[time-nuts] Selecting a used HP sweep/frequency generator

[Charles P. Steinmetz]

Jim wrote:

>Some oddities [of 3325A and 3325B's]:
>you can't get non-sine at as high a frequency on the front panel.  I 
>think square waves go up to 10 MHz or so..
>The backpanel waveform is always a sine, but not necessarily a very 
>high quality one.

The triangle and ramp signals only go up to 11 kHz.

The logarithmic sweep on a 3225A or B is accomplished by piecewise 
linear approximation -- 10 segments/decade in single-sweep mode, and 
only 2 segments per decade in continuous mode -- so its utility as a 
continuous, wide-range logarithmic sweep generator (for example, as 
one would generally want for audio testing) is limited.  The markers 
do not work in log sweep mode, either.

As a radio generator, not so bad.  But be aware that 1 mV p-p (~ -56 
dBm) is the minimum output, so you need good external attenuators for 
it to be fully useful in that role.

The 3324A is, in my view, the sleeper of the bunch.  First, it powers 
up the way you left it rather than set to the factory defaults 
(unless you set it to emulate a 3225).  Second, IME the output 
attenuator assembly is more reliable than either the 3325A or B.  And 
finally, they are usually much cheaper.  Of course, if you need 
frequencies down to uHz rather than just mHz, you need a 3325.

Best regards,

Charles