Spectrum Analyzer "Frequency Range" specification

[TheCalligrapher]

Hello

I was looking at various standalone spectrum analyzers available on the Net, trying to find one that can be used for spectrum analysis of audio-frequency signals. I don't have much practical experience with standalone devices of this kind. I noticed that all of them state that their frequency range begins at 9 kHz and extends into some GHz. So my initial impression was that these devices for some reason cannot analyze lower frequencies below 9 kHz.

However, while browsing through the User Manual for RIGOL 800-series analyzers I noticed that their frequency start/stop settings go all the way to 0 Hz, which apparently implies that they can analyze low frequencies. This contradicts my initial interpretation of what frequency range characteristic means.

So, these devices (e.g. RIGOL DSA 815) actually can analyze all the way down to 0 Hz, right? In that case what does that "9 kHz" stand for in their frequency range characteristic? What is frequency range? I'm beginning to guess that frequency range is the range of available user-selectable sampling frequencies for the sweep. Is that correct? And if that's correct, then the currently selected sampling frequency will define the upper limit on analyzable input frequencies (e.g. max analyzable input frequency = 1/2 * sampling frequency).

Am I on the right path?

I'd appreciate any insight.

[bob91343]

No.  The lowest frequency it can display is 9 kHz.  The span is the display width.  So you set, say, 9 kHz center frequency and zero span, and you get a display with no span that covers from 9 kHz to 9 kHz with 9 kHz in the middle.

For audio work there are very few units that can do what you want.  I had one, an HP 3589 I think but it had problems and I no longer have it.  It went down to below audio but it wasn't working well.  Look for one of those.

[rfclown]

The original answer is correct.

The zero span mode of a spectrum analyzer puts it in a mode that's differenct than usual. It is no longer measuring a "spectrum" of frequencies, but just one frequency. Where normally the X axis on the display is frequency, when you put it in zero span mode, the X axis becomes time. It's more like an oscilloscope that is measuring the detected signal at one frequency. But the lowest frequency that can be detected is whatever is specified by the lower limit of the instrument. Mine don't go below 9kHz. Years ago at work they had an HP audio spectrum analyzer, but now people probably do an FFT of the time waveform.

[james_s]

If you just want it for audio purposes couldn't you just use a small laptop with a sound card and software FFT analyzer? I don't think anyone is making standalone instruments like that these days simply because a low cost PC can do the job. On the other hand if you want to measure RF signals you really need a proper spectrum analyzer.

[TheCalligrapher]

No.  The lowest frequency it can display is 9 kHz.  The span is the display width.  So you set, say, 9 kHz center frequency and zero span, and you get a display with no span that covers from 9 kHz to 9 kHz with 9 kHz in the middle.

When setting up RIGOL DSA 815 I don't have to use "center frequency + span". I can explicitly set "start frequency" and "stop frequency" instead. These are two alternative (and equivalent) methods of specifying the display window. "Start frequency" is the left edge of display, "stop frequency" is the right edge of the display.

The manual clearly states that I can set these values as low as 0 Hz. So, what is going to happen if I set "start frequency" at 100 Hz and "stop frequency" at 1 kHz? What is this device going to display if, according to the above answers, it cannot display lower than 9 kHz? What would be the point of these settings going all the way to 0 Hz then?

The zero span mode of a spectrum analyzer...

I'm not sure why "zero span mode" is even mentioned here. My question has nothing to do with "zero span mode". When I mentioned "0 Hz" I wasn't talking about span at all.

Once again: these devices allow setting "start frequency" as low as 0 Hz and "stop frequency" as low as 100 Hz. That would be the left and the right of the display window. Why are they allowing that if, according to you, they can't analyze under 9 kHz? This is not intended to sound as an argumentative rhetorical question. I really would like to know what these devices are going to show on their screens.

[TheCalligrapher]

If you just want it for audio purposes couldn't you just use a small laptop with a sound card and software FFT analyzer?

Because I'm specifically interested in a standalone device.

[james_s]

If you just want it for audio purposes couldn't you just use a small laptop with a sound card and software FFT analyzer?

Because I'm specifically interested in a standalone device.

Ok well you're probably SOL then unless you want to DIY something, I'm not aware of any such unit being produced. There is not really any market since tablets, smartphones and laptops can do it. If you don't mind used equipment there are some vintage devices like the HP someone mentioned, or there was at least one portable unit made with a dot matrix LED display that was intended for professional audio applications like setting up sound systems in concert halls. My friend repaired one a while back and said they were quite rare and valuable. For RF standalone is available because commodity devices are incapable of doing the job, for audio range that isn't true.

[james_s]

Once again: these devices allow setting "start frequency" as low as 0 Hz and "stop frequency" as low as 100 Hz. That would be the left and the right of the display window. Why are they allowing that if, according to you, they can't analyze under 9 kHz? This is not intended to sound as an argumentative rhetorical question. I really would like to know what these devices are going to show on their screens.

Because that's how spectrum analyzers work, sometimes you want to see a specific range, say 0-10MHz so that each division of the graticule is 1MHz. My HP SA can even have the start frequency set in the negative, but the instrument is not capable of displaying meaningful information below 9kHz. These are not consumer devices designed to be idiot proof, they're professional instruments intended to be used by people who know what they're doing. They are less intuitive to use and easier to damage than an oscilloscope which itself is less intuitive to use and easier to damage than a multimeter. There is nothing preventing you from configuring a spectrum analyzer such that it is displaying meaningless information.

[rfclown]

What is this device going to display if, according to the above answers, it cannot display lower than 9 kHz? What would be the point of these settings going all the way to 0 Hz then?

The zero span mode of a spectrum analyzer...

I'm not sure why "zero span mode" is even mentioned here. My question has nothing to do with "zero span mode". When I mentioned "0 Hz" I wasn't talking about span at all.

Once again: these devices allow setting "start frequency" as low as 0 Hz and "stop frequency" as low as 100 Hz. That would be the left and the right of the display window. Why are they allowing that if, according to you, they can't analyze under 9 kHz? This is not intended to sound as an argumentative rhetorical question. I really would like to know what these devices are going to show on their screens.

We are all trying to explain to you how specturm analyzers work. I mentioned "zero span mode" because you mentioned seeing that the span was allowed to go to zero, and you confused that with being able to operate at 0 Hz. When you select a span of zero, the analyzer goes into a different mode of operation that is called "zero span mode". I was trying to explain that. It is actually a very useful mode of operation.

As the other poster says, most analyzers will allow you to set the span (or stop and start) to display frequencies that the analyzer isn't capable of measuring (including negative frequencies). You may think "Why are they allowing that...", but we are just trying to tell you: That's the way spectrum analyzers operate.

[rfclown]

If you just want it for audio purposes couldn't you just use a small laptop with a sound card and software FFT analyzer?

Because I'm specifically interested in a standalone device.

The thing nowadays that does this is an oscilloscope with an FFT function. Personally, I have been frustrated with the interfaces that are presented on scopes where I have tried this (years ago). I haven't tried it on the newer cheap Rigol and Siglent scopes.

[Matje]

Once again: these devices allow setting "start frequency" as low as 0 Hz and "stop frequency" as low as 100 Hz. That would be the left and the right of the display window. Why are they allowing that if, according to you, they can't analyze under 9 kHz? This is not intended to sound as an argumentative rhetorical question. I really would like to know what these devices are going to show on their screens.

The device may allow that because technically there is no reason not to, and it would be extra affort to restrict the frequency setting.

But the device is still specified to work from 9 kHz up, so anything it would display below that would be out of spec, e.g. I would expect greatly degrading sensitivity towards 0 Hz, non-linearities and so on. So even if it can be set to use such low frequencies it can't (and the manufacturer won't guarantee it to) produce reliable, dependable results there.

[rstofer]

It's neither a stand-alone unit nor even readily available but the absolute best way to do this (lower frequencies) is with the Digilent Analog Discovery 2.  There are some units on eBay but who knows what you would get.

Here is an FFT of a 1 kHz square wave.  No attempt was made to pick any particular span.

Alas, it doesn't serve your purposes...