Go Down

Topic: Spike in FFT results (Read 195 times) previous topic - next topic

DrDeform


I have an Arduino hooked directly up to this mic:

https://www.adafruit.com/product/1713?gclid=CjwKCAjwp_zkBRBBEiwAndwD9bhbwr9779j7eWlLUH75TkSD6b6axYE73z9Z3Q-dP8y1nImqWGO0hRoC_wIQAvD_BwE


And using code provided here:

https://www.norwegiancreations.com/2017/08/what-is-fft-and-how-can-you-implement-it-on-an-arduino/


I am trying to perform FFT on incoming audio from the mic. Now the problem is I keep getting huge spikes at the lower frequencies when there is little noise around the mic. These are usually in the first and second bins and are too high to be considered noise. I have attached images of the Serial Monitor & Plotter for reference.

What could be causing this? All the other bins seem to be working correctly.




Grumpy_Mike

Quote
What could be causing this
The first bin is the DC bias and the other three look like mains interference.

MarkT

Thats just the DC peak - unless you subtract the DC offset from your signal you'll always get some DC,
and normally you can completely ignore it.

It spills into adjecent bins in a manner determined by the windowing function chosen, which from your
link appears to be a Hamming window.

[ I will NOT respond to personal messages, I WILL delete them, use the forum please ]

Go Up