LabVIEW

cancel
Showing results for 
Search instead for 
Did you mean: 

WMS - problem with lock-in detection

Hello everybody!

 

I'm looking for some help with my labview vi for wavelength modulated spectroscopy (WMS). I'm using NiDaq (NI USB-6218) to: generate sine signal, modulate laser's voltage with it and measure signal after passing through gas to compare it with the reference sine wave. My problem is with comparing the two.

 

I'm trying to create a lock-in amplifier that would demodulate the signal at 1f and 2f. I decided to use the phase-insensitive approach to lock-in so I wouldn't have to sync the reference signal. For now, I did the 1f part but It doesn't seem to be working properly. I'm expecting a typical 1f harmonic shape but getting strange looking graph (see attachment). Am I missing something or is my method just plain wrong for the type of measurement I'm trying to do? Should I perhaps scrap the lock-in idea and try to use FFT?

 

I am sorry if this problem is very basic but I'm just starting to use labview and lock-in detection. Appreciate all your help!

0 Kudos
Message 1 of 2
(2,688 Views)

Look at the link, specifically "About Lock-in Amplifiers".

 

To get I & Q (X & Y) you need to multiply your input signal by a Cos and Sin of the same frequency as your signal. I would suggest for now, forget about the AO, acquire the sinusoidal data then multiply it by your reference. (If you want to use the AO, then you need to acquire both the reference and signal and multiply them together, that is, you need to channels of AI input. Actually 3 if you want both I & Q (X & Y), because you need a sin and cos for both.) You then need to low pass filter the data, this is your "time constant" on the lock-in. Once you do that you will have I & Q (X & Y) for your signal, you should then be able to reconstruct your WMS signal.

 

(Typically for WMS I & Q (X & Y) are considered equivalent; for 1f detection you should get a derivative of your lineshape from either I & Q (X & Y), they should be out of phase with each other. One component may be larger than the other depending on the phase angle between your reference and signal.)

 

mcduff

Message 2 of 2
(2,658 Views)