[time-nuts] Strange oscillations

[Richard (Rick) Karlquist (N6RK)]

> I should add one more thing:   If this OCXO uses an AT cut
> crystal, the CRYSTAL will have some
> hysteresis and 1e-11 will be a very challenging spec.   True SC's
> do not have this effect at the
> quartz but mounting structures or other components in the

You are confusing thermal transient response with hysteresis.
The old HP10544 used an AT cut crystal and was slow to achieve
frequency stability from a cold start.  The SC cut crystal
was invented at HP Santa Clara to address the cold start problem.
It was used in the 10811 and gets to 10^-7 much faster.

None of this has anything to do with getting 10^-11 in continously
ovenized operation.  In that case, the only advantage of the SC cut
is that the second order tempco is much lower.
BTW, keep in mind that a good OCXO drifts 10^-11 in a few hours
simply due to aging.  You need to have aging prediction and removal
in the first place before you worry about oven performance.
The HP E1938A was better than 1x10^-11 over 100°C temperature
change, once aging was removed.  In fact, its tempco was better
the HP5061B cesium frequency standard and most rubidium standards.
However the E1938A was not really a faux atomic clock because
every so often the crystal frequency would "jump".

Rick Karlquist N6RK