O-Solution Sound Quality Evaluation Function OS-0525

Generally, the measures and evaluations of sound generated by equipment are based on sound pressure levels, FFT analysis, and 1/3 octave analysis.
However, such analyses do not sufficiently consider human auditory characteristics, and the impression of the sound can differ even with the same results.
Additionally, when a person hears sound, various sensations arise regarding the sound, such as "loudness", "sharpness", and "roughness".
The OS-0525 sound quality evaluation function takes into account human auditory characteristics and makes it possible to obtain sound quality evaluation indices that correspond to various sensations related to sound.

*Sonoko Kuwano* This is one of the sound evaluation metrics based on the research of Professor Sonoko Kuwano, which adds the "sharpness value" and "1/10 of the A-weighted time-averaged sound level." There are good examples that correlate with the discomfort of environmental noise and mechanical noise. The scope of application is still under consideration, so please use it as a reference value.

Sonoko Kuwano* (Ph.D. in Engineering) Emeritus Professor of Osaka University, President of the Acoustical Society of Japan, Vice President of the Japan Society of Noise Control Engineering, and Vice President of the International Commission on Acoustics, among other positions.

The figure below shows an analysis example of six machine sounds with different loudness levels. The left figure (in blue) analyzes the sounds based on A-weighted sound pressure levels (noise levels), where all six sounds show the same value. In contrast, the right figure (in red) analyzes the sounds based on the evaluation measure of "loudness," revealing differences among the six sounds. When we actually listen to these sounds, they are perceived to have different loudness levels, consistent with the results shown in the lower loudness analysis. By using loudness, we can evaluate differences in perceived sound levels that cannot be assessed solely based on sound pressure levels like A-weighted sound pressure levels.

*Equipment for recording sound (high-performance sound level meter, FFT analyzer, etc.) is required.

Please be careful with the volume. Sound will be played.

Even if time-varying noise occurs before warm-up, it may not be heard after the warm-up (when the engine is warm). By conducting a fluctuation intensity analysis focusing on the time-varying components, the nature of the noise that was not visible in octave analysis can be revealed.

*Equipment for recording sound (high-performance sound level meter, FFT analyzer, etc.) is required.

*Equipment for recording sound (high-performance sound level meter, FFT analyzer, etc.) is required.

The sound generated by motors and inverters tends to be perceived as unpleasant even with a small amplitude because it contains pure tone components. Indicators for quantifying pure tone perception include Tonality, TNR (Tone-to-Noise Ratio), and PR (Prominence Ratio). By analyzing TNR and PR, we can investigate the frequencies of problematic pure tone components. Below is an example of the results of analyzing Tonality and PR. When comparing the Tonality and PR of a problematic product with an annoying sound "bee" and an improved product regarding the driving sound of two motors, it is evident that even with equivalent sound pressure levels, the problematic product has higher values for Tonality and Total-PR (the sum of PR), indicating a higher purity of tone.
From the PR graph, we can confirm the values of the prominent pure tone components of the problematic product and their frequencies. We can also check whether the PR values exceed the standards for significant discrete pure tones in ISO7779 Annex D.

defective product

improved product

Please be careful with the volume. Sound will be played.