earth voltage recording

Hi

I'm looking for a way to record fluctuations in earth voltage. We
currently are using laptop soundcards, but it suffers from a lot of
noise. Thinking of going to a USB type device for recording to a
laptop.

Data is recorded for 1second after triggering at a sampling interval of
50ms. The sampling frequency is therefore 20kHz. Typical signal
measurements are on the order of 1mV to 0.001mV. Any sample code
available for signal recording, stacking and filtering?

Regards

 

[Robert Baer]

probe@execs.com wrote:

Hi

I'm looking for a way to record fluctuations in earth voltage. We
currently are using laptop soundcards, but it suffers from a lot of
noise. Thinking of going to a USB type device for recording to a
laptop.

Data is recorded for 1second after triggering at a sampling interval of
50ms. The sampling frequency is therefore 20kHz. Typical signal
measurements are on the order of 1mV to 0.001mV. Any sample code
available for signal recording, stacking and filtering?

Regards

The sampling frequency is not "therefore" anything determinable by

any other parameter you mentioned.
And if data is recorded for 1 second, then the sampling interval is
one second, not some other value.
A sampling *rate* of 50 *microseconds* would be the same as saying
the sampling frequency was 20KHz.
Code is simple: set up an array of the appropiate size and start with
the pointer to the array (or index if you prefer that terminology) to
the first element.
Take a sample and advance the pointer (or add one to the index);
repeat until the total desired number of samples has been taken.
Stacking? What is that?
Filtering? What do you want? What kind(s)?

 

The duration of measurement is 1 second, with a voltage and time value
recorded every 50ms for a period of 1second. During stacking, the
values obtained during different 1second measurements are averaged to
remove noise. With filtering further noise removal can be accomplisehd
through the removal of e.g. powerline noise 50hz / 60hz.

 

[Robert Baer]

probe@execs.com wrote:

The duration of measurement is 1 second, with a voltage and time value
recorded every 50ms for a period of 1second. During stacking, the
values obtained during different 1second measurements are averaged to
remove noise. With filtering further noise removal can be accomplisehd
through the removal of e.g. powerline noise 50hz / 60hz.

If you keep all of the data acquired during a given period, then

transform it from the time domain to the frequency domain via a FFT,
then one could easily remove the line frequency "spikes" (fundamental
and harmonics) without disturbing the signal data.
Furthermore, if the signal has a repetitive waveform, those aspects
could be enhanced and non-related aspects could be de-emphasized.
However, care must be taken to preserve modulation in the carrier
frequencies.