Mailing List flyrotary@lancaironline.net Message #65398
From: Neil Unger 12348ung@gmail.com <flyrotary@lancaironline.net>
Subject: Re: [FlyRotary] Re: Mufflers
Date: Mon, 25 Nov 2019 07:48:27 +1100
To: Rotary motors in aircraft <flyrotary@lancaironline.net>

ED,

             What you did reminds me of a paper published in OZ many years back, where the "expert" suggested that all that was required to stop the pulses was to "Rub" the exhaust along a tube to even out the pulses.  What he designed was a muffler with many internal 180 degree bends that sent the gas many feet by "rubbing" against the wall of the tubes in the 180 bends.  Your design is the same principle as the swirling effect is rubbing the  gas against the wall of the pipe.  Your two 2.5 inch pipes amount to an approx single pipe of 3.75 ".  So a 4 " pipe would be even bigger.  Back pressure?? can only be measured by a manometer possibly?

The idea is very interesting and I wonder if the gas was reversed at each "turbine"  ( the washers direct the gas in opposite direction each time) would help to quieten the noise even better or faster meaning less washers ( weight) and back pressure?.  So many unanswered questions that can only be answered with experimentation and time.  My "you beaut" decibel meter may actually get some use if it actually works.

Thanks ED and if I can find time will experiment with black steel to start with.

Neil.

On 11/25/2019 4:48 AM, eanderson@carolina.rr.com wrote:
Sorry, Neil,

Been a while, it was two exhaust pipes, one for each exhaust port.  Mistyped, error on the drawing the disc were 2 1/2" dia inside a 2 1/2" dia pipe.  The discs fit snugg enough that when one broke loose and spun on the rod,  it scored the inside of the exhaust tube.  I got the discs from McMaster Carr.  Try oversize SS washers.   The tubes were actually glass pack mufflers which had 2" dia inlet/outlet.  I cut the front off and then after inserting the discs use wide metal clamps to put the tube back together (welding would of course have been better).

I do not believe that any of the dimensions are critical except the discs should fit touching the inside of the tube if possible.  I called them discs but they were actually oversize SS washers with a hole in the center for the rod.  I ended up with 5 discs inside each tube.  I used the spacing for the shorter 36" long tube.  The idea was to have a large portion of the shock wave dissipated inside each segment of the tube while keep the gas flowing freely.  

It was truly quite and like I mentioned I could get 6000 rpm with my 13B on take off.  Large diameter tube/discs would likely permit more/easier gas flow, but since I can not weld, just finally gave up and went on to other things.

Ed

 





------ Original Message ------
To: "Rotary motors in aircraft" <flyrotary@lancaironline.net>
Sent: 11/23/2019 10:35:27 PM
Subject: [FlyRotary] Re: Mufflers

Ed,

                  On rereading your post, am I correct in believing that you had one pipe per exhaust?  A total of 2 pipes?  How many discs were in each pipe?  I may have misunderstood, but you say you had a 2.5 inch disc fitted inside a 2 .125" exhaust pipe.  If correct measurement how did you manage to force 2.5 " inside  2.125" pipe?  perhaps the other way around measurement wise? 

Thanks,  Neil.

On 11/21/2019 3:08 AM, eanderson@carolina.rr.com wrote:
Given the recent interest in mufflers, I thought I would throw in my experience with home made muffer.  I could hit 6000 rpm on take off with my  13B with the 2.18:1 Gear ratio.  You could stand within a wing width with the engine idling and bearly hear it.  At 6000 rpm it was about 1/2 as loud as an unmufflered lycoming.  

The basic idea was to break up the shock wave while minimizing gas flow impedeance.  I used 1/8" thick 2"1 /2 Dia stainless steel disc inside of a 2 1/8" dia stainless steel tube.  The disc has slits cut from the outer edge approx 1/2" toward the center in 8 regions as shown in the diagram.  Then each "tab" was twised 45 deg from the plane of the disc.  If you looked at the disc front on it look almost as a solid disc with just a small slot area showing through.  The idea was that the shock wave would basically see a "solid or mostly solid" disc where as the gas could still flow through the slots around the tabs.  I had a threaded rod extending the length of the tube with jam nuts on each side to hold the disc in place.
The reason for the rod and nuts was not being a welder-  I used jam nuts - welding it turns out would have been much better as the nuts eventually became loose.  That was not good, when one of the disc came lose it started to spin and greatly  impeded the gas flow.   Although folks told me it sound cool - like a turbin winding up.

In any case, I decided that not being a welder there was no way I could remedy the defects.  I think if there were some way to "spot weld" the outer part of the tabs to the tube and perhas to a rod in the center the muffler would have extend it useful time.  I flew approx 12 hours 
including one trip to Tracy Crook (first extended flight with muffler) by the time I got there at least one disc had broken loose and was spinning.  Tracy was kind enough to use his welding skills to weld the discs to the rod on one of the mufflers (had two one for each exhaust), but within another 6 hours or so disc in the other muffler started to spin.  Also I found that the shock wave pounding eventual would break off a tab or two since they were not anchored and could flex.

Just thought I would throw the idea out there in case it has any merit.

Best Regards

Ed






Image
bnownp1a.jpg
Image
22gpummc.jpg
Subscribe (FEED) Subscribe (DIGEST) Subscribe (INDEX) Unsubscribe Mail to Listmaster