Acoustic pipe insulation/lagging radiation efficiency

[LROGDA]

Hi,

I have tried to find information about the radiation efficiency of acoustic pipe insulation.
Thing is I have lots of good measurement data of the vibrations of insulation covered pipes, from which I need to calculate the radiated sound power. The insulation is ISO class A, it was not covered by an aluminium layer.
I presume the frequency of coincidence is higher than that of steel, but other than that I do not know what it would be.

Any suggestions?

/GDA

 

[GregLocock]

As I'm sure you know, the coincident frequency is when the speed of the bending wave in the structure matches the speed of sound in air. In the case of a lagged pipe I would have thought that the bending wave in question would be almost unaffected by light lagging.

I don't really know how to estimate this stuff. If I were to measure it I would do a modal on the pipe to find the modes of interest, and then use an accelerometer and a close coupled mic to examine the radiation efficiency.



Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376

http://eng-tips.com/market.cfm?

 

[LROGDA]

Hi Greg,

Thank you for the answer.
I presume that since the pipe is sort of driving the lagging, and the vibrations are measured on the lagging, that it is the lagging, which is radiating into the air and not the pipe.

Your suggestion for measuring is also the way I would have done it myself. Unfortunately I cannot just go back on site and do it, and the reason I measured using accelerometers was that the noise field was too diffuse for measuring intensity.

You wouldn't happen to have any suggestions as how to find more information about the subject?

Your inputs are very much appreciated.

/GDA

 

[Tmoose]

ASHRAE handbooks have a batch of info on duct insulation performance.

Is there really a problem with flow noise coming thru the pipe wall? Or is equipment vibration (at frequencies high enough to be audible) being transmitted by the pipe due to improper isolation?

 

[LROGDA]

Hi,

Thanks for the suggestions regarding literature.

I am inclined to believe I am dealing with flow induced noise due to a very broad band noise spectrum. However when looking at a FFT of the noise tonal components are showing. The components can also be heard quite clearly. These I believe are originating at the pump.

I think I can nail the problem, and come up with a solution, but it would be nice to have some precise calculations of the radiated sound powers.

/GDA

 

[Tmoose]

Can you "feel" vibration on the walls or ceiling of the problem areas, or on the surface of the lagging.

I'd try to understand the problem more completely before gathering my possible solutions.

Even if the outer metal skin was installed the insertiona loss of any of the "A" acoustic insulation appears to be minimal below 500 Hz.

http://www.kflexenergy.com/products/acoustic-insulation

And if the whole pipe is not "panting" but wiggling, the lagging has pretty much zero affect.

If the radiated power based on the lagging'surface vibration does not support it as part of the problem, then it would be useful in the early stages.
If the frequencies in the FFT are related to pump impellar "blade pass" The resulting pressure pulsations can make the piping straighten and wiggle if there are any elbows, even if fitted with and if the piping touches any drywall or other interior panels drive them like a huge loudspeaker. Lagging does not reduce that. Sometimes pipe geometry is just right to create organ pipe resonance and over-react to the blade pass pressure variations, and make the straightening/wiggling worse, and even more able to re-appear well up or downstream despite proper isolation and anchoring at the pump.

 

[GregLocock]

Well, as one expert in the field once said, you might as well assume a radiation efficiency of 1 if you are looking at A weighted noise.

That is to say, for normal engineering materials and structures the coincident frequency is going to be in the 500-2000 Hz range, and RE is 1 above that, and A weighting kills everything under 200 Hz, so don't try and finesse things too much. Bear in mind your surface velocities are already attenuated by the lagging, I think the argument is reasonable, if not subtle.



Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376

http://eng-tips.com/market.cfm?

 

[LROGDA]

Hi,

Tmoose: I could clearly feel vibrations in the pipes. The pipes were supported on the floor through vibrations dampeners. I measured the frequency response of these, but there was too much other noise/vibrations to get a good result. I have compared radiated sound power (efficiency = 1 above 1000 Hz) with the measured noise level in one-thrid octave bands, and the shape of the two are quite similar. I cannot compare levels since one is sound power and the other is pressure.

Greg: That is a very good point you are making there. It was also what I ended up doing.

Thanks for all your help and support guys.

/GDA