Jump to content

Phantom_Lord

Members
  • Posts

    5
  • Joined

  • Last visited

    Never

Phantom_Lord's Achievements

Newbie

Newbie (1/14)

0

Reputation

  1. QUOTE (Anders Björk @ Sep 16 2008, 07:39 PM) How would you implement it? I used this because it was the first to come into my mind, given my limited knowledge of labview
  2. QUOTE (Anders Björk @ Sep 16 2008, 10:41 AM) Yes, in fact I've done in a similar way; I've downloaded the circular buffer VI but it doesn't work with my labview version. This is however a simplified version of what is going on in my VI (the FFT here is done on 10 samples and is updated every sample).
  3. QUOTE (jdunham @ Sep 15 2008, 09:15 PM) Circular buffer.. I didn't know about it (I'm quite new to labview). Very interesting topic, basically is what I was doing with the array but here I can get some priority guarantees which I was not considering in the other way. I'm acquiring data from a sensor via the CANopen protocol with the NI usb-8473. When the sensor is operative it sends messages at a fixed rate (up to 1KHz) and to acquire this data I set up a while loop in which each message is read and decoded when it is ready in the buffer of the NI device. QUOTE It's somewhat unusual to get one sample per loop, and then perform an FFT, because usually interesting FFTs come from pretty fast data streams, and you are often getting many samples with every iteration. In the starting post I haven't been clear about it, actually, I'm performing FFT at a fixed rate, with a window size dependent on the resolution I want to get. I'm going to try the circular buffer, Thank you very much
  4. QUOTE (Anders Björk @ Sep 15 2008, 08:48 PM) I think you are referring to the point by point section: I tried that, but I would prefer using spectral measurement because it supports windowing and averaging without further coding... Thanks for your reply!
  5. Hi all! In my VI I've got a while loop in which a double is acquired at each iteration. I need to do a FFT on this sequence of double. I'd like to use the "Spectral Measurements" VI, but I need to convert my sequence of double into a signal and the only way I found here is to store the samples into an array and then use "Convert to Dynamic Data" (here I didn't find any way of passing the sampling rate, the program takes 1Hz so I have to modify the multiplier in the frequency scale of the spectrum). A problem here is that Spectral Measurements takes all the samples that build up the signal and does FFT on these, instead I need only a certain number of these (which is dependent on the frequency resolution I want to obtain) to be FFTed while the acquisition goes on; so at each iteration I throw away the first element of the array and append the next, then convert to dynamic data, then FFT with spectral measurements. Is there a simpler way to do this? Thanks
×
×
  • Create New...

Important Information

By using this site, you agree to our Terms of Use.