Rigol MSO5000 FFT not filling out graph in ultrascope, but showing on the scope

[ajc9988]

So, I'm attaching a number of pictures to show what I am talking about. The scope seems to be able to display a larger sampling on the scope as compared to smaller sampling on the ultrascope utility. I am hoping to find a way to have the more detailed FFT as seen on the internal display be put onto the computer for a deeper analysis. I'm not yet able to afford a good spectrum analyzer to supplement my equipment. I am just starting my electronics lab and am currently applying this in an audio setting, although my interests in electronics extend far beyond audio. So be gentle, this is my first oscilloscope.

As you can see, the ultrascope display for 25kHz is only 2ms, and above that it only shows half. But, as seen in the other images, the scope seems able to have data above that, with 100ms being 1MSa/s with 1Hz delta f, and it still collects and shows information above that point on the scope (it cannot do 2 seconds as it shows clear issues, and 1 second it starts to show a couple points I don't trust, meaning 500ms is the highest I would even attempt with this scope, although the 1M samples suggest 100ms may be the mark with this range (edit: 100ms if using the sample rate in the FFT section, if using the 1MSa/s in the horizontal box, that is 2s, and at 2s, there are aberrations in the data which makes me question it, but that means the 500ms should be well within range for usable data if that is what the manual meant by 1M samples)).

Any insight into why the PC software will not display the information would be greatly appreciated. I'm also open to scripts that can pull the data and have it usable, similar to the PyDSA for the Rigol DS1000 scopes. This is to examine distortion. Thanks in advance!

[ajc9988]

So, here is the answer. You have to use the memory tabs in the Ultrascope program.
https://beyondmeasure.rigoltech.com/acton/attachment/1579/f-0695/1/-/-/-/-/DS1000Z%20DS2000%20DS4000%20DS6000%20UltraScope%20Getting%20Started%20Guide.pdf

On page 6 et seq., you have a situation where you click stop, then press memory, select the channels, then click update. Doing it this way, this scope has great images at 2 seconds for the time domain when using 25kHz or less for the frequency.

Now, I am looking for a decent way to convert the FFT chart from that to a .csv file for conversion from dBV to other dB scales, but this is solved for those wondering how to do it with the MSO5000.

[ajc9988]

Also, here is the email contents I got from Rigol support:

It looks like you're running up against the display data limit on Ultra Scope. When running Ultra Scope normally, it's only taking a certain amount of points from the display and building the FFT with that. It's actually separate from the FFT calculation that takes place on the scope.

Here's Ultra Scope reading a 1kHz square wave:
Image_2022-07-12_11-12-50.png

If you try to build the FFT with just this data, it will only go up to 500 Hz:
Image_2022-07-12_11-17-16.png

Instead, you will need to turn on the Memory setting at the top of the Ultra Scope menu. If your memory depth on the scope is too high, the calculations can be lengthy so mine is set to 1 Mpts. Check the M1 box, and press update:
Image_2022-07-12_11-19-28.png

Building an FFT from the memory data will now provide a much better trace:Image_2022-07-12_11-12-20.png

[thm_w]

You may be better off downloading the data from the scope and performing your own FFT on it, using pyVisa or similar.
Then you have full control over what parameters you want to use.

It can be a slow process though, can't recall if anyone figured out how to speed it up. 5MB/s was reported.