Sound Quality Noise Measurements
The need to properly identify and measure "noise levels" in some objective way has become a more necessary task for many acoustic engineers. The overall frequency weighting networks commonly used such as "A' or 'C' are at best an approximation of the effect heard by a human being to complex sound patterns. A more thorough means is required to document subtle changes or differences in sounds between two or more sources. Sound quality measurements are becoming an increasingly more important part of our lives. complex sounds from many sources
 multiple noise sources inside a modern vehicle
Traditional "single figure" measurements of the overall effect of various sounds use the 'A' weighting frequency response in the majority of cases. This is at best a simplified method and at worst can hide some quite important differences in our hearing ability. The main problem with the A scale is that it is a fixed correction that is applied to all spectral sound levels regardless of the absolute level of the frequency content in the sound. Our ears are not linearly responsive to noises or sounds as the level changes. Indeed, there are very dramatic changes to our hearing sensitivity as the noise levels increase. This was the original justification for the development and use of the three early broadband curves known as the 'A', 'B' and 'C' scales on sound level meters. Over the years many regulations and standards have been introduced that require only the 'A' level to be measured but then such small differences exist between noises that are considered acceptable and un-acceptable that it is very difficult to quantify what is considered the limit of acceptability. the non-linearity of the human hearing process
 equal loudness curves showing the non-linearity of the human hearing process
The problem arises typically in such situations as a customer's response to the "sound quality" inside the latest high cost SUV or car that they have just bought. The slightest squeak or hum or rattle that can be heard prompts complaints and endless searches and investigations by the specialists in the vehicle company to try and solve a seemingly un-measurable problem. Similar problems exist in the world of general consumer white goods and other other appliances that we bring into our homes. We all expect to come home from a hard day at work to a peaceful and relaxing environment without the dishwasher making an annoying noise or the latest power tool causing a nuisance to our neighbor when we have to put up a new shelf in the kitchen. In all of these (and many other similar) situations the problem is the complex nature of sound when we try to take into account the annoyance or the underlying quality that we hear. an example of a typical noisy sound in the home
 nuisance noise sources in the home
Straight Third octave band analysis will provide discrimination in the frequency domain but will still give us at least 30 answers to try and compare between noises that are from sources that are acceptable and maybe not so acceptable. This is not an easy task. The concept of the "Loudness" of the sound is a very important feature that has traditionally been confined to the well equipped acoustic research laboratory up until now. It involves measuring the frequency response in either octave or third octave bands and then combining the band levels depending upon the relationship between the dB levels and their position across the frequency spectrum. In essence it produces a single figure number like a variable 'A' weighted reading might do but based on many years of research into the complex nature of people and their reaction to tones and the level of some sort of "equality" of the sound. The units produced by the loudness calculations are called "sones" and they are linear such that 10 sones sounds twice as loud as 5 sones. The benefit of measurements in sones in the practical world is now that a sound can be described "objectively" rather than "subjectively" by an expensive trained listener. Phons are another unit that is commonly used for this type of work. measuring the loudness of a sound
 measurement screen from a CEL-500.C1 RTA fitted with the Loudness option
Measurement equipment to produce Loudness results has traditionally been expensive and laboratory based requiring that the noise is brought into the lab to analyze from high quality tape recordings. This is not always possible when results are required on the spot to identify sources of annoying noises in vehicles for instance. Since measurements require third octave band spectral details (or at least full octave band data) then a real time analyzer is necessary in practical situations to capture any subtle changes that may be present in the sound source. The CEL-500 family of real time analyzers are ideally suited to the measurement of the  loudness level since they are lightweight, fully portable, and can be fitted with the Loudness Option in any C version instrument. This provides the acoustics engineer or anyone charged with identifying the sound quality of a noise a practical solution to getting a field measurement without having to have a full computer based laboratory to call on.

 
time variation of the loudness value in sones to show steadiness of the source
   
Add-in options available for the CEL-500 Series Real time analyzers
Fastore Option Building Acoustics Option Loudness Option Logging Option