CCS/LAUNCHXL-F28379D: FFT magnitude function calculation: CFFT_f32s_mag_TMU0 function

Ethan1

Tool/software: Code Composer Studio

Hi all,

I'm using CFFT_f32s_mag_TMU0 function to calculate the FFT magnitude of a pretty simpe waveform which contains 1Hz and 3Hz components. As attached below, the graph shows a clear result:

And what the function gives me is:

In the function documentation it says:

The output from CFFT_f32s_mag matches the magnitude output from the FFT found in common mathematics software and Code Composer Studio FFT graphs.

But it seems not. So my question is:

Why the result is inconsistent?
Why would negeative number appears, since the magnitude should calculate sqrt(Xre[i]2 + Xim[i]2).

Any relpy is appreciated.

Best Regards

Ethan

over 4 years ago

0 Ethan1 over 4 years ago

Prodigy 90 points

Hi all,

The first quesiton was resolved. I did not pass the imaginary part into the CFFT. But I'm still curious why would the magnitude function gives negative result.

0 Himanshu Chaudhary60 over 4 years ago in reply to Ethan1

TI__Expert 3720 points

Hi,

Few assumptions that needs to hold true before call this function are:

CFFT_f32() has been called prior to this function
Stages must be at least 3 (FFTSize >= 8)
FFTSize must be at least 8 and even.

Please check all the above three conditions are valid. Also note that the magnitude result pointer after computation is stored in CurrentOutPtr so make use of CFFT_f32_getCurrOutputPtr() to access the result pointer.

If my reply answers your question please click on "This resolved my issue" button located at the bottom of my post.

Regards

Himanshu

0 Ethan1 over 4 years ago in reply to Himanshu Chaudhary60

Prodigy 90 points

Hi Himanshu,

I've checked all above points and the size of FFT is 32. And I do not think this is related to the second question. Anyway, thanks for the response.