NI virtual Lock-In Amplifier problem

[edvpc2]

Hallo,

 I am using the NI Virtual Lock-In to measure the optical behaviour of a rubin crystal. Because of the optical generated low frequency (70 Hz)  reference signal (chopper wheel), a multiplexed AD Converter (myDAQ) could be used (very small multiplex error) to capture signal and reference. The results are in very good agreement with the measurements with a SR510 lock-in amplifier

 

I tested the NI-Lock-In intensively and I found the following problem, which I cannot understand, because important parts of the NI-Lock-In are secured by a passwords and there seems to be no way to get around it (several questions to the NI support):

Consider a square-wave 4/π*(sinωt+1/3sin3ωt+1/5sin5ωt+…) is applied to a square-wave responding lock-in amplifier, each of the harmonics contribute to the output signal

                      0.5*16/π² ((1/1)²+(1/3)²+(1/5)²….) = 8/π²*(π²/8) =  1

The same with a sine-wave signal results in

                        0.5*4/π = 0.63662

A sine-wave responding lock-in amplifier shows the same result for a square-wave signal (2/π) and 0.5 for a sine-wave signal. I found some differences between the theoretical and the NI Lock-In results.

Signal           Reference        NI-Lock-In     theoretical

sine	sine	1	0.5
square	sine	1.2732	2/π =0.5*1.732
sine	square	1	2/π
square	square	1.273	1

These results can be verified by a simple Lock-In Labview example (see attachment).

Can somebody explain the differences, especially for the square-wave reference (for the sine-wave reference it is only the factor 2) ?

[NinaM]

Hi,

thank you for request.

I can´t run your VI because I don´t have SubVI´s LockInDemodulatorMC.vi, LockInDemodulatorSettings.vi and LockInPLL.vi.

Where can I find them?

Best regards,

Nina

[edvpc2]

Hi Nina,

sorry, forgot to add NILockinStartupKit.llb, where you will find all the missing SubVis. Once this library could be found for download on the NI pages, but this is no longer possible (don't ask why). I actually work with labview 2013, but I have saved the files back to 2012.

By the way, there is a little error in the table (2. row), must be

square

sine

1.2732

2/π =0.5*1.2732

if there are any further questions, please feel free to contact me

Best wishes 

Wolf

[NinaM]

Hi Wolf,

thank you for the VI´s.

I asked one of our specialists. Unfortunately it is not possible to get the password of this VI´s.

I´m sorry I can´t tell you, why there are different factors.

Why is it so important to know that? The result is the same as with SR510 right?

Maybe I can ask them something else?

Best regards,

Nina

[edvpc2]

Hi Nina,

the difference is important, because I want to use the NI Virtual Lock-In in my lectures and in some experiments for students. To explain the principels of a (virtual) lock-in, you start with some simple examples as you can see in the spreadsheet ( sine/square-wave responding LIA to a sine/square wave).

In the meantime I have written my own virtual LIA which is relative simple, only the PLL part is a little tricky and I guess, that the NI LockInPLL.vi is much better than my solution. Unfortunately it is impossible to learn from the NI LIA, because important parts (like the LockINPLL.vi) are closed, which I personally regret very much.

Best regards

Wolf

[NinaM]

Hi Wolf,

I understand what you mean. Unfortunately I can´t tell you anything else.

The VI´s are no longer supported and I have no chance to get a password for that.

There is no possibility for me to get more information about that.

I´m really sorry for that.

Have a great weekend!

Best regards,

Nina

[joe14]

Hi Nina,

We've a simple set up of a white light source and a photodetector, in between is a mechanical chopper with a window frequency of about 280 Hz.

We used one of NI LIA vi programs, MutiChannel LockIn-DAQmx, extracted from the zip file 25010 which is attached here. Our hardware is a NI 6236.

The signal stream( which has chopping/modulating signal superimposed in it ) is coming out of the photodetector and connected directly to the input channel( Ai0 ) of the NI6236. Our understanding is that for the above setup, there is no requirement for an additional physical chopper frequency signal to be connected to another input channel of the hardware; is that right?

We configured the Ai input parameter and ran the MultiChannel Lockln-DAQmx, the PLL is able to output the correct chopper frequency( reference frequency ), the lockerror indicated 0, but there is no value or activity on the X, Y, R or Theta output. We tried sine or square input on the PLL and different combination of the filter parameter without avail. We also amplified the photodetector signal before the channel input and it's the same story.

We tried differ NI LIA program, MultiChannel LockIn - both DAQmx & Simulated Data, from the attached lock_in_amplifier_example file; the simulated data part of the program works, but the daqmx real time part has the same problem, the pll is able to output the rigth reference frequency number but there is no outputs on the X, Y, R & Theta.

Would you be kind to show us some way to do the fault finding?

If anyone has an example simple vi on real time daqmx lock in amp program that works, we would appreciate your help.

Joe

[edvpc2]

Hallo Joe,

"Our understanding is that for the above setup, there is no requirement for an additional physical chopper frequency signal to be connected to another input channel of the hardware; is that right?"

No, you need the signal and reference (chopper signal)  as input.

Look at the Help for the LockInPll.vi

"Signal(In): A 2D-array. The Data Column control (see below) is used to specify which column of the matrix contains the reference signal data."

Then the NI-LIA should work as you can see in the first picture (right side) of the first contribution on this page.

Wolf

[joe14]

Hi Wolf,

Many thanks for pointing us to the right direction.

We were really "goose chasing" in the past.

When we get the program working we will post it here for everyone.

Regards

Joe