Effective muffler for a two-stroke model aeroplane engine.

[Owen_N]

I have an existing muffler for a 70cc engine, which is about 65mm diameter and 300mm long.

Output when running is roughly 95 dB when I run it in the garage- blowing the exhaust out the main door.

Can I make a muffler that fits in a similar space but is quieter?

The outlet pipe is about 16mm diameter, and 2x inlet pipes of about 20mm.

There is not enough room for a tuned pipe with a muffler.
A tuned pipe for an engine like this is getting on for 3 feet long- 900mm or so?

The engine has to peak around 6000-6500 rpm.

I can go slightly longer and wider with the muffler size.

This one is all aluminium, so it will be hard to match the weight.

I don't want internal packing, but external wrapping can be used.
The engine chucks out too much oil to use packing internally.

It possibly gets a little hot for Styrofoam, but is OK with plastic wire ties.

Attached-some muffler design images.

The "simple perf" looks buildable, but may not be as effective as some of the others.

Any ideas or experience?
I will look for a cross-section of the existing muffler type, and post that, too.
The RC plane forums don't usually get down as far as muffler internals.

 

[Owen_N]

additional mufflers-
my muffler, typical internals, muffler size.

The internals are fairly simple- I can fit a lot more in that amount of space.
The engine will tolerate quite a bit of exhaust restriction.

Any ideas on a lightweight external can design?
I don't have any sheet rolling equipment here.
I would think about 0.2mm stainless would be good.
That is what the header tubes are made with- about 8 thou?
Tinplate would be good, but stainless is easier to buy.

 

[Owen_N]

I found a good source of 0.2mm stainless- $35 nz for 300 x 1000mm, ex AliExpress.
Actual tubing in this material would be good.
A simple design is to extend all three tubes the full length of the muffle, blank the ends, and then dsrill holes in them, choosing the smaller output tube cross-section as a guide.
If it is 15mm ID, then area = 177 sq mm
a 2mm hole has an area of 3.14 sq mm, so this would need 56 holes.
the larger tubes could have half that number, or 28 holes each.
What do you think?

1) would sound attenuation be sufficient?
2) would backpressure be low enough?
<edit>
Suitable tubing is hard to find.
How about division into three sectors, with 2 walls having 28 holes each?
Would this be more effective than the current arrangement?
do I need more intervening sectors with perforations?

The two inlet tubes could be allocated to a single sector.
The current assembly could be sliced until the 3 tubes appear, then a new end wall can be bonded in using J-B Weld.

I haven't found anything better that is actually currently available. Goodwood make some epoxy-based materials, but
those are unavailable even in USA, and freight is a problem with resins.

J-B weld has good temperature and gas/petrol resistance.
I use it in two-stroke transfer ports, and it stays put.

A new cannister could be square or rectangular, and brazed together.
Rivets and sealant could be used for heat-sensitive areas- as with the J-B weld.
It is good up to about 280 degrees C, but I find it tends to slowly degrade on the outside of cylinder heads.
Spot temperatures sometimes hit 240 degrees C on my old heads.

With my new home-built air-cooled cylinder head, temperatures should stay down around 160 -180 degrees C, I think.
Where peak temperatures are below about 150 degrees C, J-B Weld is quite stable.

 

[Stefan-K]

Owen
From my own experience i can Tell you that building a good silent muffler on your own for an RC Airplane is not worth the time, effort and money. The only way to make a muffler silent, lightweight and durable is to make it out of Aluminum and solder the parts together. This needs special Tools to Form the parts. Special flux and solder and a lot of experience to solder (or braze?) Aluminum. No professionel Manufakturer uses Rivets or Jb-weld to build their Mufflers and that's for a good reason. Did you have a look at the Website of Krumscheid-Metallwaren in Germany? They really make the Best and most silent Mufflers. They are expensive But they are really worth the money. You can hear the difference too on YT.
Greetings Stefan

 

[Owen_N]

Owen
From my own experience i can Tell you that building a good silent muffler on your own for an RC Airplane is not worth the time, effort and money. The only way to make a muffler silent, lightweight and durable is to make it out of Aluminum and solder the parts together. This needs special Tools to Form the parts. Special flux and solder and a lot of experience to solder (or braze?) Aluminum. No professionel Manufakturer uses Rivets or Jb-weld to build their Mufflers and that's for a good reason. Did you have a look at the Website of Krumscheid-Metallwaren in Germany? They really make the Best and most silent Mufflers. They are expensive But they are really worth the money. You can hear the difference too on YT.
Greetings Stefan

I have the material on order.
The problem with using thin stainless is that it needs a glasspack wrap around the outside, as it tends to vibrate more than thicker aluminium.
I will have a go.

all-welded construction is a matter of cost, if you have the gear.
I can build a structure similar to a tinplate fuel tank no problem.

The J_B weld is to combine existing parts and new parts, and block all the gaps, as a caulking agent.
Mufflers do not get that hot, compared with the header pipes.-just burned myself on one today!

that is likely 120 degrees C plus.
Not as hot as a steam pipe I grabbed one day. - that left a burn print on my skin!
You would think they would lag something that hot!.

A manufacture would not do that because welding a full unit is cheaper and faster, if you have the gear.
I will check your reference.
<edit>
the base price is 188 euro, about $320 nzd .
This is OK, but freight is likely to be quite high to New Zealand.

Typical ebay freight is about $65 nzd for something this size.

that would be about 358 eur or $406 usd.
Probably not excessive, I suppose.

 

[xpylonracer]

Checking the noise level from a running engine is best done over grass in an open area, there is just too much noise reflection inside a garage.

xpylonracer

 

[dnalot]

ceramic fabric insulation is very lightweight.

Mark T

 

[Owen_N]

Checking the noise level from a running engine is best done over grass in an open area, there is just too much noise reflection inside a garage.

xpylonracer

I do not have a noise meter to check it, but I would expect it to not make more noise than a modern two-stroke scooter of about 100ccs.
There seems to be more of a "crack" or "pop" to the exhaust than necessary.
Well out in the open air, the noise energy is probably not that high.

 

[Owen_N]

ceramic fabric insulation is very lightweight.

Mark T

The temperature range of the actual muffler canister is suitable for Jute or Hessian, wound with plastic packing tape.
At the cannister in open air, it is barely 50 degrees C at the most.
Even the header pipes are below 150 degrees C for most of their length.
Possibly with more insulation it will get hotter. I will try it out, and see if the packing tape melts.
I am not insulating the main header pipes.
I think external insulation will cause a substantial noise reduction.

The main question is, of several stages of 2mm holes will cause any noise attenuation if I have 50 or so of 2mm holes. per stage.
I plan to have at least 2 stages of holes, but I can insert another layer.
This matches the standard outlet cross-section area.

A problem is that vibration of internal barriers will transfer sound energy from one stage to another.

It would work better if 3mm thick internal aluminium barriers were used, with the same number of holes.
3 layers 60mm square are not that heavy.
say 2710kg/cu m
60x60x3x3 = 10,800 cu mm, 2.7 x 10^3 x 1.1 x 10 ^ -9+4 (-5)
= 3 x 10^-2 kg, or 30g, which is a reasonable amount.

I need to ensure that external panels can be removed so that alternative internal structures can be trialled.
possibly a flanged wrap of top, end, bottom can be used, and side panels retained by J-B Weld.
this can be levered apart with a sharp blade, but is strong with a well-spread load.
running the 16mm tubing right to the end chamber could be an option.

 

[moose4621]

I used to make silencers for 2 and 4 stroke F3a aerobatic models as a commercial business. I found that the first chamber closest to the engine needs to be at least 10 times the capacity of the engine. The bigger the better. Then two more smaller chambers of unequal size to prevent synchronized resonance. The aim is to reduce the pulses from the engine to a continuous stream of gas at the outlet. If you use any form of fibre within the muffler, it will become saturated with oil and gain a LOT of weight.
A sketch is worth a thousand words.

This was to achieve a noise level of less than 94db.
" The maximum sound/noise level of the model aircraft and its propulsion device, shall be 94 dB(A) measured at 3m from the centre line of the model aircraft with the model aircraft placed on the ground over concrete, macadam, grass, or bare earth at the flight line."
You dont say how you measured your 95db so I am guessing it would be noisier than 95db if measured as per above.

 

[Owen_N]

I used to make silencers for 2 and 4 stroke F3a aerobatic models as a commercial business. I found that the first chamber closest to the engine needs to be at least 10 times the capacity of the engine. The bigger the better. Then two more smaller chambers of unequal size to prevent synchronized resonance. The aim is to reduce the pulses from the engine to a continuous stream of gas at the outlet. If you use any form of fibre within the muffler, it will become saturated with oil and gain a LOT of weight.
A sketch is worth a thousand words.
View attachment 134057

That looks good. 10x the engine capacity is quite large- that is 700cc, or about the size of a wine bottle.
A 300mm length of 16mm (5/8") tube is 40g.
I probably don't need that much.
the current total volume is 848 mls at 60mm dia x 300 mm long.
Once running, this engine can stand heaps of outlet restriction without substantially affecting rpms, so I suspect that the initial volume can be reduced to about 300 mls without a lot of loss.
This is 4x the engine volume.
If exhaust pressure is, say, 200 psi, then cleared pressure will be about 40 psi above atmospheric,
less pressure drop due to heat loss and leakdown through the divider vents.
Having every chamber a different volume is a good point.

I would possibly make the first row of the 3 dividers say a ring of 6mm holes, and not have the first 2 tubes extended at all,
just dump into the 300 ml chamber .

I have more of the 5/8 tubing on order, so it could be extended with some holes.

The best deal I found on the aluminium tubing was Ali Express, and it doesn't get here until Mar 29, which is quite a while.
In the meantime, I can put up with the less effective cannister muffler.

 

[Owen_N]

Correction:

A muffler hotspot was found of 135 degrees C after 10 minutes running.
Organic wrapping material like pure cotton should be OK at that kind of temperature.

Now, exhaust gas is hot enough to melt steel at the exhaust point- see 4-stroke exhaust valves, so temperature at the peak value must be well over 1300 degrees C , as steel melts around 1200 degrees C.

I will see if the muffler "can" gets much hotter when it is insulated a bit.

Catalytic converters can get hot enough to burn grass, (300 deg C?) which is an indication of what can happen.

The engine is acting lean and losing power at 4500-5000 rpm, but it is not fully run in, as, once I got it to 5500 rpm for a while and
It started to run out of fuel, it sped up quite a bit.

This is run number 3 , 10 minute runs, so it is due to be run at 5000 rpm.

It needs more running around the 4-5000 rpm mark before any more rpm increase.
eventually, it should easily go to 6500 rpm with this prop.

 

[Robsmith]

I made heaps of muffler designs for my 393 cu in V8 . I eventually found the best design was the one called the dual flow spiral muffler. It initially was too restrictive in flow. So I played with the spiral part and removed the pipe inside the spiral. I then put the spiral inside a perforated 2" tube. That tube was covered with stainless steel pot scourers. ( they are a donut shape when you untangle them) The whole thing was 8" long. It worked brilliantly. The open pipe(without the spiral insert) was 110 Db . With the insert. 93 Db I'm sure you could scale it down to suit. Mine is all stainless but still very light. The completed one is behind the spiral

 

[Bentwings]

Owen
From my own experience i can Tell you that building a good silent muffler on your own for an RC Airplane is not worth the time, effort and money. The only way to make a muffler silent, lightweight and durable is to make it out of Aluminum and solder the parts together. This needs special Tools to Form the parts. Special flux and solder and a lot of experience to solder (or braze?) Aluminum. No professionel Manufakturer uses Rivets or Jb-weld to build their Mufflers and that's for a good reason. Did you have a look at the Website of Krumscheid-Metallwaren in Germany? They really make the Best and most silent Mufflers. They are expensive But they are really worth the money. You can hear the difference too on YT.
Greetings Stefan

I would nt even consider o waist time thinking about soldering aluminum regardless of ll the recent hype about aluminum alder exhaust temp gets hot enough to melt electrical sold as it is . I’ve Tig welded for myself and professionally for many yers welding mos anything weld me thin aluminum is n way to weld in th best conditions I’ve made dozens of two stroke timed pipe and muffled tuned pipes there n nature arm of technology in design of muffles then getting into jt fabrication mother thing basic round canister i the jeneral at dad usually two exit pipes fancier be have cone shaped ends to deflect spud waves. But there is scieneinjus significant the too. I’ve made primavera hydro formed custom thing f my c pane. Even evolving not hydro fed tanless steel scale functioning nes on my scale warbirds. It took a lot hour and some hit and miss to make thes work .019” stainless steel is in another word to Tig weld there is sicker sders and silver brain stuff available that the back yard peri can do with simple torch and hand ful of dollars . Iv made a bunch of nice scale fnctioningtwo stroke exhausts using copper fittings and copper tubing. I silver solder and praisptty easily . By thinking a littl and doing some tedious cutting and trimming you can take over half of he weight ou f copper system. Copper also th welds ver asilyif yo keep it clean and understand some of us material properties. If you are running gas for fuel yo can see exhaust gas temps up to 900 deg F it get real hot glow fuel is much ower but hot enough melt almost any solde o don’t eve think of soldering copper . You can get copper fittings down to 1/4” inch all are quite expensive today . Rediculous actually most weld shops won’t mess with hobby stuff as time is money I just gave 3 complete Warbird scale exhaust systems to my son for his Warbird builds I’ll probably do at lest another . My vision has deteriorated severely as you can see her in text as I have to use stylus to hunt and peck I see two of every thing so two targets eye patch does not help. Welding is extremely difficult too. I’m doing instruction with my sons now. They are getting better all the time
In reality attempting to balance cost of your project it might be better to get someone to model in 3D cad then have it 3D metal printed . I’m considering this now as I’m really not supposed to be in the shop medically it being a slow process as I’ve been away from cad modeling for a long time and just managingvhe keyboard an screen is very difficult .
Slim line still makes two stroke exhaust systems there may still be a couple others still around if you look at motor cycle two stroke exhaust you can see construction things . Many are hydro formed the process is fascinating . I never got into it much in engineeringvascitvwascemergingvtech that did not apply to my areas .

Byron

 

[Steamchick]

Aha! Another area of Engineering where I have no expertise! But
I did make silencers for air powered circuit breakers.... discharging 28 bar air to atmosphere through a fast sleeve valve. 9 in diameter piston, 5 in stroke, without silencer made 135dB (@ 1m.) The previous 3-phase expansion boxes brought this down to 132dB. Which in the open air was too much at the regulation distance (I can't remember, 85dB. at 75m?).
So I followed a 4-stroke motorcycle design ("Annular discharge" silencers for NOT commando in the mid-1970s). This had 3 Helmholtz resonators in series. But I mis-read the design and made a wave phase interference design that reduced the 135 to 127dB at 1m, and easily met the regulation test in the field.
No idea how that would work with 2 strokes....
K2

 

[Owen_N]

I would like to combine stainless 0.2mm ( 8 thou) sheet , 3mm aly plate, and 5/8 aly tube.
possibly tabbing and screwing or rivets to attach the aluminium.

- a bit labour-intensive, but I don't have a tig welder, or metal forming gear.

Peak surface temperature is looking about 140 deg C at the moment, but if I wrap the exterior, it will rise.

I am planning to cut the end off the existing aly cannister muffler and tab and rivet it to the stainless.
It will need some jointing material or cement though.

I can use muffler cement if I can get facing surfaces I can clamp together,
plus I have standard brazing rods, flux, and a propane torch for joining.

The stainless is a bit tricky, as it expands a fair bit when heated.
I will see how I get on.

I think 2x perforated discs in 3mm aly, and 2x "V" deflectors in stainless, should be enough.

The idea behind using 3mm is to cut down surface vibration and sound transmission.

The first plate can have 6 x 1/14" holes, and the second can have 50 x 2 mm holes (80 thou).
The 5/8 tubing just needs an angle cut end, and no holes.

Fully brazing the box can sides is probably premature, so I will look at something less permanent until I have tested the muffler.

 

[Steamchick]

In designing a muffler, be careful to arrange the sets of holes so that you do not choke the intake. The expansion should have holes in an expanding series, so the total cross-section of the intake effectively expands as you get to the total cross-section of the next "barrier", etc.
It is relatively easy to look at the constant mass flow, as the product of P x V, and any text book (or intrnet?) on the subject can easily guide you through the pressure drop at each stage. the gas from "1 cylinder" or "1 revolution" can be used as the starter point: Expand this from the engine (manifold CSA) into the volume of the first chamber (exhaust pipe), then through the change of CSA into the Silencer first chamber, through the first chamber exhaust CSA (= second chamber inlet CSA).... etc. to atmosphere. Then you can see how the CSAs progress and pressure drops with successive expansions. Each expansion loses temperature, as the gas expands. (using PV = RT). Get it wrong and the silencer will choke the engine at higher throttle settings.
This makes for a reasonable working silencer. But the use of Helmholtz resonators is even better. The chambers each allow some gas to expand as the main stream passes the resonator, after which the resonator dumps the gas into the vacuum that follows a pulse. This smooths the gas flow to have lower peak pressure waves reaching the exhaust end at atmosphere and making noise...
A "divided tube" (sometimes concentric) will also split the incoming energy shock wave from each exhaust pulse, allowing the mixed lower pressure pulses to interfere and thus reduce the max pulse (noise) at the end of the exhaust system. (This was my strategy for the circuit breaker silencer: 9dB reduction, compared to expansion chambers 3dB reduction).
"Clever" silencers are a bit more complex though, and I do not have the expertise to explain.
Car exhaust pipes have a series of silencers, a mixture of Helmholtz resonators , expansion chambers, and resonance interference. After-market silencers are often not as good in design, and significantly louder as a result - to make them cheaper! Then your "next new car" sounds much quieter!
K2

 

[Steamchick]

Try this?
Method for design and evaluation of ICE exhaust silencers (lboro.ac.uk)
Or
the Annular Discharge silencer (attached doc.). - Designed for motorcycles, the engine speed is similar to your engine. Just the capacity is different.

Or
Helmholtz resonance - Wikipedia
123909.pdf (pan.pl)
K2

 

[ajoeiam]

Try this?
Method for design and evaluation of ICE exhaust silencers (lboro.ac.uk)
Or
the Annular Discharge silencer (attached doc.). - Designed for motorcycles, the engine speed is similar to your engine. Just the capacity is different.

Or
Helmholtz resonance - Wikipedia
123909.pdf (pan.pl)
K2

Tried to look at your suggested docs.
#1 - - - nothing useful shows up (doc in embargo - - whatever that means!)
#2 - - - the holes move from larger to smaller - - - somehow that seems backward to what has gone on in this thread before
#3 - - - one I didn't check
#4 - - - lots of theory but I'm not that into the theory - - - just wanna build a bloody muffler (LOL)

Can you help?
#1 - - - text would be useful
#2 - - - is that correct - - - holes get smaller and less numerous from intake to outlet?

TIA

 

[Steamchick]

How about this?

• Method for design and evaluation of ICE exhaust silencers | International Journal of Powertrains

I don't understand why US and UK can't see the same documents on't web... maybe something to do with publishing rights? Most addresses work for me, but I do get the occasional one from US or Canada that simply won't give me anything.

#2 is a series of Helmholtz resonators along the tube: The gas basically passes along the outer annulus, not through the chambers. It is a bit like an electrical circuit, with capacitors spaced by various impedances to smooth-out electrical noise...

#4: I am only trying to help.... proper design isn't a case of stuffing bits together that look right, but calculating the correct sizes of "bits" so the result is what you want. I am offering suggestions, not telling what you must do. Life is a free choice, so do what you want. The modern technical world isn't "by chance", just calculated to be what it is. The most complex (calculations beyond my knowledge) is obviously the best solution. But my "simple" versions - calculated, not by chance - gave some progress in the right directions. Therefore I insist "I am not an expert" - but "I have been a drip under pressure" (Ex-Spurt!).
K2