how exactly does the spectral measurements express vi calculate the power spectral density of a signal?

[katiemae]

I am trying to compare the output of the Spectral Measurements express vi set to PSD mode to a spectrum analyzer that computes the PSD. They seem to be off by a factor of between 5 and 10, and I'm trying to figure out where this is coming from by looking at exactly how the PSD is computed in the express vi.

[MDC]

I believe you have found a bug! Looks like the Spectral Measurements express VI does not scale appropriately for non-uniform windows.

Take a look at the attached example VI, which demonstrates how you can correct the scaling problem. As another workaround, you could use the FFT Power Spectral Density VI directly (also see the attached diagram). It's located in the palette Analyze>>Waveform Measurements.

[katiemae]

Thanks! I need my results in V^2/Hz rather than dB, how do I change the example vi to get a PSD in V^2/Hz?

[MDC]

I've added some code to the example to convert the results to V^2/Hz.

The linear form of the PSD VI (both Express and standard) will compute all power quantities in units of Vrms^2. With PSD our unit is in Vrms^2/Hz. To convert to V^2/Hz (which I interpret to be non-RMS or Vpk - Volts peak), we need to apply the scale factor of (Vpk/Vrms)^2, which is 2.0. See attached.