WatchUSeek Watch Forums banner

Don't exercise with an mechanical? (How my Seiko 5 gained 20sec in 40 minutes)

19K views 107 replies 37 participants last post by  Jet Jetski 
#1 ·
My Seiko 5 gains 20 seconds each time I swim with it; it's a brand new SKN607, received and set its time accurate to the second this past Friday.

I first noticed this on Monday when I took it for a swim, so I duplicated that workout today (500m each of front crawl, butterfly, breast + 200m back) but with a before-and-after-swim pic. (The gain seems to be exactly 20 seconds each time; I've been taking nightly pics to check the watch's precision since Friday.)

I'd like to know:

Is this normal for automatic, mechanical watches? Or
Is this just a Seiko 5 issue and not applicable to the better-made automatic watches?

TIA.

Pre-swim: Atomic 14:21; Seiko 14:45; delta = 24 seconds


After 40 minutes of swimming: Atomic 56:35; Seiko 5 57:19; delta = 44 seconds
 
See less See more
2
G
#3 ·
No, not normal to gain that in such a short time.

I have a 7S26 movement Seiko but I can't say I've done much vigorous activity with it. Mine loses a few a seconds day if I'm wearing it and gains a few seconds if it's wound and sitting face up.
 
#4 · (Edited)
Since this is a new watch, I've been taking nightly/daily pics against my atomic Casio. Here's the history:

Fri Set to atomic time
Sat atomic = 40:01; Seiko 5 = 40:02; delta = 1
Sun atomic = 25:35; Seiko 5 = 25:37; delta = 2
Mon atomic = 49:02; Seiko 5 = 49:24; delta = 22 (swim)
Tue atomic = 18:54; Seiko 5 = 19:19; delta = 25
Wed atomic = 56:35; Seiko 5 = 57:19; delta = 44 (swim)

So, if I don't swim, the watch seems to gain 1 - 3 seconds a day, but I noticed the 20 sec. jump on the days I swim.

This watch is only WR (and not WR-50m), but it hasn't leaked yet. Yes, I'm taking a risk swimming with it, but it is my experimental watch (it was only $60) to see if I might like a nice mechanical.

Here's larger pics ; it gained 20 seconds between the 2 pics (the 40min swim took place between the pics):
 

Attachments

#5 ·
This is because its an automatic watch. Imagine if you were walking with someone behind you pushing you to go that little bit faster. Its the same principal, the oscillating weight is winding the mainspring, although the mainspring has a built in slipping device its still recieveing the extra push from the weight. If the oscillating weight is consistantly winding then the watch will slightly gain. Also there needs to be enough lubrication on the barrel for it to slip properly, otherwise the mainspring can have a little to much friction causing a gain in amplitude resulting in a gain in time keeping... *takes breath* if not your watch is posessed :)

Sent from my GT-I9505 using Tapatalk
 
#6 ·
Booooooooooogus! :) Well, sort of - if your watch has MAJOR isochronism problems (difference in rate from fully wound to fully empty), then yes, this could be a problem, but my guess is your regular motion keeps it fully wound, and you can't explain this difference based on that.

It is more likely caused by the impact of each stroke entering the water. Mechanical watch accuracy is affected by impact/vibration etc, whether it is hand wind or automatic. The cheaper the watch, and the less positionally adjusted (like a Seiko 5), the more prone you are to seeing variance. Ball has invented "SpringLOCK" to eliminate (or at least greatly reduce) the problem.

You could test this theory by doing your swim with a less impact stroke, like breaststroke.
 
#8 ·
Those gains are certainly noteworthy, but you might want to just wear the watch normally for a while and see how it settles in. Constantly winding the watch shouldn;t make it run faster - if that were true, than everyone using a watch winder would have a pile of very fast watches.

Some of these Seiko movements require a bit of a break-in period, and then after that see if you can identify a more predictable daily loss or gain, and then think about having it properly regulated. Which doesn't have to be expensive.
 
#10 · (Edited)
I've regularly been playing sports with my Seiko 5, since I bought it in 2007. No problems whatsoever. Maybe your piece is faulty?

With that said, I would *NEVER* take a mechanical swimming. No matter how cheap, water resistant, or how little you appreciate the watch, I feel that it is a recipe for problems, that can be solved with a $10 Casio quartz.
:rodekaart
 
#22 ·
It could be faulty, and I'm still within Amazon's free return period so I need to make a decision.

I bought this watch mainly as a test to see if I'll like wearing an automatic with bracelet band (both new experiences for me) before getting a nice one. I've worn quartz digital watches all my life (except for a mickey mouse analog watch that I don't even remember wearing); but after reading this forum, I thought maybe I should replace my Casio with something different. So, this test watch must go though all the normal things I do with my watch during the "evaluation" period.

I am beginning to feel that perhaps this watch is faulty or that it is just too low end to give me a fair assessment of automatics.

Does anybody near Redwood Shores CA want to rent me an automatic for a swim?
You can stand next to the pool if you like (or get it and have a workout); I think I'll just do 10 minutes of swimming (no need for my normal workout).
 
#16 ·
That an automatic watch gained some time while exercising is not terribly surprising but I think strange things are afoot at the Circle K if this happens every time it's used. This is why I run and workout with an Ironman; it's built for this type of activity. I'm certainly not knocking you or the quality of your watch, but you may want to better match the watch to the activity. I wouldn't run in my vintage le Coultre, and I wouldn't wear my Ironman with a suit.

-hayday
 
#23 ·
Proper tools for the job; this I get. I'm also considering that a mechanical (automatic or else I'll definitely forget to wind the watch) may just not fit my life style currently. I'v always only owned 1 watch at a time and currently thinking about replacing my (very new) Casio g-shock with a nice watch. <-- A side effect of reading this forum.
 
#17 ·
Isn't this a rather notorious side affect of the Seiko movement, and the reason that Swiss fanboys opine about their properly adjusted movements? I don't know, I'm not an expert and I read too much here to properly digest any of it. :)
It's not "notorious" -- it's just what happens with a low-end movement. A Seiko that costs as much as a good Swiss watch will have a movement comparable to that good Swiss watch (quite possibly better, actually). I doubt any "fanboys" are really bragging about how their $2000 watches perform better than something that costs $48.

Given the choice, I'd take a high-end Seiko movement over most others, frankly.
 
#46 ·
It's highly unlikely that you're walking fast enough to observe this effect. Your hand motions has to match the 3Hz (6 half vibrations) resonant frequency of the watch. You have to be probably running/jumping as fast as you can before it speeds up. Again, this happens to the best mechanical movements, even a 21,600bph Patek will speed up after a vigorous run or exercise involving lots of arm movements. A higher beat rate watch would less susceptible but i guarantee you that even if you strapped a 28800bph watch to Ursain Bolt as he ran 200m, it would speed up by as much as 5 seconds.
 
#21 ·
Seems strange.
I run and bicycle (rough roads) with my SKX009 regularly and it doesn't gain/loose any noticable amount. It's not the most accurate watch by a long shot (-15sec/day roughtly) but that doesn't change much no matter what I do with it.
The only thing I do avoid are actions that have high shock (chopping wood) or very strong high frequency vibration (chain sawing) as that could cause problems. My old Wenger quartz gained 5 or 10 minutes in one day when I wore it riding my old, very buzzy, motorcycle... I wouldn't want to put a mechanical through that!
 
#24 ·
I see dudes at my gym working out with the submariner all the time.
I dont think normal people take gaining or losing a second or even a minute on their watch matters.
At least for me, my watch has the "about right" time, while my phone has the real time.
Especially that I dont put my watches in the winder and keep it running when I'm switching them out day by day.
 
#26 ·
This is supposedly a weakness with "low beat" movements in general. Although really, it's only Seiko 7S26s, 6R15s and a variety of Orients, I think, that actually see much action in the low frequency world. The Pateks, CFBs, Nomoses etc, I suspect, don't get worn during a ton of exercise, so it's difficult to compare.

I've always been kind of suspicious of the shock/vibration resistant theory regarding the frequency of a movement. It's plausible to me, but I'd like to see it demonstrated in a real experiment. The theory goes that each individual shock that disrupts the balance wheel is limited to a decreasing amount of time as frequency increases, i.e., 1/10th of a second on a hi-beat versus 1/8th of a second on a 2824 etc, and over the course of many vibrations, this can result in substantially more stable time keeping. Another way of looking at it is that it can decrease the length of a shock. I also wonder how the energy invested in the balance wheel affects this, i.e., it seems to me that a fast moving wheel will take more to significantly disrupt than a slow moving one, although that's holding all other variables constant, which they're not, particularly since the mass of the balance wheel is different between various calibres.

It is highly unlikely that the automatic winding system, or the power reserve, would cause this, as generally a full wound watch will run slower than a less wound one, although this isn't always true (but if it weren't, you'd observe it on the same watch at the end of a day wearing it anyway).

I guess I would test this with a 6R15 and a 6R20, the latter being derived from the former, but radically updated with a higher frequency escapement. No doubt the balance wheels etc are different, but then, they would be different because that's pretty much requisite. Then I'd take a 9S65 and a 9S85, both of them derived from the 9S55 and thus quite similar. I can't iron out all the variables in the real world, but I think these two head to head comparisons in a vibration test might tell us whether, and how much, the frequency benefits shock resistance. Anyone have $15,000 laying around for an experiment?

I find it all somewhat difficult to conceptualize abstractly.
 
#27 ·
Way to use my "tongue in cheek" statement to make me look like a dick. Kudos! Also, an inexpensive ETA isn't $2000, and isn't as well known to have this problem as this Seiko movement.
Huh? Well, if you can find a comparable ETA watch for $48, then I guess feel free to compare. My point was just that I doubt anyone is saying this movement has a problem because it's not Swiss (or that it's specifically from Seiko). It's an inexpensive movement in an inexpensive watch, and that's really all there is to it.

Seiko 5 watches are amazing value, but if you're comparing anything meaningfully to a Swiss movement you need to go a notch or two up Seiko's product ladder. If I'm misunderstanding you in making that response, then my apologies.
 
#29 ·
Thanks for responding, and excuse my sarcasm. I was parroting a common forum gripe about the lower end Seiko movements, and combining it with another common forum notion that a properly adjusted ETA won't necessarily have some if the same shortcomings.

Forgive my rudeness and happy Halloween!
 
#28 ·
Just from my experience working in the service center in Australia, ive personally worked on 1000s of 7s26's and theres a few things that I observed over the time I was there. The barrel lubrocation from factor is poor and not consistent. Its sufficient for 6 months to 1 year, but then you see changes in amplitude and also rate.
rhe automativ system in the 7s movements is a poor design. The automatic winding gear wears out, and the pawl ends wear also. Its a 2 way winding system so its constantly winding the barrel.
In regards to the swimming I think because you are getting a lot of degrees of movement, the oscillating weight is winding the poorly lubricated barrel and mainspring all the time.
I was just thinking when I swim that im not really shocking the movement in a way that rhe balance would be disturbed causing an Amp drop. Possibly if I was in my boxing class.
Anyway I think its good that we all can nut it out and pribably come to some conclusion.
7s mivements are great movements, they just need a proper service.
If you want value for money then Nomos couldnt be better.
Cheers guys

Sent from my GT-I9505 using Tapatalk
 
#34 ·
The barrel lubrocation from factor is poor and not consistent. Its sufficient for 6 months to 1 year, but then you see changes in amplitude and also rate.
I think that happened on my watch... My 7s26 rate changed from +12 s/d to +23 s/d in a few years and consequently I somewhat hate that watch... this hasn't happened in ETA powered watches.
 
#32 ·
This is extremely normal for a 21,600bph (or lower) movement. During vigorous exercise your movements match the beat rate of your watch, effectively reducing the amplitude delivered to the hairspring, making every 'beat' shorter, and making your watch run faster. If you want to exercise with a mechanical, wear a 28,800bph (or higher) movement, it is much much harder for your movements to be resonant unless you are Ursain Bolt. Or just wear a quartz. This is one of the reasons i don't like mechanical chronographs, especially ones at lower beat rates, since after a run the timing can be half a minute off, making it functionally useless.
 
#33 ·
If you shake vigorously a watch powered by a Valjoux 7750 (including IWC's) you get the same effect. I believe that yifu got it right with his explanation above.

I have found this many years ago, you take your 7750 chrono out of the safe, a few turns of the crown to set it in motion, you adjust the time then if you shake it to add more charge you find that the watch has gained some seconds in the procedure.
 
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