Jump to content
  • 0
mcc

SNK809/7S26C with fluctuations in accuracy

Question

Hi,

17 days ago I bought a SEIKO SNK809 with a 7S26C movement. Since I read, that it is good practice to wear the clock 24/7 for the first three weeks (break in) I did
exactly that :)
With the app "WatchCheck" I checked its accuracy. Summed up, its about 5s/day. That's absolutely ok.
What triggers my attention/curiosity is the differences in the accuracy: I usually make a measurement in the morning right after I woke up and in evening after work.
Work does not involve a lot of arm-movements...
After work the watch is running faster up to 12s (max) and right in the morning it has lost up (down) to -4s (min). So the accuracy peak-to-peak in its extrema is 16s/d.

Is this high amplitude only triggered by moving/not moving the watch...or there any trouble ahead?

Thank you very much for any help! :)

Cheers!
mcc

 

 

Share this post


Link to post
Share on other sites

17 answers to this question

Recommended Posts

  • 0

it's normal. It is a mechanical device. it's oscillator , the balance staff  is affected by movement and gravity . Also of effect is how fully wound the mainspring is.at full wind it has more tension and will supply more torque than it does towards the end.

Edited by yankeedog

Share this post


Link to post
Share on other sites
  • 0

What sort of timekeeping were you expecting from your Seiko?

Then accuracy timing machine or 24 hours is not the same thing. Timing machines give you instantaneous in that one position of what it is doing. Watches do a nice job of averaging imperfections over 24 hours. Then if you really want to do the timing machine thing you can't just do it in one position unless you keep your watch only in that position when you wear it all through the day. So you need to do multiple positions and a little math found at the PDF below. But no matter what that's still only tells you what the watch is doing at the moment you had it on your timing app/timing machine. So ideally every 24 hours you compare your watch to something that keeps better timekeeping than the watch.

Then what affects timekeeping simplistically everything. The biggest influence will be amplitude fluctuations both short and long-term. Long-term effect is as the watch is not moving or not moving enough to wind the mainspring the mainspring less power, amplitude will drop timekeeping is affected. Short-term the gear train depending upon its quality doesn't necessarily transmit the power in a nice linear fashion the wheels bind up and unbind and you can get some short-term timing fluctuations. Positional errors in the balance wheel cause timing problems and they get worse with amplitude.Then things like external influences I wasn't thinking about the moon and gravity more along the lines of vibrations & temperature extremes conceivably.

https://www.witschi.com/assets/files/sheets/X-D-DVH-Di-Im-N_EN.pdf

Share this post


Link to post
Share on other sites
  • 0

Hi John,

thank you for your explanations! I am quite happy with the accuracy of my SEIKO...no question.
The linked PDF is interesting! Thank you! :)

The only thing, which drove my curiosity was the value of the standard deviation from the mean value of the accuracy
(that is: the resulting offset of the clock from the nominal value in seconds per day), which is acceptable
in the sense of: The movement need special attention or not. Or to be more precise:
A loss of 1000 seconds at night and a gain of 1005 seconds at day results in the same inaccuracy per day as
a loss of 10 seconds at night and a gain of 15 seconds at day.
For the first case I would tend to say: "This movement needs special attention...there is something horrible wrong."
Since I am a newbie to this topic (watches) I need to develop a sense for "sane values" first...

Cheers! :) mcc

Share this post


Link to post
Share on other sites
  • 0

Seiko and their tech sheets regarding timing specifications finding out what they are can be challenging. For instance for your watch there's a technical sheets for servicing with no timing specifications. So physically the 6R15  is supposed to be similar to your watch not sure if the timing specifications are I'm almost think they might be better but the tech sheet does have specifications which I snipped out and put below.

Then note a difference between the PDF link above? Usually the higher the grade watch the more positions their time and in. So typically eight positions is for a chronometer grade watch. So they give you two sets of numbers what the watch does on the rest and what it should be doing on a timing machine. Then a clarification for the timing machine the definition of fully wound up not specified here is you let the watch run about 15 minutes up to one hour.  Usually the watch needs a little bit a stabilizing time when it's wound up. Then the isochronism number at the end of 24 hours only in one position.

 

Seiko 6r15B.JPG

Share this post


Link to post
Share on other sites
  • 0

Hi,

oh, THANKS! Another "Red Pill" ... I like that kind of stuff... :)
For the 7S26C I read somewhere, that the expected accuracy is between -20s/d and +40s/d but most movements of this kind are
better. Do I interpret T0 as "start of measurement" and T24  as "end of measurement/new cycle of measurement" correctly?
Another question: If a movement has a greater/bigger/larger/huger (choose one...sorry, no native speaker) power reserve ...
does this may imply, that the difference between accuracy when fully wound up and quasi empty is greater as with a watch, which
-- say -- 30 h power reserve ? ....because the force (not Luke's one!) is stronger in the beginning ?
Cheers!
mcc

Share this post


Link to post
Share on other sites
  • 0

I'm not 100% sure what your question is so I'm taking a slightly different approach. I consulted with one of my coworkers as I did not think that the 6R15 timing wise was really equivalent to your watch. So he seemed to think that the 4R25 was more closely related it's a slightly more expensive movement. Then Seiko has an OEM division called time module they make basically the same movements with a slightly differing numbering scheme so the 4R35 is equal to the NH3x series. I'm attaching the timing sections out of each the reason I wanted the time module sheet is it specifies in greater detail how the why and just better at explaining things.

For your mainspring there are technical complications? I've attached an image of mainsprings the lower mainspring is an American pocket watch spring the upper mainspring is a modern mainspring. Notice the back curve? The modern spring is at least some of them have been designed with these interesting back curves to hopefully equalize the force throughout its winding.  So the more modern watches conceivably will have a much more equal force running time through their entire running time or at least a much bigger section then watches that don't have these more interesting mainsprings. So the simplistic of the question is you look at modern watches their designed to run at least 48 hours so that the first 24 should be relatively even in the power curve. If you look at Rolex specifications they go over 48 hours their timing specifications are at Fully wound up and half power which of course is 24 hours.

 

Seiko NH3x timing.JPG

4R35 timing.JPG

mainspring shape.JPG

Share this post


Link to post
Share on other sites
  • 0

Hi John,

thank you VERY MUCH for your explanation, John!

there is nothing wrong with my watch... :)....which I know now from this thread. My initial thought/feeling was, that something /may/ be wrong because the watch had uptp +10 s win at day and -10s at night.
All following questions originated from pure curiosity and interest...nothing dramatic like "my watch has died".
Next question is such a question:
On the internet I read from people, who "has never give my SEIKO 5 to a watchmaker to service it and it still runs" and from other watches, which were way more expensive, which needs to be serviced ever two (?) years."
I am a person, who like to preserve things. I don't like to throw away thing, which do work and I try to repair things, which does not work.
So...what is a reasonable interval/period to give my SEIKO SNK809 (a SEIKO 5 with a 7S26C movement) to a watchmaker to service it?

Thank you very much for all these great information I got from this forum!

Cheers!
mcc

Share this post


Link to post
Share on other sites
  • 0

As far as interval of service is concerned , I am of the opinion that you have to consider the cost versus the benefit.If for example  a new watch costs $100.00 and a full service costs the same , most people are apt to by a new watch.If a watch costs $20,000 and a service costs $2,000 most people are apt to have it serviced. Your Seiko, while a fine timepiece would not be considered by most people as worth the cost. The average person would just wear it until it dies and buy a new one. This period of time is often decades ! If however you have that $20,000 watch, you are more likely to see it as an investment whose value you wish to protect. the expensive watch will eventually wear out as well, but lavishly cared for it could last for centuries.What I would suggest is that you learn how to do this yourself!

Share this post


Link to post
Share on other sites
  • 0

Hi Yankeedog,

thank you for your help!  :)

The SEIKO SNK809 has been discontinued already. As said: I like to preserve things and repair them if it needs to.
Assuming, cost will not matter (which does of course...but just for this moment it will not) I will come back to my
initial question: What is -- technically wise -- a reasonable period of time to let  a watchmaker check the heart of
this little gem? Servicing too late is always much more expensive than doing it just in time.

Cheers
mcc

 

Share this post


Link to post
Share on other sites
  • 0

You're asking in impossible to answer question?

Impossible to answer because the materials used to make the watch were they designed to last forever? I know you're not asking forever but how long do the rubber gaskets last before they let moisture in. Moisture is a great source for generating rust which is a grinding compound. The lubricants how long were they meant the last. Even the metal that they made the watch out of a lot of times Seiko's use softer metal than they really should things have a habit of wearing out faster. 

 

 

 

Share this post


Link to post
Share on other sites
  • 0
On 2/12/2019 at 10:38 PM, yankeedog said:

As far as interval of service is concerned , I am of the opinion that you have to consider the cost versus the benefit.If for example  a new watch costs $100.00 and a full service costs the same , most people are apt to by a new watch.

An intermediate approach when a full teardown service would cost more than the watch is to just replace the movement.  If you have the Seiko 5 (7s26 or 36 movement) then for really just a few dollars more you can upgrade to a similar one with hacking and the option of hand winding (still a self winder as well).

 

Even the high end Swiss brand Jaeger LeCoultre used to advertise movement swaps so you could get your watch serviced and ready to go in a few minutes.

Edited by Doninvt

Share this post


Link to post
Share on other sites
  • 0
On 2/13/2019 at 5:30 AM, mcc said:

What is -- technically wise -- a reasonable period of time to let  a watchmaker check the heart of
this little gem? Servicing too late is always much more expensive than doing it just in time.

Normally 5 - 7 years between service is recommended. And normally, most owners (no matter the value of the watch) double that interval, or even don't get any service at all, without suffering dire consequences. A Seiko 7S26 can be running just fine after 15 years, or show signs of trouble after few months. It's a bit of a lottery with it being the most mass produced mov.t.

Share this post


Link to post
Share on other sites
  • 0

Instantaneous inaccuracy integrated over a 24hr period can amount to a given value through infinate possible paths, since each piece is subjected to different condition, such as life style, altitude, temperature, etc. Typical of random distributions.

Gausian analysis applies to such behaviour of a movement, therefrom, standard deviation or other indicators is calculable or ascertainable digitally.Standard deviation for seiko movements of a certain caliber has certainly be calculated at seiko. 

It is a subject of watch design.

Regards 

 

 

Share this post


Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Answer this question...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Loading...

×
×
  • Create New...