sound Power Conversion for Large plant

Thread view:

Enable EMail Alerts

Start New Thread

Thread rating:

Tom Harper - 23 Oct 2004 09:02 GMT

I would like to ask about methods that are used to convert measured
SPL of a large plant item to a SWL for that item. For eg a generator
or chiller.

In a perfect world the manufacturer would be nice enough to give you
results in SWL, ahhh how I wish for the perfect world.

often all that arrives is measurements at 1 metre from the side of the
plant item. So it is usually move on with some calcs or stop and argue
for a month and then get no more info.

As the plant can't be a point surce at that distance my usual method
is to base the SPL to SWL correction on SWL = SPL + 10 x Log (Surface
Area). However, this always (in my opinion) overestimates the actual
SWL.

Does anyone have an alternative method that arrives at a more
reasonable correction?

Regards,

Tom

Reply to this Message

Brian Marston - 23 Oct 2004 16:31 GMT

Tom

Sound pressure levels at 1 metre are usually a hearing conservation level
measurements (only). Unless the object is fully defined, then a sound power
level calculated this way will be a crude guesstimate. I've had these arguments
with suppliers on a number of occasions ~ and usually won.

If you over-estimate ~ how much extra are you costing the client?

If you under-estimate ~ how much extra to rectify the mistake?

Once had a client who wanted to use buffer zone instead of noise controls. Land
acquisition costs to cover the 5dB(A) excess past his boundary came to
$5,000,000. (Yes, he did install the equipment mufflers).

Another client decided to design and build his own machine enclosure ($20,000),
only to find the noise increase. It cost him a further $40,000 to remove his
bargain enclosure & build the enclosure initially recommended.

If the supplier won't or can't supply what you want, the alternative is to pick
up your noise meter and go forth and measure.

Brian
CPEng MAAS

PS Are you in still in the Singapore branch office of V....

> I would like to ask about methods that are used to convert measured
> SPL of a large plant item to a SWL for that item. For eg a generator
[quoted text clipped - 18 lines]
>
> Tom

Reply to this Message

Angelo Campanella - 24 Oct 2004 22:26 GMT

> often all that arrives is measurements at 1 metre from the side of the
> plant item. So it is usually move on with some calcs or stop and argue
> for a month and then get no more info.

ISO TC43 SC1 in all its wisdom is now crafting ISO 3747 which will wrest
the sound power value from such nearby SPL measurements in a reverberant
space.

The precision and bias statement for 3747 is suitably (very) loose,
being several dB, which I think we all agree is about as 'good as it
gets' under the circumstances.

> As the plant can't be a point surce at that distance my usual method
> is to base the SPL to SWL correction on SWL = SPL + 10 x Log (Surface
> Area). However, this always (in my opinion) overestimates the actual
> SWL.

This is how 3747 does it, but

> Does anyone have an alternative method that arrives at a more
> reasonable correction?

The correction is in the form of "K2" which utilizes the reverberation
time of the test room to generate the K2 value. Where I ran amuck of
this is in deciding the real value of the test space volume, since it
was BIG (200'Lx20'H, x 60' or 120' depending on how one interpreted
'coupled spaces'; a factory assembly floor). I think that my 60' or 120
quandary places an added 3 dB uncertainty...

I think that, otherwise, 3747 can be as tight as probably +-2 dB if you
have a clear volume to deal with.

Ang. C.

Reply to this Message

Natural Science Forum / Physics / Acoustics / October 2004

sound Power Conversion for Large plant