WatchUSeek Watch Forums banner

1 - 20 of 27 Posts

·
Registered
Joined
·
3,886 Posts
It's like a radio controlled watch. Unbelievable!
 

·
Registered
Unabashed Casio G-Shock fanboy
Joined
·
8,237 Posts
Get out of here! It hasn't gained a second since then!? That's crazy!

And cool, very cool. b-)
 
  • Like
Reactions: serdal23

·
Registered
Joined
·
6,462 Posts
Didn't you say somewhere that it's actually gaining a little bit during the day and losing at night, and it just cancels out? Not that it matters much, though, because it's still bogglingly impressive. An auto that basically keeps perfect time... damn.
 

·
Registered
Joined
·
6,386 Posts
My SARB is that good...for about 12 hours, LOL. That's insane. Please update us in another 2 weeks and lets see where its at then...
 

·
Registered
Joined
·
10,228 Posts
Discussion Starter · #11 ·
It's been really accurate, but it hasn't always been dead on. It seems to run VERY slightly fast every day (under a second, most days it's not visible), and lose a slight amount crown up. On a couple days I've found it lost a second, on a couple others I saw it gained a second, so it's not perfectly stable (but still very stable). Most days, it holds, within my ability to see a difference anyway, the time it started the day as. At the moment, I've found it's best to just let it gain a tiny bit of time a day and let it sleep dial up. Crown up for too long makes it go negative (i.e., 12 hours or so seems to have a visible negative result). But I guess that's to be expected when a watch is so zeroed in like this--it doesn't take much to tip the scale slightly one direction or the other. Overall, it's very impressive. I'll probably begin the power reserve test tomorrow. After that I think I'll get creative and do some automatic winding tests between it and my Omega 8500, like letting them both die, setting the time the same, taking a 25 minute walk in one direction with one watch on each wrist, then at the end of 25 minutes, switch the watches to even it out, and walk back 25 minutes, and see which one runs longer. Of course, I can't isolate the automatic winding system, but I can say which watch, as a whole, is more efficient in ordinary circumstances. I might repeat the test at a jog and average the results, I'm not sure.
 

·
Banned
Joined
·
2,545 Posts
That's absolutely wonderful accuracy, my friend.

My ex military modded SKX173 lost a total of 1 second in 5 weeks. I wore the watch 24 / 7 to avoid the positioning effect, and it worked perfectly.

Seiko movements are very robust and reliable and, if regulated well, keeps perfect time as well.

Wear it in the best of health . . .

Capt. Serdal
 

·
Registered
Joined
·
966 Posts
Excellent results you have, congratulations :) There are many who seem to miss that one well-adjusted mechanical movement by high quality actually is able to keep time better than quartz within a period of time. A quartz are more constant while a mechanical movement can lose a second and then go back to zero.

Your GS seems to be within zero tolerance.
 

·
Registered
Joined
·
10,228 Posts
Discussion Starter · #14 ·
well, now the real challenge begins...the power reserve test, which not only tests how many hours it'll run, but how much the rate varies by amount of power in the mainspring. A lot of accurate watches go crazy here.
 

·
Registered
Joined
·
994 Posts
My GS also doesn't gain or lose any time if I wear it daily for 8 hours and then rest the watch in the dial up position. I think dial up runs 2 to 3 secs fast, and then wearing it will compensate for it by slowing it down a bit.

I will be very interested in your power reserve test, as my watch seems to go really slow near the end of the power reserve (like the last hour or two).

GS Accuracy == |>|>|>|>|>|>|>|>|>|>|>|>|>|>|>|>|>|>|>|>|>|>|>|>|>|>|>|>|>|>|>|>|>|>|>|>|>|>|>MAX!!!
 

·
Registered
Joined
·
15,029 Posts
Hyeah, right!
And Photoshop'd monkeys just might start flying out of my arse!

JK.
Am floored.
 

·
Registered
Joined
·
10,228 Posts
Discussion Starter · #17 ·
The power reserve test is winding up now and it's done well, but I can confirm the super slow finish. The rate of loss more than doubles in the last 8 hours of running. That's a pretty unusual result (normally a watch would speed up here). Well, I don't know how long the watch will last, but I take the last accuracy measurement at the official rating (i.e., if a watch is supposed to last 60 hours but ends up lasting 63, the last recorded accuracy test is just at 60). Here are the results from 0 through 55 in 12 hour increments (whole numbers are 24 hours, .5 represent 12 hours, "final" is not in a decimal equivalent because it's 7 hours since the last number). The +/- number represents the TOTAL gain or loss at that point. Time 0.0 represents the time it is placed in the watch case and when the watch is "perfectly" in sync with time.gov.

Grand Seiko 9S85

0.0: +0
0.5: -1
1.0: -1.5
1.5: -3
2.0: -4
Final: -8


Omega 8500 (comparison)


0.0: +0
0.5: +2
1.0: +3
1.5: +5
2:0: +6
2.5: +8

Interestingly, despite the watch running ever so slightly fast on wrist, it seems like it will lose time in any position off wrist. The 9S85 is actually outperforming the 8500 until the very last part of its life, where it loses 4 seconds in only 7 hours. That seems bad, but it's not like the trend could continue for 24 hours--the damage is effectively limited to 7 hours, so it's basically impossible to lose more than 4 seconds at that point (because the watch will die and it'll be irrelevant). The end results are eerily similar: -8 for the GS, +8 for the 8500. Granted, the 8500 was tested at a full 5 hours later than the 9S85. Which is better? I think this is a beauty in the eye of the beholder sort of thing. Up through the first 48 hours, the GS is clearly the star performer. The watch is a full 33% more "on target" than the 8500 at that point. In the last 5th of either power reserve, the race becomes close and then it becomes whether or not you have a preference for a slow or fast running watch. I'm a fast guy, but this GS' tendency to run slightly fast on wrist means that, in theory, I will gain that time back over the next few weeks of wearing, assuming no extended time off wrist. At least, I'll be getting increasingly close to 0. The 8500 will always be running away from 0, because again, in an eerie parallel, in no position can the 8500 LOSE time (it gains under 1 second/day on wrist). So it's a tough call. Running fast is preferable to slow, but it means you're going to have to set your watch sooner, which isn't the case with the GS.

I don't guess there's a way for me to ever really know, but the far superior ending performance of the Omega may be due to the result of the free sprung balance...it is reported to be better at isochronism than a regulated balance.

Another interesting result in this test is that Omega claims to have used a double barrel mainspring (in sequence) to achieve better isochronism, yet, at least until the very end, they are clearly bested by Seiko's simpler single barrel design. But then again, perhaps it's that end that the double barrel is visibly improving.
 

·
Registered
Joined
·
136 Posts
Interesting results. My Hi-beat runs pretty much the same way. If kept wound and worn during the day, then off the wrist and placed to rest at night, it comes pretty close to a "0" deviation. If I just keep it at "rest" but wind it fully / by hand and let it run at rest, it runs about -2 seconds slow per day.

I also owned an Omega 8500 (newer planet ocean in Ti and liquidmetal). It was scary consistent, but not as accurate as yours. Mine ran about +2.5 per day no matter what. It was a very good watch, with a great new caliber. My new Rolex Explorer II 42mm with their Rolex's newest movement runs pretty much the same. Consistent as all get out, at +2 seconds per day, again.....no matter what. Both the Rolex and the Omega use specialty materials for the balance springs, and employ free-sprung balances, so you are probably on to something there!

Cheers,

Joe
 

·
Registered
Joined
·
10,228 Posts
Discussion Starter · #19 ·
Seiko actually uses a special material for their hairsprings and mainsprings too, the spron alloy. Actually, Seiko uses a custom spron mainspring alloy on just the 9S85, so it gets a special material they don't debut on the other movements.

I'm not totally sure why Seiko, and Japan, are so fully dedicated to smooth balances. As far as I can tell, Citizen, Orient and Seiko have never made a free sprung watch, and obviously (at least with Seiko) it's not because the prices were so low. Historically, the free sprung balance was considered the best option if you could afford to adjust it, but there's an emerging trend towards smooth balances. Breitling surely had enough money to design any sort of movement they liked, and they chose to use a smooth balance on their new B01 (and Watchtime reports excellent performance). Nomos is also an extremely good performer who had the recent option (last 10 years that is) to go free sprung and decided to stay smooth. And of course, there's the elephant in the room, Zenith, which uses smooth balances on the legendary El Primero. So, based on what these respected movement makers are choosing, I think that they know something we don't and the regulated versus free sprung debate has a lot more to it than forumites think.

Also interesting is that virtually all modern high-beat movements use smooth balances. This may have to do with the increasing problem of air resistance at those speeds and the effect that the screws (or whatever they use) on an adjustable inertia balance might be interfering, or it may more simply be the result of them trying to avoid too much weight on the balance itself since the energy cost is much higher on a high-beat and thus it's avoided for efficiency reasons. It's hard to say. As far as I know, the Breguet XXII is the only current production high-beat that uses a free sprung balance:



But this watch is so expensive that it's difficult to apply their techniques, I imagine, to "cheap" watches between $5k and $8k like Hi-Beats and Zeniths. Note, for instance, the full silicon escapement wheel and pallet fork.
 
1 - 20 of 27 Posts
Top