Official Luthiers Forum!

Yes, that 58 Hz peak is probably electrical noise. It seems pretty low, but the dB level depends on what the software uses for a reference and how everything is calibrated. You could be sure by playing/singing a note well above 60 Hz, and looking how the 60 Hz peak compares to an actual sound. It is probably so low it's not an issue.

That broad first peak could be a number of things, either real noise (HVAC rumble?) or an artifact with the measurement system. If it is a real sound, it is too low in frequency (and probably too low in level) for you to hear. My guess is this is all so low in level (would be the case if the dB level is referenced to the standard pressure level) that it is not an issue at all.

Brad, regarding your setup, it might have been a ground loop. If you were plugging both the mixer and the PC into AC outlets for power, that can happen. Who knows? It could be the ghost of Nikola Tesla, messing with us from the great beyond.

As you need a nice clean background in this low freq. range here's a little trick to try. Record the background as a noise reduction
profile in Audacity and then load it for NR and remove it in any future plots. Just record a 10 sec or so sample and save it, then after recording your taps go to Effects and Nose Reduction and click Load Profile. Load your sample, then remove said sample from your tap recording. I use this method when I record anything using Adobe Audition. I just checked and this feature is also available in Audacity.


As an edit:
With your noise freq.being at or near 60hz , a tap measurement just might land on the same number (murphy's law)
thus preventing it from showing in the final plot. If outside the noise range it would be fine... thinks I.
Funny, what crosses one's mind driving to the grocery store.

The DAW software Reaper has a noise reduction function like what Ken has described, too. I have not tried it for frequency analysis purposes, though.

Try removing the Hanning windowing , you may get a better spectrum of the actual noise.

Great idea on the noise reduction. To keep this thread informative to anyone in the future here is how to do it in audacity: https://www.podfeet.com/blog/recording/ ... -audacity/

Having said that, the graphs that I plotted pre and post noise were identical. So apparently I don't have a noise problem.

As for the Hanning window, I cannot remove it per se. In audacity there is no option to NOT use a window however they have several options for other windows such as Bartlett, Blackman, Gausian, etc...