Noise measurement of equipment.

[alansimpson]

I am working on the design of office computer equipment with a view to reduce noise level.
I think noise levels would be 30-60 dBA. Not sure of frequency levels.

I would like some advice on choosing a noise meter and what features to go for.
There are many meters on the market ranging in price from 50 to over 1000

Thanks in advance

 

[MikeyP]

If you want to quote noise levels for your product, then you have to be able to measure the noise to some recognised standard. The standard will tell you the quality of measurement equipment you require and unfortunately it is likely to be at the expensive end of the market (probably "type 1")

You will also need a suitable environment in which to measure the noise - this probably means renting an anechoic chamber.

M

--
Dr Michael F Platten

 

[SomptingGuy]

... or on an exterior hard surface with no nearby buildings and a low background noise level. Type 1 Sound Level Meters are not cheap to buy, but they can be rented.

- Steve

 

[GregLocock]

Depends on whether you want to quote noise levels, in which case the standard you are working to will tell you what you need, or are trying to improve the design and undertsand what is going on, in which case you need to consider the frequency content and the directivity of the noise the thing makes, not just quote a sound pressure level at 1m, or a sound power in dBA.


Cheers

Greg Locock

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

 

[alansimpson]

I am still working on this problem.

It has been decided to look at noise levels of components individually. This means measuring even smaller noise levels.

I do not need to work to a specific standard. I just need to judge if I am reducing acoustic noise level and comparing performance with other similar products.

So far testing is very subjective. I record sound levels under similar conditions. Do spectrum analysis with Acoustic software and judge if there is any improvement.

Could anyone recommend a more quantitive test method?

Thanks again

 

[GregLocock]

Sounds to me like an acosutic intensity survey would be a logical next step. The greta thing about that is that witha bit of diddling you can get useful results even from a pair of cheap mics, although it is much nicer to use phase matched mics with a flat frequency response. It all depends on whetehr your time is worth more than the instrumenation. You should be able to hire an acoustic intensity analyser if you don't feel like rolling our own.

On the other hand just working out what frequencies dominate and then searching for the source with the mic is a perfectly sensible approach that avoids all the maths and many puzzles.

Photos, plots etc would probbaly yield more helpful advice.





Cheers

Greg Locock


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

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

 

[reidh]

Depending on your location, it may be easier and cheaper to send your components to a test lab with an in-house anechoic chambers and staffed with engineers. I know that many of the test labs that formerly catered to the automotive industry are more than willing to branch out to other industries to keep their business going.

Also, a comment on the OP. Noise signature levels are typically specified as XX dB at YY distance (e.g. 60 dB at 10 meters.)

Regards,

Reidh