Mark Steitle wrote:

Paul,
This may be a question for Tracy, but I was wondering if a PIC could be
used to monitor for detonation and then retard the timing if/when
detonation occurs?  Obviously, an aural warning would be possible, but I
would like a system that would give warning and retard the timing before
permanent damage to the engine could occur. A "RETARD DISABLE" switch could
be included for those times when you absolutely don't want to turn over
control to a microchip.
Mark Steitle

Well I'll be interested to see what Tracy says about it
as only he can incorporate this into his hardware and software.

I guess one could build a box and feed the crank angle
sensor data into it and delay the pulse before being sent
to Tracy's box. Common in after market knock sensors for
the rotary. There is some evidence out there to the effect that 
this is not easy to do. The knock sensor itself is tuned but 
some sort of analog or digital signal processing might be required
to separate the knock signal from the engine back ground
noise. There certainly is a need for it. 

A real big problem is developing the signal processing software
with out destroying several engines in the process. Looks like a job
for Mazda engineering. I tested my pure audio
version by hitting the junk rotor housing with a hammer.

Perhaps what one could do, and this is a BIG sledge hammer approach,
is to run the stock third gen RX7 computer in parallel with
Tracy's computer and monitor the pulses going to the coils
for evidence of retardation when knock occurs. Just an idea.

Paul Lamar
 
The AirCraft Rotary Engine NewsLetter.              Powered by Linux.
ACRE NL web site.
http://home.earthlink.net/~rotaryeng/            http://www.linux.org