<snip>
First a collector system won't work on a three rotor.
Tuned pipes will help of course.

The exhaust pulse travels down the pipe at the speed of sound.
As the RPM changes the distance between the pulse changes.
Higher RPM shorter distance. Optimum length occurs as one
pulse is leaving the collector and another is being emitted by
the engine.

The speed of sound is about 880 feet per second.
Each rotor fires once per rev. At 6000 RPM or 100 RPS
pulses are every 1/100 of a second apart or .01 second.
It takes the pulse about .003 seconds to travel 32 inches.
That is an exact multiple of .01 second. So the pressure
wave is bouncing back and fourth in the pipe three times
and the low pressure part of the wave arrives back at
the exhaust port just as the port opens again. This
also sync's in the collector enhancing the low pressure
part of the wave in the other pipe. Multiples of 32 inches
also work such as 16 inches at 6000 RPM.

Paul Lamar

Paul,
Attached is a quote from an old message dealing with optimum exhaust runner
length.  I am beginning my 3-rotor exhaust system and would appreciate a
clarification on a couple of points.  I will be using Tracy's 2.85:1 psru.

1)  A collector won't help any on a 3-rotor.  If this works on other
multi-cylinder engines, then why not on a 3-rotor wankel?

Long story. The pulses are 120 degrees apart and not 180 degrees as in many
multi cylinder engines.

2)  If I understand your sound wave analysis correctly, then based on
distance the sound wave will travel in .001 seconds would result in a
runner length of 10-1/2".  From what I've read, this would be the length of
the initial 2" dia. runner tube, before it dumps into some larger sized
pipe, 3" in my case?

3)  How much difference will the intake runner size make in determining the
optimum exhaust runner length?  The graph you refer to in the past are for
a 170mm intake runner length.

My head is beginning to throb!

Thanks,
Mark S.

As it should. Most of the exhaust data from Mazda was for a peripheral port
engines. There is a strong interaction on PP engines. For a side port engine you 
can probably treat them as separate systems.

Paul Lamar
 
The AirCraft Rotary Engine NewsLetter.  Powered by Linux.
ACRE NL web site. http://home.earthlink.net/~rotaryeng/            
Copyright 1998-2003 All world wide rights reserved.