Feedback system with fourier transform

[havenoidea]

Hi!

I have a project in which I need to make a distance measurements in terms of frequency signals (from a LC circuit, so the frequency is AC voltage analog signal). These signals are eventually used to regulate the PZT elements, so that a system of two planes maintain its parallelism. In fact, I am making "distance" measurements at three different spatial points on a plane. 

My question mainly concernes about converting the AC signals into a fourier transformed spectrum. I succeeded obtaining signals out of an electronic through RS 232, which are varying at 1kHz - 100 KHz.

1. How fast would I need to extact the samples, so that the Labview does a meaningful transform? What's the sampling rate limit by the way? THe pzt elements has a resonance frequency around KHz range as well, so I need to avoid giving "feedback" at that frequency.  What other times scales should I be considering?

2. How can I determine the bandwidth on the Fourier spectrum? Can the Labview handle this? Also, how do you extract the peak spectrum (resonance) in the presence of multiple noise peaks?

3. I am very interested in knowing various time scales involved surrouding the labviews. Is there any graphical hierachy available, starting from the limit given by TTL rising time (a few tens of nano sec, right?)  

4. Any possible short and quick algorithm one can think of ? for distance vs pzt voltage.

Thanks a lot!

[AESulzer]

I am by no means a signal processing guru, so for your questions specific to that, I'll have to defer to others on this forum and to this tutorial. As far as signal sampling, according to the Nyquist criterion, the sampling frequency must be at least twice the maximum frequency component in the signal. I don't know what the sampling limit is on RS232, but that is an awfully strange way to acquire a signal. Have you considered using one of our data acquisition (DAQ) boards? I think using a DAQ card would make the whole acquisition process much easier and might even make the analysis process easier by getting you a better signal, and it integrates very easily with LabVIEW code.

Kind Regards,