Frequency response analyzer should make it obvious that the SQUARE of the response function is plotted

Support & Troubleshooting Moku:Go
1

Hello! I’ve been using the frequency response analyzer function on the Moku:Go, and it’s pretty great! Today, I noticed that I was seeing twice the phase shift on the Moku software compared to when I looked at the voltage on the oscilloscope. It seems that the Moku software is plotting the response function squared (H^2) rather than H.

Is my understanding correct?

I suppose this makes sense if you want to look at the response to the power rather than to the voltage. I’m not enough of an electrical engineer to know which plot is more common, but I assumed that most similar plots are of H (not squared). Anyway, it might be useful to label the Y-axis of the plots or otherwise indicate what is actually being plotted. I suppose a toggle switch to plot either H or H^2 would be the ideal situationn, since it makes it clear what is being plotted and lets uses to select the style of the plots based on their needs. What do you think?

2

Hi Dan,

You’re right that H is plotted in the bode plot but can you please confirm for me whether the discrepancy is in the phase or the voltage/magnitude so that I can try reproduce your issue?

If you’re generating the signal from the Moku:Go and using the BNC outputs just make sure that you have impedance matched as the FRA has a 1 MΩ input but outputs at 200 Ω. This will have an affect on the magnitude, but you can use Multi-Instrument Mode to avoid this.

3

So, is the magnitude plotting 10log10(Vout/Vin) or 20log10(Vout/Vin)? It seems that you’re using the 20log10, which I think is standard, since many people are interested in the power. Can you confirm?

I notice the that the magnitude is twice the dB that I expect and that the phase is 2x what I expect when looking at the voltage.