Vibration order analysis: I want to convert angle-based data into order-based data by applying FFT.

I'm working on order anlysis but I don't know how to devleope a script to scale my order axis (x-axis) so that each frequency magnitude fall on the correct order. Also you can see my angle based data in the attachment.

Thanks

[Mathieu NOE]

hello

could you more precisely describe what the attached plot data are ?

can you export this to the workspace and share it ?

[Adeel Ashraf]

Bascially I have incemental shaft encoder which give me a square puls train signal. I find out the "delta t" using trigger function in b/w every rise of pulses and delta t show the time b/w one dgree of angle. So that I can do the order based fft analysis. You can find the attachment of angle based time data basically this data is for 8 complete cycles divided into 360 increments in steday state.

AA = delta_T.signals.values(402396:429096);

save ('delta_t.mat','A')

%% The extracted data is in delta_t.mat file.

FFs = 290; % sampling rate

tt = 0:1/Fs:1-1/Fs;

s = AA;

xxdft = fft(s);

xxdft = xxdft(1:length(s)/2+1);

DDF = Fs/length(s); % frequency increment

ffreqvec = 0:DDF:Fs/2;

plot(ffreqvec,abs(xxdft))

form this script I plot the above result.

[Mathieu NOE]

hello again

sorry, in your code Fs is not initialized - what is the relation with FFs ?

so what you have in the mat file is only the signal of the encoder, but when you do order analysis , you must acquire the encoder and some noise or vibration signals together, because what is the purpose of the analysis is the noise / vibration signal orders - not simply the fft of the encoder output.

You have to acquire synchronously the noise / vibration data and the encoder (ADC is triggered by encoder signal)

[Adeel Ashraf]

Hello Matheiu

Thanks for correcting me. If it's possible once again can you go through my data. I attached the wrong file last time. I'm new to matlab that's why doing lot of mistakes.

Fs = 285; % sampling rate

t = 0:1/Fs:1-1/Fs;

z = B.Data(:,1);

xdft = fft(z);

xdft = xdft(1:length(z)/2+1);

DF = Fs/length(z); % frequency increment

freqvec = 0:DF:Fs/2;

plot(freqvec,abs(xdft))