Exhaust within subcowl

[pictsidhe]

I was actually being sarcastic...
My truck was godawful loud when I got it, thanks to a muffler 'upgrade'. Wasn't too long before I put a proper muffler on. As a bonus I got an instant power boost. I could tell as the 3-4 clutch had never flared before...
I've seen the same thing with bikes on an actual dyno. The aftermarket tuning trick there is to remove a bunch of torque from the midrange so that the top end power, which is actually much the same, feels like it is hitting much harder. The reality is the bike is not only slower but harder to ride smoothly. A little fiddling of 'correction factors' and a dyno curve then boasts superior power.
The trick is to only use a dyno outfit to measure changes that you paid some one else to do...

 

[MolsonB]

So I haven't even be able to try any of this out yet. Spent the last year getting the airworthiness. Have 5 hours of flight and the exhaust it just to plain loud. Following https://www.homebuiltairplanes.com/forums/threads/quiet-flight.6560/page-3#post-69866 has lead me to change my exhaust around at the engine.

Everything is 321ss (Even the flex bellow).
Primary tubes are 2" double slip joints.
Outer can (CA - Centrifugal Accumulator) is 5" (0.05")
Inner pipe / down pipe is 2.75" (0.035")

I drilled the inside pipe with a total drill area of just under double the area of the 2.75" pipe. Hopefully in a few days will get it all welded up.

Once the exhaust is manageable, I can play around with different options under the plane.

 

[jonlaury]

FYI, there are dB meter apps for cellphones, so you don't have to go deaf researching this
Like your SOP engineering.
John

 

[MolsonB]

FYI, there are dB meter apps for cellphones, so you don't have to go deaf researching this
Like your SOP engineering.
John

I got one off Amazon for 15 bucks. The cell phone mic's really aren't worth a poop for dBA measuring.

The previous exhaust with just an Aero Turbine 2525XL, measured ~113dBA. With it removed, was closer to 120dBA. Funny around 2000 prop rpm (6340 engine rpm) was louder then at 2300 (7291 engine rpm) full static.

 

[rv7charlie]

Agree that cell phone mics are of little value when doing spl measurements, especially at higher levels.

But. Take it from an old live sound engineer; don't depend on the dBA scale to protect your hearing. dBZ is what you want, but it's hard to find on inexpensive SPL meters. At least, find one with a dBC scale. Take a look here:
https://www.noisemeters.com/help/faq/frequency-weighting/
and
https://www.tinnitustalk.com/threads/dba-vs-dbc-vs-dbz.28971/
The A scale (curve) ignores a significant amount of energy below 1 kHz. That curve means that a 250 Hz tone would register on the meter as 10dB lower than the actual SPL. That means the energy is *10 times* the energy shown on the meter. At around 90 Hz, the indicated level would be -20 dB, or *100 times lower* than the actual energy.

Now, a lot of the rotary's most irritating energy is up in the midrange and above, where the A scale is relatively flat, but there's no good reason to use the A scale for *anything*, except perhaps when fitting a person for hearing aids. The A scale has been mis-used by businesses for decades as a way to hide how loud their products or work environments really are. Low and high frequency energy will still damage your hearing, and it can damage your hearing in the critical midrange; not just at the extremes of the frequency spectrum.

Honestly, a more useful tool when dealing with the rotary would be a real time spectrum analyzer (RTA). There are RTA cell phone apps that could have some value, as long as you back away from the source far enough to keep overall *levels* down under 100dB or so. Strip out the high order harmonics between 3kHz & 10kHz, and you'll likely find the rotary's sound to be much more pleasant. (And believe it or not, investing an extra 10 pounds in the stock cast iron manifold (for a 13B) is by far the easiest way to do that. The stock Renesis manifold does a good job too, at less of a weight penalty.

Charlie

 

[MolsonB]

Success with the header muffler. Design was a bit tricky to come up with a design plan with the welder but we finally got things going. I got about 1.5hrs on it so far and I'd say it's comparable now to a typical aircraft engine. I can talk to the test pilot even during take off. Even can start to hear the prop noise. Using an spectrum analyzer app on phone, it peaks are below 320Hz. Sadly I didn't take samples of it before, but I'd guess they would be extremely higher in freq which makes it so hard on the ears.

I checked backpressure on the Centrifugal Accumulator can, I was getting 3.5psi @ 2200rpm prop. I'd like to sample putting in another 'inside center tube' with less holes drilled to see the backpressure vs noise difference. The inside center tube is 2.75", and I drilled about double the area of it. I'd like to try the same area next.

I ran out of time, but phase two was adding the underbelly pipe with holes drilled in it.
http://www.rotaryeng.net/coax.html Quickly I mocked something up and put it in place by hand and magically a lot of engine noise disappeared. All I could hear was the prop, it was amazing. In a few months, I'll get back to the plane and test it out more. I want to measure the backpressure of the underbelly pipe and find a happy medium between psi vs noise.

 

[MolsonB]

After 15hours, my 321ss slip joint bellow cracked at the accordion folds. I think there was too much movement in the long exhaust under the plane, guessing my rubber exhaust mounts could be better. I ordered a new slip joint bellow from SPD exhaust which is 625 inconel for the accordion folds.

I'm wondering if the original spot (actually where the blue arrow is pointing) is the right location, or if I should move the slip joint down under the plane (or even have it)?

 

[Map]

It looks to me that the slip joint should be where the blue arrow is pointing. That is the spot with the highest stress if there was no flexible joint, so any cracks would show up there first.
With such a long pipe, the other place that could use a flexible joint is right after the 90 deg second bend, downstream.

 

[rv6ejguy]

The acoustical signature of Wankels has been known to crack exhaust piping in race cars in short order sometimes. Will be interesting to see if the Inconel holds up longer here.

Great work getting the noise level down which is often a challenge on Wankels without losing too much power.

You have limited space here but if the new bellows doesn't last, you might try a slip joint in each plane to deal with the thermal expansion.

 

[MolsonB]

Sorry my posts are so sporadic on this. I live in Canada but built the plane in FL, so I'm not around it much (specially now). When I do go down to work on it, it's 16hrs in the hanger, sleep and repeat. No computer time. I'm hoping to fly it back to Canada this spring, where I can go for an afternoon, tinker, and report back on here.

Map, the 3d drawing is outdated and in fact the slip joint bellow was where the blue arrow was pointing. I do agree with you that if I add a longer pipe underneath, I should have another one down there too with better exhaust mounts.

Ross, I'm pretty sure it was the vibration from prop, exhaust heat and not enough tension exhaust mounts underneath, that let the slip joint bellow crack from moving around too much. My prop isn't balanced yet to the engine, which is the first order of business when I get down there in a few weeks. Found a place that rents a prop balancer, so will remove the psru, balance the engine, install psru then balance the prop. I did add double slip joints from each header tube to the Centrifugal Accumulator. Then also the internal perforated pipe is slip joint inside. I do have a bunch of tabs cutout, to hold the exhaust pipes together if I add slip joints along the way. Do you think V-Bands add any thermal expansion relief?

The Centrifugal Accumulator (CA) did a great job reducing the noise. By taking the sharp acoustical exhaust pulses, the CA pushes them through the center perforated pipe which turns it into one solid note. I want to double check on my back pressure measurements in flight though. When I tested adding the coax muffler underneath the plane, I was amazed how quiet it went. Just need back pressure numbers to confirm I'm not choking the engine.

 

[rv6ejguy]

I don't think V bands move much or at all. I use multiple slip joints on my exhaust in 2 planes. No cracks in over 445 hours but my EGTs are probably lower and I have .065 wall tubing on the inlet side of the turbo, .058 on the outlet. Seen tons of light wall stuff crack on planes and race cars over the years.