1675 accuracy survey

[shedlock2000]

I know a lot of you don’t care about this at all, but I’m curious what your accuracy rates are for a recently serviced (RSC) 1675. My AD (also an RSC watchmaker) says that they rarely come back from RSC running anywhere near to COSC. Mine runs -8.8spd daily over the last week.

Photo for attention.




Sent from my iPhone using Tapatalk

[flyinghell34]

Thanks for asking the question because I've always been curious myself what others experience.

My 1675 was serviced about 5 years ago after I noticed it loosing about a minute a day. After I had it serviced, which included a new mainspring, with daily wear mine currently runs about -4 seconds per day.

I have to say my 1675 is probably one of my most accurate watches that I own. Not bad at all for a 50+ year old watch.

[shedlock2000]

Originally Posted by flyinghell34

Thanks for asking the question because I've always been curious myself what others experience.

My 1675 was serviced about 5 years ago after I noticed it loosing about a minute a day. After I had it serviced, which included a new mainspring, with daily wear mine currently runs about -4 seconds per day.

I have to say my 1675 is probably one of my most accurate watches that I own. Not bad at all for a 50+ year old watch.

Yours looks sharp on that strap!!

As an diesel engineer (originally), I get that the shifting tolerances due to wear are harder to account for so deviation will likely be greater, but I’d hoped they could still be kept within the broader COSC (not Rolex’s +2/-2.

I’ve just let my 116710 go (I can’t justify two $10k watches), so it’s jarring to come back from -0.4spd over 5 months; -10spd is a minute a week out!


Sent from my iPhone using Tapatalk

[TimeLord2]

I recently had mine serviced by Phillip Ridley in 2020 and it was running about .5 seconds fast per day after! And that was after it being in a sock drawer for about 20 years, However, after two weeks on the wrist I reset the date for a short month and it is at maybe +1 second over the two weeks. I will mention that I set the watch to my iPhone and wear it all the time even sleeping but not in the shower.

[springer]

50-year-old Rolex watches are far and away from still being certified as within COSC specifications but if your watch still is, that is a bonus. The COSC certification was not a lifetime guarantee of accuracy.

[shedlock2000]

Originally Posted by springer

50-year-old Rolex watches are far and away from still being certified as within COSC specifications but if your watch still is, that is a bonus. The COSC certification was not a lifetime guarantee of accuracy.

Yeah, I expected this to be the case in virtue of the wear on the individual components, John. I did wonder how much of it could be tweaked out with judicious attention to the balance wheel, though. I’m assuming if it’s constantly -10spd, for example, that some adjustment would resolve some of it.


Sent from my iPhone using Tapatalk

[shedlock2000]

Originally Posted by TimeLord2

I recently had mine serviced by Phillip Ridley in 2020 and it was running about .5 seconds fast per day after! And that was after it being in a sock drawer for about 20 years, However, after two weeks on the wrist I reset the date for a short month and it is at maybe +1 second over the two weeks. I will mention that I set the watch to my iPhone and wear it all the time even sleeping but not in the shower.

I miss my 16753 — a lot. Yours looks lovely; and 1spd is outstanding.!


Sent from my iPhone using Tapatalk

[1675-David]

Mine runs around +90 seconds a month, so well within COSC spec. I take it off at night and regulate by placing it crown up on the bedside table. Last service was 4 years ago done by an AD.

[Swearengen]

Mine is +5 a day

[timnik]

Originally Posted by 1675-David

Mine runs around +90 seconds a month, so well within COSC spec. I take it off at night and regulate by placing it crown up on the bedside table. Last service was 4 years ago done by an AD.

That is handsome.

May I ask what year it is?

[jedione65]

I got mine about 3 years ago and was running about -6 sec/day and not sure when it was last serviced. I got it serviced at a local watchmaker and was running about -1/day then had it regulated at another watchmaker for and was running about +4 or so. My last measurement was +2.8 over 9 days. I use WatchTracker app to track the timing. Anyway short story is I stopped caring. It’s my designated travel watch and always seems spot on.

[shedlock2000]

Originally Posted by jedione65

I got mine about 3 years ago and was running about -6 sec/day and not sure when it was last serviced. I got it serviced at a local watchmaker and was running about -1/day then had it regulated at another watchmaker for and was running about +4 or so. My last measurement was +2.8 over 9 days. I use WatchTracker app to track the timing. Anyway short story is I stopped caring. It’s my designated travel watch and always seems spot on.

I could live with mine if it were not 5 minutes a month out, as I never take my watch off and I do need to start teaching at the correct time in a morning (5-10 minutes out over a month or two means I can be late starting class by quite a bit). But 2.8s over 9 days is excellent!


Sent from my iPhone using Tapatalk

[shedlock2000]

Originally Posted by Swearengen

Mine is +5 a day

Gaining is certainly easier to manage than losing. That’s a handsome piece, Gabriel.


Sent from my iPhone using Tapatalk

[shedlock2000]

Originally Posted by 1675-David

Mine runs around +90 seconds a month, so well within COSC spec. I take it off at night and regulate by placing it crown up on the bedside table. Last service was 4 years ago done by an AD.

I never take my watch off, I always feel a bit uncomfortable without it on. Time keeping’s a pretty big deal for me because I’m a bit OCD and I don’t really have the time (nor inclination) to pull the crown and re-set the watch every week — but accumulated error is problematic when I’ve to be bang on time for things like class!

I could live with 90 seconds a month, but a minute a week is a bit much.


Sent from my iPhone using Tapatalk

[Swearengen]

Thank you Steve, gaining time is preferable

Its a Mk3 Radial dial

Originally Posted by shedlock2000

Gaining is certainly easier to manage than losing. That’s a handsome piece, Gabriel.


Sent from my iPhone using Tapatalk

[1675-David]

Originally Posted by timnik

That is handsome.

May I ask what year it is?

Thank you! It’s from 1970

[TimeLord2]

Originally Posted by shedlock2000

I miss my 16753 — a lot. Yours looks lovely; and 1spd is outstanding.!

Thanks! Phillip said I was quite lucky that it was running within that range.

[jedione65]

Originally Posted by shedlock2000

Gaining is certainly easier to manage than losing. That’s a handsome piece, Gabriel.


Sent from my iPhone using Tapatalk

I’m definitely at stickler for accuracy at least for any new (to me) watch I purchase. I’ve always wondered why gaining is preferable to losing time especially for these old gmts without a quickset date. Wouldn’t it be easier to adjust the time forward rather than backwards when it deviates?

[Dr.Smellody]

Originally Posted by jedione65

I’m definitely at stickler for accuracy at least for any new (to me) watch I purchase. I’ve always wondered why gaining is preferable to losing time especially for these old gmts without a quickset date. Wouldn’t it be easier to adjust the time forward rather than backwards when it deviates?

I think running slightly fast is more indicative of a healthy movement. Where as running slow is more indicative of drag, dirty, etc. Of course not always.

[shedlock2000]

Originally Posted by jedione65

I’m definitely at stickler for accuracy at least for any new (to me) watch I purchase. I’ve always wondered why gaining is preferable to losing time especially for these old gmts without a quickset date. Wouldn’t it be easier to adjust the time forward rather than backwards when it deviates?

Well, it’s not a matter of vintage-only or health of the movement (it’s not a case of drag or anything — the accuracy in a healthy movement is simply a function of properly adjusting the microstellar screws on the balance wheel — assuming, of course, the movement is healthy).

It’s ‘better’ for a movement to gain time because it’s just easier to re-sync with real time: instead of hacking the movement, adjusting the time, waiting for the seconds to sync, and then pushing in the crown — you can just pull the crown for a few of seconds and it falls back into sync with real time!


Sent from my iPhone using Tapatalk

[Kongelunden]

Beautifully watches in this thresd - wow! I unfortunately cannot chime in on the OPs question, but the 1675 is very high on my grail list - so hopefully will be able to comment in the future :p

[jedione65]

Originally Posted by shedlock2000

Well, it’s not a matter of vintage-only or health of the movement (it’s not a case of drag or anything — the accuracy in a healthy movement is simply a function of properly adjusting the microstellar screws on the balance wheel — assuming, of course, the movement is healthy).

It’s ‘better’ for a movement to gain time because it’s just easier to re-sync with real time: instead of hacking the movement, adjusting the time, waiting for the seconds to sync, and then pushing in the crown — you can just pull the crown for a few of seconds and it falls back into sync with real time!


Sent from my iPhone using Tapatalk

I suppose that makes sense too, adjusting the time by pulling out the crown and letting it set itself to resync. Does not seem the same same for vintage gmts since these were non hacking movements though….

[Swearengen]

I don't think that is 100% correct, from early/mid '70s 1675 had a hacking movement. I'm sure someone will chime in with the exact date this happened

Originally Posted by jedione65

I suppose that makes sense too, adjusting the time by pulling out the crown and letting it set itself to resync. Does not seem the same same for vintage gmts since these were non hacking movements though….

[shedlock2000]

Originally Posted by Swearengen

I don't think that is 100% correct, from early/mid '70s 1675 had a hacking movement. I'm sure someone will chime in with the exact date this happened

I’m no expert, but I think you’re right: the early 1575 was non-hacking, and the later 1575 was a hacking movement. The quickset date appeared with the 3075, I think.

I’m not sure of the date when the 1575 started to hack, though.


Sent from my iPhone using Tapatalk

[Dan S]

My records show that the hacking/non-hacking transition happened in the 2.5M-2.9M range. I haven't been able to pinpoint it more precisely yet. If anyone has a watch in this range, please post or send me the first 4 digits of the serial and specify hacking or non-hacking. I'd like to get more data in this range.

[shedlock2000]

Originally Posted by Dan S

My records show that the hacking/non-hacking transition happened in the 2.5M-2.9M range. I haven't been able to pinpoint it more precisely yet. If anyone has a watch in this range, please post or send me the first 4 digits of the serial and specify hacking or non-hacking. I'd like to get more data in this range.

Interesting. I don’t think mine is that early; I’ll check.


Sent from my iPhone using Tapatalk

[jedione65]

Originally Posted by Dan S

My records show that the hacking/non-hacking transition happened in the 2.5M-2.9M range. I haven't been able to pinpoint it more precisely yet. If anyone has a watch in this range, please post or send me the first 4 digits of the serial and specify hacking or non-hacking. I'd like to get more data in this range.

Mine certainly does not hack but it’s a very early matte dial 1.6m serial. Very rudimentary movement.

[00Seven]

[shedlock2000]

Originally Posted by 00Seven

Good info! Yours is a bit better than mine! Thanks for the data set. Here’s mine:





Sent from my iPhone using Tapatalk

[Tools]

COSC is not criteria wherein which a watch should run, even though that seems to be Internet Lore. Additionally, COSC did not even exist prior to 1974 when it was standardized as the Swiss Chronometer testing organization.

-4/+6 is the 10 second swing that a watch under stress testing can swing in order to pass the COSC criteria.

I would be happy if a pre-1974 watch ran within 5 seconds. If it did not I might ask the watchmaker to regulate as best he can for me.