FFT and time signal condition

[Eng P&D]

I have some doubts about conditioning signal in FFT analysis in Labview.
Like this, I would like to know when vi's that use FFT gets time sinal to calculate power spectrum they take a range of numbers to make it numerically, but is it this range so long to guarantee that at low frequencies it make it right? The sampling period that FFT gets
I have made some tests, when setting (Number of samples = Frequency sampling) or some to increase frequency resolution the FFT calculation give different results.
Look at picture when frequency resolution it's lower and when it takes higher.

I try to normalize condition of tests to get better results.

I read some about in ni site but always rest some doubts.
I know about AC cuttoff frequency and it's limitations.

Sorry for the english. 😃

[Jarrod_S.]

I'm not quite sure I understand your question. In general, there is quite detailed documentation regarding LabVIEW analysis functions, the algorithms they use in various situations, and how to implement them optimally. In particular the FFT function is widely documented.

You might check out this document, LabVIEW Analysis Concepts, if you haven't found it before. Chapter 4 is on frequency component analysis.

[Eng P&D]

Thanxs..
I think it's a good material, don't have found none like it yet.
After read I'll post again.

[Eng P&D]

Dear Jarrod, I have a new doubt, gonna try to explain it.

 When Labview make daq to process FFT, it sends a block of Ns( number of samples) to vi. This block is acquired with Fs(Frequency sampling) and it takes dt = Ns*(1/Fs) seconds to acquire this block. Then if the whole block it's acquire in dt seconds the minimun frequency we can observe is Fmin = 1/dt. But when labview make FFT, the power spectrum frequency scale starts at 0. Here is the doubt, from zero to Fmin, is the FFT valid?

[Jarrod_S.]

As I understand it, there are no intermittent values between the minimum detectable frequency, Fmin, and zero. If you graph it, you will see a line between f(0) and f(Fmin), but that is only because the graph is interpolating these intermittent values. The value at frequency zero is called the DC component of the signal, and is explained further in the LabVIEW Analysis Concepts document.