We have the FEA on the new bell housing design.
Hopefully I can publish it shortly.
Only minor mods are required to put ALL the bell housings stresses
down in the blue.
This is the same high loads we subjected the
stock plate and our combined adapter plate motor mount
system to a few years ago. The load represent what happens
when you power on stall an aircraft with a 50 pound prop rotating
at max RPM and the pitch angle changing at one radian per second.
Just about enough to bend a Lycoming engine prop flange
according to the Sky Ranch Engineering manual. This is
the worst thing you can do to a PSRU or a direct drive
aircraft engine. BTW If you go over a bump with your Cadillac
500 cubic inch V8, direct drive powered air boat the crankshaft
will break. Same principle.
Here is the static test rig again and the old FEAs on the adapter plates.
Bob White crashed his airplane with this adapter plate and it did not bend
but the lower motor mount tube buckled and the blades of the prop where
sheared
off by contact with the ground. Nothing like a crash to test one's
designs.
The new bell housing has lower stresses and deflections
than this plate. However Larry is adding a tiny bit of meat here
and there to put it all down in the blue. There will be little or no
weight increase.
Paul Lamar ...No rotor no motor.
Paul,
Does that 600 lb bending load include assymetric prop thrust, or is it
just gyroscopic loads?
Regards,
Gordon.
I think you mean asymmetric prop thrust.
No but I can't imagine that exceeding plus or minus 50 pounds
unless you are interested in running a one blade prop.
Do you have some data that suggest otherwise? If you get 50 pound
variation on blades that are generating only 200 pounds each you have a
very poorly made prop indeed. Also the moment arm is a max at 1.5 feet.
If the prop throws a blade a separated PSRU is a blessing
as that means the weight of the engine stays on the airplane.
If the engine falls off you die for sure due to severe aft CG.
Thrust is evenly distributed in the bell housings so that adds
little stress to the material. Same goes for torque.
Gyro reactions are by far the worst type of loading the bell housings
will see.
Can you think of any other horrible scenarios? :)
Paul Lamar ...No rotor no motor.
Paul,
I'm talking about the asymetric thrust due to P-factor.
If you're flying at an a 10 degree angle of attack, the downward blade
will make considerably more thrust then the upward blade, due to a
difference in the angle of attack between the two blades of 20 degrees.
The resulting difference in thrust between the left side and the right
side results in a bending moment about the vertical axis. The moment arm
can be up to 3 feet on a 6 foot prop. This bending moment could actually
be higher than the gyro moment of a 1 radian per sec noseover.
The thing I'm wondering is if there is any flight maneuver where the two
moments could be cumulative. I don't think so because if you are flying
at a high angle of attack and then enter a stall, the nosedown will get
rid of the asymetric thrust. I'm not sure though?
I will put together a spreadsheet to figure it out. It is a more
involved calculation than gyro moment.
Regards,
Gordon.
The thrust is centered on the blade more or less so the moment
arm is about 1.5 feet. Not to argue but I doubt the angle attack approaches
ten degrees unless you are in a slow flight mode near stall.
Also the slip stream is always faster than the air speed.
Show me the 3D CFD and I might buy it.
Paul Lamar ...No rotor no motor.
Paul,
You are correct that high angles of attack will be seen in slow flight
mode, near the stall. However, this is a vital part of the airplane's
performance envelope so it cannot be ignored.
In fact, depending on the airplane and wing, you could see quite a bit
more than 10 degrees AOA just before the stall.
We cannot assume that the total thrust acts at the blade halfspan point
for the purpose of calculating bending moments -- like we do in
calculating wing bending moments.
Unlike a wing which has more or less the same lift all along its span, a
prop has much higher airspeeds at the tip, hence higher lift. Most of
the prop's thrust comes from the tips -- actually a few inches short of
the tip, since flow reversal at the tip deflates lift there.
I'm not sure I follow you on the airspeed of the prop slipstream. How
does that affect anything? The difference between slipstream speed and
aircraft speed is greatest at slow speeds. Thrust is also greatest at
slow speed.
The result is that asymetric prop thrust is a real force that produces a
real bending moment at the propeller shaft axis. This load will
naturally be transferred to the engine via the reduction drive and its case.
Let me work out the calculations and then we can argue about it...
Regards,
Gordon.
You can't extrapolate wing theory too far to prop
blades. Air has viscosity, adhesion and mass that causes it to do
complicated things when twirled around in a circle. As I said
I will believe it when I see the 3D CFD.
I have plenty of blade element spread sheets for props. Tons! I don't need
any more of them. Here is a good one. It does a fair job but
simplifying 2D assumption were made.
We don't want to argue about it on here. Nothing will
be resolved.
If you want to pop for strain gage installation
all over the bell housing with high speed data gathering I will listen.
If you want to do a wing angle of attack plot of an airplane through out
the speed range I will publish that providing you take into
consideration the aspect ratio. That should be pretty accurate.
Paul Lamar ...No rotor no motor.
The Rotary Engine NewsLetter. Powered by Linux.
ACRE NL web site.
http://www.rotaryeng.net
Copyright 1998-2006 All world wide rights reserved.
