Harmonic Distortion Analyzer VI

[Amirber]

Hello all,

I am analyzing data from an ADC with some of the waveform measurement pallet vis. The "Harmonic Distortion Analyzer vi" seems every useful but is giving me some heard time. I compared the fundamental and overtone amplitude of a signal in two ways: from an FFT, and by converting the amplitudes to dBV (I am measuring Volts). Both are outputs of the same vi, but there is an offset of about 3 dB between the two. Am I missing something basic?

Hope you could help me out with this.

Amir

[Blizzard]

Hi Amir,

I am a little unsure abuot the set-up of your waveform analysis. Can you attach a sample VI to show how you are using the Harmonic Distortion Analyze VI and how you are comparing amplitudes?

Thank-you. 

Best regards,

Anna L

[Amirber]

Hi Blizzard,

I'm away from my computer right now, so I'll post it on Sunday. I connected a WFM signal to the HDA vi and inspected it's output in two ways:

1. Converting the extracted frequency components to dB (20*log10(Amp)).

2. Inspecting the amplitude of the same peaks in the FFT exported form the same vi.

By the way, are the frequency components amplitude shown as V RMS?

Thanks,

Amir

[Blizzard]

Hi Amir,

Are you doing any windowing on the peaks from the FFT and if so what type? Are you applying a correction factor in association with this wave?

Thank you!

Regards,

Anna L

[Amirber]

If I remember correctly, the HDA vi usses Haning window as default, I used the vi in it's basic fetures without any additions.

Amir

[xavvvv · Accepted solution]

Hi Amir,

I've tried replicating what you described, and I am able to see a small (about 3 dB) offset between my two outputs. This is when the FFT is set to Peak-to-Peak, with Hanning windowing (which is indeed default for the Harmonic Distortion Analyzer VI).

This offset disappears when the FFT is changed to RMS, which indicates that the Harmonic Distortion Analyzer VI outputs RMS amplitudes. To make sure this is correct, try changing the settings of your FFT and see if the offset disappears.

If this isn't the case, please go ahead and post a sample VI of your procedure, as Anna suggested, so that we can troubleshoot this further.

Thank you,

[Amirber]

Looks like this is the source of the discrepancy, Thank you both!