I have two analog AC inputs through a DAQ. There is a phase and aplitude relation between the two that I am interested in obtaining. How can I do that?

[MaddenS]

I have a question concerning signal processing. I  have two AC voltage channels that I am capturing using a DAQ device and the Acq&GraphVoltage-IntClk.vi example. One of the signals is an applied potential wave which I control. The other is the response of a circuit (it is proportional to a current). I would like to be able to extract the amplitude of the response and find the phase shift between the two AC signals. Basically I need to do electrical impedance measurements on the circuit.

 One complicating factor is that the response signal has a DC decay that I would like to be able to remove to be able to look simply at the AC component. It is a exponential DC decay. Are there any signal processing VIs that would assist in this task or act as a starting point for me to build my own VI.
 
Thanks,
 
Sam
 

[ABCPrograms]

if you want to filter out the DC signal, you could use a high pass filter.  Example can be found here

[Michael_78]

Hi Sam,

I can't help much at the moment but I am about to start a project which sounds very similar to what you are describing.  I am planning to apply a band pass filter to the signals then do a complex fft on each signal.  V/I in complex form will then give me a complex impedance which will include amplitude and phase information.  The idea of the band pass filter is to filter out the dc offset as well as the high frequency components that are introduced by the digitized sine wave I am using as the excitation.

Good luck,

Michael

[MaddenS]

I have a question about the FFT. My understanding is that it will only work on a static signal. If the signal is dynamic, that is if the phase and amplitude change, the FFT will lose all that information? Is that correct? Can you set up a FFT to look at a specific number of wavelets?

[Michael_78]

Take a look at this article: FFT Fundimentals it may help.  I think the section on spectral leakage is important, select the excitation frequency and sample rates to give an integer number of samples over one cycle. I have no idea how well this will work yet, but later this week I plan to start taking measurements and doing the analysis so watch this space- it will be me asking the questions then!

Michael