I thought Rock Wool was volcanic ash - mostly Aluminium doixide or sulphite or some such (not carborundum) that can be spun like silica sand at temperature - the same way sugar is spun to make Candy-floss ("Sugar-candy" for the Colonials?). It is pretty inert isn't it? Otherwise we would not fill houses with it as insulation? Not at all like Asbestos!
But car silencers use a nickel chrome stainless wire wool I think? - I buy it cheaply from E&@y for making radiant elements for enhancing power from gas burners.
I am still confused by the noise reduction required here. Is it on an aircraft? A bench engine? or what?
I think we have a 70cc 2-stroke, planned to run at 6000rpm. Is it a single cylinder? or multi? Or Wankel? Atkinson? Or what? I have missed something here...
I thought you started off talking just silencer. but not I am confused as to whether you want a "reflecting wave expansion box"? = The complete opposite to a silencer.
Incidentally, You ask about "where does the energy from the pressure waves (noise) go?" - It goes as heat in the gas expelled from the exhaust, through side walls, and through the emitted gas. But it is not a lot of heat compared to the waste combustion heat and pressure when the exhaust valve/port opens.
I think you need to consider pressure dropping through the system, to understand that.
When the exhaust valve/port opens, you have the expanded cylinder of gas at some pressure (You postulate 18psi? - call this P1, T1, V1) and temperature (>600deg.C?....?). This expands into the first chamber - the exhaust pipe volume - and as it does it heats the exhaust pipe metal and the gas expansion causes some "adiabatic" cooling. But there is still some combustion (CO to CO2) until the temperature drops below 300C... -ish. So maybe this is more like Isothermal expansion until the temperature has dropped enough. Have you run the engine without an exhaust pipe to see the flames from the port? This exhaust pipe combustion can be guessed at by seeing how much "blueing" you get on a polished steel tube as an exhaust pipe. The blue and yellow colouring - like tempering colours - relate to metal temperatures, not burning gas - but are indicative of how far the combustion travels in the exhaust. In pressure and temperature terms, where there is steel discolouration, the gas is expanding isothermally, due to the remaining combustion keeping the gas hot. Where the discolouration ends, the combustion has stopped, so from there onwards the combustion is adiabatic. It makes a difference as to the pressure, = speed of sound (for expansion box reflected wave analysis). and for a starting point to determine the pressure at the intake to the first chamber of the silencer. The exhaust pipe expansion (and pressure drop) is the pressure and temperature and volume at the input to the silencer first chamber.
http://edge.rit.edu/edge/P11221/public/Exhaust Muffler Design Principles
(This link may help).
You next consider how the total volume of hot gas expands into the total volume of exhaust pipe + first chamber of the silencer. This becomes P2, V2, T2. This has to get through the hole(s) CSA into the next chamber, if you are using multi-stage expansion silencing. (like cars etc.). In multi-stage expansion, the gas flow is usually split, so some gas goes a higher pressure route, and some a larger chamber lower pressure route, thus splitting the pressure wave energy into 2 bits travelling at different speeds, (speed of sound varies with pressure in the chamber) and when re-combined later will be of lower amplitude. But if you use a series of Helmholtz resonant chambers, the first should be tuned for the primary shock wave frequency. (Near 95Hz, you reckon? = engine valve/port opening frequency). Then the second chamber to another frequency (1000Hz - you think?), and possibly a third chamber for another frequency? - The the gas passes out of the tailpipe. This is most effective for constant power/engine speed engines, such as a car with A/T where it sits at the normal max. torque rpm during acceleration. For M/T where the revs are constantly changing, this is alternatively only used for "Highway cruising speed/engine power condition". - I have a car that is loud at ~63mph, but quiet at 70~75mph (= max Highway speed in UK/Europe) due to inlet and exhaust resonators. In resonators, you do not need packing: the sound resonance with the empty chamber takes some energy and dumps it later where the after-wave pressure drop occurs, so flattening the noise peak. Alternatively, a resonant expansion chamber (on racing 2-strokes) is used to reflect the pressure wave so it pushes gas back into the cylinder just before the exhaust port closes. This increases the mass of gas in the combustion chamber (and heats it a bit) and returns any new fuel-air mix that has come out from transfer pressure of the new charge. Helmholtz resonators, sound absorbers, and "exhaust gas splitting" kill this performance enhancement. But many engines are tuned to have some back pressure or reflected waves, to increase torque like the expansion box reflected wave method.
In silencing "car" engines, there are often multi-pipes (different lengths) inserted at the first phase barrier of the silencer, to split the pressure wave into 2~6 elements that are slightly out-of-phase. This reduces the peak pressure wave downstream, and a series of split paths through sound absorbing materials reduces these pressure wavelets further.
Car silencer internals often have a helmholtz resonator - as a dead-ended chamber - within the multi-chambered and gas path construction.
That's all I can think of for now - rememeber - I Am NOT and expert! - so where I am wrong, don't shoot me!
K2
1) rock wool: some people want to sell you a lot of it- you can't just buy it from the hardware store.
- it seems to be available as huge slabs or as flexible bales like glass wool.
2) I am assuming exhaust release must be at around 200 psi, to make such a "crack" sound.
if I expand this to 11 times the volume, it comes out at 40 psi.
This is still too high.
the 1000 hz is just my estimate of rise time and duration of the sound from the exhaust event.
This occurs over a very small part of the whole cycle. - significant pressure drop is required at 18% of the cycle. or
1/6.
If we are down to 10 psi above atmospheric pressure by this stage, then noise production has about ceased.
the 95 hz is the actual firing frequency.
Possibly "significant" volumes for parts of the muffler would help.
I an thinking about 1/2, then about 3:2 ratio for the remainder- that is 3/10, and 1/5.
having whole number ratios could be a bit of a trap, though.
The 45 degree deflectors block the straight passage of shockwaves, and direct them at the side walls, where the external packing can
absorb them.
I think 6 x 1/4" holes is not going to provide much obstruction to the amount of gas, velocity, and pressure at the first barrier.
it will help break up any directed stream flow, along with the first deflector.
The last set of 50 small holes should stir the passing flow up more, without actually providing a lot of actual resistance, as well.
The final stage is a 380mm "stinger" at 5/8". This is probably a bit large in diameter for the engine.
The idea us to pull the muffler pressure down, but it conflicts a little with it being a good silencer.
3) The main engine I am running at the moment is a "big bang" two-stroke twin, which is effectively a single for silencing purposes.
I want the silencer to be small enough to fit on a model aircraft.
This won't be happening until I get the electric starter, and a contra-rotating gearbox, sorted out.
The aircraft I am looking at is tail-sitter, and I need the contra-rotating props to cancel out the prop torque reaction.
Before I build something like that, I need to learn how to do RC flying properly.
For now, I like mucking about with engines in my workshop more than actually flying model planes.
I also have a single cylinder two-stroke in a partly built stage that I am trying to get to run more like a four-stroke,
with a fairly smooth fast idle, and smooth acceleration without misfires.
The last running version seemed a bit "cammy" if I may use a 4-stroke term, with a big jump from idle up to
4,000 rpm or so, with a propeller load.
It will get a new head, and I am trialling reed valve transfers, when the parts arrive.
This engine will only be a test-bed engine, but a bit more muffling would be nice.
4) I am not getting into a complex, car-like muffler design. I want to see what will happen with the simple elements I have described.
I will make the can easy to disassemble, so I can try different things,
I will read your reference.
