how to measure a FFT of 50Hz signal?

hi

I want to measure the FFT of 50HZ signal. I collect the data from oscilloscope which is also attached herewith. the total sample points are 10000. I need the FFT in bar graph as like in attached picture ( mag (% of Fundamental) vs Harmonic Order )but I am unable to do this. kindly help me I will be highly thankful to you.

[Star Strider]

Your .csv file is a (10000x2) matrix. What are the two columns?

If the first column is your independent varialble, it needs to be saved in much greater precision since there are multiple instances of identical values. They each need to be unique.

[Mudasir Ahmed]

Dear sir,

One is the time and the second column represents the amplitude of the wave

[Star Strider]

The same times apply to several different amplitude values.

Is that correct? I would expect the times to be unique.

[Mudasir Ahmed]

Dear Sir

actually, I take this data from an oscilloscope. I am not confirmed but I think this is because the signal is 50HZ but the sampling rate is 10KHZ which is high compared to the signal. due to this it is overlapped and same amplitude comes for the range of time samples.

[Walter Roberson]

Looking through the data, I can see that the times given are stored to only two digits after the decimal place, which is a problem because at the initial range the change is all in the third decimal place. You would be better off ignoring the times and re-synthesizing them knowing they are 1E-5 apart.

[Walter Roberson]

"but the sampling rate is 10KHZ"

Your timestamps do not make any sense unless the data points are 1E-5 apart, which corresponds to 100 kHz not to 10 kHz.

Answer 1

[Image Analyst]

Did you use fft() or pwelch()? If not, why not? If you did, what did you do? Where is your code?

[Mudasir Ahmed]

dear sir,

first I load excel file in Matlab. then I give the command

[R, HARMPOW, HARMFREQ] = thd(signal,10000,100) ;

then I convert harmonic power in percentage

h=100.*(10.^(HARMPOW/20));

t=1:1:100
bar(t,h)

but the graph does not look like as attached in a picture.

[Walter Roberson]

The graph uses 40 harmonics separated by 1/2 rather than 100 harmonics separated by 1.

Also as I indicated above, your sampling frequency is surely 100000 rather than 10000 .