[time-nuts] Austron 2010B possibilities?

[Chuck Harris]

John Ackermann N8UR wrote:
> Hal Murray said the following on 02/11/2007 01:32 PM:
>>> I had a thought, though -- would it be a worthwhile exercise to use
>>> the 2010B as a cleanup oscillator following my HP 5061A cesium?
>> ...
>>> By the way, the 2010B has three loops, presumably phase, frequency,
>>> and aging.  The first loop has selectable time constants of 5, 250, or
>>> 5000 seconds; the second loop is 50,000 seconds, and the third loop is
>>> 200,000 seconds. 
>> What sort of crystal is in the 2010B and how does it compare to the one in 
>> the 5061A?
>>
>> Is the cleanup coming from a better chunk of quartz or better software/PLL?
> 
> The 2010B has an Austron 1150 oscillator, the same one used in the 1250A
> standard and the 1210D portable clock.  It has very respectable
> specifications (similar to 10811).
> 
> I think the interesting question is whether the short-constant loop is
> clean enough to give an improvement over the Cs oscillator with its 1 or
> 2 second loop to the Cesium beam.  I suspect that maybe the 250 second
> first loop might be the optimum for this situation; the 5000 second
> probably makes more sense when hooked to a noisier LORAN-derived signal.

I used to use a 2010B as a disciplined standard to a loran receiver.  It
really didn't help much.  The problem is it's crystal was a good one, but
the step size of the servo loop that controlled the crystal's VCO input was
way too coarse.  I recall it was as bad as a few parts in 10E10 per step...
So, on average, you were right there with the USNO's C-Beam standards, but
on the short term, you were dithering about some point with a few hundred
second period on the dither.

-Chuck Harris