Subject: Spark plug area cooling--early Powersport development
From: Rotary Engine
Date: 10/10/2008, 9:50 AM
To: AAA Put this in the To box


Paul, Doug, Bill, Rob, and all--when this issue of cooling related to
the spark plug boss area came up, Doug and others asked for photos and
information relating to the cooling on our Superlight prototype build in
the 1990 and developed through 1995. Sorry it took so long .  Each of
these photos was taken from color transparencies so they were
monstrously large.  They had to be reduced too much to send and it made
them a little ragged without enhancement.  After I bring them up, a
simple double left click restores the image quality.

A few notes--

1. Most of these photos shown are of the SuperLight engine/installation
in the RV-3; two are shown with the successor conversion engine
installed in the RV-4  but others show the two engines side-by-side on
the bench.  The Mazda conversion prototype was  subsequently test flown
in both the RV-3 and RV-4.  The conversion engine was  essentially
a Mazda engine which among other things included our ports/induction
system, mechanical fuel injection, electronic ignition and internal spur
reduction gear for the propeller.   Standard unmodified gen two water
pump worked very well.

2. On the Superlight,  the coolant "in" is shown with the hose connected
to the center manifold on top of the engine (this engine was indexed 90
degrees).  This cooling manifold had cooling jets directed at the spark
plug bosses.  The return fitting and hose can also be seen. The Aeroquip
fitting on the center housing is an oil line (this was a dry sump
engine).  Some photos show components partially assembled or off.

3. We made this externally mounted and routed pump on
the SuperLight from a combination of OEM parts and others we had cast
and machined/welded in-house.

4. There two water heat exchangers are shown angled up on each side in
the RV-3 installation photos.  A third exchanger of exactly the same
size for cooling oil was mounted between them on the horizontal.  All
three of these were identical size/model Serck plate-finned, furnace
brazed heat exchangers which are sold for use as race car oil coolers.
Very efficient. Because these have no header when used for water,
we fabriacted the one shown in the photos.  It was also a tangental
swirl pot for coolant.  The configuration for this was from
a declassified WWII technical paper.  Worked great.

5. Included is a photo of our aluminum end housings prior to application
of the wear surfaces to give you a better idea of the cooling flow
through them.

6.This engine mount of the RV-3 shown in these photos was modified for
the conversion engine.  We swapped the prototype conversion engine back
and forth between the RV-3 and the RV-4.  To save work when converting
the RV-3 engine mount, we simply left the heat exchangers and baffling
used on the SuperLight installation you see here which had an aluminum,
annular-inlet cowling.  It was in this form that we raced the RV-3 at
Sun'n Fun in 1997 with Alan up in time-to-climb and cross country 100 mi
race.

7. For the Mazda conversion engine, we found with sustained high power
settings without the cooling jets in the manifold shown here atop the
Superlight engine,  it was necessary to modify the standard rotor
housings to increase surface area and coolant turbulence around the
spark plug bosses.  And we also found the second gen water pump the best
up to that point, and adequate in a 100 mile race and other testing.
while these results were very encouraging, they are by no means was
definitive long terms at the same and/or higher power.    .

8.As I look at these photos now,  they reflect the observation which
Paul made a few days ago relating to the cost and complexity of altering
the basic Mazda system. Most of this kind of complex one-off stuff such
as our SuperLight has little applicability to the collective advancement
of rotary aircraft engines to the largest number of enthusiasts.  Our
goal was ultimately limited production.  We did most stuff in house--the
exception was some foundry work, some gear cutting, heat treating, wear
surfaces, etc.  But even with doing most everything in house, it was an
obscenely expensive program, and the inherent complexity a siren's
song.  We both spent ourselves into the poor house.  We found we could
eventually develope these complex systems for cooling, torsional
vibration control and others work but the costs and complexity put the
finished piece(s) were beyond a reasonable point for target consumers.
That's why we ultimately focused finally on refining the basic Mazda
engine with our basic systems evolved with the earlier
prototypes.  Although the Mazda iron end housings imposed some
performance penalty due to weight, but rugged as hell.  We tweaked the
ports and runners and turned up the rpm and both the RV-3 and RV-4 were
faster with the conversion engine than the SuperLight.  But the
Superlight was a labor of love at 167 pounds including all accessories.
But from it and other proceeding engines, we were able to develop
systems which were simplified for use on the Mazda conversion engine
with little or no performance loss.

Spending so much time and money on R&D with our earlier rotary aircraft
engine programs ultimately led us to conclude it is much more
cost-effective to optimize the basic Mazda configuration.  That includes
the basic cooling configuration and components.


Steve Beckham


Steve,

Thanks a lot.  WOW!!!  The SuperLight design is simply amazing.  Well
done. Tell me one thing.  On the Mazda conversion engine, what is the
device that is sticking out of the location where the CAS  or
distributor would normally go?

Doug in Japan


Steve

On closer look it appears to be a magneto.
Doug


Thanks Doug.  Regarding your question, what you are looking at is the
electronic ignition, built inside the basic shell of an old Mazda
distributor.  Below are a few notes to hopefully make sense of this.

1. This electronic ignition was adapted from the system we did for the
earlier SuperLight prototype engines.  These used an eccentric shaft trigger
wheel for the ignition.  For the conversion Mazda prototype we basically we
fit this system inside the Mazda distributor. We did that because the design
goals of the project were to work within the stock Mazda configuration and
at that time the vast majority of cores/housings available were of the
earlier type aluminum front cover with provision for distributor or other.
So we redesigned our system to work with those and the existing drive
mechanism.

2.  The basic ignition system design--
The electronic iginition was dirt simple, probably quaint by comparison
today's coil-on-plug computer driven systems. It was battery powered, with
spark plug firing accomplished optically by infared triggers.   Each module
has built-in primary switching ciruitry--switching one of the three small
motorcyle ignition coils.  Fully redundant. Eliminated the need for a remote
amplifier box.  Dwell (coil saturation time) controlled by the length of the
timing segments on the trigger wheel.  We machined the trigger wheels in
house. Conventional, non-amplified, ignition coils provided ample spark
energy and a long duration spark.  Three coils firing four spark plugs; one
coil with a secondary winding firing the two leading spark plugs
simultaneously.  This fired a wasted spark on the other rotor where the
spark plug exposed to burned mixture.  This allowed use of only one coil to
fire separate spark plugs.  Trailing plugs fired independently.   The
electronic switching and overall design made it possible to use normal
supression type sparkplug wires to prevent radio interference.  No shielding
of wires necessary.

3. Converting from SuperLight to the Mazda conversion engine--
On the Superlight we used  an eccentric shaft ignition trigger wheel.
Machined a steel billet to include the trigger wheel segments on the back
side of the eccentric shaft counterweight machined on the other side with
"ring" gear teeth machined on the periphery for the starter.  That engine
was a counter-rotator with the propeller, so the eccentric shaft was
reversed; facing to the rear of the aircraft, and the tri-function ignition
trigger wheel/eccentric shaft counterweight/ring gear was consolidated there
at the rear of the engine with the starter, dry sump pumps, etc.

4. The ignition was checked on runup by switching each of the "leading" and
"trailing" units just the same as you would to check the right and left mags
of a recip.  These early Mallory Unilite modules would occassionally fail
(we used to carry a spare set) but the later models were very robust and
worked flawlessly.  The photos of the two engines on the bench shows the
Mazda conversion engine with the aluminum cover of the "distributor"
removed.  These covers were tapped for allen drive screws which when
tighened secured the cover to the unit.  We machined the covers by recycling
worn out diesel engine pistons in our scrap bin.  Look at the right side of
the engine installed in the RV-4 and you will see the unit with the cover
installled.

Steve Beckham


-- 
The Rotary Engine NewsLetter. Powered by Linux.
ACRE NL web site. http://www.rotaryeng.net
Youtube key word UTUBPLEASE
Copyright 1998-2008 All world wide rights reserved.