Enclosure noise attenuation assessment

[ebarba]

Greetings,

We must assess the noise attenuation of a sound enclosure designed to contain a 135 dBA noise source with a reduction of 50 dB. The noise source is not yet active and won't be until late September, but an assessment of the noise enclosure is necessary for payment purposes.

A preliminary noise survey measured the following levels:
[ul]
[li]Noise OUTSIDE the booth: 65 dBA[/li]
[li]Noise INSIDE the booth (door closed): 35 dBA[/li]
[li]Total reduction: 30 dB (nowhere near the 50 mark).[/li]
[/ul]

Here goes the question: should I worry about the above, or is the apparently poor performance due to the booth not being designed to attain low noise levels inside?. What I mean is that the booth doesn't have the kind of provisions you would see in a semi-anechoic chamber, such as floating floor, resilient mounts between main structure and concrete floor of the warehouse, etc.

At the moment there is no way to test the booth with a noise source inside, unfortunately.

Some facts about the noise enclosure:
[ul]
[li]The booth has only two apertures, 24x12 in each. Each aperture leads to a silencer rated for 55 dB, one silencer is for the inlet air flow and the other one is for the outlet air flow.[/li]
[li]There are no windows.[/li]
[li]All walls and the roof are made of a double layer of special acoustic panels. The combined weight of both layers is 10 psf[/li]
[li]The panels have one noise-absorbent face, both layers have the absorbent side facing inwards the booth.[/li]
[li]There is a 1" cavity between both layers of panels and they are fastened to the main structure with C profiles with resilient neoprene between the panels and the profiles (there are no studs in the middle of the walls)[/li]
[li]The smallest wall is 13x10 ft, the largest is 16x10 ft[/li]
[li]The roof is 16x13 ft, double layer, made in two sections 8x13 ft each. It too has no studs separating the inner and outer layers[/li]
[li]The combined wall materials have a rated noise reduction in excess of the 55 dB as per lab certification[/li]
[li]The door is also rated more than required, lab-certified.[/li]
[/ul]

 

[IRstuff]

It's worse than that. Your booth is possibly not even related to the noise source you are trying to contain. Since you have your design specification (presumably), you should be able to make measurements to determine whether the booth even remotely represents the sound environment you are trying to attenuate.

How did your company go on contract without any independent means of verification?

TTFN
I can do absolutely anything. I'm an expert!
faq731-376 forum1529

 

[Strong]

"At the moment there is no way to test the booth with a noise source inside, unfortunately."
Can you use speakers inside enclosure? The outside-to-inside measurements are useless!

Walt

 

[ebarba]

It's worse than that. Your booth is possibly not even related to the noise source you are trying to contain. Since you have your design specification (presumably), you should be able to make measurements to determine whether the booth even remotely represents the sound environment you are trying to attenuate.

Can you please elaborate a little further? Why you suspect the booth is not related to the noise source it will contain?

How did your company go on contract without any independent means of verification?

The customer is (very) late with the plant that goes inside the booth, so it was agreed that if we provide reasonable evidence the booth works, then they will fulfill the payment.

Can you use speakers inside enclosure? The outside-to-inside measurements are useless!

Technically, yes... but it becomes a task in itself: we have to find and rent a very powerful amplifier and speakers, make an "artificial" noise signal that has the spectrum of the real source and finally play it at (or near) the level of the real source.

We've done it previously, but It will take time and we don't have it right now.

 

[IRstuff]

"Can you please elaborate a little further? Why you suspect the booth is not related to the noise source it will contain?:

> You've answered that question in your last response; since you are not simulating the noise frequency content nor its amplitude, what proof could you possibly offer that the system meets its requirements? Moreover, you don't meet your design requirements, so that's a bust to me. Your only hope is that your frequency content is radically different than what you designed to and that the correct frequency content might show better performance. Moreover, the usage of an external source is only marginally useful, since you cannot duplicate any sort of resonance or other contact phenomenon that the actual source might have.

"The customer is (very) late with the plant that goes inside the booth, so it was agreed that if we provide reasonable evidence the booth works, then they will fulfill the payment."

> That wasn't my point. My question was how you were going to prove to yourselves that the system was working before you laid everything on the line and potentially failed the acceptance test.






TTFN
I can do absolutely anything. I'm an expert!
faq731-376 forum1529

 

[GregLocock]

It sounds like lawyer time to me. Specifically are you required to deliver -50 dBA from the source, or 85dBA outside the enclosure, or what?

Cheers

Greg Locock


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

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

 

[ebarba]

Greg,

Max noise outside the enclosure must be 85 dB

 

[GregLocock]

In that case an operational test is the only realistic way forward as the sound source needs to replicate the impedance as well as the spectrum of the final equipment. I must admit if you need 85 dBA outside then 55 dB attenuators sound a bit marginal to me as it ignores leaks and structure borne noise.

Cheers

Greg Locock


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

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

 

[ebarba]

> That wasn't my point. My question was how you were going to prove to yourselves that the system was working before you laid everything on the line and potentially failed the acceptance test.

There is no practical/cheap way to test walls or roof this large, before they are actually built on site. We rely on past experience and good engineering practice to design our noise enclosures. To be certain that the performance will be as expected, we use certified materials, so it all boils down to make sure everything is mounted and installed the way it was designed to be.

There is however something that's bothering us: this time we decided to outsource the door, but normally we build our own. I have a gut feeling this one is not what they promised. Mind you: it is a third-party certified acoustic door... but my noise meter showed 5dB more noise in front of it than in front of the walls. The door in question is very large, taking up about 9% of the total enclosure surface, if it is indeed flawed, the whole enclosure performance will be much lower than expected.

Then again, the door side that faces inward the enclosure does not have acoustic absorbing materials, as opposed to the walls. This could explain the 5 dB difference in attenuation.

 

[ebarba]

I must admit if you need 85 dBA outside then 55 dB attenuators sound a bit marginal to me as it ignores leaks and structure borne noise.

This issue was addressed by making both layers of the walls/roof as structurally independent as possible. Originally we offered an enclosure within an enclosure, with the inner one being mounted on air springs, but the manufacturer of the noise source showed that a floating inner booth wasn't necessary: they have the same equipment in their facility with walls made of a single layer of 4" acoustic panels around it, no silencers in the inlet and outlet streams and they meet the 85 dB requirement.

We have a composite wall made of two 2-1/2" panels and a cavity between them. On paper we're good to go, but the door gives me a bad feeling...