[AMPS] Spectrum Analyzers

Dave dhaupt@bewellnet.com
Fri, 01 May 1998 03:03:50 -0600


Oh, dear, I've kept my mouth shut so far on this reflector, but now
you're on my primary business turf, so I'll butt in.

We application engineers at HP entertian a whole lot of questions about
spectrum analyzer dynamic range, and what's the best input level and so
forth.  The answer to every proper technical question, of course, is "it
depends.".  However, putting forth a baseline observation is worth
doing.

The actual level with which you drive the front end of a spectrum
analyzer is not very important.  What's important is what level is seen
by the first non-linear device in the instrument.  In a spectrum
analyzer, that's the front-end mixer.  In every current HP analyzer, and
all the way back to the HP 141T mainframe series, the front-end mixer
saturates at betwen -5 and 0 dBm.  If you're using a resolution
bandwidth of 1kHz, then the "best" level to drive the mixer with is
-30dBm.  That "best" level means that you can put two signals into the
instrument whose total power is -30dBm, and the third order intermod
products will be at the noise floor of the analyzer.  If your concern is
fifth order products, the level is higher.  If you use a narrower
resolution bandwidth, then the level is lower.

Note that the input power to the mixer is easy to determine: you can
manually select the input attenuation.  Therefore the power to the mixer
is simply equal to the power delivered to the analyzer's input connector
minus the attenuator.  Sometimes there's a lowpass or YIG tracking
filter between the input connector and the attenuator, but such a device
has very low loss, and is distinctly lower distortion than the mixer
itself.

With proper selection of resolution bandwidth and input attenuator,
modern analyzers can measure third order products as far down as -100
dBc.  Those hams who are fortunate enough to own a 140/141-based system
can expect this performance.  Since such systems are available on the
surplus market for ca. $2000 I tend to assume any ham who focuses on RF
design issues owns one.  Of course, I could be wrong.

Online application notes on this, and similar topics, will be made
available within the coming few weeks at:

http://ots.external.hp.com

and yours truly has been informally appointed as the webapprentice for
the site.  Who says you can't learn anything new after 40?

Dave
W8NF



--
FAQ on WWW:               http://www.contesting.com/ampfaq.html
Submissions:              amps@contesting.com
Administrative requests:  amps-REQUEST@contesting.com
Problems:                 owner-amps@contesting.com
Search:                   http://www.contesting.com/km9p/search.htm