Email: Password: Remember Me | Create Account (Free)

Back to Subject List

Old thread has been locked -- no new posts accepted in this thread
???
11/10/06 06:51
Read: times


 		 #127687 - Alfredos Code Doesnt seem to work properly.
Responding to: ???'s previous message
Dear Steve,

I tried to simulate and even emulate Alfredos FFT code available on your home page. The code first generate a data sample with following components,
[pre]
component amplitude Phase
f1 256 0 (cosine)
f2 128 0 (cosine)
f3 85 aprox. delayed 90 degrees.

[/pre]

Then it does FFT on this data and store amplitudes of various orderes in amplitude array. I have tabulated the results of amplitude array after FFT.

[pre]
Amplitude[0] 393
Amplitude[1] 32119
Amplitude[2] 24989
Amplitude[3] 26980
Amplitude[4] 20988
Amplitude[5] 23216
Amplitude[6] 24899
Amplitude[7] 20041
Amplitude[8] 28644
Amplitude[9] 3725
Amplitude[10] 14440
Amplitude[11] 29572
Amplitude[12] 21141
Amplitude[13] 10579
Amplitude[14] 8424
Amplitude[15] 16881
Amplitude[16] 390
Amplitude[17] 27176
Amplitude[18] 7337
Amplitude[19] 26350
Amplitude[20] 29026
Amplitude[21] 5426
Amplitude[22] 6190
Amplitude[23] 27775
Amplitude[24] 28283
Amplitude[25] 18472
Amplitude[26] 10698
Amplitude[27] 19362
Amplitude[28] 31545
Amplitude[29] 32454
Amplitude[30] 22025
Amplitude[31] 27156
[/pre]

The above results seem to be something not expected. Has anyone tried alfredo's code before. What could be wrong with this code.

Thanks & Regards,
Prahlad Purohit


List of 32 messages in thread
TopicAuthorDate
Measuring THD [ Total Harmonic distortion ]            01/01/70 00:00      
   ANalogue method.            01/01/70 00:00      
      Re: Analogue method.            01/01/70 00:00      
         fundamental flaw?            01/01/70 00:00      
            Re: Fundamental Flaw.            01/01/70 00:00      
               Digital filter            01/01/70 00:00      
                  Re: Fundamental Frequency.            01/01/70 00:00      
               FFT            01/01/70 00:00      
                  mains frequency            01/01/70 00:00      
                     Why rely on an assumption?            01/01/70 00:00      
                  Re: FFT            01/01/70 00:00      
                     FFT            01/01/70 00:00      
         What use?            01/01/70 00:00      
   Goertzel filters, maybe ?            01/01/70 00:00      
   FFT is the first choice            01/01/70 00:00      
      Is it?            01/01/70 00:00      
         If you know the fundamental.            01/01/70 00:00      
            Re: Fundamental            01/01/70 00:00      
               Multiple cycles            01/01/70 00:00      
                  The 50Hz fundamental must dominate            01/01/70 00:00      
               Fundamental and harmonics!            01/01/70 00:00      
                  Re: Fundamental and harmonics.            01/01/70 00:00      
                     You're getting into theory            01/01/70 00:00      
                     That is very simple!            01/01/70 00:00      
         I think, yes            01/01/70 00:00      
   Alfredos Code Doesnt seem to work properly.            01/01/70 00:00      
      Do you want it to write in "C"?            01/01/70 00:00      
         Yes Kai,            01/01/70 00:00      
      Eh? Endian issue.            01/01/70 00:00      
         Tags            01/01/70 00:00      
   How about this method for THD.            01/01/70 00:00      
      Mix of numerical calculation and FFT            01/01/70 00:00      

Back to Subject List