You are here: NI Home > Support > Product Reference > Manuals > Sound and Vibration Measurement Suite 6.0 Help

Use the Weighting and Filtering VIs to apply acoustic weighting to time-domain signals, FFT-based spectra, or fractional-octave spectra.

Traditionally, weighting filters are built using analog components. If you use an external weighting filter, use the weighting filter parameter in the channel info control of the SVL Scale Voltage to EU VI to ensure proper display of the selected units. Also, the Weighting and Filtering VIs use the channel info parameter to report an error if the application attempts to apply additional weighting to a previously weighted signal.

Note  The weighting filter parameter in the channel info control of the SVL Scale Voltage to EU VI assigns the correct units to the waveform, but this parameter does not cause the VI to perform any filtering.

Applying Acoustic Weighting to Time-Domain Signals

Use the Weighting and Filtering VIs to apply acoustic weighting to time-domain signals. The Weighting and Filtering VIs use pre-designed filters to apply the desired psophometric weighting. Therefore, the Weighting and Filtering VIs support only a finite set of sampling rates. The following table lists the sampling rates that the Weighting and Filtering VIs support.

Sampling Rates	Supported Filters
	A-, B-, or C-weighting	ITU-R 468-4/Dolby	CCITT/C-message
4 kHz to 20 kHz 4 kHz, 8 kHz, 10 kHz, 11.025 kHz, 12.5 kHz, 12.8 kHz, 16.666 kHz	Yes	—	Yes
20 kHz to 1 MHz 20 kHz, 22.05 kHz, 25 kHz, 25.6 kHz, 33.333 kHz, 40 kHz, 44.1 kHz, 48 kHz, 50 kHz, 51.2 kHz, 80 kHz, 96 kHz, 100 kHz, 102.4 kHz, 192 kHz, 200 kHz, 204.8 kHz, 500 kHz, 1 MHz	Yes	Yes	Yes

Note  Use the Weighting (Arbitrary Rate) VIs to apply A-, B-, or C-weighting for sampling rates not listed in the previous table.

NI recommends using the fixed-rate weighting filter VIs if the VIs support the desired sampling rate. The fixed-rate weighting filter VIs offer two advantages over the Weighting (Arbitrary Rate) VIs: compliance with the appropriate standards over the entire frequency range, and slightly faster execution because of precomputed filter coefficients.

Note  The filter design algorithms that the fixed and arbitrary rate weighting approaches use are different. Using a fixed-rate weighting filter with a supported frequency or using the equivalent arbitrary rate filter at the same sampling rate achieve different results. Each implementation offers compliance with an appropriate standard over the frequency range.

Performing A-Weighted Sound Level Measurements

The following block diagram shows an application that uses acoustic weighting filters to perform an A-weighted sound level measurement with a slow time constant (L_AS) on a simulated signal.

The SVL Scale Voltage to EU VI scales the time-domain signal and sends the scaled signal to the SVT A, B, C Weighting Filter (Fixed Rates) VI. The application applies acoustic weighting to the time-domain signal with the SVT A, B, C Weighting Filter (Fixed Rates) VI because the signal is simulated with a sampling rate supported for A-weighting. The SVT A, B, C Weighting Filter (Fixed Rates) VI then sends the weighted signal to the SVL Exp Avg Sound Level VI.

The following front panel shows the time-domain input and output waveforms when a 250 Hz sine wave is sent to the SVT A, B, C Weighting Filter (Fixed Rates) VI using the A-weighting filter.

A phase difference exists between the input and output signals because the A-weighting filter applies the time-domain weighting. The transient behavior at the beginning of the filtered waveform corresponds to the filter settling time.

Applying Acoustic Weighting to FFT-Based Spectra

The most efficient way to compute a frequency-weighted spectrum is to apply weighting in the frequency domain, especially when you need to compare the power spectrum of a signal with the power spectrum of the same signal after applying a weighting filter. You can use the SVT Weighting Filter (frequency) VI to apply acoustic weighting in the frequency domain.

The following block diagram shows using an acoustic weighting filter in a frequency measurement.

In the previous block diagram, a single power spectrum is computed.

The following block diagram shows a different implementation based on applying the weighting filter on the time-domain signal and then computing the power spectrum.

In the previous block diagram, the power spectrum is computed twice, leading to more CPU usage and increased processing time. By applying the acoustic weighting filter in the frequency domain, you can decrease CPU usage and processing time.

Applying Acoustic Weighting to Octave Spectra

When performance, such as CPU usage, is an issue, applying acoustic weighting in the frequency domain can improve the efficiency of the measurement process. The following block diagram uses the SVT Weighting Filter (octave) VI to apply acoustic weighting to a third-octave spectrum.

Errors Due to Uniform Corrections

When you apply acoustic weighting to a fractional-octave spectrum, a continuous frequency response function defines the attenuation of the weighting filter. When you mathematically weight a spectrum consisting of data from a fractional-octave spectrum, the correction values applied for the weighting are equal to the theoretical values at the center frequency of the fractional-octave band. Applying correction values creates an essentially rectangular filter that does not have a continuous response. The filter applies the same correction to all energy within each fractional-octave band. Applying the correction values to a signal containing a pure tone near one of the fractional-octave filter edges might introduce a measurement error. The error is usually negligible unless an A-weighting filter is used at frequencies below 500 Hz. At frequencies below 500 Hz, the slope of the A-weighting curve is steep. The following illustration shows how the slope of the A-weighting curve can contribute to potential measurement errors at low frequencies.

Note  The same type of measurement error as in the previous illustration can occur when applying weighting to FFT-based spectra. However, the error is almost always negligible as long as the frequency resolution of the spectrum is reasonable. For example, the error is negligible with a frequency resolution of 10 Hz or finer.