Take off 6dB (or more) from a noisy source 1

[Highspeed]

I have a 0.5 liter volume available to take off 6dB (or more) from a noisy source (motor+fan)with the following constraints/environnement.
-Noise to be taken off at air inlet
-Pressure drop < 2hPa for 1L/s
-noisy source : 42 dBA of broadband noise + F0 + 2xF0 (not dominant)
-F0 can be anywhere from 250Hz up to 450 Hz
-The small available space doesnot easily allow any Hellmoltz or 1/4 wavelength based design.
So far,I have tried a perforated panel design with only a 4dB absorbtion coefficient.
Any tips or litterature that could help will be welcomed.
Many thanks.

 

[GregLocock]

What shape is the spectrum? what frequency range? Post a plot if you can.

Failing that try a labyrinth of semiporous foam. At least make sure that there is no direct path from outlet to inlet.

Cheers

Greg Locock

 

[Highspeed]

Thanks Greg,
How can i post a spectrum plot easily?
I'd like to avoid classical semi-porous foam as this device can be used in high RH environements
are you aware of any material that can have some sound absorbing propery and still be waterproof?
Cheers

 

[GregLocock]

Post a jpg on a website somewhere, then check out the instructions for posting it which can be seen if you preview a post instead of submitting it. or just post a link.

I don't know much about specific absorbent materials, it has been a long time since I had the dubious pleasure of playing with that sort of thing.





Cheers

Greg Locock

 

[Rob45]

For posting jpegs, one free site is &quot;photoisland.com&quot; primarily used for digital photo storage, but they'll &quot;give&quot; you 10 MB of free doc or jpg storage that you can link to.

 

[sirmick]

Why do you think a 1/2 liter volume is too small for a Helmholtz resonator? What frequencies are you trying to reflect? With a Helmholtz its my understanding that the dimensions are to be substantially less than the wavlength for it to be effective.

 

[Highspeed]

In fact time has spent from the original post and you are 100% right a helmholtz resonator actually works, but it's a bit hard(for me) to tune ( by damping) to cover the wide operating frequency range but I am working on it and it's mostly experimental, currently 3 designs are competing :
-Classical silencer (expansion chamber) + few damping materials ... 4dB
-Helmholtz : 4 dB
-Active noise cancellation coming from an old design : 7-10 dBs
As there is some cost involved I am hoping to have one of the 2 first guys work (2dB's left...they are the toughest)
In parralel the noise source is also worked out.
Cheers

 

[sirmick]

I have just started reading on acoustics so I may be totally wrong, but wouldn't a Helmholtz tuned for the lower frequencies combined with an expansion chamber tuned for the higher frequencies work? Though it is my understanding that these elements must be some difference apart.

BTW, what are you making?

 

[Highspeed]

It's an xtra small medical ventilator
Currently in the enclosure it's doing 27dBA noise @ 1m (not that bad though)
The deal is to go down to 25 dbA where 80% of the people and 99% of the patients will not hear it.
The other part of the deal is not to have any open cell foam in the air path.
The expansion chamber after the Helmoltz is hard to fit due to the tubing size requiring at least 0.2L or I'll have to give up a few hPa in pressure drop thus efficiency...
Cheers

 

[sirmick]

No open cell foam? How is the device being filtered? With pleated paper or something?

 

[Highspeed]

Sorry, There is a disposable air filter at the inlet.
BUT I am trying not to have any acoustic foam in the air path inside the device and that is where the challenge is....

 

[GregLocock]

Highspeed - check that you are not getting case radiated noise from your silencers. This can be a significant issue.


Cheers

Greg Locock

 

[Highspeed]

You are right Greg, I dealt with that a week ago, a good suspension of the silencer helps a lot, what I need to do is to qualify the air gap really needed between silencer and housing
optimized air gap ... more volume ... more efficiency in theory ..
And suspension + a sealed housing is a lot better than trying to calm down the silencer case with foam as this one is loosing efficiency.
1.5 db to target---Close to absolute silence

 

[Highspeed]

I finally got the design to work over expectations
You cannot tell when it is on.

I did find a really great enclosure damper somewhat &quot;magic&quot; as opposed to what I tested before and wanted to let you know:
Manufatured by TMAT it's named P15/D05.
Very usefull for plastic enclosure
HS

 

[dpeters13]

Armaflex makes a closed cell insulation that will not absorb water or moisture. It has some sound attenuating charagteristics but it is not as good a sound attenuator as fiberglass

 

[Rob45]

Sounds like your problem is solved, but Sirmick raised a question about Helmholtz tuners that bears answering:
many texts don't mention that there's a minimum volume requirement for a Helmholtz tuner to work:
in Imperial units, this minimum volume must be such that the product of the tuning frequency and the tuner volume is at least 18,000 "inch-Hz".

So for the 250 Hz firing order noise in "Highspeed's" case, the minimum volume requirement would be 72 cubic inches, or just about 1.2 liter.

Just something to keep in mind.