WatchUSeek Watch Forums banner

1 - 18 of 18 Posts

·
Registered
Joined
·
2,221 Posts
I can’t wait for this tech to be < $2000
 

·
Registered
Joined
·
229 Posts
I am very impressed
Most watch-manufacturers try to brain-wash me with stuff about horses or what watch Churchill wore. They invest in the perceived value of the brand-name rather than the quality of the product.
Citizen list some expensive procedures which make me think that they have engineered the watch to meet 1spy with a high degree of confidence -
> pre-aging of xtal & selection of best-performance examples
> individual measurement of temperature characteristic
I also like the achievement of precise hand-alignment (emphasised by a simple dial without date-window).
 

·
Registered
Joined
·
886 Posts
I am very impressed
Most watch-manufacturers try to brain-wash me with stuff about horses or what watch Churchill wore. They invest in the perceived value of the brand-name rather than the quality of the product.
Citizen list some expensive procedures which make me think that they have engineered the watch to meet 1spy with a high degree of confidence -
> pre-aging of xtal & selection of best-performance examples
> individual measurement of temperature characteristic
I also like the achievement of precise hand-alignment (emphasised by a simple dial without date-window).
to be honest, these practices are already done by seiko and citizen itself and for models already for sale in the last 20 years

for the 0100, citizen did a good job but is also marketing it with "already seen" stuff

of course all of this doesn't diminish the effort by citizen but the true "secrets" of the caliber 0100 are yet to be truly unveiled
 

·
Registered
Joined
·
229 Posts
Except for the doubling of the TC table, I did not see anything that was not already known.
What struck me was the statements :-
a) Every oscillator behaves differently (because of temperature change)
b) Citizen was already unique in temperature-testing every oscillator

My reading of threads on the Longines VHP was that a significant proportion of the production did not meet 5 seconds per year 'off the wrist'.
This would suggest a weakness in the thermal-correction, probably becuase a 'median' correction-curve is used, rather than measuring the actual curve for each watch.
To temperature-cycle each watch, measure the variation in rate and then use the measured rate to generate a custom compensation-curve must be expensive (either in initial test-cell (capital) cost or per-unit cost (testing time) ).

I wonder if they double the TC table by doubling the number of test-points, or just by doing more interpolation.
 

·
Registered
Joined
·
18,167 Posts
...

I wonder if they double the TC table by doubling the number of test-points, or just by doing more interpolation.
Assuming the compensations needed on the crystal are fairly linear over the desired temperature range, you need two things to make TC more accurate: more significant digits in calibration (by doing longer calibrations) and finer increments of adjustment. This will give you a more accurate inhibition count for each individual crystal.

If the crystal is not linear over the desired temperature range, then you need a table of inhibition counts. But on-the-wrist temperature ranges are actually pretty small. And most accuracy claims seem to be based on wearing the watch every day for most of the day.

If the crystals are sufficiently uniform, you don't even have to do individual calibrations. But traditionally volume production of crystals show sufficient variation between individuals that you need individual testing if you want super high accuracy.

And, of course, you need to allow newly made crystals to age to 'iron out' their internal stresses. As I remember our data, it seems age is not a factor even at high accuracy after a year or two.

It's not magic.

I confess I haven't spent much time on WUS so honestly don't know. Are people finding these new movements are actually meeting the specified +/- 1 spy?
 
  • Like
Reactions: DaveM and Tom-HK

·
Premium Member
Joined
·
8,926 Posts
I can’t wait for this tech to be < $2000
Don't hold your breath. Obviously, we don't know, but I use the Eco Drive One as an indicator...similar custom tech, high end, limited market. 6000 for the initial LE, only dropped to about 2600 or so with the latest models. (Which can sometimes be found for < 2000.) But the black dial 0100 LE is 20% higher. Worse, there's nothing about that model that suggests there will be a big price drop, other than perhaps a less expensive bracelet.
 

·
Registered
Joined
·
1,604 Posts
As I remember our data, it seems age is not a factor even at high accuracy after a year or two.
Loki posted a long term accuracy graph about his Chronomaster.

The aging disappears so easily inside other temperature noises, but he put some extra effort in to avoid that.

BTW Good to see you back!

What was that Eagles song again?
 

·
Registered
Joined
·
18,167 Posts
...

What was that Eagles song again?
“Life in the Fast Lane” (1976)?? Nah, that's not me anymore! (Just turned 72.) After a life of collecting sports cars, I now own the fastest car I've ever had and I've never had it more than 20mph over the limit LOL

Loki's data is interesting. I remember 4-5 spy was about the best you could really expect on a good watch-grade quartz crystal oscillator. This data doesn't contradict that remembrance.

I pulled my Citizen E760 out a couple of months ago. I remember it as quite accurate (appx 5-10 spy) but messed up the cold start three times and put it aside. (I'm wearing a giant heavy TAG 50 ATM-rated diver [an ETA 2892A2 module with a DD Lemania-5100-like chrono complication in a really heavy case!]. Half a minute/day is acceptable on a watch of such a brutish nature!)

I also remember most vendor accuracy claims don't stand up to scrutiny.

I've spent most of my time helping run a local hippy/anarchist/feminist local middle-American FM station. (WCRS LP Columbus Community Radio | 98.3 and 92.7 - a service of The Neighborhood Network and a Pacifica Affiliate)... I program the Old Time Radio stuff... I often run vintage Elgin American adverts! LOL That's mostly my watch related stuff.
 
  • Like
Reactions: tomchicago

·
Registered
Joined
·
1,604 Posts
Loki's data is one of the pearls of the internet.

Sabresoft is another long term one.

About the caliber 0100: It all depends on the fine print. 1 second per year can mean absolutely anything.

But I admire Citizen for daring to go all nerdy. I mean, gears accurate to one micron?

I am interested to see how far one can get, if no expense is spared.

I am now testing a $1 HAQ chip. A DS3231. With, and that fortunately brings us back to the subject, an aging register. I still need to read up on what it does though.

A vintage watch case advert? That's as close as you can get?
Close enough. An advertisement Google did not get their hands on.
 

·
Registered
Joined
·
18,167 Posts
...

I am now testing a $1 HAQ chip. A DS3231. With, and that fortunately brings us back to the subject, an aging register. I still need to read up on what it does though....
So the Maxim folks are spec'ing to about 60 spy for their chip... (Did I read the spec sheet right?) And the aging corrections are externally generated?

Seems the major advantages of the chip are the wide temperature range for that accuracy and the lack of need for initial calibration. I remember you were handy with a soldering iron. How is it going? You have a separate thread on this?

My last pre-retirement job was helping run cell phone networks. We used GPS-based clocks in our switches to set time (for the radio equipment that is --- we never displayed to the users time anywhere near that accurate).
 

·
Registered
Joined
·
1,103 Posts
And a good Mole-man to you!

Loki's data is one of the pearls of the internet.

Sabresoft is another long term one.

About the caliber 0100: It all depends on the fine print. 1 second per year can mean absolutely anything.

But I admire Citizen for daring to go all nerdy. I mean, gears accurate to one micron?


I am interested to see how far one can get, if no expense is spared.

I am now testing a $1 HAQ chip. A DS3231. With, and that fortunately brings us back to the subject, an aging register. I still need to read up on what it does though.

A vintage watch case advert? That's as close as you can get?
Close enough. An advertisement Google did not get their hands on.
 

·
Registered
Joined
·
1,604 Posts
The interesting thing, if you're into that, is that that chip has a bank of capacitors to match the temperature.
Last image on this page

No fancy inhibition counting here. The quartz frequency stays the same, no matter what. Maybe your GS, in my earlier testing, does its TC like this too. A fresh part to the puzzle.

I only have the thing running for a week, and it look good so far.

It has a glamorous role in the near future. It needs to switch on my electric blanket every night.
Electric blankets are only allowed to automatic switch off. Not on. So I need to make a timer myself with an Arduino.
I don't see the need for time synchronization, so it needs to run reasonably accurate for a few years.

Otherwise not a chip for a watch. It needs to stay in deep sleep to match an ETA movement's power use. It makes me appreciate member schröder's fight for every µA.

The aging correction is a 3 sec/yr rate change. Across the board? Not too sure. Still not read the manual too closely.

GPS is the atom-clock in the sky. Why not use that one? You'd be silly not to.
 

·
Registered
Joined
·
348 Posts
at this price point they should think more about design. I like minimalistic and bauhaus watches. But this is a truly uninspiring model. Unfortunately.
 

·
Registered
Joined
·
1,243 Posts
at this price point they should think more about design. I like minimalistic and bauhaus watches. But this is a truly uninspiring model. Unfortunately.

;-) only ONE of many opinions ...
 

·
Registered
Joined
·
227 Posts
When I was making a jump from automatics to quartz I was torn between Citizen and GS. I eventually went with GS and the 9F movement. Granted only good to 10 sec a yr, I know they make a * version rated at 5 sec a yr.

I have found my 9F to be out 1 second at bi yearly time changes, good enough for me. The GS's looks better too, SBGX119 with doom crystal. Granted I have a battery to change every 3 yrs.

If I had to do over I wouldn't change my choice.
 
1 - 18 of 18 Posts
Top