WatchUSeek Watch Forums banner

Tell Us About Your Automatics UnReal Accuracy.

3K views 23 replies 20 participants last post by  HilltopMichael 
#1 ·
Howdy everyone.Up until 2 weeks ago I was strictly a poor mans watch collector.Mostly Seiko's,Orient's & Invicta's.About 3 months ago I sold off 20 some odd watches to fund my grail.When I got the watch on Aug. 21st.I hack set it with "time.gov".I managed to get it perfect & checked it a couple of times to make sure I wasn't 1 or 2 seconds off.10 days later I checked the watch against "time.gov" to see how it was doing.In 10 days the regulated(NOT COSC),ETA 2824-2 movement had gained a total of 3 seconds!NO B.S,+3 seconds in 10 days.It's now been 13 days & it is at +4 seconds!When I ordered my watch I expected +- 3 to 6 seconds per day,I never dreamed it would run like this.So tell us,am I the only person to ever score an automatic with amazing accuracy?
 
#2 · (Edited)
It will change. It is still in the running in period!
but do not worry, even if the watch performa +-6-10 sec a day, it is quite OK!
And yes, I have a couple of watches with a better than COSC.
One JLC Master Moon, and two JLC Reverso.
They have never been COSC tested, but all went through the 1000 hour testing by JLC.
 
#3 · (Edited)
I assume you are talking about your Doxa (very nice, btw). I like autos because of their engineering and functioning. So long as they're within about 10 seconds per 24 hours, I'm happy (and most of my autos are well w/in that range). I don't live my life to within tolerances of 10 seconds per day anyway (hell, not even 10 minutes per day, for that matter).

At any rate, congrats on your upgrade! I'm sure there will be more to come. . . .

D
 
#5 ·
...In 10 days the regulated(NOT COSC),ETA 2824-2 movement had gained a total of 3 seconds!NO B.S,+3 seconds in 10 days.It's now been 13 days & it is at +4 seconds!...
That's quite good of course, but those movements are actually quite good. I don't know what grade of movements all of my watches have, but I can get most of them to be very accurate - at least for how I wear them. My manual wind ETA 6497/8 movements are always within a few seconds of perfect - they're a bit sensitive as to how they're set down at night and I can get them to gain or loose a bit as needed.

I also have a few watches with the ETA 2824 movements and while they're not as positionally sensitive as the 6497/8 movements, they are more consistent from day to day. A Debaufre ran about +1 sec/day right out of the box. A Glycine Lagunare 1000 ran better than 1 second slow/day and after I regulated it, it ran +0.4 sec/day over 45 days (and since it ran a couple of seconds slow the last couple of days I could say that it was exactly +0 seconds off for the last 30 days I wore it) My Kobold LSD has been running at +3.8 seconds/day for the last few months and it's very consistent from day to day. I'm sure that when I get around to opening it I'll be able to regulate it to around +0.5 sec/day or better.

I really wish that one could regulate the watch without having to open the case. I know of only one watch that has this feature, although I suspect there are some others. That way, if one really cared to keep good records it would be trivial to get the watch to run almost perfectly for however you wore it.
 
#17 ·
Well, a COSC movement shouldn't run more than +6 sec/day - it comes with a certificate stating so...

It's not like I've owned enough brand new watches to be anywhere near statistically relevant, but I've never noticed any change in how a new watch runs. Watches don't "break in" as far as I've seen. COSC especially shouldn't change as they've been run for a while just to get that certificate. It would seem pretty silly to spend the money and time to test a movement to some specification and then expect it to change as soon as the customer buys the watch.
 
#13 ·
An Orange Monster that runs 1/2 second fast per day(without regulation).Now that is VERY impressive!Anyone who's ever owned one knows they usually run like a race hourse unless regulated.
 
#8 ·
My Oris Titanium diver runs +3-6 daily depending on position at rest.
I don't think that is Unreal just average. Interestingly though it runs faster when I don't wear it for a few days and hand wind it to keep it going.
I observe it's rate but don't get too hung up about it.
 
#9 ·
My Damasko DA36 is just ridiculously accurate.
It will gain/lose a second or 2 per day and end up within +-3 or so over a month's time with no positioning tricks.
Like that from new and it's a few yrs. old now.
 
#10 ·
Speaking of which, I just finished checking the 24 hour period gain / loss for my Hamilton Khaki - also a 2824 equipped watch - gained 3 seconds over the last day.

Had it for almost a week as of today - I'm quite impressed. We'll see what it averages out to after a week of use.
 
#11 ·
I currently only have two autos...

Seiko Black Sumo (6R15A)-- Runs about +/- 2-4 seconds per day depending on how it is stored. When worn, it's even more accurate. (This watch is less than three months old.)

Hamilton Jazzmaster (ETA 2824-A2) -- Runs about +/- 1-2 seconds per day, again depending on how it sits in the box. But, it is virtually dead on when worn. (This watch is a couple years old, but was just adjusted a couple of months ago).
 
#14 ·
My Tissot PRS516 Valjoux 7750 is dead on out of the box and for the last 5 months. When I sit lazily at home doing nothing, it loses 1 second, and after a very busy and long day it gains 1 sec. That's all. I guess main spring tension plays a role, but a minor one at that. But I don't share the idea of a watch having a break in period. I fail to observe this in any of my mechanical watches.
 
#18 ·
My Prometheus CR1 with a Shanghai 3LZF2 runs 1 second a day slow if left in the draw with the face up, and 1 second fast if worn all day vigorously. I'm on day nine of tracking it and its running 6 seconds slow, so it's losing 2/3 of a second a day.

Tuesday, August 25 7:44 AM: Mark

Wednesday, 8/26 7:44 AM: -1 second

Thursday, 8/27 8:20 AM: -2 seconds

Friday, 8/28 7:32 AM: -3 seconds

Saturday, 8/29 8:00 AM: -4 seconds

Sunday, 8/30 8:00 AM: -3 seconds

Monday, 8/31 7:20 AM: -3 seconds

Tuesday, 9/1 7:45 AM: -4 seconds

Wednesday, 9/2 7:30 AM: -5 seconds

Thursday, 9/3 7:30 AM: -6 seconds
 
#19 ·
Over the last 30 days, I have lost 33 seconds:
 
#24 ·
Yes, it's certainly possible for a watch to vary from day to day yet the differences more or less cancel each other other out over the long run. My manual wind watches are like that. They can vary a bit depending on how I wear them, and then they can also gain or loose a bit depending on how they're set down overnight. With my automatic watches there seems to be less differences from day to day, but there still could be some difference between the fastest and slowest running days.

What I do is simply wear the watch for a week or two and note the exact time roughly every 24 hours. From this I can get an idea of the differences from day to day, and I can calculate the average daily error. If the watch is pretty similar from day to day I can then regulate it so that it's pretty close to perfect for how I wear it. So far just about every watch I've owned could be regulated to be quite accurate for me - certainly well within the stated COSC specs of -4/+6 sec/day. (and yes I know that my watches might not strictly pass the COSC tests, but I could really care less on the movement behaves on a lab bench - I'm more concerned with how it behaves when I wear it.)
 
#22 ·
My Black Monster ran about 10 seconds fast per day. Certainly something that I could live with. I am the opposite of the OP. I was not a "poor mans" collector and had several high end (expensive) watches. I realized that a few seconds a day didn't matter a fig to me and I wasn't going to overpay and coddle some more high priced watches just because they had a perceived value. I am now a huge fan of Seiko, Citizen, Orient watches that keep very good time, are durable, and look great. Love them on my wrist.:)
 
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Top