Hi everyone,
The message below was my reply in a private discussion regarding cooling
of the Rotax 914. I thought that maybe some other people are interested
in this as well, so I decided to share this contribution on the list.
Just a reminder: After initial problems with the cooling of the 914 in
my Europa, I decided (after many hints from other list members to lower
the oil cooler) to take the oil cooler completely out of the cooling
duct. At the moment I have flown 50+ hours with the oil cooler on the
passenger foot well, fed via a 2 inch air opening next to one of the
original 3 inch openings. The 3 inch openings are closed. That solved
most of my problems. The discussion below is about a further
optimization of the cooling, what I consider the ultimate cooling
solution. Ultimate, because it offers good temperature control in all
circumstances, in summer and winter, and comes with the least "air
consumption". The less air is taken in for cooling, the less cooling
drag. Cooling drag is one of the major sources of drag of the Europa.
==
Hi Paul,
> Well I test flew my airplane this evening and disappointingly, my set up
> over heats as well.
Yep, I feared that this was going to be the case. The stacked radiator
setup reduces the air flow too much, no inlet or diffuser can change this.
I have now flown quite a lot with my split radiator setup, including in
mediterran countries with temperatures around 100F and prolonged climb outs.
My findings are:
1) The water temperature is no longer an issue. Never. Can't get it over
100 Celcius no matter what I try. This includes long slow speed climbs
in high ambient temperatures with the cowl flap closed! It is a dramatic
difference. Before I splitted the radiators, the water had a tendency to
reach the boiling point, even in low power level flight.
As I did not change anything else regarding the water cooling, it is an
obvious conclusion that the oil radiator was just blocking the air
stream too much. With the oil radiator out of the way, the water
radiator gets enough air flow and even appears to be oversized. I fear
that in the winter I won't be able to get the water temp up high enough,
unless I change my cowl flap design.
2) The oil temperature takes a long time to climb up high enough to
allow me to take off. Far too long. I have to wait for a long time, and
take off as soon as the oil finally reaches 60 degrees Celcius. (which
is much lower than I actually want, but I can't get the oil temp up
higher than that in a reasonable time. Off course, once take off power
is applied, the oil temp comes up in a short time). Water temp comes up
almost instantly after starting, but oil temp takes a very very long time.
It is obvious that I won't be able to fly this winter, unless I find out
a way to heat up the oil.
Again, it is obvious that impeded air flow was the problem in the
stacked radiator design. The stacked radiator design had one advantage
though: the oil radiator was receiving pre-heated air, allowing for a
reasonable warm up time.
3) The oil temperature still climbs too high in prolonged climbs. It
more or less stabilises just a tad before the red line, in warm weather.
This indicates that the oil cooler is not setup efficiently enough. The
air inlet is a meager 2 inch via a scat tube, and the diffuser is a
simple wedge shape, and testing shows that most of the air leaves the
radiator near the far end, and the front end is not doing much. Still,
it is impressive that an air inlet of just 2 inches provides better
cooling than the standard huge XS "mouth".
4) Forget about directing air through the cowling. The cylinders /
engine core are not a great heat source. I have completely closed off
the round air inlets since I splitted the radiators. All mentioned
testing has been done with these inlets sealed. Attempts from other
people to solve the cooling problem by increasing the air flow over the
engine are just ill attempts to air cool an engine which is designed to
be cooled by liquids instead, just like a car engine. Blowing loads of
air over it might work but it is inefficient and comes with a lot of
cooling drag. It is simply the wrong route.
Needless to say, with 85 hours on the hobbs, I have no signs of elevated
cowling temperatures. The spark plug labels are still bright yellow,
cable ties have not melted, there is just nothing wrong with the cowling
temperature. Note that I still have the Naca ducts in the top cowling,
which are my only cowling air sources. Gills were never opened at all. I
have no shroud over the cylinders.
Keep in mind that I have the oil cooler mounted on the foot well. All
hot air from the oil cooler is dumped inside the cowling without any
routing. If I optimize this, cowling temperature will drop even further.
Cooling air for the cowling is just not an issue once the radiators are
de-stacked.
So.... After thinking a while about it, I'm going to try the following
solution:
I will reduce the oil cooler radiator to a fairly small size. Yes,
something that has equal dimensions as the intercooler would be
sufficient, although my intention is to put the oil rad somewhere on the
port side. This allows for an easier routing of oil and air.
Then, the oil that leaves the radiator is fed through an oil-to-water
heat exchanger, fitted in the hot side of the water radiator hose.
I figure that this will give me optimal performance in all situations:
1) During startup, the oil radiator isn't doing much as there is hardly
any air flow, so the oil temperature is dominated by the heat exchanger.
The warm water is heating up the oil, so it will reach take off
temperature much sooner.
2) During slow hot climbs, the oil radiator is probably not sufficient
to keep the oil cool. Here, the excess heat is transferred to the water,
and as the water radiator has now surpluss capacity, I have no doubt
that this isn't going to have a bad influence on the water temperature.
(Remember, max heat dissipation of the oil is 7kW, while the max heat
dissipation of the water is over 30kW. The pre-cooled oil maybe needs to
exchange just an additional 3 kW or so to the water, which is only 10%
of the water cooling capacity. )
This setup also prevents over cooling of the oil. The water stabilizes
the oil temperature to a comfortable 100 to 110 degrees Celcius.
3) During the cruise in winter time, the heat exchanger prevents the oil
---From cooling down too much. Several people have reported a problem with
the oil temp in winter, and I can easily believe it from what I have
seen now. Of course I have to close the cowl flap to prevent the water
temperature going down too much.
4) This setup eliminates the need for an oil thermostat. I hate oil
thermostats with their complicated routing, lots of connections, and
valves waiting to fail. Coupling the oil temp to the water temp is a
more elegant solution, and allows for a faster warming up of the oil
than a thermostat could achieve.
Of course I have to change my water radiator inlet and outlet once
again. The whole duct has previously been optimized for more and more
cooling, and with the oil radiator out of the way, the thing simply
cools too much and the cowl flap can not be closed far enough. I foresee
a reduction in cooling drag as well with my new setup. :-)
Maybe you wonder why I still want to have a small oil radiator, as the
whole setup is going to work probably well enough with the heat
exchanger alone. There are a few reasons for this:
1) reduncancy. I don't want to fry my entire engine if I loose the water
cooling, because of a coolant leak or whatever. It is acceptable if I
bake my cylinder heads, but as long as I keep some oil cooling I can
spare the rest of the engine during the flight to the forced landing spot.
2) I'm not sure how high the oil temperature is at the exit of the
engine. Oil temperature is measured at the oil inlet, AFTER cooling, so
I don't know what it is before the radiator. At least I want to get the
oil temperature down to 130 Celcius before it is allowed to go the heat
exchanger, to prevent localised boiling of the water. I know, some
Finnish folks fly with the heat exchanger alone and report good results,
even in hot weather, but I think it is safer to pre-cool the oil a bit.
I figure that a small radiator is enough to get the oil temp down just a
tad, as the temperature difference between oil and air is very large at
that point. The water takes then care of cooling the oil further down.
> I have a TIG welder and I can custom make my own but I have not had any
> luck finding a suppler for blank cores. Have you had any luck in
> finding different radiator cores ?
No, but I haven't tried very hard, as it is not my goal to get the
radiator stacked with the intercooler. This would make routing of the
oil line too long, given the fact that it has to be routed back to the
in line heat exchanger as well. And I have developed an aversion against
stacked radiators. ;-) It didn't do much good for the water cooler, and
I fear that a stacked oil cooler will also spoil the air flow through
the intercooler, which isn't great to begin with.
What do you think about all this?
BTW, I still have no luck with the intercooler. Airbox temperature is
still high. Optimizing the air flow through the intercooler is one of my
other winter projects. ;-)
Frans
|