Harmonics, Interferences, Autocorrelations, FFT

AI Thread Summary
The discussion revolves around FFT spectrum analysis and the presence of harmonics and interferences in signal processing. The user questions why there are no harmonics at 120Hz, 180Hz, or 240Hz despite a strong 60Hz peak, and seeks clarification on how FFT can separate frequencies amidst interference without saturating the amplifier. There is also inquiry into whether a peak at 295Hz is considered noise or a signal, and how autocorrelation functions in identifying signals hidden in noise. Participants emphasize the importance of understanding the mathematical principles behind signal analysis rather than relying solely on simulation software. The conversation highlights the complexities of analyzing real data and the challenges of distinguishing between noise and meaningful signals.
Ephant
Messages
147
Reaction score
2
Hi I have some questions about the following:

60hz ac fft 2.JPG


1. It is an FFT spectrum analysis output from the Sigview software. First. The large peak at 60Hz is AC interference on air (maybe capacitive coupling) as the amplifier is completely battery operated. How come there are no 120 Hz, 180Hz, 240Hz harmonics? What situations when harmonics can be created and not created?

2. If the purpose of your amplifier is to get FFT, then there are no problems getting interferences as long as it won't saturate the amplifier, because the FFT can separate the frequency and you will still see the frequency even with all the interferences, right? This is the main question I need to know.

3. Is the peak near 300Hz (the 295Hz) considered part of the white noise or do you consider it a signal, in your experience?

4. When I pressed Autocorrelation, the following is the result. So Autocorrelations can't produce anything if it is only single frequency hidden (such as hidden harmonics) in the noise floor and it only look for signal that has reverse phase or something? What does autocorrelations exactly look for?

60hz ac fft 2 autocorrelation.JPG


I'm learning about harmonics, interferences, autocorrelations, FFT at the same time for general knowledge so as not to be ignorant. Thank you.
 
Last edited:
Engineering news on Phys.org
@Ephant Yet again you are proving that it's not possible for you to learn anything from fiddling about with a simulation package. That package uses Mathematics to produce its results. I am confident that you are not capable of reverse engineering it to reveal the basics of the Maths behind signal analysis.
Avoid that simulation until you have some idea about the basics.
 
  • Like
Likes phinds and DaveE
sophiecentaur said:
@Ephant Yet again you are proving that it's not possible for you to learn anything from fiddling about with a simulation package. That package uses Mathematics to produce its results. I am confident that you are not capable of reverse engineering it to reveal the basics of the Maths behind signal analysis.
Avoid that simulation until you have some idea about the basics.

It's not a simulation. I'm using a real amplifier with data saved with Audacity wav. Then played the wav later at SigView and running the FFT spectrum analysis of real data.

I was asking if the amplifier was able to get the 60Hz signal on air or via capacitive coupling (as the amplifier is completely battery operated). Why doesn't it show the harmonics. If the harmonics is hidden in the noises. How come FFT didn't produce the harmonics, and neither did the autocorrelation.

But most important is. If the end result is FFT, where you can display all frequency. What is the problem with no shielding against interference as long as it won't saturate the amplifier. You can still acquire any signal amidst the noises using FFT.
 
Buy and read a text book, dude.
 
  • Like
  • Haha
Likes berkeman and sophiecentaur
OP is on a 10 day vacation from PF for multiple reasons, so we can all take a break in this thread. :smile:
 
Usually harmonics are present only in sound or music. Power from an outlet or an AC line is kind of "dirty" so there is going to be quite a bit of noise.
 
  • Skeptical
Likes sophiecentaur
El hilo está cerrado por moderación.
 
Lordy. OP is banned for cause.

Ephant said:
Whatever, now I don't have to dive below the noise floor to search for signal. I can simply scout the surface of the noise floor ocean. Now my Dark Lab will be more active and hopefully replicate a Sigmal 5 experiment that if verified can change everything, our view of the universe and more.
 
  • Like
Likes Tom.G
Back
Top