From: Marc Thomas [mmthomas@ix.netcom.com]
Sent: Friday, December 29, 2000 12:41 PM
To: 928
Subject: [928] Exhaust systems and more...

Tom, Mark, et al

I will be as brief and as factual as possible.

1   all venturis are bad from a pure mass flow standpoint and whatever is
going through is going to lose energy many different ways.

2   Merge collectors are not venturis, and are nothing more than gentle
"anti reversion" cones  used to keep the pulses going in the same direction
as they hit a different size tube and produce a standing wave that travels
backwards and can inhibit flow.  Different tapers and lengths are used to
tune this effect.minimize this effect.

3   Anti-reversion is best accomplished at the head, but can also be
effective at the collector

4   The merge collector acts as a "gate" for each pulse, and you will notice
that any merge collector is ALWAYS larger than any single primary tube....

5   Cam overlap and timing has as much to do with gains from exhausts as the
exhaust itself.  the higher the overlap, the greater the scavenging and the
greater the gain.  Scavenging comes mainly from the primary, and can be
assisted by the exhaust.  It is possible to have too much scavenging on a
fuel injected engine as the FI pulse is fixed, unlike a carb!  This will
lean out the engine.  Injector timing is critical in highly pulsed tuned
engines.

6   Head flow at a given lift has a lot to do with the effectiveness of an
exhaust and intake system.

7   Dragsters use open headers as they produce the most power at a given
rpm.. accept it.  Speed boats are the same.

8   NASCAR teams  have different exhausts for different tracks depending on
where they need bottom end hp/tq,or more top end/high rpm power.

9   Visit DRGAS.com as they are the one who developed the "X" style
crossover for NASCAR

10  The BEST header system is one that produces good power gains at all
rpms, is cost effective, durable, easy to install, allows servicing the rear
main seal or clutch easily, does not add additional heat to the drive line
components or fluid lines. does not need additional eng management control
to get 85% of the potential, etc
   
11  DEVEK headers are built to satisfy #10, they are not equal length,
although close, however, they are very close to equal flow.  They are tested
on S4's with normal outputs (in good mechanical condition) on a DynoJET.
See the DEVEK Catalog for all dyno graphs.  Same car, in some cases a day or
two apart to minimize any changes.

12  DynoJets are the single most reliable method for testing performance
products universally since the drums are very close to the same weight on
every dynojet built and as Mark K points out, it is only math thereafter
needed to calc power.

13  The DynoJet is the accepted dyno of choice for the 928 community as they
are readily available everywhere!


Regards,
Marc
DEVEK