X-Mozilla-Keys:                                                                                 
Bill Freeman wrote:

Perry,

All car engines are designed to provide *only* essentially
pure torque to the tranny input shaft.  There are small
thrust loads due to the clutch throwout bearing, and small
lateral loads due to vibration and slight misalignment of
the tranny shaft.  But - essentially pure torque output is the
design point.

This is one of the key fallacies of the whole "auto engine
in my airplane" movement, and is mostly ignored or at best
only partially understood  (i.e single bearings on most belt
PSRUs, with a lot of the lateral still in the engine main brgs).

My std recomendation has been for any direct drive auto motor --

1) take the power off of the flywheel end (you are doing this) -
    it is surprising the number that insist on taking it off the end
    that was never even designed to carry the full  *torque* - the "wrong"
    end is typically designed to only drive the fan and accessories.

2)  Build a 4130 truss structure supporting a large outboard bearing
     (a LARGE ball bearing is more convenient, and a plain brg with
      oil flow is messy, requies seals, may be hard to make, etc but will
      last a long time)   to take the thrust and lateral loads.

 (2A)  for a reduction drive, use a double-shear supported (i.e. two bearings,
     one on either side of the gear/pulley) system, since the side loads
     for gears or belts are very substantial.

So,  you really need an outboard bearing just fwd of the prop flange.  At
this point, I'm sure this is pretty disheartening - but if you machine off the
prop flange and counterbore the shaft to take an internal prop stubshaft with
a prop flange, and use a couple rows of radial bolts thru the original shaft
into
the stubshaft, you can salvage most of your system.  Use as tight a fit as
possible, perhaps a 0.001" loose, which will probably require a light press
to assemble over the bearing, which should be a light press fit onto the
shaft, probably onto a newly machined shoulder for sizing reasons.

You'll need a large diameter ball bearing capable of taking the thrust loads,
perhaps 300 lbs at the rated RPMs.  This might be hard to come by, which
is why plain bearings (oil fed, with front and rear seals, a pain) are
conceptually
nice, altho they can be difficult to find.  Pouring and scraping still
works, and
I have made engine bearings this way, but it is tedious.

Support the bearing in a housing (alum is easier to machine) which is tied
to the
main engine structure with at least three (four or five is better) 4130
steel tubes,
say 3/4" OD 0.035" wall, guesstimate w/o any calcs.  If you get serious about
this, we can look at specifics of bearings and support structure sizing.
1/2" tube
may well suffice, esp if more than three are used.

My guess is about 15-30 lbs for the whole assy.

Not good news, but the laws of physics **will** be strictly enforced - even for
Wankels.

My guess is that with your prop extension you are going to rapidly fail the
bearings in the system.

Bill

The Aircraft Rotary Engine Newsletter.              Powered by Linux.

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