I’ll start by
saying I don’t disagree with anything that Lynn says here (message
below), but I’m going to look at it a bit differently. It’s a
matter of objective and what you get for your trouble; and its importance to
flying (5000-6500 rpm) vs racing.
I studied a lot about tuning intake and
exhaust before doing my installation, and ultimately let my design be dictated
strictly by the needs of the ‘packaging’ configuration for minimum
weight and drag – and some quieting of the exhaust. The result is
shown in the attached photo. (See more at http://members.cox.net/alg3/ROTARY%20ENGINE.htm
) I have a clean short intake manifold; an inconel tangential manifold/muffler
with short headers going into the can; and a small muffler of my own design
downstream. I did pay attention to port-matching and smooth bends on the
intake for good flow.
The actual measured dynometer results give
me 280 hp (93.3 hp/ rotor) at 6500 rpm with a normally aspirated engine –
and 9.0 to 1 compression rotors; a nice flat torque curve, and a reasonably
quiet exhaust. It allows for a nice clean, compact cowl, and clean
airflow into the prop (Velocity pusher). I don’t doubt that I could
get a few more horses with optimized induction and exhaust, but from what measured
results I’ve seen, I don’t think very much. We ran with and
without the secondary muffler and found no detectable change in power, and 6 –
8 db noise reduction. The easiest gain in power would have been for Atkins to
put in 9.7 to 1 rotors – but that’s another story.
Keep in mind also that wrapping the
exhaust (or ceramic coating on the outside which would make little additional difference)
to reduce heating inside the cowl, makes the pipes run that much hotter which can
reduce their life (stress corrosion and thermal stress). Coating the
inside is a good idea because it reduces the wall temp and keeps a bit more
heat in the gases, although analysis suggests the .007 to .010 coatings would
have less of an effect than the purveyors of the coatings advertise. The
big gain would be from reduction of surface emissivity, but it’s
difficult to say what effect remains after the inner surfaces get coated with
exhaust products.
I chose to leave the surfaces bare, and
added a small scoop in the bottom of the cowl to direct some airflow around the
exhaust to keep the temps down. Also note that the expansion of the
exhaust into the tangential muffler results in some instant cooling (Charle’s
law of expanding gasses).
Having said all that relating to performance,
I don’t know yet about the reliability of the exhaust design. I
only have about 15 hours on the engine, and probably less than an hour of that
is at full power. I do have concerns about that and will keep a very
close watch on the exhaust welds.
FWIW,
Al
-----Original Message-----
From: Rotary motors in aircraft
[mailto:flyrotary@lancaironline.net] On
Behalf Of Lehanover@aol.com
Sent: Friday, August
25, 2006 5:43 PM
To: Rotary motors in aircraft
Subject: [FlyRotary] Exhaust
systems
Check for posts by Lynn
Hanover – he’s called out the exhaust lengths and diameters a bunch
of times (I can’t remember exactly otherwise I’d parrot it here for
you).
Although
the rotary is a 4 Otto cycle engine it tunes like a 2 Otto cycle engine. And it
is very sensitive to inlet runner lengths and diameters and exhaust lengths and
diameters. So it tunes like a big dirt bike engine. You cannot make a huge gain
in HP by doing it right, but you can leave a lot on the table by doing it wrong.
Except for the RX-8 engine the intake and exhaust are connected during
overlap at BDC and the exhaust can interfere with the intake to poor result.
The
number one factor in exhaust design is that both primary pipes join at the same
length and enter the collector at the same angle. Look at the Racing Beat
Catalog pictures of the street headers. Note the extra bends in the rear tube
to result in the same centerline length as the front tube. Any design that does
not collect the primary pipes is a non-starter, HP wise.
Primary
tube diameters are just a bit larger than the exhaust liner. In most cases 1
7/8" to 2" ID is what you end up with. Measure the liner ID at the
face. I gave up on the exhaust gaskets years ago, and Just lay on a bead of
100% GE silicone around each port about 5/8" outboard of the hole, and
bolt on the manifold before it sets up. Never had a leak.
Primary
lengths in multiples of 10"-12" looking for 32" to 36"
to the collector if you have the room and can stand the complexity.
Ceramic coated inside and out would be good. Heat wrap makes an
astounding difference. You can touch the exhaust while the engine is screaming.
The race car is not drivable without heat wrap on the muffler. The interior
sheet metal turns from white to brown with paint blisters. We run heat tape and
a shield over a 4" ID stainless packed Borla.
After
the collector ending at 2 1/2" then up to 3" for the run to the
muffler if any. Only round will last very long. I thought the Spintech looked
promising with all of the vanes supporting both sides. If the design allows for
a large volume to let the pulses drop to subsonic, velocity is traded for
pressure and generally the muffler is hammered to pieces in short order. After
the collector there is little affect in differing pipe lengths. In the RS
series and in the Racing Beat catalogue there are recommendations for lengths
that include having the collector right at the
muffler
102" from the flange. Good for overall RPMs, but not particularly good at
any specific RPM. Even with the Mazda racing muffler at the very end of a long
system under the rear bumper, it was a good idea to wire up both ends of the
muffler so when it comes apart you don't drop pieces on the track.
If you
have minimal back pressure, any leaks will result in air being drawn into the
system with much popping and banging on spool-down.
The
system has to be stainless. You must shield either the pipe (Easy and light) or
the equipment you don't want melted, (harder and heavier), or it will overheat.
If the exhaust system can see it (line of sight) it will melt.
If the
engine can move and part of the exhaust system is tied down tight, it will
fracture a tube and set a fire going. Add or fabricate a flex joint. Allow for
a length change with temperature changes.