Proper setting of Resolution Band Width 2

[ohmrun]

What is the proper setting of the Resolution Band Width of a Spectrum Analyzer relative to the Band Width of an amplifier? Should it be greater, equal, or narrower than the amplifier's Band Width?

 

[VEBill]

Significantly narrower.

The resolution bandwidth shows you the details of what's going on within the amplifier's bandwidth, and perhaps beyond the amplifier's stated bandwidth in case it's misbehaving (e.g. oscillating).

Here's a good intro:

http://www.ni.com/white-paper/3983/en/

 

[MacGyverS2000]

Setting the resolution of the analyzer to that of the amplifier bandwidth would give you an indication of when the amplifier was on/off... and not much else. That's an expensive power light.

Dan - Owner

http://www.Hi-TecDesigns.com

 

[ohmrun]

I'll buy that MacGyverS2000. I gather from these post that you narrow the RBW to detect frequencies within the bandwidth. So the sweep should be as wide as the bandwith?

 

[VEBill]

An amplifier's "bandwidth" is often specified at the -3dB points, so there might be plenty of interesting things going on outside the spec sheet bandwidth.

In general, one might wish to explore around a bit. Set the initial sweep to 'DC to Daylight' just to confirm that nothing funny is going on. And then narrow in on whatever it is that you're specifically interested in.

Worth mentioning that there's a trade-off with resolution bandwidth, in that a scan might take longer. Not a big deal if everything is steady state, but when trying to capture transient activity it becomes an explicit trade-off.

It's worthwhile to connect an antenna to a spectrum analyzer, and explore the effect of the various settings while viewing (for example) the FM Broadcast band.

 

[ohmrun]

"It's worthwhile to connect an antenna to a spectrum analyzer, and explore the effect of the various settings while viewing (for example) the FM Broadcast band." I never thought of that VE1BLL. Thanks for the advice.

 

[electricpete]

I can't say I've heard the term "resolution bandwidth"

However the fft "resolution" (with respect to determining frequency) is sometimes referred to as the "bin width" (it's not exactly the same due to windowing considerations, but it's very closely related).

fft Bin width is a frequency increment equal to (or less than) the Fs / (2*N) where Fs is sample frequency and N is FFT length.

This distinction in terminology (band width vs bin width vs resolution) may be important because a better frequency resolution may imply a lower bin width and bandwidth.

I'm not sure if this terminology is or could be confusing to op. (Maybe it's just me that's confused)_. Feel free to disregard if it's not relevant to what you're talking about.

=====================================
(2B)+(2B)' ?

 

[VEBill]

The explanation is probably that spectrum analyzers (of the old 'swept tune' architecture) have been around for decades longer than the practical implementation of more modern digital techniques (FFT).

At the risk of oversimplification for brevity, the RBW is essentially the IF bandwidth.

 

[electricpete]

I reread your first response and this time followed the link. It was well covered and makes sense to me now that I read in your link how the term RBW is defined (looks like it is roughly the bin width of the FFT).


=====================================
(2B)+(2B)' ?