What exactly does Dn,e W dB indicate?April 8, 2011 12:10 AM   Subscribe

What does Dn,e W dB indicate exactly? It is a measurement that refers (apparently exclusively) to the acoustic attenuation performance of "trickle vents".

I have been asking everywhere to try to find an answer to this.

It is a measurement that refers (apparently exclusively) to the acoustic attenuation performance of trickle vents, such as, for example: http://www.greenwood.co.uk/data/DN%20Vent%20data%20sheet.pdf

I suspected it indicates that it is e-weighted sound power level, but that would be a bizarre way to measure this. What the Dn stands for I have no idea.

Googling turns up nothing except more pages of manufacturer's trickle vent product pages.

In the sheet I linked to above the sound attenuation curve indicates that with this weighting curve, whatever it is, results in an attenuation of 39 dB rating at 125 Hz, which doesn't really make any sense. That's a lot higher than a "soundproof" staggered stud wall's a-weighted rating (about 20 dB).

Anyone?
posted by Nish ton to Technology (4 answers total) 1 user marked this as a favorite

This paper (page 9) defines Dn,e as the element normalized level difference, which matches the vertical axis of the plot in your datasheet. It's not a simple attenuation amount; that document goes into detail but it's a bit over my head. They give an equation for relating Dn,e to the sound reduction index on page 10.
posted by Rhomboid at 1:17 AM on April 8, 2011 [1 favorite]

On the first page of that same datasheet, they say:
The external unit, manufactured from ABS plastic has internal silicone fins that provide acoustic attenuation to a level of 37.7dB(A).
This seems to describe the graph on the second page. The units are logarithmic, decibels (dB), so a factor of 30 would be 1,000 times below ambient acoustic noise (over some assumed frequency range). A factor of 40 would be 10,000 times below ambient.

The use of W dB would just clarify they are measuring a power instead of an amplitude, which could mean the difference of a factor of 2 in the dB rating.

I think, like you do, that their rating is bullshit. Looking at the picture of that thing, I can't imagine how it would perform so well. Maybe they include the effect of some awesome window it is already mounted in.
posted by fatllama at 6:09 AM on April 8, 2011 [1 favorite]

Yeah, if only I could get access to the standard they use for doing the measurement, which is mentioned in the PDF Rhomboid posted (ISO 701-1 or something).... but standards cost serious \$\$.

In the same PDF there is a reference section below the other stuff that says Dn, e, W means the element normalized level difference, weighted. The problem is that they keep swapping *curves* for *numbers*. If you lookup sound reduction index on wikipedia they explain that the weighted level difference is a *single number* which is arrived at in a similar way to the Sound transmission class, which is used instead of STC in the UK etc.. which makes sense because if you google "trickle vent", they seem to only be used in the UK (due to differences in the building codes).

They give some equations which clarify muchly in that PDF, so that is a great document, thanks for it Rhomboid.

Basically it seems like you take the element normalized level difference curve for the test specimen, which refers essentially to the degree of attenuation of the *sound power *level** (which is watts per square meter (sound power)of sound mapped onto a logarithmic scale(sound power level), the element normalized part refers to normalizing to the area of the element compared with a reference area, which arises by virtue of the fact that it is watts per *meter*) of the sound signal that the specimen is in the path of, in decibels, for a range of frequencies.

Then you compare it to the reference curves as described in some ISO standard ( and outlined in the wikipedia article on sound reduction index), and pick the curve that "fits" according to certain criteria, similarly to how is done for STC. That does the weighing part and converts it to a single number (the number of the reference curve) in one go...

It's not that acoustics is actually hard, it's finding the documentation that describes things accurately that is the hard part...

BTW, while it is a measurement based on sound power, if you check out the equations in the wikipedia articles on sound power level and sound pressure level, you can see they are roughly the same, not half.

Thanks everyone though, both best answers.
posted by Nish ton at 3:02 AM on April 9, 2011

Sometimes with a little creative googling (and especially with the "filetype:pdf" google operator) you can find copies online. Here's ISO 140-10 for example. (It's locked behind a Flash PDF viewer but if you create an account and then upload something apparently you will then have enough credits/points for the download link to work.) I'm not sure how helpful that is as it seems relatively short and it seems to reference the other parts a lot.
posted by Rhomboid at 8:04 AM on April 9, 2011

« Older I might be wrong, but I can't imagine a banjo...   |   Grade This! Newer »