[time-nuts] FE-5680A Question
magnus at rubidium.dyndns.org
Wed Feb 15 22:37:29 UTC 2012
On 02/15/2012 11:17 PM, Bob Camp wrote:
> If you look at an LPRO or FRS or any of the other brands of Rb's their sweep
> is very different. They don't spend very much time at the ends of the sweep
> at all. It's pretty much a triangle wave sweep up and back sort of thing.
> They don't seem to care weather they catch the lock going high to low or
> going low to high. The rate of the sweep is *much* slower than what FE is
> doing on these units. Since there isn't a "brain" in most of them, there's
> no pattern watching stuff there. They just look for a lock signal (amplitude
> out of the detector) and stop sweeping when they see it.
Indeed. For analogue logics of the old dinsaurs that I hurd in the
basement a sawtooth or triangle scanning is a good strategy to track in,
but for a processor logic you can do things a bit differently. Also,
consider that the Q seems to be wide on the FEI 5680A you can sweep
faster without missing the dip. Another aspect on speed is just how
large the (needed) scan range is.
More information about the time-nuts