Need help modifying VI blocks for acquiring complex X/Y data

[VD89410A]

1: is the Analyzer's screen when performing measurement exactly as Labview configure VIs command.
2: is the XY graph when displaying the X/Y axis data from XY-axis data VIs. 
=> X and Y data VIs block just do the data read-out from Analyzer, not setup any of its parameters.

[CanadaGuy]

Ah, I see. You are looking at a log mag trace to begin with and that trace looks exactly like what you're getting for your XY graph. I don't see any issue with what you are obtaining from the instrument. What you would want to look at then is toggling between the Real / Imaginary traces (F3). If you select that, it should change the Y scale to volts (or maybe dBV). What you should be able to do is set this trace (A) to Real by pressing F3, then enable trace B (press the button probably) and then select F3 so it shows imaginary.

From there you should be able to modify your VI so that you are grabbing both trace A and trace B, and there you will have your complex data.

Are you using the tracking generator on this instrument to generate the signal? Where is it coming from? The stuff above 100kHz is just noise, the "interesting" part of the signal is the low frequency stuff below that it appears.

What are you measuring?

[VD89410A]

That is a good idea, 
by separating the real and imaginary part into different traces and call it up individually (I'm not sure if the analyzer can be setup like that)

The instrument itself does not state clearly how X and Y should be in the "Zoom" mode, X:real or Y:imaginary or somehow vice versa. It just say that the Time-domain data is Complex. And I'm not quite sure that if changing Magnitude log(dB) into mode F3 will affect the state of measurement ?. 

As I'm trying to perform: Noise and transfer function measurements

I'm not with the analyzer right now, surely I will try it out on tomorrow.  

Thank you for your help tonight!

[VD89410A]

It’s still not working. I realize that the measurement sate can only be recalled for 1 trace at a time only (trace A) and so trace B is another irrelevant measurement setting… Another thing is that the Magnitude setup is actually for Y axis so it’s important to keep it at Log(dB) to maintain the required measurement state. In the image 3, I tried switching the Y axis data format into Real (I) and the result is screen displays at different measurement mode.  

Here are again photos 4-5 I taken directly from Analyzer’s screen and Labview graph when performing measurement, also the data set I exported so you can see the difference.

[CanadaGuy]

I thought I used old equipment in the past, but this sir, beats my record 😛

4.jpg shows that you're measuring a noise level around -110dBVrms (I think that's the unit I can make out). So that's not going to produce a good measurement. I'm still a bit surprised that you can't turn on more than one trace. Can you do it manually if you aren't recalling states? You should be able to just press the trace button so that a second light comes on. If the light moves from trace A to trace B, then I guess it means you can configure multiple traces (say A for magnitude, and B for phase) and then switch between them. Can you try that? You should skip the whole state recall thing for now, just to get the measurement figured out. You can skip all the function calls, and just call the TRAC?  function to get the data, switch to trace B, then call the TRAC? function again to get that data. After that you should be able to assemble things in LabVIEW using the complex numbers blocks.

But first, you need to make sure your data is good, and right now from 4.jpg and 5.jpg, it doesn't look good.

[CanadaGuy]

Oh, and 3.jpg again shows a measurement of 5 uV, so probably not good either. However, 3.jpg with the format as Real is what you want to graph for your trace value. If you graph X (for frequency) and Y (for the real values) then configure it to Imaginary, and grab again, you should have both components and be able for form the complex transfer function from that.

[VD89410A]

Yes, of course I can do that (manually pressing the buttons) configuring multiple traces, toggling between them…I got your ideas and I think of course it would be possible to separate the measurement into 2 traces and read-out each of them as combining X and Y axis
However, the Data format I need for my measurement is in Log scale for both X and Y axis and sadly it can not be separated into 2 traces (as in the picture). And even if it can then the data read-out I think it still would be incorrect due the incomplete of X & Y data VI blocks
the data from those images 3, 4 and 5 are just noise from the Analyzer itself (It’s quite old Analyzer), nothing is measured yet.

And if non of those are possible, I wonder if Plot/Print mode can solve my data acquiring problem?
So there (still) definitely some problem with those X/Y data blocks 😞 

[CanadaGuy]

You can always modify the Real/Imaginary data that you receive in your LabVIEW code or Excel or whatever. Once you have Real/Imaginary, you can convert to Mag/Phase, then take 20*log(mag) to get your magnitude in dB. Your phase would then be in radians which you can convert to degrees or leave it as is.

For plotting frequency response in a complex manner, you need two plots...one is magnitude versus frequency and the other is phase versus frequency. These two plots completely describe the complex function you are trying to receive. Of course, you can plot, Imaginary vs Real to get a single plot in the complex plane, but that would be nonsensical if what you are after is the frequency response.

I don't believe other modes will give you "pure" complex data, unless it gave you Z (Real+j*Imaginary) versus X (frequency) in a single array, but again, that can be constructed from the two sweeps...one for real vs frequency and one for imaginary versus frequency. I've used a lot of instrumentation, and because of how the data is presented, it is very rare to get "complex numbers" straight from the instrument, again, due to the nature of the variable representation.

For completeness, I will add that you can also get magnitude (log or not) and phase (radians or degrees) and convert that (back to linear magnitude if needed) to a complex number. The result will (should) be identical to the complex sequence obtained by getting the real/imaginary traces.

Maybe you can share a bit more about what your output needs to look like?

[VD89410A]

Honestly, I don't clearly understand how producers explain the terms "zoom" and "baseband", neither how data can be represent as complex on instrument's screen. 

I'll try to perform some measurements directly by the Analyzer and we shall see through what is my output need to look like. 
Thank you!

[VD89410A]

Here are my detailed measurements of an operational amplifier:

The Gain and Noise (picture 1-2-3) I pick-up both Real and Imaginary part for both of them    

But for the complex manner, there seem to be no information tell where exactly the complex part is, but it’s definitely has something to do with FFT manipulation before the final output.

And yes, you’re right: there is no “complex numbers" come straight out from the instrument, the Zoom mode mix the input and LO in order to select narrower span at any frequency within the analyzer’s frequency range. So It’s just a buffer step that help establishing the data rather than final output data.

I’ve read a part called “Time data” wrote in Operator guide I attack below: it explain more detail about what is in the zoom mode, I don’t know if it point out any solution….

If it not duel to X/Y axis VIs that aren’t able to read-out the complex value, then I don’t know where is this wrong now? I have the analyzer fully configured by labview but not for this data read-out part...OMG