Mazda Rotary Muffler Experiment
During a discussion on one of the on line rotary
newsletter, one of the guys mentioned that he'd found a potential source
for mufflers for rotary engines. The design looks very promising and
we are now flying it on the RVotter. If you want to see details on
the design go to www.spintechmufflers.com.
The only downside is the material was not adequate for the rotary's
exhaust temperatures so we're having a prototype made in 321 stainless.
Following are some progress reports that I posted
to the "Fly Rotary" on-line newsletter. The
Spintech muffler test was funded and sponsored by members of this on-line
group. BTW, If you’d like to subscribe (it’s free), send an email
message to: Flyrotary-request@olsusa.com
December 12, 2001:
Received the Spintech test
muffler yesterday and started fabricating mounting hardware.
I'm hyper critical of weight and it felt heavy.
Put it on the scales & measured
13.5 pounds which is 6 pounds heavier than current muffler.
Looks very well made.
The flat top & bottom of the muffler caused some people to
think it would probably fail due to flexing of wall due to sonic &
pressure waves but I think this may not be a problem due to the internal
construction. There are many
baffles welded between the walls which should eliminate this.
Have not finished fabricating all the mounting
hardware so no flight tests yet but did one brief ground test.
Very interesting. Due
to the design, it is not very effective at idle to low rpm, so it was not
much quieter than my current muffler.
My testing assistant (Laura) and I had very different reactions to
the nature of the sound. She
thought it sounded nasty but I thought it was nice.
The only instrumented test I had time for today
was a ground run at near minimum cruise power setting.
The distance from plane to sound level meter was about 30 feet.
The sound level was measured continuously while walking around the
plane and noting the highest and lowest readings seen.
Ideally, the test should be conducted well away from any buildings
to eliminate reflections but time did not allow this today.
We were right next to my hangar. The maximum reading while walking
around the plane on a test like this is greatly affected by sound
reflections from surrounding buildings.
At low cruise power setting, the minimum noise
level on the old muffler was 103 db and 98 db on the Spintech, a 5 db
improvement. This is a
significant difference because the db scale is logarithmic.
(this is how the human ear responds to sound)
A 3 db increase in sound level requires the power from the sound
source to be doubled! The
peak reading was about 3 db less on the Spintech.
Another interesting point was that the maximum
sound reading on the Spintec muffler was the same at a high idle speed or
low cruise power setting. In
other words, it is more effective at higher throttle settings.
Another point to remember about these measurements
is that a significant amount of the noise generated by an airplane is
propeller noise. Since the
prop noise is not affected by the muffler, that means the exhaust noise
reduction was even greater than the measurements might indicate.
No way to evaluate how big a factor this was.
Bottom line for this test is that it sounded
substantially quieter to me while sitting in the cockpit so I can't wait
to see what it's like in flight. Stay
tuned for more.
December 20 , 2001:
Made one short flight test yesterday.
No one around to make external sound measurements but did take some
cockpit level readings.
The improvement in cockpit level exhaust noise was
immediately noticeable. At
economy cruise power, noise was reduced almost exactly the same amount as
indicated by previous ground tests; 105
db vs. 110 db on old muffler. These
absolute readings do not really indicate the relative difference in
exhaust noise. Taking off the
noise canceling headphones makes it immediately obvious what the noise
sources are. From highest to
lowest, the noise sources are:
1.
Wind noise coming through canopy and a couple of gaps between
canopy frame and fuselage. This
is the biggest noise source but headsets are very effective at blocking
this type of noise. It is not
noticeable with headsets on.
2. Structure
borne noise transmitted through airframe.
There is very little sound damping material in my plane and no
upholstery. I also found that
my quick & dirty muffler test installation caused the exhaust
collector to rub on the lower edge of the firewall which greatly increased
this noise. This will be fixed
before next test.
3. Exhaust
noise. If this were eliminated
entirely, the cockpit noise level would still be above 100 db.
It will be interesting to see the external noise
test results since these will not be affected much by the first two noise
sources. Unfortunately, they
are greatly affected by prop noise.
The biggest surprise was the effect on power.
I was expecting a slight loss of power but instead got a slight
increase. Very good news
indeed. Even more
surprising was the effect on drag which I assumed would be about the same
as the Edelbrock muffler I tried. The
Edelbrock was slightly narrower than the Spintech but it was thicker and
hung down further below the fuselage.
That caused about a 5 MPH drop in cruise speed.
The spintech seemed to result in almost no loss in cruise speed.
I think this is a result of the Spintech being thinner and staying
mostly inside the turbulent boundary layer near the fuselage.
The broad, flat surface of the muffler in close proximity to the
fuselage does radiate a lot of heat to the skin and It will be necessary
to install a heat shield on the fuse.
Something like a thin layer (1/16") of fiberfax (Sp?) covered
with self adhesive stainless steel to reflect the radiated heat.
Verdict so far is 'Two thumbs up'.
Now if it will only survive rotary exhaust for a long time.
December 28, 2001 : Received the following question
from a builder:
What is the reason for not keeping it (the muffler) under the cowling?
Mostly personal preference.
In my installation is was not practical to fit an efficient exhaust
header and collector along with the muffler under the cowl.
In addition, removing the muffler from the cowl improved cooling a
lot. This was due to
elimination of radiated heat and the improved airflow through the less
crowded cowl. It is also
MUCH easier to do exhaust & muffler experiments with this arrangement.
Safety is another factor.
If the failed Edelbrock muffler
had been under the cowl it could have been a disaster.
Someone also asked about the header &
collector sizes. Headers are
about 26" long and of 1.75" diameter .035 wall 321 stainless.
Collector is 2.25" (same material).
January 12, 2002:
Every objective measurement taken so far indicates that the
Spintech is significantly quieter than my homemade muffler, by about 5 db.
This holds true whether the reading is taken from inside or outside
the plane. I have attached two
.wav files to give an idea of what the difference is.
Recording1.wav is
in-cockpit sound at cruise speed with old muffler.
Recording4.wav is the same condition with the Spintech.
Keep in mind that the recorder has an AGC (automatic gain control)
which keeps the over-all level of the digital recording about the same.
In other words, if the sound source is low level, the recorder will
turn up the gain or turn it
down if the source is loud. To
judge the relative difference in the two recordings, compare the wind
noise level with the engine noise. The
engine noise is almost drowned out by the wind noise in the Spintech
sample.
Click on buttons to hear sound
samples.
Real World Solutions, Inc
