Hi
I need some advice, ideally from someone with experience in servicing watches.
I've just had a new old stock watch serviced with a lemania 5100 movement. The watchmaker's timegrapher shows a perfect flat line, +1spd, excellent amplitude etc. However, when I received the watch it was already running fast and gaining nearly 30 seconds per day.
I've sent the watch back and it's currently being tested, his timegrapher shows the same result as before and it's being tested on an auto winder. I've not had any explanation yet and for all I know, he might have already done some sort of intervention. It will probably be another couple of days before I get it back again.
I've been wearing mechanical watches for 30 years. My least accurate watch is currently a new Citizen Ny0040 which gains 10 seconds per day, I have a Hanhart and a Longines that are within 1 second per day. I'm not expecting chronometer standards but had hoped for better than nearly half a minute a day. The watch in question was keeping better time before it was serviced, though it was stopping (poor power reserve) and the date sometimes was sticking. It's about 25 years old and when I got it, still had the original case back sticker.
It's not as though the watch isn't being worn or is in a poor state of wind. Upon receiving it, I wore the watch every day for a week, noting the timekeeping morning and night against a Multi band 6 G-shock. Position off the wrist didn't make a difference and the timekeeping was consistent day and night - it's averaging +27.5 seconds per day.
I don't want to fall out with this guy as his service has been excellent, it's certainly not my intention here, and I've other watches that could benefit from his services - but I'm at a loss to explain the difference in performance on his bench compared to my wrist. Could it be some sort of positional error, magnatised in the post, or is there another likely cause? Does anyone have experience of servicing a watch only to have a customer send it back with this type of issue?
Sent from my JSN-L21 using Tapatalk
Last edited by AKM; 17th March 2020 at 17:46.
Magnitised seems ruled out by him testing it on the machine and getting the same result as before it was sent to you.
I'd also expect it to be running faster than 30s per day if magnetised.
A dial up reading on a timegrapher is typically c. 5 seconds faster than real world conditions. This, again, offers no explanation for the difference you're seeing.
Best thing is to have the servicer check it themselves over a real 24 hour period.
Sent from my CLT-L09 using Tapatalk
Are positive that you fully wound the watch before wearing? Might need fully winding before relying on auto wind
dunk
"Well they would say that ... wouldn't they!"
I'll see what he comes back with.Originally Posted by stefmcd
Thanks for reading my long post. I didn't want to leave anything out.
I know that one of our members, Paul AKA Walterwerk, likes to actually wear customer's watches for testing, it always seemed strange to me - perhaps there's something I don't know that it mitigates.
Sent from my JSN-L21 using Tapatalk
A simple yet very plausible explanation.
Sent from my CLT-L09 using Tapatalk
I'm as certain as I can be - I manually wound it when it arrived before putting it on the wrist. I've also taken it off for a couple of days to see how long it ran for, before stopping. Then I fully wound it by hand, before starting my timekeeping trial. It's been consistently gaining the same amount of time.
I've lots of other mechanical watches and not had any issues.
Sent from my JSN-L21 using Tapatalk
A nice steady trace with good amplitude is good, but what really matters is the delta (sometimes referred to as positional variation), the difference between the fastest and slowest rate in the commonly tested positions. It can be displaying +1spd in the dial up position but it's meaningless without the other timing data as for all you know it could be losing 30spd in the 9H position and gaining 45spd in the 3H position.
One of the aims when servicing is to reduce the delta to as small a figure as possible but there are myriad reasons as to why it isn't possible to eliminate positional variation completely, and with vintage watches that have been worn for decades there are even more reasons why this isn't practicably possible. As your Lemania is a NOS example it should be possible to get the delta reasonably tight as there should be minimal wear but this is assuming the movement hasn't been messed about with in the intervening years, an inadvertent prod on the hairspring or balance rim with a screwdriver can ruin the timekeeping!
The fact that it's gaining 27.5 spd consistently whilst being worn is a good sign that adjustment and regulation should be able to sort out the issue. If your watchmaker has a Witschi Chronoscope timing machine it can work out automatically what the predicted daily rate is going to be and will allow him to adjust and regulate around that figure.
Thanks for your detailed response.
I hadn't considered whether there could be a defect of some kind. I do wonder about NOS watches as in my mind, there's always question mark over why they weren't originally retailed and whether they were from a defective batch or assembled at a later date.
It had all of the correct retail packaging for the period including leather strap in style they haven't made for years. The photos of the strip down of the watch, revealed the movement parts to be unmarked and other than a bit of dried oil, it looked brand new as if it was made yesterday.
I'll ask if he can check it in other positions. I can't fault the quality of the work otherwise though, as from what I can see, the strip down and reassembly have been done without leaving any marks on the parts or the case. I know that there shouldn't be any marks, but it's not always like that. The power reserve and date issue where also resolved.
Sent from my JSN-L21 using Tapatalk
Watches can run fast when underwound.
Sent from my CLT-L09 using Tapatalk
Nope, I've had a watch that ran fast as it wound down. In the last hour or so of reserve the inaccuracy was relatively significant. Can't mind which one it was as it was about 3/4 years ago. Sticks in my head as it seemed counterintuitive but nevertheless.....
Isn't it +30 the op is indicating?
Sent from my CLT-L09 using Tapatalk
I vaguely remember mine being 20+ seconds out from what it had been a few hours previously. I was testing the power reserve which was why I let it run down.
Sent from my CLT-L09 using Tapatalk
I would think the issue is the movement itself.
Any moron with a screwdriver, no skill or timegrapher can regulate a movement perfectly well.
As I proved myself with a £28 Vostok with cheap hand wind movement that probably cost a few quit to make.
Here were my results comparing to atomic clock. Remember I had only a little flat head screwdriver to open the case and regulate the movement, nothing else.
I observed no detectable change at all for well over a day wearing it on the wrist and off the wrist at night to sleep.
13/02/20
23:35 -1 SEC / OFF WRIST
14/02/20
07:35
08:35 -1 SEC / ON WRIST
09:45 -1 SEC
12:45 -1 SEC
14:00 -1 SEC
15:15 -1 SEC
16:00 -1 SEC
18:25 -1 SEC
19:30 -1 SEC
20:35 -1 SEC
21:25 -1 SEC
22:00 -1 SEC
23:35 -1 SEC / OFF WRIST
15/02/20
08:50 -1 SEC / ON WRIST
09:35 -1 SEC
I've just checked my four year old Speedie before giving it it's daily wind and it was +4s/d with around half of the power reserve remaining. After a full wind it's now running at -6s/d. This prompted me to check my Seiko 4R36 which I gave a few turns on the crown to get it started and gave +34s/d dropping to 0s/d with a full wind. Seiko 6138 went from +40s/d with a few turns dropping to -6s/d with a full wind and all measurements were taken in the dial up position.
It sounds counter-intuitive but as the mainspring winds down the force it puts on the train wheels drops, the friction on the pivots and jewels decreases and causes it to run faster, so it is possible the OP has been checking the timekeeping in a low state of wind.
The reason a watch will run slower or faster as it winds down is usually to do with something called "curb pin error".
On a conventional balance (read not free-sprung) the hairspring is held between the curb pins (or a pin and a boot) on the regulator. The position of the curb pin is how the effective length of the hairspring is changed when the regulator is moved.
Generally
- if the hairspring is centred in the middle of the curbpins when at rest, the watch will lose as it winds down (as the amplitude reduces, there is less contact of the hairspring with the curbpin as it "breathes" between the curbpins, resulting in increased effective length of the hairspring)
- if the hairspring is nearer one of the pins when at rest, the watch will gain as it winds down (as the amplitude reduces, the hairspring spends more time in contact with the curb pins as it "breathes", reducing the effective length of the hairspring)
There are other factors which could affect this, curb pin error is only one of several isochronal errors in a watch.
Most watches tend to lose as they wind down, because textbook hairspring adjustment demands a hairspring perfectly centred between the curb pins. However there are reasons why sometimes a watch may not be adjusted that way.
For info there are whole books dedicated to the subject of watch adjustment, and they do not make for particularly easy reading - it is not just about moving the regulator on a watch, very far from it!
We don’t need watchmakers on this forum any more, just xellos99 with his flat head screwdriver.
Simon
Sent from my iPhone using Tapatalk
Last edited by mycroft; 19th March 2020 at 00:12.
So, what happened?
I sold the watch to an American collector with full disclosure about the issue as I was unhappy with the watch maker's curt / rude responses and didn't want to throw good money after bad, either with him or paying another watchmaker to service the watch a second time.
I don't expect much, just some basic customer service and some common courtesy. I had approached the watchmaker with the issue within his warranty but it wasn't resolved and after a year of illness with COVID, it had become the least of my worries and was no longer covered.
I've since bought a current model Sinn 140 St to replace the Fortis, as I've had good service from TWR who are the official service centre they have excellent communication and I don't have to plead with them to get them to get the job done right. I don't dispute that the watchmaker I used didn't do the work as he provided photos, but his communication and approach over the time keeping issue was really poor and I don't need hassle.
Last edited by AKM; 9th January 2023 at 14:44.
And who was the watch fixer?
"Once is happenstance. Twice is coincidence. The third time it's enemy action."
'Populism, the last refuge of a Tory scoundrel'.
Thanks for the update mate. Sounds like an ex member here I had appalling service with.
Wasn’t me!
Whoever worked on this watch has let the owner down and he’s ended up selling the watch. I’m always very wary of NOS watches, there’s always a story behind them and a reason why they ended up NOS, all too often it’s because the watch had a fault from new and was replaced.
I’ve got a NOS watch on the bench at the moment with a Chinese 7753 clone movement, it’s riddled with problems, all of which can be resolved........at a price.
Posting Permissions
Reply With Quote