Data Center Teach-in with Clean Air WNY, North Tonawanda residents, the Allies of the Tonawanda-Seneca Nation, and NYC DSA

There are different types of weighting filters and I will talk about the most common ones in use today, they are dbA and dbC.

The purposes of the filters are to match the human ears frequency response at different sound intensities.

Now clearly humans here do not have a flat frequency response they have a non-linear frequency response.

Let’s understand the frequency response of a human before diving into the filters topic.

The consequence of a non-linear frequency response at different frequency levels is that you don’t hear different frequencies as well as other frequencies so the consequence is that not all frequencies of equal sound, sound equally loud to humans. What it means is you know if all the frequencies from 20 hertz 20 kilohertz are being played at the same amount of amplitude or same loudness they’re all not going to sound equally loud; some of them may sound too loud, some sounds may sound bright, some sounds may sound dull for the human ear some of them might sound too feeble and that’s the result of a non-linear human frequency response to sound. That is because the human ear is most sensitive in the range of 2000 to 5000 hertz, that is the range where we pick up sounds easily. But outside of the range it might be a little difficult now the hearing sensitivity also drops down toward the lower and the higher so we’re not able to hear all the lowest bass or the highest treble.

The first research on the topic of how the ear hears different frequencies at different levels was conducted by Fletcher and Munson in 1933. Until recently, it was common to see the term Fletcher–Munson used to refer to equal-loudness contours generally, even though a re-determination was carried out by Robinson and Dadson in 1956, which became the basis for an ISO 226 standard. ISO stands for International Organization for Standardization, which are based on a review of modern determinations made in various countries.

Fletcher and Munson first measured equal-loudness contours using headphones in 1933. In their study, test subjects listened to pure tones at various frequencies and over 10 dB increments in stimulus intensity. For each frequency and intensity, the listener also listened to a reference tone at 1000 Hz. Fletcher and Munson adjusted the reference tone until the listener perceived that it had the same loudness as the test tone. Loudness, being a psychological quantity, is difficult to measure, so Fletcher and Munson averaged their results over many test subjects to derive reasonable averages. The lowest equal-loudness contour represents the quietest audible tone—the absolute threshold of hearing – dashed green line. The highest contour is the threshold of pain.

Churcher and King carried out a second determination in 1937, but their results and Fletcher and Munson’s showed considerable discrepancies over parts of the auditory diagram.

In 1956 Robinson and Dadson produced a new experimental determination that they believed was more accurate. It became the basis for a standard (ISO 226) that was considered definitive until 2003, when ISO revised the standard on the basis of recent assessments by research groups worldwide.

The NT city code for noise was last updated in 1988 and does not reflect this new standard, however the noise meters used to measure the sound do reflect this new standard.

The generic term equal-loudness contours is now preferred, of which the Fletcher–Munson curves are now a sub-set, and especially since a 2003 survey by ISO redefined the curves in a new standard.

The key things to take away from here is that in the bandwidth of two thousand to five thousand hertz we do observe that the curve drops down whereas in every other case it just shoots up which proves that human ear responds to frequencies heavily in the 2000 to 5000 range and poor in the treble and very poor toward the base.

However at very high decibel readings the curve flattens out and becomes more linear as at very high decibel readings, humans can hear the low frequency sound almost as well as the other frequencies.

In order to overcome those limitations, these dbA and dbC filters were developed.

Let’s dive into the difference between dB(A) and dB(C) when measuring sound.

They filter out certain frequencies to show how sound is as a human would perceive the sound, and your current noise code is based on the EPA guidance from the 1970’s.

A weighting filter; dbA.

1. • dB(A) stands for A-weighted decibels.
   • It’s a measurement scale that adjusts the decibel level to account for the sensitivity of the human ear to different frequencies at low volume.
   • The A-weighting curve emphasizes frequencies between 500 Hz and 10 kHz, which are the frequencies most relevant to human hearing.

The A weighting filter filters out significantly more bass frequencies compared to other frequencies and is designed to approximate the ear at around the 40 decibel level, a quiet night in the suburbs.

A-filters are very useful for eliminating inaudible low frequencies so what the filter does is it filters out the base region in the spectrum so it correlates with what a human ear would hear at a low volume. The North Tonawanda city code references and uses the dbA or decibel A filter in determining noise violations.

C weighting filter

1. • dB(C) stands for C-weighted decibels.
   • This scale is useful in situations where the full frequency spectrum needs to be considered, such as assessing industrial machinery noise or evaluating sound in music production and audio engineering contexts.

The c weighting filter differs from the filter in the fact that they filter less of the lower and higher frequencies.

The c filters approximate how the human ear hears at very high sound levels.

Now one thing to note is that at very high sound levels there is no disparity between the lower and the higher frequencies; all the frequencies almost sound equally loud at very high sound levels. So unlike the “a” filter, the “c” filter follows the 100 decibel loudness curve.

It is used in sound measurement of especially loud and noisy environments and it is expressed in dbc.

In summary, dB(A) and dB(C) are two different ways of measuring sound levels. dB(A) is weighted to match human hearing sensitivity, making it suitable for assessing noise exposure in low noise level human environments, while dB(C) measures sound levels across all frequencies equally, making it useful for loud noise industrial applications.

Other bits of information to consider are:

For every doubling of distance from the noise source the decibel readings fall by 6 decibels.

For every 10 decibel decrease the sound hears half as loud.

Weather, wind and atmospheric conditions play a role in how far sound travels and the quality and loudness of that sound.

If you are upwind from a sound source that sound will sound softer as the wind bends the sound waves upward over your head.

Download (PPTX, 1.87MB)

Wired Magazine – The Trump Administration’s Data Center Push Could Open the Door for New Forever Chemicals

TruthDig – The Ecological Cost of AI Is Much Higher Than You Think

The Verge – Republicans are looking for a way to bring back the AI moratorium

The Verge – I looked into CoreWeave and the abyss gazed back: meet the company Nvidia is propping up.

The Center for Public Enterprise – Bubble or Nothing

Gizmodo – US Approves $1B Loan to Restart Three Mile Island, as Microsoft Data Centers Drive Demand

Inside Climate News – Members of America’s Largest Power Grid Can’t Agree on How to Power Data Centers

Gothamist – NY talks over Tesla Buffalo plant sputter in ‘fits and starts’