Sound speed in gas pipes

[maxh]

Hi Guys,

An interesting question for you.

Does the sound speed of air (or any other gas) contained in a pipe vary from what you would calculate in from the typical formula v = (k.g.R.T)^0.5 ?

Where
v = sonic velocity
k = Cp/Cv
R = Universal gas constant
T = Absolute temp

Would the size and dimensions of the pipe make a difference to any aspect of the calculation, I know it does in liquid filled pipes for hydraulic surge calculations ?

Would pressure make any difference ?

Does anyone know of any published work on this subject ?

 

[GregLocock]

Would the size and dimensions of the pipe make a difference to any aspect of the calculation, I know it does in liquid filled pipes for hydraulic surge calculations ?

yes,

Would pressure make any difference ?

yes

Does anyone know of any published work on this subject ?

yes, anything from 1880 onwards

What are you really after? are you a student?

Cheers

Greg Locock

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

 

[maxh]

Sorry, am only a student of life these days

Let me re-phrase the question.

If I take two microphones, set them apart by a known distance and measure sound speed between them, using a single pulse from a speaker and measuring the time lag, will I get the same answer, if I then take those same two microphones and put them the same distance apart inside a straight piece of pipe and repeat the sound speed measurement ?

I can find some good info on speed of sound on the internet and in books, acoustical journals etc, but I can't find one that deals explicitly with my situation - even if its just to say - that's a silly question of course you get the same answer, or, of course you don't get the same answer !!

Someone, somewhere must have already done this work and published their results and that's what I am after, I have no wish re-invent the wheel.

 

[GregLocock]

Oh, OK.

To a reasonable approximation the speed of sound is unchanged inside a pipe. That is to say, you use the same speed of sound inside an organ pipe as in the air around it.

I don't know what I was getting at in my preceeding post, I'd obviously thought of an example where it makes a difference. Probably I was thinking about elastic pipe walls that could couple into the motion, which would affect the k/m part of the equation.

Cheers

Greg Locock

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

 

[sms]

Might try Engineering Dynamics and Southwest Research Institute, both have nice websites. Search under pulsation and reciprocating compressors.

Sound speed in gas is pretty well unaffected by pipe size.

-The future's so bright I gotta wear shades!
Please see FAQ731-376 for tips on how to make the best use of the Eng-Tips Forums.

 

[Rob45]

Depending on the length of the pipe, and to an extent the diameter, the speed of sound can vary in a long pipe discharging into atmosphere due I think to frictional effects.
I remember seeing a technical paper on this topic some years back, but it was in Japanese and I don't remember all the details.
This is the source of the exhaust pipe phenomenon sometimes known as "schoolbus bang" or some such - I forget the exact term. When the pipe is long in relation to the speed of sound, pipe frictional effects and reflected pulses can combine to produce a Mach 1 flow locally.
Certain U.S. domestic cars such as the GM 'J'-car family, when the manufacturer was too cheap to design a good exhaust system, produced this noise when cold (only), and it was very loud, very annoying.

The noise went away as the pipe warmed up and the speed of sound in the pipe increased; a good solution would have been to use a mid-system resonator - giving effectively a shorter pipe - but GM were too cheap to do that.
So their customers bought Camrys...

 

[SomptingGuy]

One of these perhaps?


Y. Nirei, T. Konomi, H. Nohira, “Study on abnormal exhaust noise caused by shock wave and its improvement”, Toyota technical review 41, 29-34, 1991.

M. Okada, S. Morikawa, T. Abe, “Analysis of the generation mechanism for abnormal exhaust noise”, JSAE Review 8, 34-39 1987.

K. TAKAYAMA, O. ONODERA, N. SEKINE, K. ITO, S. MATSUMURA 1988 Society of Automotive Engineers 880082. Shock wave development and propagation in automobile exhaust systems

 

[GregLocock]

Oh, good one. Yes, there is also a noise in the downpipe when it is too small in diameter, which we used to call the raspberry noise- typically found on small cars that we turboed. We assumed it was chaotic flow due to high M numbers, and shock waves (it sounded like a shock wave). Cheap cure was twin wall downpipe. Proper cure was bigger OD exhaust.




Cheers

Greg Locock

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

 

[SomptingGuy]

I've managed to simulate the rasping noise too, hence my interest.

 

[Rob45]

Somptingguy:
Yes, those were the papers I was referring to: the phenomenon is commonly called "schoolbus rap" and it doesn't have a thing to do with scowling guys wearing their hats backwards...

It was as I recall a result of relatively small diameter pipes that were long with respect to the speed of sound; as the exhaust gases heat up (the vehicle warms up) the threshold pipe length for this to occur becomes longer until (hopefully!) it exceeds the system pipe length, at which point the "rap" will no longer occur. The common solution is to shorten the effective pipe length by inserting a silencer of some sort or simple expansion chamber.

I don't know if the JSAE paper by Okada et al is available in English, but it gives a good explanation of the phenomenon - or rather, Matsuyuki Okada himself gives a good explanation of it!
Regards,
- R

 

[Strong]

Regarding the two microphone method. This method works well for pipe diameter small enough to ensure plane wave sound propagation. If the pipe diameter is large, then the sound path may not be direct from one microphone to the other, and you will get a false sound speed value. Sound speed physics is well understood, but it is easy to screw up an experiment or physical measurement.

Walt