LabVIEW

cancel
Showing results for 
Search instead for 
Did you mean: 

Regenerate time domain waveform

Highlighted

Hi all, 

 

I am trying to compare waveform of two-tone signals that being generated ( form RFSG & RFSA) with the oscilloscope. I am unable to regenerate a signal having all the information such as, amplitude (in voltage) , freq. and phase to be compared with oscilloscope. The vi contains all the data are needed with the photo of the scope for the generated two tone. Thanks in advance

Download All
0 Kudos
Message 1 of 9
(620 Views)
Highlighted

Hello Al-rawachy,

I am not sure how to interpret the data in your vi:

LabVIEW_2018-10-03_13-41-20.png

I understood that your voltage array contains the definition of two waves ("tones") in complex number representation. Using math, I find that the first one has a max. amplitude of ~0.069771 and a phase at index 0 of 37.93°. The second wave is ~0.068835 and 37.87°.

Note: amplitude = sqrt((-0.055...^2)+(0.042...^2)); phase = arctan(-0.55.../0.042...)

 

 

I interpreted 1G and 1.001G as the frequency of the both waves. I then assumed that you added both waves to get the one shown in your jpg. So far this seems to be close to correct, as the maximum shown in the jpg fits the sum of the two waves' amplitudes.

 

Fs would be a sampling frequency. However, as 30E6 is much smaller than 1E9, I have no idea what this value is meant for.

 

When creating two waveform with the values I derived from you VI, then adding these, my result looks different from what you expect. Mainly because the difference in frequency is so low (1E9 vs. 1.001E9) that the shift does not happen as fast as seen in your jpg.

 

Edit: Attached my vi, just in case this was about how to create the code.


Ingo – LabVIEW 2013, 2014, 2015, 2016, 2017, 2018, NXG 2.0, 2.1, 3.0
CLADMSD
0 Kudos
Message 2 of 9
(577 Views)
Highlighted

Hi Kaiser,

 

Thanks for your feedback, yeah, all the information that you mentioned for interpreting my data is right. I have added the sampling rate just for clarity.

 

Regarding, the small shift in frequency, Is there any way to capture it in NI PXIe-5663E as happening in the Scope DS 8220 or not? Thanks in advance

0 Kudos
Message 3 of 9
(575 Views)
Highlighted

Actually, I do not  understand how the the small shift in frequency (1 GHz vs. 1.001GHz) given in numbers shall fit the image (scope.jpg) you posted. In there, I see three peaks (@ 2.95E-8s, 3.06E-8s, and 3.35E-8s) that seem to be the superposition of the waves. These are almost consecutive, there is almost no change in width and height, so the two waves seem to be mostly in sync here. However, there is one double-peak in the middle, that clearly shows two peaks that far apart that they are almost to single peaks instead of one superpositioned (@ 3.15E-8 to 3.25E-8). I have the feeling I am missing some information here.


Ingo – LabVIEW 2013, 2014, 2015, 2016, 2017, 2018, NXG 2.0, 2.1, 3.0
CLADMSD
Message 4 of 9
(557 Views)
Highlighted

@ikaiser wrote:

Actually, I do not  understand how the the small shift in frequency (1 GHz vs. 1.001GHz) given in numbers shall fit the image (scope.jpg) you posted. In there, I see three peaks (@ 2.95E-8s, 3.06E-8s, and 3.35E-8s) that seem to be the superposition of the waves. These are almost consecutive, there is almost no change in width and height, so the two waves seem to be mostly in sync here. However, there is one double-peak in the middle, that clearly shows two peaks that far apart that they are almost to single peaks instead of one superpositioned (@ 3.15E-8 to 3.25E-8). I have the feeling I am missing some information here.


My guess is that the OP is not setting up the scope properly. It appears that the scope is undersampling the actual waveform and is displaying a decimated waveform instead. I simulated what the actual and the decimated waveforms look like. It's not an exact match but very close to the OP's scope display. If you want the original waveform displayed on the scope, then it must be set to sample at least to the Nyquist rate.

 

FP.JPG

Message 5 of 9
(549 Views)
Highlighted

Hi Jamiva,

 

Thanks for your valuable information. Actually, the setting was set to default except some parameters were changed to be able to get the signal while the sampling frequency was overlooked. I am just wondering whether the single shot acquisition causes this issue, especially the continuous acquisition is not considered in this measurement. Thanks in advance   

0 Kudos
Message 6 of 9
(541 Views)
Highlighted

@jamiva, thank you very much for mentioning the undersampling, that was very clever! I noticed it but did not draw the right conclusion.

 

@Al-rawachy, I am not sure if I understand your question:


@Al-rawachy
I am just wondering whether the single shot acquisition causes this issue, especially the continuous acquisition is not considered in this measurement.

For a single shot acquisition you most likely have to define a suitable sampling frequency as well. "Most likely", as you did not explain where and how you acquire your data, so I can just guess you acquire more than one sample and do it with a piece of hardware that allows to set a sampling rate...


Ingo – LabVIEW 2013, 2014, 2015, 2016, 2017, 2018, NXG 2.0, 2.1, 3.0
CLADMSD
0 Kudos
Message 7 of 9
(518 Views)
Highlighted

Thanks iKaiser,

 

I have mentioned in the top the current hardware that is being used to generate the waveform. I have been told that the problem due to the one acquisition and I have to run multiple acquisition to be able to get a nice and correct waveform. otherwise no accurate waveform will be detect. Thanks again

0 Kudos
Message 8 of 9
(508 Views)
Highlighted

Hello Al-rawachy,

Is this question answered for you?

 

Actually, the setting was set to default except some parameters were changed to be able to get the signal while the sampling frequency was overlooked. I am just wondering whether the single shot acquisition causes this issue, especially the continuous acquisition is not considered in this measurement.

I am not sure if I understood this comment correctly. I understand that you are referring to the (heavily under-sampled, as jamiva showed) external scope in the first sentence. This device is of course fully independent of your PXI hardware.

 

Regarding this:


I have been told that the problem due to the one acquisition and I have to run multiple acquisition to be able to get a nice and correct waveform. otherwise no accurate waveform will be detect.

Could you please elaborate what you refer to with "no accurate waveform"? Is the accuracy too low? Or the sampling rate?

 

In your first post you only stated two driver types and a general device name: "( form RFSG & RFSA) with the oscilloscope".

In your second post, you stated "Is there any way to capture it in NI PXIe-5663E as happening in the Scope DS 8220 or not?". For the following I'll assume that this question from your second post is related to the first comment, and not independent. Furthermore, I could not find any information on an oscilloscope named "DS 8220". Is it an CompuScope 8220 that you have, maybe?

Very generally, there is a limitation on the hardware in use: The PXIe-5663E you mentioned in your second post consists of a PXIe-5601 RF Signal Downconverter, a PXIe-5622 IF Digitizer, and a PXIe-5652 RF Analog Signal Generator. The Digitizer has a bandwidth that is much higher what the PXIe bus and/or PXIe controller or you PC can handle. Therefore you can either measure continuously, but undersampled (useless data in most cases), or do a high-speed "one-shot" acquisition, that acquires into the 5662's internal RAM until it is full. Afterwards, this data is "slowly" transferred to the PC. In case you have a repetitive signal, you can of course measure it multiple times to increase the accuracy of the measurement. Is that what your colleague ("I have been told") was referring to?


Ingo – LabVIEW 2013, 2014, 2015, 2016, 2017, 2018, NXG 2.0, 2.1, 3.0
CLADMSD
0 Kudos
Message 9 of 9
(439 Views)