Variable Restrictiveness Exhaust - A concept

[Robert Harris]

A simplified approach - to get the idea into play.

Let us look at exhaust gas flow.  Consider the normal mass of exhaust gas as
thin molten lava.  When the valve opens, it behaves as a high temperature
viscous liquid and flows out of the cylinder, down the headers and out the
tail pipe.  This is essentially a liquid flow subject to all liquid flow
rules.  As long as the cylinder pressure is significantly greater than
atmospheric, this flow is essentially unaffected by low order acoustic
variations.

As the piston approaches TDC on the exhaust stroke, hopefully most of the gas
is evacuated and the pressure is falling.  It is in this region thru intake
valve opening "overlap" and the closing of the exhaust valve that resonant
tuning can help.  The "help" can be significant.  If the intake "see" the
chamber as a negative pressure, the initial air flow into the cylinder will be
greatly accelerated.  Ideally, the pressure wave will cause enough depression
to totally clean out the chamber "scavenged" but not excessive draw intake
charge into the exhaust.   This effect is strictly dependent upon the exhaust
pressure vs intake pressure.  If the exhaust pressure is positive to the
intake, exhaust will flow up the intake - no exception to the rule - path of
least resistance, pressure differential and all that rot you know.

The pressure wave arrival is a function of the acoustic characteristics of the
exhaust.  Its frequency a function of the tuned lengths and volumes, its
amplitude a function of the shapes "Q" for the RF types.  It is possible to
achieve significant negative or positive values even with an above atmospheric
average pressure.   Remember this pressure wave is only significant when the
exhaust gas is below critical pressure and until the valve closes.

Currently we cut into metal something that experience or calculation show us
will function in a specific rpm region.  We then reduce the Q of the system to
broaden the effective range - trading some reduction in peak benefits to
provide some of them over a more usable range.  We extend the overlap to widen
the range where there will be significant negative pressure conveyed to the
intake charge to cause it to accelerate.  

What if it were possible to "tune" the negative going peak to optimally arrive
thru a relatively broad RPM region?   We could then sharpen the response -
raise the Q - and get even more effect.  

Or, during cruise, we could tune for a positive pressure in the chamber.
Since we know the majority of the exhaust flow is unaffected by the acoustical
tuning effects and simply follows the kinetic rules, tuning the exhaust
pressure wave to be positive will have little overall flow effects - unlike
placing a restriction would have.  

The positive wave would not only stop the scavenging effect ( depending on
phase and amplitude ) but could be beneficially used for massive EGR at
cruise.  ( Another thread - simply put - there are strong arguments and
evidence that massive EGR can nicely increase economy at light to moderate
loads ).  Without the accelerating effects of overlap, the throttle would have
to be significantly more open to flow the air needed - thus reducing the
throttling losses.

In summary, being able to detect and tune the phase of the exhaust acoustical
pressure wave can result in significant performance and economy enhancement.