Spectrum analyzer measurement

[Leiser]

Hi,

well, I´m going to ask a basic question but I´m finding some difficulties when measuring power levels with the spectrum analyzer (SA).

I will reduce my problem to a basic question:

If I connect a signal generator to my SA with a 200MHz sine tone what should be the SA settings to get the most accurate measurement? specially, RB, VW. I have seen big differencies depending on the settings.

What I have tried is to do a 0 span measurement and use a marker. That gives me just the cable losses, around 0.2dB.

 

[VisiGoth]

The analyzer may be giving you power density readings. If you are not putting in a power density but rather a power at a specific frequency you want to convert the power density of your reading into an absolute power by integrating over the instruments bandwidth. The manual should explain what values and factors to use.

In zero span you are creating a downcoverter and a scope so you are looking at teh time domain.

 

[Leiser]

Hi,

Well, I wanted to know how to set the RBW and VBW properly. I am aware that the power measured depends on RBW a per 10logBW2/BW1.

Well, I´ll check it again!

Thanks.

 

[logbook]

Every SA for the last 20 years has an automatic setting for RBW and VBW. If they are set to "auto" you will get a correct reading. If the cursor is set to power density you will get a readout in dBm/Hz or a similar unit. You need to set the cursor so you get a readout in dBm, for example. In other words you need a single unit not something per something-else.

In any case you should be able to read off the screen. The top of the screen is the "reference level". If you are on 10dB per div, the reference level is 0dBm and your signal is one box down from the top, the level is -10dBm.

Set the centre frequency to 200MHz and the frequency span to 10MHz to get a sensible picture.

 

[biff44]

The important thing to ask you is: is this a sine wave tone, or is there some sort of modulation on it? A secondary thing, although less important, is what sort of phase noise is there on the tone (i.e. does it jump around in frequency a lot).

In general, if you make the RBW (maybe 3 MHz)as wide as possible, you will get an accurate measurement. But, obviously, if the signal is jumping around +/- 5 MHz, then a 3 MHz RBW is not wide enough!