PXI4462 slop samples and ringing when sampling fast switching signal

[cupid]

Hi,

I am using PXI4462 dynamic signal acquisition card in Labview8.6.1 enviroment. I use the card to sample at a rate of 1000SPS.

I found that when I sample a fast switching signal, such as a square wave, there are always samples on the slopes of the switching signal. Besides this, there are also very big ringing in the samples which are near to, both before and after the slopes.

Is there anyone who can explain this to me?

Thanks

Shaobo

[Seth_B.]

Hello Shaobo,

I've linked a KnowledgeBase Article below that explains the phenomena you are seeing. Because DSA device, such at the PXI-4462, were not designed to measure signals such as square waves the on board signal conditioning that helps make them so accurate for measuring dynamic signals relating to sound and vibration applications causes this ringing behavior.  The KnowledgeBase article explains the reason in more detail.

KnowledgeBase 2OIFLC2H: Why Do I Get Ringing When I Output or Measure a Square Wave with a DSA Board...

Regards,

[cupid]

Hello Seth,

Thank you for your answer. From your answer, I learned that PXI4462 is not proper to measure fast swtiching signals and the ringing is a very common phenomenent. Although I don't know exactly why there is ringing, I can reduce its effect simly by digital filtering.

Here I am more interested about how to avoid having sampls on the slopes of a square wave. To understand the reason of this phenomenent, I need to know clearly how the PXI4462 works on the input voltage, especially the parameters of the anti-alias filters. I hope you can explain it as clearly as possible.

Thanks again

Regards

Shaobo  

[Seth_B.]

Hello Shaobo,

Basically, a square wave is really an infinite combination of odd harmonic sine waves, each of a different frequency.  Because the DSA boards use anti-aliasing filters to help ensure the best performance on frequency related applications, some of these sine waves are filtered out.  The result is that you start to see ringing that was previously damped by the now-filtered sine waves.

The parameters of the anti-aliasing filters are listed on page 3 of the PXI-4462 Specifications and depend on the settings you choose for you application.  It should be noted that in that chart, fs is the sampling frequency you set.

In order to avoid recording samples on the slope of the square wave, the best method would be to sample at your frequency of interest, then do analysis on the data and discard samples that show the traits of being on a slope.  Specifically, you should compare each sample with the one before and the one after.  If the sample before and the sample after differ by more than some threashold, discard the current sample.

I hope this helps.