Bill Freeman wrote:

A venturi works because air is compressable.

This is absolutley  NOT correct.  A venturi works because it changes the flow
velocity, and Bernouli's Principle says that pressure and velocity are inversly
related -- this is what holds up your airplane.  It also holds up submarines
by their diving planes on the sides of the sail.

If you put a "T" fitting with a slight venturi in the fuel line just
upstream (low
pressure side) of the fuel pump, the fuel flow will cause a local pressure drop
which will suck in oil from a metering jet in the side port of the "T".  If the
metering jet is suitably sized (which may be difficult to estimate and expensive
if too small and you fry your engine) the oil will be sucked from the oil tank
(assume a gravity feed to the "T").

A venturi after the buterfly in a peripheral port would allow for a vacum
signal in proportion to RPM.  This would also keep you from streaming oil
all over the inside of your manifold.

Please reread my "Clarification"  which I posted about 1 minute after the
first post
when it occurred to me that I had not explained what I was thinking very well
at all, and it might be assumed that I meant dribbling oil into the intake
manifold.

Dribbling into the manifold would be bad because the driving vacuum is in
opposite direction to what we need, i.e. max vacuum at idle so max oil at idle,
which is wrong.  So, self-feeding into the manifold is out.  I did say in the
clarification that I suggest metering fuel into the fuel line, but perhaps
you were thinking on the high pressure side of the pump.

I don't know if this would mix very
well with the fuel/air, and you still have to figure out the ratio.

I wonder if you could use a fuel injector, just use oil instead of fuel.  A
bigger injector would allow for the viscosity difference, and you could run
the mixed fuel until you get the ratio figured out. (when it smokes, back
off on the fuel/oil mix until you are not adding any oil to fuel).  You
might be able to slave the oil injector off of the fuel injector, and the
timing would be taken care of.

I considered the oil injector idea, but it brings back all the
complexity,weight
and cost problems that we avoid with adding fuel to the tank.  I think my
auto-feeding system would work well, is extremely simple and has *no* moving
parts (my favorite number of moving parts) and the qty of oil injected would be
proportional to the fuel flow, which is what we want and the engine requires.
This does probably require a manual or automatic oil shutoff valve to prevent
drippling your fuel line full of oil when the engine is shut down.  This is
a VERY
critical failure mode, either due to human error failing to open the valve at
start, or having an automatic system fail closed. The valve would need to
be located in the pilot's scan, or at least an idiot light for valve
position. Since
a "oil valve off" while running is an engine failure ( I assume, it would
seize --
might it just wear out the seals very rapidly, but run OK, albeit damaging
itself??)  Anybody know?

A valve that opened up the oil line any time there was fuel pressure would be
ideal.

Bill

The Aircraft Rotary Engine Newsletter.              Powered by Linux.

http://home.earthlink.net/~rotaryeng/            http://www.linux.org