"Salvador Ferrairo" wrote in message
...
Rigth, I have seen that info, (0.01% of center freqency + 2% of freqency
span + 5 x N MHz), but I quite don't understand it ...

If say I am measuring a 300 Mhz signal, and I have a span of say 500 Khz..

- What is the precision or error that I can expect ?

Any hands on experience with that spectrum analizer, and in special
regarding frecuency marker precision, will be welcomed.

I haven't used that particular spectrum analyzer, but I've used similar
ones.
Let's deal with the first two terms:

0.01% of center frequency ( 300 MHz x .0001 = 30 KHz )
+ 5% of frequency span ( 500 KHz x .05 = 25 KHz )
= 55 KHz

Note: The catalog states that spans 10 MHz have 5% accuracy, not 2%.

This means that the readout may be 55 KHz off and still be within
tolerances. Added to this is the term: 5 x N MHz. What this is isn't
explained in the catalog, but there are other things which affect the
accuracy which may be included in this term. i.e.:
timebase accuracy/stability, readout precision and method of digitization.
Timebase accuracy (0.007%) and digitization (span/intervals) are probably
included in the first two terms. The last term might be concerned with the
frequency displays. Marker readout precision is a function of the number of
significant digits in the display. The display is essentially a frequency
counter and they typically are affected by the time base and have an
uncertainty of +/- one on the least significant digit. i.e.:
If the display shows 300, it may be 299 to 301. (3.33% error)
If the display shows 300.0, it may be 299.9 to 300.1. (0.333% error)
If the display shows 300.00, it may be 299.99 to 300.01. (0.0333% error)
.... and so on.

Hope this helps, I'm not sure I can do more to help.

-NM