PicoScope 7 Software
Available on Windows, Mac and Linux
how to measure noise spectral density in V/sqrt(Hz) vs frequency
how to measure noise spectral density in V/sqrt(Hz) vs frequency
Hi,
I have picoscope 6 and am going to measure the electrical noise spectral density in V/sqrt(Hz) vs frequency. I have already measured the noise spectrum in V vs frequency. how can I convert my result to V/sqrt(Hz) vs frequency?
Best,
Shadi
I have picoscope 6 and am going to measure the electrical noise spectral density in V/sqrt(Hz) vs frequency. I have already measured the noise spectrum in V vs frequency. how can I convert my result to V/sqrt(Hz) vs frequency?
Best,
Shadi
Re: how to measure noise spectral density in V/sqrt(Hz) vs frequency
Hi Shadi,
We don't perform that in PicoScope 6. So, to convert the components of the Spectrum plot you would need to export the csv file, so that you can work on the data set (e.g. in a suitable tool such as Excel, Matlab or Python) and then do the following:
First you need to detrend the spectrum (remove any trend or drift in the Spectrum), and then remove any applied Window function. Then you need to convert each |V/Hz| linear (rms) component to a |V^2/Hz| power component by squaring it. Finally, you need to apply the scaling for your required window function. Also, we ignore the first few bins as they tend to be inaccurate for a linear FFT.
If you need more information on Windowing functions and the process of implementing the calculated FFT coefficients, then this document is one of the best references on the subject: https://holometer.fnal.gov/GH_FFT.pdf, and there is some discussion on the subject here: topic39794.html.
Regards,
Gerry
We don't perform that in PicoScope 6. So, to convert the components of the Spectrum plot you would need to export the csv file, so that you can work on the data set (e.g. in a suitable tool such as Excel, Matlab or Python) and then do the following:
First you need to detrend the spectrum (remove any trend or drift in the Spectrum), and then remove any applied Window function. Then you need to convert each |V/Hz| linear (rms) component to a |V^2/Hz| power component by squaring it. Finally, you need to apply the scaling for your required window function. Also, we ignore the first few bins as they tend to be inaccurate for a linear FFT.
If you need more information on Windowing functions and the process of implementing the calculated FFT coefficients, then this document is one of the best references on the subject: https://holometer.fnal.gov/GH_FFT.pdf, and there is some discussion on the subject here: topic39794.html.
Regards,
Gerry
Gerry
Technical Specialist
Technical Specialist