Reflective Noise

[redgdon]

Is it possible to more than doule the noise gain from reflective sound.

We have some plant housed in an enclosure. Manufacturer rates the equipment at 78dbA +/- 3dbA at 1 meter free field. The measured value inside the enclosure is in excess of 85dbA.

Does this mean that the plant is generatin more than 81 dbA or is it possible for a gain of more than 3dbA from reflective noise.

Any help would be appreciated.

 

[SomptingGuy]

Absolutely. Each reflection plane can be treated as another source. Total sound at your ear (or at your microphone) is the linear sum of the real and reflected sources. Note that each of the reflected sources will have a different path length to your ear, so there will be interference between the direct and reflected waves.

 

[MikeyP]

You can also think of it in terms of resonance. The space inside the enclosure has many acoustic resonances at many frequencies. Resonances (acoustic modes) amplify the sound beacuse they store the sound energy. The amplification factor of each resonance can be considerably greater than 2.

M

--
Dr Michael F Platten

 

[phanikumar]

Dr. Platten, Assuming the Sound energy is exciting resonances at many frequencies in a particular structure, how many resonant modes can we expect it to excite?, since a structure will almost have infinite modes the Sound power level should be able to excite only first couple of modes. Since higher frequencies need significant energy to excite even to get small amplifications, beyond a point modes become irrelevant as there will be hardly/no amplification observed. So how many modes should we be concerned about?. If hysterisis damping of structure and viscous of the media are a factor how can we claim the resonance to be considerably greater than 2? should we consider only all the modes whose amplification would be >=2 ?

 

[GregLocock]

The sound power from the source will excite resonances at every frequency in its spectrum.

I strongly dispute the first couple of modes figure you suggest, indeed one way of thinking about the high frequency response of a room is to consider the modal density, ie the number of modes per third octave. Even though we treat that as a lumped repsonse, in reality each resonance is still being excited.





Cheers

Greg Locock

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

 

[GregLocock]

Also the original poster is in danger of confusing sound power levels and sound pressure levels. What size is the enclosure? Where were the microphones?



Cheers

Greg Locock

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

 

[redgdon]

Thanks guys, appreciate the help.

Greg,

The enclosure is approx. 10m x 5m x 4m, with duty/standby compresors inside. I am not sure of the construction of the enclosure, but it is skined with sheet steel, inside and out.

The microphone was approx. 1m from the duty compressor.

 

[SomptingGuy]

Practically speaking, it sounds like you need some foam lining in your enclosure if people need to go in it.

 

[MikeyP]

phanikumar,

The factor of amplification by a resonance is the same whether you excite it with -160 dBA SPL or 160 dBA SPL. If there is energy at the frequency of the resonance then it will be excited (except in the academic case of a point source positioned at a nodal point).

How many modes should you consider? Well, all those whose natural frequencies fall within the range of your SPL measurement. At higher frequencies where there are many modes it no longer makes sense to consider each mode individually. You can get around this by treating all the modes in a particular 1/3 octave band as a single degree of freedom whose resonant frequency and damping properties are the mean of all the modes in the band.

redgdon,

The plant rating of "78dbA +/- 3dbA at 1 meter free field" is effectively a measure of the sound power of the plant. The average SPL in your enclosure will depend on the sound power (more power - more SPL), the absorption inside the enclosure (usually expressed in terms of the reverberation time, more absorption or shorter T60 - less SPL) and the volume of the space inside the enclosure (a bigger space means that the modal energy is spread out over a larger volume and hence the energy at a particular point is less, however a bigger volume also implies more resonant modes!)

The +/- 3 dBA bit is there to account for the fact that the plant does not emit sound equally in all directions. When you take your SPL measurement, you need to measure the SPL at a number (10 say) of RANDOM positions in the enclosure (steering clear of places close to walls or the plant) and then average them logarithmically. Don't feel that you have to take your SPL measurement at 1m from the plant.

M



--
Dr Michael F Platten

 

[Rob45]

Think of the compressor as a source of sound power; when it's in a free field, you are measuring principally the sound coming directly from the source, but all the rest is escaping. When you enclose it and prevent the energy from escaping, the sound pressure level within the enclosure increases.
So as suggested above, in order to reduce the sound level within the enclosure, you need to introduce some means for dissipating the energy; foam or fibreglas for example.