Question on sampling frequency/low pass filter

  • Thread starter Thread starter hob
  • Start date Start date
  • Tags Tags
    Filter Sampling
AI Thread Summary
Sampling at a high rate of 100Hz can effectively capture and filter out high-frequency noise (1-15Hz) when analyzing lower frequencies (0.004Hz). If sampling at 0.5Hz, the data may primarily reflect noise, risking aliasing issues when filtering. A low pass filter at 0.25Hz could mitigate some problems, but higher frequency sampling allows for better noise management. Implementing an analog filter before sampling is recommended to prevent aliasing, making lower sampling rates more feasible. Ultimately, the choice of sampling frequency should consider the power spectrum of the data and the potential for aliasing.
hob
Messages
6
Reaction score
0
Hi all,

I have a question relating to sample rates and data filtering.

I have a (relatively) high sampling rate of 100Hz, there is some high frequency "noise" present that acts over the range of 1-15Hz.

My question is, given that I am interested in much lower frequencies (slow movement around 250 seconds /4x10^-3 Hz) is there anything to be obtained from higher frequency sampling?

I.e provided my sample rate is sufficiently high (say 0.5Hz) and I apply a low pass filter (at 0.25Hz) to prevent aliasing, is there any more to be gained compared with higher freq logging with a low pass filter at 1Hz?
 
Physics news on Phys.org
The answer depends on the power spectrum of the measured data, and the effects of aliasing when you sample the data.

As an extreme example, suppose the signal you are interested in at 0.004 Hz has an amplitude of about 1, but there is a noise signal at about 10 Hz with amplitude 10.

if you sample at about 0.5Hz, the sampled data will be mainly the noise, and the 10 Hz will be aliased to some frequency between 0 and 0.25Hz (the Nyquist frequency). That may cause problems when you filter the data.

On the other hand if you sampled at 100 Hz, if would be trivial to filter out the noise at 10 Hz.

Actually, in real life the best technique is often to use an analog filter before sampling, to make sure there that aliasing won't happen. If you did that, a sample rate of 0.5 Hz (or even lower) would probably be OK.
 
Thread 'Motional EMF in Faraday disc, co-rotating magnet axial mean flux'
So here is the motional EMF formula. Now I understand the standard Faraday paradox that an axis symmetric field source (like a speaker motor ring magnet) has a magnetic field that is frame invariant under rotation around axis of symmetry. The field is static whether you rotate the magnet or not. So far so good. What puzzles me is this , there is a term average magnetic flux or "azimuthal mean" , this term describes the average magnetic field through the area swept by the rotating Faraday...
Back
Top