[time-nuts] 5370B practical question

[Bruce Griffiths]

Christophe Huygens wrote:
> Hi,
>
> I am trying to start some Allen Deviation measurements on some VCXOs
> I have built those for 117, 106.5 and 123MHz (amateur radio use).  The
> phase noise
> it not my primary concern (for now) but short term stability is
> (systems are locked
> to GPS etc).  I would like to more formally characterize the
> performance hence
> my interest in AD.
>
> I am using the 5370B. For the moment, to get my feet wet, I am just
> measuring
> frequency - I don t want to deal with TI yet. I think I understand the
> limitations of
> both the method used and the counter.
>
> Upon measuring the 106.5MHz results are more or less as expected: I
> see excursions
> around center of +/- 20milliHz... (1second measurement time) I think
> these are due
> to hitting the frequency performance limit of the instrument.
>
> But at 123.5MHz, about 1 in 5 measurments, there is a 1Hz jump... the
> instrument
> showing 122.999998005 in stead of 999005. I never notice 100, 500 mHz
> jumps,
> only +/-20 and (+/-20 - 1000).
>
> I don t know much about the internals... is this the counter hitting
> the 100MHz limit,
> is something broken, or not likely is my VCO jumping by 1Hz?
>
> Thanks for any insight,
> Christophe
Christophe

The performance of the 5370B as a frequency counter isn't stellar, its
much better at measuring time intervals.
Have you checked the alignment of the internal frequency multiplier chain?
Adjustment of this is critical to the 5370 performance.

What is the input signal level?
It is possible that the amplitude of your input signal is marginal at
123 MHz.
Is the 5370 set for 50 ohm or 1 Megohm input impedance?
Is the 5370 input set for 1x or 10x attenuation?
What trigger level setting are you using?

What type of crystal oscillator does your 5370 use?
Are you are comparing your VCXO's against this internal oscillator or
are you using an external standard?

Bruce