lock in amplifier

[akodu]

Hi,  How can I use a PXI 5922 card to implement a lock in amplifier. Can it be done with external reference imput. Please advise.

Thanks.

[Paul_C.]

Hello Akodu,

The way we implement a lock-in amplifier with a PXI-5922 is using software. We simply acquire the signal using the high dynamic range and resolution of the 5922 and apply the bandpass filter in software.  The frequency of the filter is selected in software.  You won't be able to use an external reference input unless you acquire the reference and measure the frequency.  

I was able to dig up some examples that I found in our database. The Lock-In Amplifier with NI-SCOPE and NI 5922 (original).zip folder contains a modified version of the Lock-In Amplifier toolkit that the you could reference. I have never used this modified version but it is very similar to the Lock-In amplifier download on our website but uses SCOPE functions instead of Traditional DAQ. 

In the LLB which is attached, there are demos and examples that are included for review that use the NI SCOPE driver such as LockInDEMOSimple.vi or LockInDEMO.vi. You will notice though that when you try to open some of the other .vis that you will see a red "Censored" note. This is because these VIs can not be modified or viewed in this toolkit. You are free to use the VIs but not look into them to view the contents. 

Below are some links that might also help you get started:

How to Measure Small Signals Buried in Noise Using LabVIEW and Lock-In Amplifier Techniques
http://zone.ni.com/devzone/cda/tut/p/id/5613

Lock-In Amplifier Startup Toolkit Compatibility with National Instruments Hardware
http://digital.ni.com/public.nsf/allkb/0556EAC2FD846DE5862570C0006FCE21

Creating a Lock-In Amplifier with LabVIEW
http://zone.ni.com/devzone/cda/tut/p/id/4026

This example below uses DAQmx with the Lock-In Amplifier toolkit. It is a good example to review to see how the toolkit can be used and could be easily modified to work with NI-SCOPE functions.

Multi Channel Count Lock-In Amplifier with NI-4472 and DAQmx
<link no longer exists> 

Hope this helps! Let me know if you have any questions. 

Regards,

Paul C.

Message Edited by Paul C. on 01-21-2010 10:44 AM

[akodu2]

Paul , Thanks for the prompt response:  Here is my dilemma..
 I have a PXI 5922 and  USB - 5133 that I want to use as a lock in amplifier (LIA) for spectroscopic measurements. Currently we are using a standalone  LIA unit from Stanford research (SR844). I wanted to implement the LIA in a more compact sensors. I have few options:
1. Use PXI 5922 to implement the LIA as you suggested.
2. Use NI USB - 5133 to see if this can be implemented as LIA. Since its a USB unit it can be easily portable. If it can be used as an LIA, what sort of software will I need and whether this unit will perform as good as PXI 5992.
 We already have hardware for 1 and 2 mentioned  above. I wanted to use these and compare their functionality as an LIA with Stanford research LIA unit and make sure that NI acquisition will be more efficient and equally precise, which I think it will be... but I have to experiment that. If this works out we can get either one of the two NI units mentioned below for enhanced measurements (if this is true)
 a. NI DAQ 4472 , PCI version DAQ and use this as an LIA.
b. NI  Dual Phase virtual LIA     <link no longer exists> .    
I think a. and b. are same units except other one is a USB interface, please correct me if I am wrong. Again all the other options have to be tested and compared with the other to get an optimal solution. Please advice.

[Paul_C.]

Hello Akodu,

The software should be the same regardless of the hardware used.  The performance of the LIA is dependant on the specifications on the device used to acquire the samples. The reason the 5922 would be better is because it has higher resolution / AC accuracy.  The 5922 uses a delta-sigma ADC which also improves the dynamic range.  Here are links to the specifications documents: PXI-5922 and USB-5133.  In regards to your questions about the 4472 LIA vs the dual phase virtual LIA, they are the same except one is just packaged with LabVIEW.

Regards,

Paul C.

[akodu]

Hi Paul,

I am having a bit of problem using 5922 (NI scope) with LIA, probably its my VIs. The LIA works fine with a simple sine input but not for my application. We have a continous data inout from a photodetector with 30 KHz modulation. I am using 1 MS/s at 300 us. Not quite sure what could be the problem?

Does, NI 5992 works with DAQ assistant ?

[Paul_C.]

Hello Akodu,

The 5922 is programmed with the NI-scope driver.  It has its own express vi,which can be found on the NI-SCOPE palette. Just navigate to Measurement I/O » NI-Scope.  I'd have to know more about the signal and what is wrong for me to help.  Perhaps you can post some screenshots of what you are seeing and what you expect to see?  Also, please include the code you are using.

Regards,
Paul C.

[akodu2]

Hi Paul, I have attached a folder that contains the VIs and word dicument explaining our results and brief info on how we are using the Lock in. The word document should give a fair idea so I suggest reviewing that first. The main VI for the application is "DM_VirtualLIA_NISCOPE", other files are the sub VI's used with the program.

Thanks.

[Paul_C.]

Hello Akodu2,

Unfortunately, the lock-in part of the code was written briefly for one of our NI-Week Tradeshows.  I'm afraid I don't know who actually developed the example vi's.  It does look like you are getting good data just not enough points.  All of the examples that I have seen using the same lock in code acquires data continuously and feeds it into the lock in vi's. Have you tried running it multiple times and overlaying the graphs? For example, with the graph you sent me, how many times did you receive the reference trigger?  I'd like to know if this is just a problem of not acquiring enough datasets.

Regards,

Paul C.

[akodu]

Hi Paul,

 The data set in the excel file in the folder  attached in my previous post contains data with several cycles of the acquisition. One could overlay this dataset (of several cycles) and still get similar results (similar to one cycle). I have used exactly the same VI (couples of months ago) with NI 4472 and DAQmx and it works perfectly fine with lock-in. Do you suspect the way NI scope feeds data into lock-in VI compared to the 4472 DAQcould be the difference ? If this is true could you please explain the difference in the way NI-scope feeds the data and NI-DAQ would?

Finally, I didn't quite understand the "reference trigger" you mentioned in the previous post, could you please explain that as well?

Thank you.

[ColeTrain]

Hello akodu,

Are you using the following instance of the NI-Scope Fetch VI?:


This wfm output as a double should be all that you have to wire into the Lockin PLL VI to adapt the example for NI-Scope.  The reference trigger that Paul referred to is what is often used with Scopes to tell us what data to return from the buffer. 

As Scopes can often measure signals at a very high rate of speed, the scopes will look for a reference trigger to decide when to return data to the host. The reason Scopes do not stream all of their data back si the PCI backplane often does not support data transfer rates as fast as the Scope can sample.