We use them a lot in vibration testing/measurement. If you consider the sampling rate of a sine wave, and the apparent measured frequency of high frequency signals measured too slowly, Nyquist is a good place to start, but we typically acquire data at much higher speeds than 2X our filter cutoff frequency.
Execellent explanation on using filters and the Nyquist-Shannon theorem . This will be of great use to every engineer as Nyquist-Shannon theorem was and is realy the backbone of communication engineering. Thanks Jon!
Very nice explanation on using filters and the Nyquist-Shannon theorem - something every test engineer needs to keep in mind when determining their sampling rate. Thanks for another great article, Jon!
I recommend using an anti-alias filter whenever you must measure anuthing more than a DC, or near-DC signal. Some companies include them in data-acquisition equipment or on analog-to-digital-converter (ADC) boards, and some don't, so it pays to ask. If a board or system includes a filter or filters, find out how much control you have over it and get a plot of frequency vs. attenuation (a Bode plot) and a plot that shows phase vs. frequency. I didn't get into phase changes in this column, but people should know that filters change phase relationships of signals, too. Those changes could affect measurements when you must correlate signals in the time domain.
Engineers can build their own anti-alias filters, but I don't recommend that course unless they have filter-design experience and plan to build a lot of them. Commercial filters are the way to go in almost all situations.
Switched-capacitor filters have a few disadvantages. They exhibit greater sensitivity to noise than their op-amp-based filter siblings, and they have low-amplitude clock-signal artifacts -- clock feedthrough -- on their outputs.