What is Nyquist Frequency?

Nyquist frequency, also known as the Nyquist-Shannon frequency, is a foundational concept for operating any digital system that records data or regulates processes — like a vibration controller or data acqusition system.

The Nyquist frequency is equal to half of the sampling rate (number of measurements per second) — The formula of Nyquist frequency

In the case of RULA vibration controller and data acquisition system RL-C21M, the maximum sampling rate of the device is 216 kHz — this means the Nyquist frequency is equal to 108 kHz. The Nyquist frequency will decrease together with the sampling rate.

The principle of Nyquist frequency helps engineers design ADC and DAC converters and set appropriate sample rates. And this is not just about converting signals from continuous (analog) to discrete, but also when you want to reconstruct a smooth, continuous signal from a set of discrete data points.

To further emphasize the importance of this principle, let’s consider what is going to happen if we ignore it. If a signal is sampled at a frequency less than twice its highest frequency Relation between frequencies per Nyquist theorem, two phenomena called 'aliasing' and ‘distortion’ occur. 

Aliasing is the effect that happens when high-frequency information is incorrectly identified as low frequency — it's similar to the effect seen in films where wheels or helicopter blades appear to be turning backward.

Aliasing phenomenon as seen on the camera

Figure 1: 'Aliasing' as seen on the camera

Aliasing is generally undesirable because it introduces errors or distortions that are not present in the original signal.

This is why, in practice, an anti-aliasing filter is used before the sampling process to remove frequencies above the Nyquist frequency (half of the sampling rate), thereby preventing aliasing.

The addition of filter is the reason why the sampling rate is usually chosen in the range of Values of Nyquist frequency with the anti-aliasing filter.  With this in mind, the maximum frequency RL-C21M can measure is 80 kHz.

The figure below is an example of bandlimited signal and frequency aliasing.

Relation between Nyquist frequency and signal frequency on the graph

Figure 2: Correlation between Nyquist frequency and the resulting signal (red line – original analog signal, blue line – reconstructed signal)

1) Fmax < 1/2Fs. Signal without any artifacts
2-3) Fmax = 1/2Fs. Different phases of sampling frequency
4) Fmax > 1/2Fs. Signal aliasing. New frequency = Fs - Fmax

If you have further questions about the concept or would like a consultation on choosing a suitable data acquisition system for your task, please contact us in any convenient way:

contact@rula-tech.com
+371 6610 2166