If by "central part" of the spectrum you mean the center bin, that is the value of the spectrum array that is in the middle, you have to be careful how you interpret that value. Normally when you analyze a signal using FFT you should only use the spectral values that are below the Nyquist frequency (half the sample rate). I'll suggest you to use the high-level Power spectrum or FFT VI(s) in the main palette Analyze>>Waveform Measurements instead of the low-level FFT VI. These VIs do some "clean-up" for you and only return the relevant values.
If your number of samples is even (2*N) (not just only power of two), the value at index N (called Nyquist frequency) will be totally unreliable. Think of it as a sine tone that is sampled exactly twice per period. If yo
u are "unlucky", you may sample exactly at the zero-crossing points and the resulting value will be zero, or you could sample exactly at the positive/negative tops, in which case the value is maximum. So don't count on that value.
If the number of samples is odd (2*N+1), then the values at index N and N+1 will be equal in magnitude, (N-1) and (N+2) are also equal in magnitude and so on. Is this what you are seeing? If yes that is normal.
The VI you attached can not be open because it is missing sub-VI(s). Use the Save with Options features and select "Development Distribution", and remember to save default data if needed.
