Thanks. Setting SP145 to auto indeed fixes this.
I wanted to get the absolute minimum noise (not having any attenuation or amplifier in the input path) but this is fine for my use case.

Thanks Justin.
So SP145 is less suitable to do harmonic analysis on a signal generator’s output? (let’s say if I need to check the 3rd and 5th harmonics of a 2.4GHZ signal)
Because that’s a use case that I can’t put a bandpass filter for.
I’m curious: how would SP145 behave in this case? (e.g. would it show false artifacts like images that don’t exist?)

[Mehdi]

Thanks Justin. It actually moves, as I move around the center frequency.

Does this happen on BB60D only, or SP145 and SM family could also show similar behaviour?

EDIT:
I played a bit with span settings, and now I can see other spurs (depending on the span, and center freq)
It’s a bit concerning, because then from now on, every time I have to check if the spurs I see are artifacts of the spectrum analyzer itself, or exist in the signal.
I had never noticed this (maybe because I rarely work with frequencies this low).
Could it be something more likely to happen in the lower frequencies? According to BB60D’s diagram, this specific signal I tested, pass through a different signal path (<130MHZ) than what I usually use.

• This reply was modified 1 month, 2 weeks ago by Mehdi. Reason: Added more details

To provide additional details: I connected my VS60A to my SA124B option 1 (OCXO) and generated a 1GHZ -20dBm CW. Initial offset was -293 Hz. I let them run for 1 hour. And then the offset was -223 HZ (so around 70Hz drift over 1 hour)

But then provided a reference input to VSG60A. Initial offset: -284Hz. Already after 10 minutes, the offset is -224Hz (so, 60Hz drift in 10 minutes)

So, there’s no way for me to tell if the external clock is correctly detected and in-use by VSG60A based on these 2 scenarios

Hi Andrew. I was going to open a topic to ask the same question, and found this existing one.
Could you please add this as a customer feature request for the future updates? Spike already provides similar indication for external reference. Would be nice if we can have a similar feature for VSG60A too.

Thanks

Hi Andrew. No, actually I was just comparing their built-in reference.
It was clear to me that SA124B with its built-in OCXO would have accurate reading after the warm-up, and also not much drift over time.
However, BB60D surprised me with not having any drift over the 2 hour period.
Thanks for your hints.

Ok, so I did another experiment, and the result was a bit surprising.
I connected the output of a GPSDO to SA124B and BB60D.
The initial drift measured by “Frequency Difference Meter” was 0.1Hz and 34Hz respectively.
I let them run for 2 hours. I expected some drift on BB60D, but to my surprise, they both still had the same drift as before!
Maybe I should measure over a much longer period like 24 hours or so.

Thank you.
I redid the tests, with a GPSDO (output set to 100MHZ)

In this test, SA124B showed smaller offset/difference:
SA124B -0.1Hz (BB60D: 34Hz)

I also played with the Sweep Recorder. If I understand correctly, upon Playback, I can only check the spectrum and read the power by Marker, as other parts in menu are greyed out in the playback mode (e.g. the Analog Demod or Freq diff meter), but then as you explained, that won’t be an accurate method to check frequency reading accuracy/stability. Do you know what I can do for this?

Also, FYI I am only curious to test the SA124B’s stability over time and compare it against BB60D for experimentation, otherwise, as you mentioned, I will use an external reference for more accuracy/stability in my real world measurements.

One more question: would I get the same results, if I use the regular SA124B with my own OCXO providing its external ref input?

[Author]

Posts