Paul,
  How about a dual ignition systems?  Assume coil-on-plugs design
 and that we would use a two stroke marine engine magneto and
 coils incorporated into the starter gear or torsion plate area
 of the housing.  Also assume we would have  capacitor driven
 ignition modules like the new solid state digital programmable
 MSD, 6AL-2.  Check them out.  Small and compact.  One for each
 rotor.
  First some givens about our rotary engines with the single
 leading plug:  The engine doesn't run very well on trailing
 plugs alone, as quite a lot of the fuel goes out the exhaust
                 and
 is burned in the exhaust system.  Said another way, with a two
 rotor engine and a leading plug fouled, the exhaust system may
 well overheat, or the engine would run rough enough that we
 would want to shut it down after a short time.  That means the
 Achilles heel of the Mazda rotary is the leading plug.   As
 quite a few members have pointed out though, a spark plug is
 pretty basic technology.  Yes I agree  but I also can't count
 the number of times I have had them crap out on a  2 stroke oil
 burning engines.  Prudence would dictate changing them quite
 regularly since it is a critical point of failure or beg Mazda
 to cast us up some two bung leading plug housings.
  Conversely if the trailing plugs fail the engine power only
 reduces about 3% if that.  So with a good leading plug and a
 magneto system,  the engine would continue to operate just fine.
 BSFC will be a little off but we won't get our feet wet and the
 plane can continue on for many many hours.
  OK we can all agree that  a magneto is basic technology and
 very fail-safe.  The magneto fires when a magnet passes by
 coils set 180 degree  (20Bs would need three spaced at 120
 degrees) apart.   These coils could be wound to produce 300
 volts.  Timing can be fixed  for a cruise rpm.
  Now for the plus alpha.  The MSD ignition devices can be
 programed to fire the leading (and trailing) plugs many times
 and at moments when the magneto isn't doing that.   Each MSD
 power pack could  get its signal to prepare for  a firing
 sequence from the magneto or modified coil voltage pulse. Since
 the MSD devices have intelligence they can vary the timing with
 rpms. and other inputs received from sensors or even a
 MEGASQUIRT engine control board.
  The question is,  How do we isolate the MSD power from the
 magneto power because they are both hooked up to the same coil?
 Diodes are the easy answer but I wonder if that is a robust
 solution?  In the scenario above  power unit could be charging
 the leading plug coil.
  A FET switch  might be an answer but the fail safe would have
 to allow the magneto to continue firing the leading plug coil
  FETs on the two MSD outputs to the leading and trailing plugs
 would allow split timing if necessary at the higher rpms.  For
 instance the magneto fires the first burst on the leading plug,
 followed immediately by two stronger pulses from the MSD
 Capacitor while perhaps a third pulse from the MSD is directed
 only to the trailing plug.
  Your thoughts?
  Doug in Japan
 I think you are making a mountain out of a mole hill here Doug.
 A phenomena call sag happens when the plugs get to be about 50
 hours old or so.  Minor loss of power.
  The solution is put in new plugs on a regular basis. The plugs
 catch hell in a rotary as they spark twice as often as a 4
 cycle piston engine.  The same as a two stroke.  Now this has
 been reported with inductive type ignition only but as far as
 I know nobody is using a CD system in a flying aircraft.  CD
 systems have a proven track record of making plugs last longer
 in piston engines.
  No telling what the CD spark plug life would be in a rotary
 until we get some good data.  The MSD ignition apparently only
 fires more than once up to some RPM which I know not.  I have
 talked to their engineers about it and they don't seem to
 understand the rotary so they cannot give me a good answer.
 One MSD system per plug would give a lot of redundancy.
 We don't have any data running on trailing plugs alone for a
period of time so it is hard to say what will happen.  Ignition
failures of any kind are rather rare with the possible exception
f 2nd gen igniter ballast resistors which can be traced to
continuous high RPM use.  The wattage rating of these resistors
just be increased.  Tracy also had problems with early Corvette
oils over heating but I suspect that was caused by excessive
well and the fact they were being triggered twice as often.
ne unknown is the charging coil in an outboard motor may not be
ble to charge more than one capacitor at a time so the out board
type ignition system may require two magnets, two capacitors and
two charging coils per rotor for firing both leading and trailing
oils.  Some experiments need to be made to check and see if this
s true.  You will also need outboard type CD coils as RX8 or
Corvette coils per plug will not work with CD systems.  The
thing about 2 cycle outboard coils is they are designed to run
t high RPM and CD systems are actually easier on coils.  They
become mere pulse transformers and not a device to store energy.
n that case they should be far more reliable

  You will also need outboard type CD coils as RX8 or Corvette
 coils per plug  will not work with CD systems.
  The thing about 2 cycle outboard coils is they are designed
 to run at high RPM and CD systems are actually easier on coils.
 They become mere pulse transformers and not a device to
 store energy. In that case they should be far more reliable
  Paul Lamar
  Hello Paul,
  Attached please find the wiring diagram for a 1977 V4 2 cycle 85
 HP outboard.  I do believe the stator and circuitry is more
 complex than the application we are looking for due in part that
 it is working 4 cylinders instead of two.  If perhaps someone
 had a wiring diagram of a 2 cylinder 2cycle outboard it would be
 even better.
  With regards to the throat diameter of the carbs I was
 considering, I was going to use Tracy's estimate of 1 square
 inch per 50 horsepower.  Is that in the ball park or should I go
 larger?  I have not thought about a P-Port yet as I'm still
 learning and the thought of taking a grinder to the engine is,
 at this point, beyond my ability.

  Regards,
 FranklinHello Paul,
  Attached please find the wiring diagram for a 1977 V4 2 cycle 85
 HP outboard.  I do believe the stator and circuitry is more
 complex than the application we are looking for due in part that
 it is working 4 cylinders instead of two.  If perhaps someone
 had a wiring diagram of a 2 cylinder 2cycle outboard it would be
 even better.
  With regards to the throat diameter of the carbs I was
 considering, I was going to use Tracy's estimate of 1 square
 inch per 50 horsepower.  Is that in the ball park or should I go
 larger?  I have not thought about a P-Port yet as I'm still
 learning and the thought of taking a grinder to the engine is,
 at this point, beyond my ability.

 Regards,
 Franklin

 Nothing was attached Franklin. Try again.
  1 square inch is on the small side for 50 HP. 1 square inch
 is more like 45 HP
  Jeff can do the p-port thing for you.
 jeff.doddridge@gmail.com
  A lot easier than making a manifold however you must
 be willing to disassemble and reassemble your engine.
 Mazdatrix will also build a P-port engine for you with
 Jeff's modified p-port rotor housings.
  Since you are using carbs you won't need the slide throttle.
  Read this. http://www.rotaryeng.net/intake.html
  If you still want to use a side port a lot depends on how
 much room you have under the cowl when building manifolds.
 What kind of airplane are you building?
  Is your engine a 4 port or six port?
 Are you good at tig welding aluminum? Steel welding?
 Steel tubing is the easiest to use as the welding skills
 required are minimal.
  Here is a trick for the transition of round tubing
 to oval intake ports. Don't try to flatten a round
 tube. The area will be much reduced which will seriously
 affect the HP you get. We have been all over this in years
 past. A lot of people have been disappointed in the results
 when building side port manifolds.
  Paul Lamar
  Let me try the attachment again.

 Regards,
 Franklin

 There seems to be enough parts for a two rotor but they
 must be rearranged. I am not sure as there is no picture of
 the flywheel showing the position of the charging coils
 and firing coils. Do you have  a digital camera or a cellphone
 with camera?

 BTW I just found a source for magnets. This is a major
 breakthrough for us.
  http://www.maxrules.com/fixomccdi11.html#MAGNET
  I can now go wild designing a magnet holder which is the key
 to the whole works.
  Paul Lamar
  Attached are pictures of the flywheel from the Johnson.  As you
 can see there are sets of magnets embedded into the flywheel and
 then machined smooth.  There are 6 sets of magnets or 12 magnets
 all together.  I'm not certain but I do believe that the sets
 are opposite in polarity.  I'm fairly certain I read that
 somewhere.
  I'm planning on building an aerocanard and I'm trying to keep it
 as simple as possible.  Don't believe I will be able to do the
 gravity feed fuel system, so I'll have to go redundant on the
 low pressure fuel pumps.
  Franklin Canosa
  Lots of magnets in there. I think I have found a picture of
 what goes with it which I included. Do those look familiar?
 The two light colored coils are the charging coils for the
 CD ignition. The other two are the generator coils.
  The big challenge now is making a light weight holder for all
 those magnets that will fit the e-shaft. Do you have a friend
 with a lathe? You will get an alternator out of the deal.
  Paul Lamar

 Paul,
  Yes those are part of the CD system.  The blue ring device is the
 'spark advance mechanism' that we can probalby do without as you
 suggested we just time the engine to 22 BTDC.  The stator is
 correct and you are correct (I believe) about the different coil
 purposes.
  I don't have direct access to a lathe, but would be willing to
 send down the parts from my old outboard for experimentation
 purposes.  I still have the stator, flywheel, a couple of good
 coils and perhaps even the power pack.  I was going to try and
 sell them on ebay, but I would probably get 10 bucks for the lot,
 and this is a far more fun and useful application of the parts.
  Let me know your thoughts.
  Franklin

 I would like to take a close look at that flywheel.
 Most of the other parts are interchangeably or can be bought
 used or new.
  I would like to put it in a lathe can cut it down
 to drastically cut the weight and then make an aluminum
 hub for it to fit the flange on the e-shaft. The flywheel
 aspects of it are not needed and in fact it is not good
 to add rotational inertia to the front end of the e-shaft.
  This. I think, is the key to making this system practical.
  With the flywheel I have all I need do is drill out
 the rivets to remove the magnets. I can then mount
 them in an aluminum hub.
  Paul Lamar

 Paul it may be simpler than that. Machine the outboard flywheel
 to fit under
 the damper, machine it lite and use as is. a stupid suggestion by
 milo!
  I'll investigate that option shortly. I have one of Tracy's
 dampers sitting right here.
  -------------------------------
 Here is what I have so far. I just had an idea this morning
 on how to make this a bit cheaper and better.
  If I chuck the front hub with
 the bolted on pulley in the lathe I can take an small internal
 cut out out of the steel pulley to insure it is concentric.
 Then press in something that looks like this.
 The groove will be filled with uni carbon to cut
 the rotational inertia and increase the burst strength.
 Important at 7500 RPM.
  Paul Lamar

Paul
 "More than one way to skin a cat."  BTW, historically that phrase
was about cat fish not the long felines.
 Generators or magnetos coils work in the radial plane like most
alternators and the featured motor boat stators as you have
posted.   Another way to do it is counter sink circular magnets
into an aluminum starter gear or perhaps Tracy's Al. flex plate
and place the coils pointing in the axial axis.   Coils and
magnets can be mounted on either side (or both) the flex plate/
starter gear in a circular pattern.
  However just slapping in a outboard motor stator and coil
components  is no guarantee it will work as the operating design
rpms are quite different and the speed range on the rotary quite
wide.  Output voltage and efficiency varies with rpm so voltages
will be all place.  You may need a DC- DC converter circuit board
to smooth out the voltage.

Doug in Japan

Good idea Doug. Circular or cylindrical magnets are much easier
to buy that is for sure and much easier to implement. Maybe no
cheaper. We are looking at $100 for just four. The cheapest is to
find an old outboard motor and scavenge the magnets out of the
flywheel.
 I'll do a dwg using the round magnets.
 Voltage variation on the CD ignition system will not be much of a
problem. Outboard motors have to have spark at starting
RPM's too.
 You may not get 12 volts at start up but 6000 RPM at cruise
will be no problem.
 -------------------------------------------------
 To make a magnet carrier out of this will require a new plate
and cutting down the Ford spline flange. The number of
rivets Tracy used is way over kill. My RX8 clutch disk
has only four. The rivet circle diameter will have to be reduced
to make room for the magnets. Round or other wise.
 Paul Lamar
 --
Paul,
I appreciate the circuit diagram.  Thanks for posting it. Instead
of another triggering coil weren't you considering using a hall
sensor circuit?
I personally wouldn't mess with trying to make an alternator and
just concentrate on two [ignition-charging] coils.  Alternators
are cheap and well tested with high regulated amps in a small
package.

Doug in Japan
I tend to agree with you Doug on that. In fact one might
spend less money just using an alternator.
I just weighed Tracy's damper plate and was shocked to find it
weighs 5 pounds. Here are some places to machine to save a pound.
I also placed 4 round 1.5 inch  magnets but I don't think it is
practical to install the charging coils here without a major
redesign of Tracy damper plate. Just not enough space
between the damper plate and the PSRU adapter plate for both
agnet and charging coil. Perhaps with large 2 inch dia. holes in
Tracy's PSRU adapter plate that the coils could poke through.
Paul Lamar

Paul,
While I like the magnet research, why not use the flywheel from
the outboard?  If you use the stator and flywheel you get the
entire system.  I mentioned this a couple of years ago when
researching the Mercury system with Orbital direct injection
my 20B.  The orbital system requires an air pump to assist the
injection, but as a bonus you get an oil injection pump that you
an use to feed 2-cycle oil to the apex seals.
I turned back
because I was laid off and couldn't afford to buy the stuff at
the time but it appears very workable.  Even the typical RPM
atched up well as 6200 is normally max but the racers turn much
higher than that.  The 6 cylinder or 3 cylinder 2 cycles would
work for me.  The 4 cylinder for the 2 rotor.  You might need to
machine the flywheel for a new center to work on the rotary but
that wouldn't be real hard, simple turning.
Bill Jepson

Too heavy at 6 pounds for what it does. We also do not
need more rotational inertia on the front end of the e-shaft.
Paul Lamar


Paul,

Ok Paul 6 pounds, but don't forget that that includes an
integral alternator. BTW my e-mail is messing up the posts
sorry for any foul up. The motorcycle stator systems from old
2 cycles would also work. The outboard systems are designed for
continuous high RPM though, and are generally tougher.
Bill Jepson

What email program are you using Bill? Any idea why it is
doing that. Rather strange. I have to go through everything
and re edit. It looks like it is cutting off the first
character in every line that has no space character in
front of it. Do you have it set to plain text?
Some times html screws things up big time and it is
not needed. A word is a word regardless of the color
or font.

Paul Lamar

Paul,

I understand about the weight of the flywheel housing on the end of the e-shaft where the PSRU is located so what about putting it on the other end (i.e. where the pully's are)?  I know space may be an issue on some designs, but there has got to be a way to do this.
I'm also going to try and make a test rig using a high speed electric motor.  Unfortunately the sensor coil / spark advance mechanism on my Johnson outboard is totally shot, so I will have to try and pick one up for testing purposes.

Franklin

The pulley end is the right place for it IMHO.

If I chuck the front hub with
the bolted on pulley in the lathe I can take an small internal
cut out out of the steel pulley to insure it is concentric.
Then press in something that looks like this.
The groove will be filled with uni carbon to cut
the rotational inertia and increase the burst strength.
Important at 7500 RPM.

Paul Lamar