I’ve heard that the service interval for a new watch is typically a decade. At that point the mainspring likely needs to be replaced and the watch needs to be lubricated.
I have seen some very old automatic watches of people around me. Namely from Movado, Rado and Rolex. All of them are way older than a decade, and their backs have not opened yet, and they keep time very well within their specs.
I think it was pretty common in old days to damage the balance wheel assemblies and the mainsprings since the shock absorbers were not that well, or non-existent. However, modern/upscale watches have much more robust assemblies and alloys.
I may send my watch in for regulation and check in a decade, but I don't expect that its mainspring will be replaced at that interval. Even my entry level Seiko 5 is pretty much spot on after 5-6 years.
On the other hand, my first mechanical, a Swatch Diaphane automatic has died because it had no shock absorber and fell to a soft carpet from my hand, approximately from 1M above ground. It can no longer keep accurate time if it's not flat on a table.
My understanding from listening to watch people is that old lubricants tended to degrade over years which meant that watches back then needed to be serviced every 5-10 years regardless of use. Modern chemistry means they're much more stable, so a watch can last indefinitely on the shelf, but will still have tiny amounts of wear while running (thus hard-on-hard bearing surfaces - jewels - are used to extend the life). I find it a bit ironic that these advances were made after mechanical watches have become close to deprecated for actual timekeeping duty.
the effects of errosion can be mitigated by using a lower beat too - this is why some of the big names in independant watchmaking have been pushing for lower frequency watches - to extend the service interval, I believe George Daniels wrote about it before his death.
It's pretty much physically impossible for a Rolex to still be within spec (+-2 sec / day) after a decade of use.
If the only reason you take your watch in for service is because the time is drifting a lot each day, you've waited too long. At that point the lubricants have almost certainly broken down and the metal contact points are grinding away. The metal that's worn off ends up on other parts of the movement as dust and can be abrasive.
Your Seiko 5 can be cleaned and serviced (you can regulate it yourself with a Timegrapher you get on ebay for $100), but I'd probably just replace the movement with a new one. They are essentially disposable.
> It's pretty much physically impossible for a Rolex to still be within spec (+-2 sec / day) after a decade of use.
Neither confirm, nor deny. It's not my watch, I'm just the messenger of the owner. He said it's not opened. I can't say anything else.
> If the only reason you take your watch in for service is because the time is drifting a lot each day, you've waited too long.
For me a drift for service is a minute/day at most. Also, these are the words of the retailer I bought the watch from. They carry and service from Tissot to Omega, and everything in between for the last 20+ years, so I'm assuming they at least they know what they're talking about (Longines official recommendation is 6-8 years on their website, BTW).
> Your Seiko 5 can be cleaned and serviced (you can regulate it yourself with a Timegrapher you get on ebay for $100), but I'd probably just replace the movement with a new one. They are essentially disposable.
I'd rather not replace a japan made movement which constantly hits +/- 4 seconds every day, esp. if it's designed to be very low maintenance and repair. Of course I can upgrade it in myriad of ways, but I bought it for what it is, and I'd rather keep it stock. A 7S36C can be a very simple movement, but it's not exactly disposable.
I'm aware of the depth of the rabbit hole, and I'm making a conscious choice of staying at the surface, because I rather have less number of dependable items rather than a fleet which needs constant love. I've found the depths of proverbial rabbit holes in other ways, so I'm pretty experienced about the consequences.
