[time-nuts] FE-5680A Question

Chris Albertson albertson.chris at gmail.com
Wed Feb 15 20:03:45 UTC 2012


I can see the programmer writing code and he knows he has to wait for
warm up.  So he does a sweep then calls wait().    Or it could be the
software is doing more than one function and the dead time is used for
measuring temperatures or currents, self tests or whatever.

I think they do a wide bottom up sweep because they might be looking
for "landmarks", some dips that occur below the one they want.
Possibly this is a cost saving thing.  If you come up and are centered
on exactly the dip you need you have t be accurate.  But this one
hunts by looking at nearby dips.   We really can't know why it is
designed this way and not some other way but I'd bet the reason is so
they could use a cheaper less accurate crystal.

It's like asking why did they change the design so that 5V is
required, saves the cost of a regulator chip and some caps.  They seem
to be concerned with saving even $2 on a $1K device.

On Wed, Feb 15, 2012 at 9:49 AM, Bob Camp <lists at rtty.us> wrote:
> Hi
>
> They do indeed need to know that they start low and go high. They do need to
> sweep to either side of the resonance. Every Rb I've ever torn into does
> those things.
>
> Why sweep *way* low, and almost not sweep high enough? Normally the sweep is
> centered.
>
> Why spend most of the sweep time not sweeping and then sweep real fast? I've
> never seen one that stops at the ends.
>
> Wouldn't it be better to spend the same time sweeping slowly? That's what
> the other designs all do.
>
> We're not going to do much about the dead time, but the centering of the
> sweep is something you will impact if you re-tune the VCXO. When you do
> retune it, it would be nice to know if the centering was dead on when
> manufactured and it's drifted,  or if it needs to be off center in order to
> lock properly under all conditions.
>
> What could happen? There are multiple transitions in the Rb atom. Some of
> them are pretty close to the one we use. If you lock onto the wrong one, you
> are on the wrong frequency. Different designs get around the problem
> different ways.
>
> Bob
>
> -----Original Message-----
> From: time-nuts-bounces at febo.com [mailto:time-nuts-bounces at febo.com] On
> Behalf Of Chris Albertson
> Sent: Wednesday, February 15, 2012 12:36 PM
> To: Discussion of precise time and frequency measurement
> Subject: Re: [time-nuts] FE-5680A Question
>
> On Wed, Feb 15, 2012 at 9:12 AM, Bob Camp <lists at rtty.us> wrote:
>> Hi
>>
>> Which still gets us back to - why the really odd sweep on the FE's? and
>> should you center the VCXO as a matter of routine maintenance?
>
> I think a very asymmetric sweep makes the most sense.
>
> First some history of sailing ships.   Back in the 1600's navigation
> was not perfect and you never knew your exact position on the open
> ocean.   Knowing within 30 miles was hard to do.  So to get to a given
> location in North America from Europe they typically would aim about
> 50 or more miles to the north of their intended destination and then
> when they reached land would sail south until they found the
> destination.   This added a day or more to the trip.   If they tried
> to hit the target dead-on they would likely miss but then they'd have
> to literally guess wetter to go North or South and if they guessed
> wrong it could really be bad so they always headed for enough north of
> the target so there was no guessing about which way to turn.
>
> I think the sweep is done the same way.  If you start way low you know
> 100% which way to go, the lock has to be up.  Seeing as we know they
> are using software in the loop this makes sense.
>
> So on pwr up the uP starts looking with bottom up sweeps.   The sweeps
> fail to find lock because the temperatures of the crystal and Rb are
> to low but after some minutes, finally a sweep finds the lock
>
> Chris Albertson
> Redondo Beach, California
>
> _______________________________________________
> time-nuts mailing list -- time-nuts at febo.com
> To unsubscribe, go to
> https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
> and follow the instructions there.
>
>
> _______________________________________________
> time-nuts mailing list -- time-nuts at febo.com
> To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
> and follow the instructions there.



-- 

Chris Albertson
Redondo Beach, California




More information about the Time-nuts_lists.febo.com mailing list