[time-nuts] Any guesses as to how Citizen is claiming ±1 second/year with using this AT-cut 8.4MHz XTAL?

[Bob kb8tq]

HI

> On Apr 11, 2018, at 5:38 PM, Richard (Rick) Karlquist <richard at karlquist.com> wrote:
> 
> The aging spec on the 10811 is 5 parts in 10^10 per day.
> After 60 days, it could be off 30 ppb.  So what we
> have here is a non-ovenized AT cut that is better
> than an ovenized SC cut.  I'm sure.

As a practical point - they didn’t leave the factory until they hit that spec, Manufacturing
being what it is, having OCXO’s hang around was not a popular thing. The 5x10^-10
per day at 5 or 10 days might well drop by quite a bit a month or two later. Back in the
day, this was a “who knows / who cares / meets spec” sort of thing. 

I’m not sure that everybody buys the “log rule” for aging. I think there are a lot of reasonable
objections to it. That aside, if there *is* a rule that applies to 80% of your crystals …. that’s 
doing very well. By watch standards that is “perfection”.  Measure something over a 10 day
period and plug that into the firmware. Yes, the watch sits in test for 10 days … .and you 
charge for that. 


> 
> I am reminded of the old Accutron ads.  The headlines
> guaranteed so many seconds a day or whatever it was.
> The fine print says they don't actually guarantee that.
> The only remedy under that guarantee is that they
> agree to adjust the watch to be in spec at the
> watch repair shop and hand it back to you.  Thus,
> they didn't have to worry about aging.  Just come
> into the shop as often as necessary :-)

I wonder how many watch shops are set up to re-shoot the memory in a watch CPU? 
Most of them seem to have a hard time swapping a battery and keeping things dry enough 
that the watch doesn’t fog up in the cold ….

Bob


> 
> Rick N6RK
> 
> On 4/11/2018 9:26 AM, tnuts at joshreply.com wrote:
>> That comes out to about 30ppb, and this is a pocket watch so they don’t seem
>> to depend on the temp stabilization of being attached to a human wrist.
>>  https://www.ablogtowatch.com/wp-content/uploads/2018/03/Citizen-Cal-0100-Eco
>> -Drive-Movement-04.jpg
>>  I’ve been reading about the new watch that contains this crystal for about a
>> month, but just saw some more detail today
> 
>> ---
>> AT-CUT QUARTZ CRYSTAL OSCILLATOR
>> While AT-cut quartz crystals have indeed been in production and use since as
>> early as 1934, the technology is more common in larger applications and not
>> necessarily wristwatches. To address the needs of individuals seeking only
>> the most accurate performance in a wristwatch, Citizen sought to apply and
>> optimize this available technology in a way that could serve watch consumers
>> on a more direct and personal level. When working to reach the accuracy of
>> the Cal.0100, Citizen opted for an AT-cut quartz oscillator instead of a
>> more traditional tuning fork shape (XY cut). Perhaps most notably, AT-cut
>> variations allow for greater temperature tolerances, specifically in the
>> range of -40°C to +125°C. Additionally, this configuration allows for
>> reduced deviations caused by wearer orientation, which can cause significant
>> changes in accuracy that aren't negligible when attempting this kind of
>> performance. As a result, wearers will not have to worry about errors caused
>> by spatial orientation and positioning becomes less of a concern. The same
>> can be said about durability, which Citizen also improved upon in
>> conjunction with the AT-cut oscillator. After all, shock experienced in
>> day-to-day situations could easily prove detrimental even for quartz
>> movements. And when the goal is an annual accuracy of ±1 second, that just
>> isn't acceptable.
>>  https://www.ablogtowatch.com/citizen-cal-0100-eco-drive-watch-movement/
>> ---
>>  Is this possible with an MXCO running across this wide temp range? How are
>> they compensating for aging at this level of precision?
>>  Thanks!
>> _______________________________________________
>> 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.