Effect of exhaust tube thickness on vibration

[EngForm78]

I am looking for a general relationship between the thickness of an exhaust tube and the amount of vibration and sound. Does anyone know where I can find equations relating these. Will other factors (length of tube exhaust pulse frequency and amplitude) effect vibration/sound in a way that compoundeds the thickness of the tube in a non-linear relationship. Any help is appreciated.

Thanks

EngForm78

 

[GregLocock]

The thickness of the tube is important, but it is not the prime determinant of the noise radiated.

Really you are asking a very complex question, it would need a large textbook and several years study to answer it.



Cheers

Greg Locock

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[Warpspeed]

Tube wall thickness will probably be much less significant than the method of supporting the pipe, and avoiding unfortunate resonances. The type size and location of muffler will also be fairly important.

A quick look at the exhaust system on any production car will tell you that it is all far more complex than just following a few simple mathematical rules. If there was one simple best solution, all exhaust systems would have evolved towards a single common design.

 

[MikeHalloran]

Here's a general relationship:

The radiated exhaust noise goes up when the tube fractures from vibration or corrodes away.



Mike Halloran
Pembroke Pines, FL, USA

 

[SomptingGuy]

The only significant noise you might get from an exhaust (not including the big hole at the end) is "shell noise" from cheaply manufatured mufflers. In that case, double-skinning is the normal approach. I wouldn't worry about radiated noise from the pipe sections.

 

[EngForm78]

Thanks for the help.

 

[Warpspeed]

I agree, radiated noise direct from the pipe sections may not be a problem, but vibration transmitted through any pipe mountings could well be.

Acoustic resonance, and mechanical resonance are not the same thing. Both need to be avoided.

 

[GregLocock]

SomptingGuy

The tube thickness will have a significant effect on the frequency of the overall flexural modes. At least on our engine, they occur in the same frequency range as firing order at idle.



Cheers

Greg Locock

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[SomptingGuy]

Greg,

I realise. But what sort of sound radiation efficiency does a thin tube have? My comments only applied to the sound, not the vibration. I've never heard of noise radiation from the surface of an exhaust pipe being a problem, whereas shell noise from a muffler often is.

 

[GregLocock]

The thin tube will only radiate high frequency (say 1 kHz plus) effectively, for instance, the downpipe is a thin walled tube and is directly responsible for some high frequency noises.

When I was developing mufflers I did use an SLM with linear weighting for idle noise, and when we had a boom at idle it was due to the bending mode of the exhaust. As I remember 85 dB lin was customer complaint level, and you could see the tailpipe vibrate.



Cheers

Greg Locock

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[Tmoose]

Round/circular ducting has way less "breakout" than rectangular or even oval section ducts. The ASHRAE handbooks offer some frequency vs loss info

Pipe elbows respond differently to pressure than straight sections. Internal pressure or even A nice pressure pulse tends to straighten out an elbow like a party favor.

http://images1.comstock.com/Imagewarehouse/CR/SITECS/ThumbnailVersions/15404/15404-47DG.jpg

Greg, was the bending mode an ovalization or panting of the circular pipe, or a curves section/elbow tending to straighten out?

 

[Warpspeed]

It is not difficult to imagine how every bend in an exhaust system could act like the Bourdon tube in a pressure gauge, and either curl more, or straighten out with every cycle of internal pressure.

This slight pipe movement will set up a vibration along the whole pipe that may be amplified by any tendancy towards mechanical resonance. The whole mess may then flex and vibrate in various modes at different frequencies.

The simplest cure is probably to keep the whole pipe as straight as possible. The problem is going to be worse closest to the engine where pressure pulses are going to be more violent.

 

[Rob45]

"The only significant noise you might get from an exhaust (not including the big hole at the end) is "shell noise" from cheaply manufatured mufflers. In that case, double-skinning is the normal approach. I wouldn't worry about radiated noise from the pipe sections."

This and several similar posts are very much mistaken, and the error is compounded when very sharp exhaust pulses are involved, e.g., from naturally-aspirated diesel engines.

This can be proven by substituting what is generally called "laminated" pipe for conventional tubing in an exhaust system. (Laminated pipe is made on a conventional tube mill using two separate rolls of strip stock, having a common weld seam but otherwise separate. The two layers are of course in intimate contact, and vibration of one layer relative to the other produces coulomb damping.)

For example, I designed a front pipe, a 'Y'-pipe assembly when diesel engines were first introduced in light trucks at Ford. In addition to making a dramatic improvement in interior sound quality, this reduced passby or exterior noise by 2 dB(A), permitting us to get below the legal limit. I designed a laminated pipe for an NA diesel 4-cylinder engine in the same time period, and got similar improvements in noise.
I also used laminated pipe successfully to reduce pipe ring in SI engines, but the improvements weren't as dramatic, and the cost increase, though small, wasn't justified for light-duty trucks at that time.
By specifying different materials for the inner and outer layers, since the inner of course didn't require salt corrosion resistance, I was able to limit the cost penalty of this laminated pipe assembly to less than 5% as compared to one made with homogeneous tubing.

It is for similar reasons that several luxury cars use laminated pipe, to control "pipe ring," and that's why at least one major tube producer in the U.S. is expanding their capacity to produce laminated pipe up through 4" diameter.

Finally, the use of laminated pipe is generally limited to the front part of the system, ahead of mufflers and/or catalytic converters.

So, in answer to your original question, I haven't the foggiest notion how to model the pipe wall behavior, and as Greg pointed out, it may be enormously complex, but experience has shown that thinner walls are more likely to transmit higher frequencies and have a "tinnier," i.e., cheap, sound. And experience has shown that noise radiated from pipe surfaces is not to be ignored if you're interested in producing a quiet vehicle.

Regards,
- Robert

 

[SomptingGuy]

I stand corrected. I guess I've spent too many years synthesising and analysing tailpipe noise!

 

[Warpspeed]

Tailpipe noise (from the open end) is a whole different can of worms. That is mainly concerned with acoustic design, pipe lengths, decoupling volumes, muffler construction and placement, and so on. Almost a completely different field of study, and a very valid one.

The issues being discussed here are more about mechanical resonance and damping of the pipe itself, rather than acoustic resonance and damping of the gas within. The nature of what you hear and feel inside the vehicle can be totally different to the noise spectrum on the outside, especially at a distance.

At a distance the nature of the sound will be almost totally due to what comes out of the open end. But inside the vehicle there can be a whole lot of other close proximity effects that can come from movement of the pipe itself.

 

[GregLocock]

That's a good point. I can't remember for the life of me whether the sound was acoustically coupled, or via the hangers, for the idle boom. I remember doing the test, but not the answer.



Cheers

Greg Locock

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[Warpspeed]

How was the test done ? Was the sound level being measured inside or outside the vehicle ?

If inside, it is probably more direct resonant mechanical coupling into the floorpan. Earlier, you mentioned severe pipe shake associated with this exhaust boom phenomena. That leads me to suspect direct mechanical coupling may be the greater cause. Very difficult to avoid at very low frequencies.

If outide, it would probably be more likely to be an acoustic resonance of the whole gas column within the pipe.

 

[GregLocock]

The SLM was inside the car. I have had problems with acoustic coupling from the tailpipe to the rear hatch on another car, so I'm not going to discount the possibility of acoustic coupling.

Cheers

Greg Locock

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[bukac1]

Try just recently published book by Prof. Wermer Soedel of Purdue University. The name of book consist word Shell(s). Sorry I don't know more details.