Custom muffler design. Formulas etc???

[TERRA Operative]

Ok, I'm sick of the buzz and drone of my exhaust, so I have measured the thing with the spectrum analyzer from work.

Here are the results:

AE101 GTZ Levin (4AGZE), blacktop extractors and stock cat with 2.5" mandrel bent straight through exhaust, hot dog resonator, split at the rear axle to two mufflers as per the stock system.

SPL: 99.9dBA (as measured to EPA standards)
Drone: 63-125Hz (depending on speed, 63Hz at idle, 125Hz in 5th at 110KPH)
Buzz: 8000-10000Hz (That damn Honda like buzz...)



So, I've come up with a design for a muffler that will prevent the buzz getting through, like a baffled muffler does, but is still free flowing. Kind of like a hybrid between a baffled and a straight through.
What I'm looking for is some formulas to allow me to work out how big to make my chamber (the total size of the muffler) to tune it to the frequency of the drone.

I assume that the formulas will be similar to those used for Helmholtz resonators, but if anyone can provide actual formulas, charts and/or know-how, that would be awesome.

[JustenGT8]

I did this exact exercise in designing the rear muffler for the GT8....worked a treat too. At the time i had just done a noise regulation course for work and the workshop manuals had all the info needed. Unfortunately i have no idea where the manuals have got to (over 5 years ago now) but it wasn't that hard and i suspect any text on 'noise' would have the theory.....wiki for starters?

[TERRA Operative]

The stuff I've found so far has either been crazy theory and big formulas, or hopelessly vague.

I just need some basic formulas to put me on the right track.....

[Billzilla]

Don't copy my design, it doesn't work.



Still have yet to fiddle with the internals to take the raw edge out of the sound at times.


Anyway for more conventional mufflers you need as few flat surfaces as possible to stop drumming and passing on noise and a non-direct path to help break up the pulses.

[JustenGT8]

The stuff I've found so far has either been crazy theory and big formulas, or hopelessly vague.

I just need some basic formulas to put me on the right track.....

search harder

I had a look through my training file but no joy sorry, i can school you in the policy process or water sampling protocols if you like?

It was pretty simple from memory though and a bit rule of thumb. Just a diameter related to the amplitude that will work for a certain for (freq). You'll never be spot on regardless.

I can tell you they generally need to be big. Mine came out at 220mm dia!!

Seeing you sound committed to putting the effort in to making something, you might want to try this? The best muffler i have seen operate (and the theory was good as well) is essentially set up like a gun silencer. Just a central tube with many small cavities along it's length. 2 issues to overcome....dia again, as the bigger the dia the more effective it will be; and weight. Once you put all those baffles in it could be quite heavy. If you can stamp out thin sheet and find a way to attached it to the muufler body then you are on a winner for sure. You see this design on stationary engines as can be tuned for the narrow operating range and size isn't an issue...they are super effective.

Dedicate some more search hours or call a noise consultancy firm...you might get lucky with some advice from a geeky engineer happy to part with info?

[TERRA Operative]

This is my design:




The center plug is designed to deflect any high frequency noise into the surrounding muffler (the cone is 45deg) without significantly impeding gas flow itself, it's meant to stop the high frequency buzz from 'beaming' through the muffler. The inner sleeve is perforated (just like a normal straight through muffler) as is the center tube, except for the ends to provide the deflection.
The outer tube and the inner tube are filled with fiberglass, and the supports for the center tube are angled to provide a bit of swirl to the gas so it stays in the muffler longer.
I've calculated the area in the muffler to be a little more than the area of the incoming and outgoing pipe to try to slow down and cool the gas too.

What do you guys think? Worth giving a go to see what it does?

[JustenGT8]

Complicated little sucker It'll be a bitych to make. Your flares in and out are pretty steep angles and the exit will be a choke point for sure.

If you reckon you entry cone deflector will work, just add a smaller one to a normal chambered straight thru and see what it does? Just step the tube dia so you don't lose x-sectional area.

[Billzilla]

This is my design:




The center plug is designed to deflect any high frequency noise into the surrounding muffler (the cone is 45deg) without significantly impeding gas flow itself, it's meant to stop the high frequency buzz from 'beaming' through the muffler. The inner sleeve is perforated (just like a normal straight through muffler) as is the center tube, except for the ends to provide the deflection.
The outer tube and the inner tube are filled with fiberglass, and the supports for the center tube are angled to provide a bit of swirl to the gas so it stays in the muffler longer.
I've calculated the area in the muffler to be a little more than the area of the incoming and outgoing pipe to try to slow down and cool the gas too.

What do you guys think? Worth giving a go to see what it does?

Yeah it'll work, as it's been done before.
I think the mob was called Rocket Mufflers and they were on the Gold Coast, not sure if they're still around but they were briefly popular about ten years ago.

[TERRA Operative]

Complicated little sucker It'll be a bitych to make. Your flares in and out are pretty steep angles and the exit will be a choke point for sure.

If you reckon you entry cone deflector will work, just add a smaller one to a normal chambered straight thru and see what it does? Just step the tube dia so you don't lose x-sectional area.

Yea, I gotta work out optimal angles yet, apart from that, it's all to scale.
Good idea re. the test cone. Could almost try the whole plug in a large bore muffler and reduce the diameter before and after the muffler itself.....

Yeah it'll work, as it's been done before.
I think the mob was called Rocket Mufflers and they were on the Gold Coast, not sure if they're still around but they were briefly popular about ten years ago.

Sweet, at least my theory is sound then. What happened to them? Were the things too hard to make?
I might do a bit of research into them and see what I can turn up...

[Billzilla]

Sweet, at least my theory is sound then. What happened to them? Were the things too hard to make?
I might do a bit of research into them and see what I can turn up...

Not sure, I never really followed the company.
Yours is a little more complex but it'll work well.

[oldeskewltoy]

Hmmmmmmmm, I wonder.... instead of your center section(cylinder with two 45 degree ends) would using a drop tank design be more effective??



these are models of drop tanks, but you get the idea

[TERRA Operative]

Yea, I thought of that. Could be hard to design that shape with perforations etc.

Also, would I have to match the outer pipe to maintain constant area too?

[thechuckster]

are you running two or one mufflers? having a resonance killer (e.g. hotdog) before a free-flowing muffler seems to work well. Relying on the one device at the end of the zorst means you'll having to counter effects of the piping as well as the engine.

[TERRA Operative]

I've got a long hot dog resonator, then the exhaust splits to two straight through mufflers.
I think it's the hotdog that's not doing it's job....

The old exhaust was 2 1/4" press bent with a baffled oval center muffler and it didn't buzz at all.

[oldcorollas]

what is the internal design of the hotdog? many of them are.. shite