increasing of amplitude in the result (vibration analyse using FFT)

[Louay02]

Hi,

I have made a diagram to analyse vibration using FFT there’s something wrong during the working of this diagram the amplitude still increase every period I don’t understand what I missed ?

Thank you

[Louay02]

the result :

[Henrik_Volkers]

while looking at your diagram I can't figure out what you want to show/analyse...

do you know what the FFT is doing?  Seems that you don't feed the correct data 😉

[JÞB]

---

@Henrik_Volkers wrote:

while looking at your diagram I can't figure out what you want to show/analyse...

do you know what the FFT is doing?  Seems that you don't feed the correct data 😉

---

Considering your sig line I'll let that slide.  An unbundle followed by Cartesian to Polar would make it more obvious that the FFT is on rho over time.

Increase the sample rate and the FFT size.  The signal is either Nyquist folded many times OR there are several harmonics you are not seeing.

What is that coercion at the terminal? Is analyze...reel a Chart?

EDIT: silly me of course it's a Chart the default FFT size is 100.  Change it to a graph.  The amplitude will stabilize when the size is met.

[Henrik_Volkers]

Sorry JÞB,  grap some samples (block) , calc the RMS power of that block (rec window) and drip that into the ptbypt FFT (another rec window),

I (we) have no idea of the SR and number of samples .. nor the expected input 

depending on signal and SR and blocksize(duration) a lot of funny folding/leakage effects will show up.

that's why I question the signal processing..

a low frequency will show up as double the frequency, etc 

However, there are a lot of (funny) vibration analysing methods/standards i don't know ... 🙂

But the chart vs graph after the FFT output is a good catch.

[JÞB]

The array^2, sum,^-2 looks like the Pythagorain Theorm every day.

I simply assumed array was x,y and the op really wasn't insane ( risky.) And vibration analysis on displacement magnitude, rho, is common.