Noise enclosure for supersonic nozzle

[ebarba]

Hi all,

we have to build an enclosure to attenuate an SPL of 137 dBA @ 6 ft, coming from a high-temp supersonic nozzle.

A double-walled enclosure with all due precautions and acoustic tricks should provide the 55 dBA attenuation, but we need to reach 60 dBA attenuation and I'm especially worried about noise induced vibrations from the very-low-frequency components of this noise (i.e. everything below 100 Hz).

I think we need to rest the inner shell of the enclosure on low freq. spring suspension. The issue is estimating the vibration level that will be induced on the enclosure walls, floor and ceiling, but we only have octave-band levels between 63 to 4K, nothing on the very low freq. range. Noise Spectrum data provided by nozzle manuf. follows:

Freq. SPL dB
63 91
125 100
250 101
500 112
1000 118
2000 123
4000 126
ALL LEVELS UNWEIGHTED

The nozzle ejects 9 kg/min of exhaust gases (CO2, H2O, N2) @ 1500ºC

Anyone knows what is the noise spectrum shape for this kind of exhausts?

Thanks and regards!
EB

 

[btrueblood]

A fairly decent reference is

http://ntrs.nasa.gov/archive/nasa/casi.ntrs.nasa.gov/20110012036.pdf

I'm surprised the manufacturer can't give you a more complete/finer resolution spectrum. Ask them for the actual data (either time records, or psd's) from the test(s) that was/were used to generate the octave-band data.

Do you know the fuel/oxidizer combination? This can help indicate whether to expect low-frequency "pops and bangs"...hydrocarbon fuels tend to do that more than the hydrazines.

 

[ebarba]

Thanks for the reference! I'll run some calculations according to it and see what comes out.

The manufacturer won't give more info because they want to sell their (very expensive) enclosure...

The fuel is propane and oxidizer is air.

 

[btrueblood]

Ok, that explains the somewhat low temperature. Propane/air is quite a bit less noisy than lox/kerosene...but you may still have some random pops to deal with. Do realize that the knee in the frequency curve is driven by the size of the nozzle throat, roughly related to the thrust level of the rocket (i.e. smaller diameter nozzles have a higher relative frequency peak than larger ones).

 

[ebarba]

The nozzle manufacturer told us yesterday that they source the booth from another manufacturer and they don't know the constructive details. They do know the walls are 100 mm thick, no suspension at all and the noise at 1m from the walls, outside the booth is 85 dBA.

That is one remarkable attenuation for a single wall... if it was purely mass-driven, it would take something like 180 kg/m2 to get to that attenuation. If it's a composite wall (which it has to be), then probably 40 kg/m2, but it need to have something incredibly heavy and very clever elastic provisions inside.

Any thoughts? The costs of the double walled-floating booth are too high for what the customer can pay...