Tracking generator as a simple signal generator?

[Noody]

Is it possible to use tracking generator as a simple signal generator also? I have just got an SSA-3021x and as it outlputs and tracks it, I figured that it's just a 2.1 GHz signal generator. Is it possible to simply tell it to send a signal at a particular frequency?

Thanks 

[nctnico]

Did you try zero span?

[TurboTom]

If you need an accurate output level, you will have to feed (a portion of the) signal back into the SA and correct it. TG output level is only specified accurate within +-3dB of adjusted value. Moreover, signal quality is only half-way decent with the first harmonic at round about -35dB (tested 10...500MHz) and a measured THD between 2.5 and 3%. Yet, in many situations the TG comes in handy as a signal gen, I frequently use it as such.

Cheers,
Thomas

[rf-loop]

Yes, zero span and tune freq what you want.

Signal is not as from High-End RF generator but more than nothing. Also level accuracy is "poor". In normal TG use we have "Normalize" function for set correction curve.  Also level range is -20 to 0dBm.

But, so or so, here 28MHz.

[Noody]

Zero option is what does it! I knew that I must be easy to do - if only I had known what to do!

Thank you all so much and Merry Xmas! 

[jfphp]

The main problems with the TG  even mounted in costly SA systems (Agilent, Anritsu...) are 
1. spurious and sub/harmonics content : it is not important because the SA always follows the main output but if you feed the TG into another SA is is horrible ;
2. as previouly explained level precision (of course with such a  forest of signals varying with frequency) and maximum leveled power is frequently around 0 dbm. Using an amplifier at the TG output will degrade  (if still possible ! ) spurious, harmonics and, more important broadband noise (have a look at the band 0 10 Mhz-2...3 Ghz of various sweepers in comparison with the upper bands) ;
3. stability : apart some very costly stand alone TG, all the oscillators involved in the output frequency synthesis (in fact the reverse path relative to the SA) -- at least one oscillator --  are not synthesyzed and there is a frequency drift specified (X Hz in Y seconds).

[tefe]

In 0 span, the SA is not only as a simple RF gen, it can also be an oscilloscope with a narrow bw. But of course, the osc can only see an wave envelope, not the shape. You may have noticed the marker frequency became a time.
That's the miraculous 0 span!

[cncjerry]

I've done this many times but an SA (sometimes) doesn't have fine DB control of the TG output.  If you can live with that, then you have a signal, settable with some degree of accuracy when in zero span.

[outremer]

Hi Noody

I use the tracking generator from my DSA815TG spectrum analysor in zero span; spectral purity is far from being good  especially below 1 MHz. But frequency accuracy can be excellent when using an external 10 MHz reference source (I use a cheap GPDSO). The measurements I made with other means up to 165 MHz show a "not so bad" signal.
Having a true 1.5 GHz calibrated signal generator would anyway be out of reach for me.

Regards

[JohnPen]

The Scotties Spectrum analyser group use the the attached 'leveler', designed by Sam Wetterlin, to provide a calibrated o/p for up to 1G.  It will help, but not solve the harmonics problem, but does however give a an almost constant output level ~-8dBm over a wide frequency range.