[Computation of Kolmogorov's spectrum]
 

Dear all,

I'm trying to get an energy spectrum (e.g. Kolmogorov spectra) out of (*D) velocity mesurements as a function of time. I've computed the time average to get the fluctuations.

I've computed somehow the trace of the Reynolds stress tensor (without time averaging) (u_i'*u_i'). Then I took the FFT of my signal, and plot (1/2)(u_i'*u_i') and the result is somehow strange.

Any advice/references about how to compute the Kolmogorov's spectra out of a time signal??

I've found out this post:
http://www.cfd-online.com/Forums/mai...-spectrum.html

But still not able to fix my problem...

Cheers :)