very hot weather.
4) Superfast heat-up times. 5 minutes is even in the winter sufficient
to get the oil temp far enough up into the yellow arc to perform a take off.
5) Low cowling temperatures, despite the lack of ventilation.
Temperatures don't exceed 60C, except briefly after engine shut off. In
hot weather I open the oil tank access door after landing for a few
minutes to let the hot air out.
6) Weight savings. I have not weighed the difference, but I'm sure the
flat radiator and heat exchanger is lighter than the two stock radiators
and tunnel hard ware.
About the picture:
This was the unfinished design. The final version is even smoother.
The only exit opening here is the proptrusion of the "tunnel" against
the belly of the fuselage. The nose wheel opening is sealed off and
flush with the belly. The cowl flap can be extended further but this is
the typical cruise setting. Quite a difference compared to the stock
tunnel, eh?
In addition of the rectangular radiator opening I have two round inlets.
The one on the starboard side serves the intercooler for the turbo (via
a wedge diffuser and butterfly valve). You don't need this inlet if you
don't have an intercooler. The opening on the port side connects to the
cylinder wall shrouds. You could use the standard Rotax shroud with the
opening below the prop instead, but I made a round inlet on the port
side to maintain the symmetry. In most setups you can ommit these round
inlets. That would be an even cleaner nose!
The starboard inlet also facilitates the engine air intake. The port
inlet also connects to a very small auxilliary oil radiator. Without
this radiator the water-oil delta T was 10C, and with the extra radiator
I got my desired delta T of 5C. It is not really needed if you are
satisfied with the standard delta T.
On the top of the firewall you see a small servo, a similar model as the
vertical trim servo. This servo connects via a rod to the cowl flap. The
cowl flap hinges close to the radiator so even when it is fully open the
angle is very low. A simple but very efficent mechanism!
The front exhaust elbows are wrapped to save the cowling which is very
close.
If there is enough interest in this subject, I'm willing to write an
article about it. There is a lot of testing, reading, thinking, and
learning from failures behind this design.
Although I have been flying a year with it now, I have only recently
been able to test it in very hot and demanding conditions during our
just finished round trip in Corsica, Italy and Croatia and overflying
the Alps twice.
More about that trip later.
Frans
|