Why are Frequencies Lower Than A Fundamental Frequency Present In Spectrum Analyzer?

[RonnieSpectrum]

I have a question about frequency analysis in music. If I load a virtual instrument such as a piano VST and play just one note, say an A 440, my understanding of the harmonic series is that upper harmonics would show up on a frequency analyzer. However, there is one concept I am not understanding. A frequency analyzer shows additional frequencies BELOW the fundamental frequency. Can somebody explain to me what these frequencies are and how / if they relate to harmonics. Thank you so much in advance!

[Wonks]

They are simply a function of a real-time spectrum display, where the sampling frequency is too fast to accurately show the bass frequencies. So you get a form of aliasing, which shows higher frequencies as lower frequencies.

If you want to accurately look at low frequencies in your sounds, do it off line.

You should be able to highlight a small section of a waveform in your DAW and do a frequency analysis of that, which will be far more accurate.

Some plug-in real time frequency displays allow you to adjust the sampling period for the display, which allows you to display the low frequencies accurately, but it does make the display rather sluggish. But the displays in, for example, EQ plugs, rarely allow that, so you get what you get. So treat those basic real time displays as good for mid and high frequencies, but either go offline or to more specialised plugs for more detail, or use a frequency band analyser which will give an accurate indication of the energy in the different frequency bands.

Otherwise, use decent headphones and trust your ears.

[ef37a]

FYI, Right Mark Audio Analyser is not a 'real time' analyser but it does give nice clean 'snapshots' and you can change the parameters as well.

Also, if you do an RTA of single tone from a distorting source you WILL get sidebands above and below the pure tone, shows up tape intermodulation distortion very well..."Ooo! Nasty". (anyone here old enough to place that?)

Dave.

[RonnieSpectrum]

Thanks so much for the replies! I greatly appreciate it. Wonks, your response is revelatory and extremely helpful. This question arose not from a mixing perspective, but from a conversation with a colleague who was suggesting that harmonics below the fundamental frequency are sometimes generated by waveforms. However, it has been my understanding that harmonics are only produced above a fundamental frequency, not below.

Your response has helped me understand a flaw in utilizing frequency spectrum analyzers. I took a small piece of the waveform and ran the spectrum analyzer offline. Here's what it looked like. This is still me striking only A4 (440 Hz) on a virtual piano instrument. The readout looks very different. I'm still a little confused as to why the first peak is significantly below 440 Hz and more like 394 Hz, but I suppose that's a question for another day. Thanks again!

Offline spectrum analysis version:

Real-time spectrum analysis version:

[Hugh Robjohns]

Yes, it could be aliasing as a measurement artefact, but it could also be an intentional sound generation artefact to replicate the sympathetic resonances of a real acoustic piano.

[DC-Choppah]

The impulse of striking a string with a hammer contains all frequencies. This includes energy at frequencies below the fundamental as well as above. An impulse contains all frequencies.

The 'impulse response' of the instrument filters this wideband impulse.

So when you hit a middle C, the hammer striking the string is an impulse and contains all frequencies. The instrument filters this impulse and yields a sound that may have a fundamental at middle C but still contains ALL of the frequencies in the impulse, including the lowest ones. The fundamental is not even always the strongest.

Musical notes are not frequencies. They are a filtered impulse.

A flute is closer to a pure tone since its impulse response is very narrow band.

[Tomás Mulcahy]

That is very nicelt put DC, I'll borrow that thank you!

Just to add to Hugh and DC's points- your analyser looks to me to have very coarse resolution so it's a bit misleading I think. Try Rightmark (already mentioned) or even Voxengo Span which is real time but has a number of adjustments to improve accuracy, depending on what you want to see. Both let you measure the frequency precisely.