Company Events Academic NI Developer Zone Support Solutions Products & Services Contact NI MyNI

Acquiring and Simulating Dynamic Signals (Sound and Vibration Measurement Suite)

Sound and Vibration Measurement Suite 6.0 Help
December 2007

NI Part Number:
372416A-01

»View Product Info

This topic discusses methods for obtaining data and some key issues when acquiring or simulating dynamic signals to ensure valid measurement results. You can use the following three techniques to obtain data:

  • Acquire data with a data acquisition (DAQ) device system
  • Read data from a file
  • Simulate data with a generation VI or other source

The measurement and analysis VIs in the NI Sound and Vibration Measurement Suite do not compensate for inaccurate data. You must calibrate the test equipment to ensure accurate results. Generally, the test equipment must have specifications at least 10 times better than the specifications of the device under test (DUT). Use a verifiable and repeatable test procedure to get accurate results.

Whether you are obtaining the data from a DAQ system, reading the data from a file, or simulating the data, aliasing, sampling rates, and time continuity are common considerations in measurement analysis.

Aliasing

Aliasing is the phenomenon by which frequencies greater than the Nyquist frequency are shifted erroneously to lower frequencies. The Nyquist frequency is calculated with the following formula:

fNyquist = sampling rate / 2

When acquiring data with an NI dynamic signal acquisition (DSA) device, aliasing protection is automatic in any acquisition. The sharp anti-aliasing filters on DSA devices track the sampling rate and filter out or attenuate all frequencies above the Nyquist frequency.

When performing frequency measurements with an NI E Series or M Series DAQ device, you must manually eliminate aliasing using the following methods:

  • Increasing the sampling rate
  • Applying an external lowpass filter
  • Using an inherently bandlimited DUT

Simulated data also can exhibit aliasing. Simulated signals often are generated according to a time-domain expression and, therefore, have high-frequency components that are aliased in the discretely sampled data. The following front panel shows an example of the aliasing for a simulated square wave.

The only way to protect data from aliasing is to apply appropriate aliasing protection before the data is generated or acquired. Aliasing occurs when the data is generated or sampled. You cannot remove aliased components from the data without detailed knowledge of the original signal. In general, you cannot distinguish between true frequency components and aliased frequency components. Therefore, accurate frequency measurements require adequate alias protection.

Sampling Rate

The scan rate, or the sampling rate in NI-DAQmx, determines how often an analog-to-digital (A/D) conversion takes place. A fast input sampling rate acquires more points in a given time and can form a more accurate representation of the original signal than a slow input sampling rate.

Two key parameters, maximum rotational speed to analyze and maximum order to analyze, determine the sampling rate for order analysis applications. For sound and vibration signal acquisition, choose the sampling rate according to the following equation:

sampling ratesound and vibration = 2.56 * max order * max speed (RPM)/60

When using an analog input channel to acquire a tachometer signal, set the sampling rate to a faster rate. When performing run-up or run-down tests, the measurement results are highly dependent on the accuracy of the tachometer pulse measurement. You typically want to select a tachometer signal sampling rate at least four times faster than the sound and vibration signal sampling rate. Use the following equation to calculate the tachometer signal sampling rate:

sampling ratetacho = 4 * 2.56 * max order * max speed (RPM)/60

Note  When the number of pulses the tachometer generates in each revolution is very large, the required tachometer sampling rate might be faster than the one you calculate from the above equation.

When measuring high orders that run at a fast speed, you must set the sampling rate to a very high value. The synchronized analog input channels for tachometer signals and sound and vibration signals usually work at the same sampling rate. The fast sampling rate for the tachometer signal leads to unnecessary processing for the sound and vibration signals that need a lower sampling rate. Some DAQ devices might not be able to set the sampling rate to the required value. In this case, you can use a counter device to acquire the tachometer signal and keep the sound and vibration signal sampling rate at an appropriate value. When you use a counter device synchronized with a DAQ device, the counter can acquire the tachometer signal at a much faster rate than the DAQ acquisition rate. Using this combination of devices can generate more accurate measurement results.

Time Continuity

When you acquire data in a continuous acquisition, you can use the t0 parameter in the waveform datatype to ensure no gaps exist between successive blocks of waveforms returned by sequential calls to the DAQmx Read VI or AI Read VI. When you generate signals with the Generation VI, the t0 of the current waveform always is one sample period later than the timestamp of the last sample in the previous waveform. Continuity is enforced in this way until the generation is reset.

The waveform datatype is integral for testing time continuity in the Sound and Vibration Measurement Suite. If you read data from a file or simulate a signal using signal generation VIs in the Sound and Vibration Measurement Suite, wire a t0 value that meets the continuous timestamp condition to the waveform datatype wired to the measurement analysis VIs. This action prevents unexpected resets of the measurement analysis due to detected discontinuities in the input signal.


Resources


 

Your Feedback! poor Poor  |  Excellent excellent   Yes No
 Document Quality? 
 Answered Your Question? 
Add Comments 1 2 3 4 5 submit