Here is some good evidence that we will not be able to use the
 flywheel in
 place of the flex plate.

 On this page at Mazdatrix, scroll down until you see the tools labeled
 "flywheel stopper"

 http://www.mazdatrix.com/**ftools.htm<http://www.mazdatrix.com/ftools.htm

 Note that one of them fits the non-turbo flywheel BUT NOT THE AUTO.
 And the
 other fits the turbo (Which I assume is different than the n.a. auto
 since
 they don't list it there either).  We may be stuck just trying to beef up
 the existing flex plate.
 --
 David Leonard

 Turbo Rotary RV-6 N4VY
 http://N4VY.RotaryRoster.net
 http://RotaryRoster.net


 I would not look upon it as being "stuck". I would now look at it
 as being better :)

 Ideally one could make a die and form thin 4130 to nest inside
 the flex plate. Bond it in with RTV. That would allow motion between
 the plates and damp the vibration. Unfortunately sandwiching it
 between 4130 plates would help but it would move it towards the starter.
 Perhaps we could space the starter out 1/16th of an inch.

 Also lighten Tracy's damper plate to reduce its moment of inertia.
 Drill the bolts and cut away low stress parts of the plate. I will do
 a sketch.
 Let Jeff do this as he can set up a static balance rig.

 Paul Lamar

Mazda has a similar automatic flexplate ring gear locking tool listed in
the Factory Service Manuals.  I suspect the Mazdatrix version won't work
with automatics due to spacing between the rear iron and the ring gear.
Additional spacers would probably make it work.  Same goes for the
starter to engage using Tracy's redrive.

NOTE that automatic flexplates can and do crack in car usage.  I am
amazed this has not happened previously in an aircraft application.
You were also using a 20-25 year old flexplate with unknown history.  A
new one from Mazda without history may be the short term ticket.

The moment of inertia of a flexplate is very small compared to even
composite props.
You want some moment of inertia (MOI) to smooth out the torque pulses.
You may actually want more MOI than the flexplate/ring gear provides...
Gyroscopic forces due to aerobatics are even worse.
If you want to reduce the weight and MOI you could look at 5-7" race
style button clutches and starters.
Some BMW style guibo/flexdiscs could be adapted to the RX7 counterweight...

It might also be worth trying a custom 3/32" or 1/8" (4130 fatigue
resistant) flexdisk plate with a ring gear, doublers near bolt holes and
spacers to line everything up.
Anyone know the reasoning (flexibility?) behind the 86-91 flexplate
lightening holes and tabs?

Cheers
Cary

Exactly what I just proposed :) Great minds Cary :)



Paul Lamar

  Too right, I wont't be touching those piddly flex plates!!! . I think a
custom steel unit with metal where its required, between the centre ring of
bolts and the dampening plate attach bolts. The outer circumference can
taper down to reduce weight and mount the re-placeable ring gear. something
more like the manual flywheel. Maybe a 6 point dampener instead of 4. Yes
the prop MOI out weighs all other considerations, which leads me to the gyro
forces on a long thin profile such as this psru, some suring up can only
help the situation as I intend using the aerobatic abilities of my a/c
should and as I wish. When these issues are resolved I'll feel much more
comfortable about the situation.
 I'm sure you are questioning the origin of your flexplate Dave.

BTW thanks for testing the water Dave appreciate it. Definitely better the
devil you know, and Dave is helping us understand the beast. The integrity
of the sun gears is well tested so I suspect the issue lies elsewhere. This
means both sun gear failures are weren't failures but results of another
issue.
 Graeme