From: Epijk@aol.com
Message-ID: <38.24184f1b.29b3d42e@aol.com>
Date: Sun, 3 Mar 2002 14:31:58 EST
Subject: Re: DQ update
To: lancair.list@olsusa.com
MIME-Version: 1.0
Content-Type: text/plain; charset="US-ASCII"
Content-Transfer-Encoding: 7bit
Reply-To: lancair.list@olsusa.com

          <<<<<<<<<<<<<<<<--->>>>>>>>>>>>>>>>
          <<  Lancair Builders' Mail List  >>
          <<<<<<<<<<<<<<<<--->>>>>>>>>>>>>>>> 
>>
In a message dated 3/3/2002 3:18:08 AM Pacific Standard Time, 
ckohler1@cfl.rr.com writes:

<<<...With fixed magneto timing at 24 degrees BTC we have a
potential problem. If the engine has been primed, and the impulse coupling
has engaged the stop pin, and wound up the spring, it will fire nearly
instantaneously with the start  switch. At 24 degrees BTC....>>>

Charlie:

You have hit upon ONE of the potentially severe problems caused by low 
mass-moment-of-inertia propellers. 

Regarding your analysis of the timing issue, I'd like to clarify your 
statement (above). The action of the impulse coupling accomplishes two 
separate functions. Firstly, it stores energy (relatively slowly) in the 
windup spring during a few degrees of crank rotation, then, when the 
flyweight contacts the pin, it suddenly releases that energy into the mag, 
producing enough magnet velocity (insufficient at cranking speed) to generate 
sufficient primary current to induce sparking voltage in the secondary. The 
second function is to retard the spark timing from the fixed setting (say 24 
BTC) by roughly 10 to 15 degrees (depending on the specific mag). With a 
high-moment prop (lots of "flywheel effect"), firing at 10 BTC doesn't 
generally produce a kickback. But with a low-moment prop (or no prop at all,. 
see below), the kickback (as you discovered) can be devastating.

We discovered this, to our chagrin, when we attempted to fire up our first 
high-performance Lycoming on the Superflow dynomometer (pictures of that 
setup will be added to our website soon). The Superflow dyno uses a water 
brake absorber which has essentially zero moment of inertia, and of course, 
there was no prop. The engine had big compression, trick cam, proprietary 
intake and exhaust systems, modified fuel injection system, and many other 
internal improvements. For initial testing, timing was set at 22 BTC.  

Ready to start: impulse mag ON, fixed mag OFF, normal priming sequence, 
throttle cracked, starter ENERGIZED; the engine cranked two cylinders, and on 
the third, produced a LOUD noise, then quiet. Examination revealed that the 
kickback had blown the entire nose off the starter motor. (I should have 
realized that might occur BEFORE it happened, but that's how test programs 
go.)

To solve the problem I modified the impulse coupling on the mag to produce 
not 15 degrees lag, but 26 degrees of lag, guaranteeing that the starting 
spark would not occur until after TDC. 

Problem solved, simply and directly. Subsequently, we included that mod on 
ALL high-performance Lycomings (experimental, of course).

The scenario you hypothesize in which the impulse coupling is just ready to 
fire when you begin cranking, and lights off that first cylinder, seems 
unlikely to occur. If the impulse is ready to fire on any given cylinder, 
then the engine must (by definition) be positioned very near TDC-compression 
on that cylinder. If that is the case, then (a) the intake valve on the 
subject cylinder is (and has been) closed, so no fuel-air charge would have 
been inducted, and (b) there is no compression pressure (generally necessary 
for combustion) in the subject cylinder.

The solution you suggest may improve your odds, but the bottom line is, you 
still have a low-moment prop, and your ignition event will still occur at 9 
to 14 BTC (depending on your mag), same as before. By bringing the engine up 
to full cranking speed before energizing the mag, you are taking advantage of 
al the "flywheel effect" available, but the fundamental problem still exists. 
(BTW, it just occurred to me: did you verify that the non-impulse mag is 
REALLY DEAD when you turn the key into the START position??)

Also, I'm trying to visualize how many hands it will require to implement 
"Start-Ignition-Prime". Something like "Prime-Start-Ignition-after-3-blades" 
might be easier and accomplish the same thing (no fire without spark) while 
requiring fewer gymnastics. Works pretty good on the big radials I have 
flown. 

Again, congrats on no damage in the aftermath of that failure.

Jack Kane
EPI, Inc.

>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>
LML website:   http://members.olsusa.com/mkaye/maillist.html
LML Builders' Bookstore:   http://www.buildersbooks.com/lancair
Please remember that purchases from the Builders' Bookstore
assist with the management of the LML.

Please send your photos and drawings to marvkaye@olsusa.com.
>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>