The Lower Measurement Limit and Self-Noise of a Sound Level Meter
The definition of the total range of a sound level meter in the new international standard IEC61672-1:2002 and the new sound level meter calibration regulation JJG188-2002 is: the A-weighted sound level range that can be tested for the response to a sine signal, from the * low sound level on the * high sensitivity level range to the * high sound level on the * low sensitivity level range, with no overload or underrange indication and level linear error within the specified tolerance range. At the same time, it is stipulated that within any frequency weighting or frequency response range of the sound level meter, the level linear error plus the expanded uncertainty caused by measurement (0.3dB) should not exceed ± 1.1dB for level 1 sound level meters and ± 1.4dB for level 2 sound level meters at all levels of any frequency range.
Therefore, in order to ensure the requirement of level linear error, after deducting the influence of uncertainty, the self generated noise of level 1 sound level meters should be at least 8dB lower than the measurement lower limit, and level 2 sound level meters should be at least 6.7dB lower than the old standard, which is at least 5dB lower than the requirement.
However, many manufacturers currently set the self generated noise (background noise) value as the lower limit for measuring sound level meters, which is clearly misleading users. Users should pay attention when selecting, as the actual measurement lower limit of these sound level meters is 6.7dB~8dB higher than what they provide. Some manufacturers still measure the lower limit of the noise level 5dB higher than
the background noise according to the national and international standards of the old sound level meter, which is not accurate enough.
The measurement lower limit of a sound level meter mainly depends on the sensitivity of the microphone and the self generated noise of the sound level meter. To reduce the measurement lower limit, we need to start from these two aspects. In the new international standards and regulations, manufacturers are required to provide * high self generated acoustic noise and self generated electrical noise respectively. It is required to place the sound level meter in a low-noise sound field to measure the self generated sound noise. As some only have a low-noise sound field for the A-level, only the A-level of the self generated sound noise can be measured at this time. Self generated electrical noise is measured by using equivalent impedance instead of a microphone. We know that microphones also generate self generated noise (thermal noise), so the self generated sound noise of sound level meters is usually greater than electrical noise. The equivalent impedance of a microphone is essentially a capacitor, with a capacitance of approximately 50pF for a 1-inch microphone and 15pF for a 1/2-inch microphone.
Different capacitance measurements will result in varying levels of self generated noise. When testing self generated electrical noise, matching devices used for electrical signal conversion should not be used. The capacitors inside these matching devices are 0.01 μ F or 0.1 μ F, and the electrical noise measured with them will be significantly lower. In addition, when measuring self generated noise, the arithmetic mean of 10 randomly read readings of the weighted sound levels for F and S time should be taken within 60 seconds, rather than the maximum reading. For the time averaged sound level, the average time should be at least 30 seconds.
