François Badoux wrote:
The pump has a 31 tooth sprocket and the shaft has a 23 tooth
sprocket.
Lynn E. Hanover
Francois Badoux wrote:
No, sorry Lynn, on our engine, the pump has a 36 tooth sprocket and
the shaft a 24 tooth sprocket.
Francois
Which would slow your pump even more making cavitation even less likely.
24/36 = .66 e-shaft speed verses 31/23 = .74 e-shaft speed. If what you say is true Francois it would be a major breakthrough in Mazda rotary engines because
it would allow a drastic increase in the 10,000 RPM redline allowing more power to be generated. The red line of the engine now is based on a combination of
high oil pressure and rotor weight.
Paul Lamar
<< Lynn,
>
> What is your take on the apparent cavitation mentioned by Franois in
>
> the previous post?
>
> Kelly Troyer
The problem seems to exist only on their test stand not in the
aircraft. So as an engine problem it does not exist. The pump ring
and rotor are made of powdered metal and will tolerate cavitation
only for a few minutes. They have much more time on the engine than
that and even more flight time, so it is a test stand thing they ran
into. The pump, or should I say the pickup tube does run into a bit
of a vena contracta that limits the upper end of pump performance
before the top output can be achieved. They have made the change to
the pickup (adding a bell shape to the end of the tube, but that was
apparently no the cause.
It is a kind of thing you would only see at the end of a long
straight-away on a really hot day if you had used a multigrade
nonracing oil. The engine will foam up the oil after a long run
above 9,500 RPM. Multigrade oil is the worst at doing it, and
nonracing oils have very little anti foaming agent included. So, the
oil foams up, the hot oil is transformed into a higher viscosity
from the heat, entrained air displaces the oil and the vena
contracta is maxed out and pump output sags. So you start the lap
with 100 PSI and end the lap with 80 PSI. No big deal at all, as
once below 9,000 in a few places, the windage tray starts to burp
the air out of the oil and things are good again. This is not true
cavitation because there is no damage to the pump even if you do
this for years.
With racing oil there is no problem. With synthetic oil there is no
problem. With a low viscosity dino oil there is no problem. With Max
revs below 8,000 RPM continuous there is no problem.
If I had to put the pump back inside the engine, I would use the
racing beat pump with heat treated rotors and the "O" ring groove at
the rear of the pump housing. There will always be some suction
leakage and pressure leakage from the pump body iron junction, so
that would help a bit. You can adjust the relief to give well over
100 PSI and what more would you need from a pump? Anything over 65
pounds of oil pressure is just fine below 9,000 RPM.
So, they did something on their test rig while measuring pump output
that gave them a start when it cavitated. And now they are reviewing
their setup to see why. It is not a problem in the engine. It is on
a test stand.
On the bigger picture, I feel that it is important that these people
are successful in getting this certified. There have only been
experimental advances in aviation for ages in this country and
around the world because of lawyers. The government has to step in
and limit liability claims of these nut cases who sue because of red
hair. Or because Dad flew his Piper into cumulogranite.
We should have spent the last 40 years developing newer better
aircraft engines not dying behind the 1920s designs our great
grandfathers trained on. Part of every penny you spend on aviation
and related businesses is a tax that goes to the trial lawyers of
this country. You didn't get to vote on it, but taxed you are
anyway. Every product has a hidden charge for liability insurance.
We must reform tort law in this country or the whole world will pass
us by. The first step would be to never vote for a lawyer for
anything, not even dog catcher. But I run on again.
Lynn E. Hanover
Lynn,
I like-- no, I really like your tone!! It's not just product
liability-- the high cost of medical care is testament to what lawyers
are doing to all of us.
Bob Lemings
From Francois:
I fully agree with both of you, but I don't have the privilege to vote in the USA, so it's better for me to shut up... at least on US domestic political
issues! J
Back to the pump, please:
1) Lynn said: ". The pump ring and rotor are made of powdered metal and will tolerate cavitation only for a few minutes."
Wrong (sorry Lynn... nothing personal...): our engine pump's inner and outer rotors are made of solid steel and will tolerate cavitation for tens if not
hundreds of hours, despite increasing pitting and slowly eroding performance. We may all be talking of different engines with different oil pumps... are
we???
2) The problem is not whether cavitation is likely or not: we HAVE cavitation, it's an observed fact. We have it on the bench engine, on the Arrow's engine,
and we are able to reproduce the problem at wish on a specially made test rig.
3) I accept the fact that we may have the wrong oil. We are going to test several oils next week. But it's only a contributing factor. The reason for
cavitation seems clear to me: our georotor pump has two 4-lobe inner rotors (the outer rotors have 5 lobes), one on-top of the other, with an Al spacer plate
between them. They are offset at a 45° angle. If you look at this arrangement, it seems pretty obvious that, at high RPM, it becomes nearly impossible to
feed the second rotor THROUGH THE FAST CLOSING CHAMBER OF THE FIRST ONE. Simple fluid dynamic calculations show that you just cannot achieve that within a <1
bar head loss. Avoiding cavitation on this particular pump model is impossible past a certain RPM which depends, primarily, on oil viscosity (as a matter of
fact, we had much less of a flow problem on the Arrow when we switched to 20 weight oil. But then, the prop governor didn't like such fluid oil). And sure
enough, in the '93 model oil pump, Mazda added on the side of the pump's body a passageway to feed the second rotor. No coincidence there, knowing Mazda,
they had reports of a problem and made the necessary change. They didn't change the pump's housing and the geometry of the pump's pad on the side-housing
just for the heck of it.
4) Despite cavitation, and even air in the oil, we NEVER observed less than 85psi and always had plenty enough flow to lubricate and cool the engine. The
problem is to maximize flow through the oil cooler to make it as small and efficient as possible.
5) Air is entering the pump through the chain lube hole when the pump is cavitating, and ONLY THROUGH THIS HOLE. When plugging the hole shut, we still have
cavitation all right, BUT NO MORE AIR IN THE OIL AT ALL.
6) The fact we blew up a tube on our first Griffin oil cooler is testimony that the pump is giving plenty of pressure despite cavitating. The reason for this
happening was MY GIVING AN INCORRECT PRESSURE SPEC to Griffin. Griffin delivered a custom-made replacement cooler that was designed to the right pressure
spec in A RECORD-BREAKING 36 HOURS. I wish to say here that we are perfectly happy so far with the product quality and the GREAT SERVICE offered by Griffin.
I know it's a puzzling problem. But it doesn't help to deny it. It's there, with brand new stock Mazda parts. Lynn mentioned several times that the oil may
be a factor and I believe he has a point. I believe that the right oil quality will help but not make the problem go away. We'll experiment with several
oils and keep you posted.
Anyway, it's pretty much an academic problem by now, we found the fix, it works beautifully. We now just have to machine a new, airworthy, pump body for the
Arrow.
I thank all contributors for their support.
Francois Badoux
Describe your test set up Francois and we will try to duplicate your results.
I have some oil pumps lying around.
BTW you won't know whether those rotors are powdered metal until you hit one
with a large hammer and it shatters.
Paul Lamar
The AirCraft Rotary Engine NewsLetter. Powered by Linux.
ACRE NL web site. http://home.earthlink.net/~rotaryeng/
Copyright 1998-2003 All world wide rights reserved.