My take on watch accuracy

Lately, we have been discussing a lot about Seiko accuracy problems, and the 6R35 movement is always mentioned.

In my case, I have only one 6R35, in my SPB207J1, and while it might not represent anything other than my own experience I think it is still as valid as any other sample watch.

So here's my experience:

I tested it positive for magnetization not long ago and I demagnetized it (I had 2 other vintage Seikos that were also magnetized).

I have started tracking the accuracy through a mobile app, I am measuring the accuracy twice a day, mornings and evenings. This takes some time and also keeping the automatic watches running with a reasonable power reserve, so I have to manually wind the ones I have not worn that day.

The 6R35 is the one with the longest power reserve, 70 hours, while the rest are all in the 40 hours.

What I have noticed so far is that the 6R35 varies "more" depending on if I have worn it (movement throughout a day and got fully wound) or if the power reserve is low.

The King Seiko from 1971 is more stable, the Citizen a bit less.

None of these watches have been regulated by me (the King Seiko is second-hand so I don't know if has ever been regulated).

Here is a screenshot from the app:

Image

You can see from the screenshot, that the MM200 after 28 measurements is doing an AVERAGE of +0.85 spd, with the Citizen (400 euros watch) doing -1.48 spd with 92 measurements (awesome in my opinion). The King Seiko from 1971 is doing +0,51 spd.

Does this mean every day they make the same difference? No, they vary each day (again the 6R35 more than the rest).

If I drill down on the MM200 registers, you see there are differences between 24 hours (morning to morning, evening to evening) of +4, -4, -10, -2, etc. but I look at the last streak, I set the time on the watch on 23/09, and today, 5 days later it is -7 seconds behind, so the average is -1.38 spd.

Image

The King Seiko oscillates from -3, +2, 0, +1, so since the last time I set it on 17/09, it has lost 4 seconds with an average -0.46 spd.

Image

So in conclusion, if you want to track accuracy you have to do it regularly like this, not just one day or two, you might be as surprised as I am with the "long" term result.

Also, for reference, here is the definition of accuracy from Grand Seiko website:

Mean daily rate is a mean value of daily rates in a condition where the movement before assembly in a case is measured in six positions in a fixed manner under artificially controlled environment for 12 days.

Reply
·

I have the 6r35 in an Alpinist. I did buy it as a weekend watch so didn't notice the shocking accuracy at first. At the moment its over 90 seconds plus a day. Its an expensive watch for me, I will get it regulated then its off to ebay and no more Seiko mechanics for me unless cheap.

·
weedge

I have the 6r35 in an Alpinist. I did buy it as a weekend watch so didn't notice the shocking accuracy at first. At the moment its over 90 seconds plus a day. Its an expensive watch for me, I will get it regulated then its off to ebay and no more Seiko mechanics for me unless cheap.

Have you tested with a compass? Maybe it's magnetized.

·
antoniogcar

Have you tested with a compass? Maybe it's magnetized.

I haven't tbh and don't have a compass or demagnetiser, another thing on the list !

·

I have an SARB035 that runs at -16spd. I set it a minute fast and live with it. I rotate watches frequently so it's no big deal.

I had a gen. 3 Monster that was -8spd and sold it, but not for that reason.

·
weedge

I haven't tbh and don't have a compass or demagnetiser, another thing on the list !

I didn't have a compass, so first I tried some mobile phone apps that detect magnetic fields using the phone sensors, but I couldn't understand the readings, so I went on to the Internet and found how to build a makeshift compass using a needle, a glass of water and the magnetic seal of the fridge to magnetize the needle, like a school science project .

It worked as a charm, I could clearly make the needle turn by moving the watch around the glass, and it only happened with three of my watches and not the rest.

So I bought the demagnetizer, compass included, for around 12 euros. Fixed the issue in two minutes with the demagnetizer.

·
antoniogcar

I didn't have a compass, so first I tried some mobile phone apps that detect magnetic fields using the phone sensors, but I couldn't understand the readings, so I went on to the Internet and found how to build a makeshift compass using a needle, a glass of water and the magnetic seal of the fridge to magnetize the needle, like a school science project .

It worked as a charm, I could clearly make the needle turn by moving the watch around the glass, and it only happened with three of my watches and not the rest.

So I bought the demagnetizer, compass included, for around 12 euros. Fixed the issue in two minutes with the demagnetizer.

It has to be worth a try ! I think I will just stump up for a demagnetizer anyway. Thanks for the advice. I later found out that Seiko do not regulate at the factory. I didn't expect this on a 5 but on a close to £700 watch I did. I am a Seiko fan but this has put me off them a lot.

·

I have 2 6r35 movement watch's

1 is running at +1sec per day, the other -4sec per day

No issues here

·

Thanks for sharing this! 👍

·
weedge

It has to be worth a try ! I think I will just stump up for a demagnetizer anyway. Thanks for the advice. I later found out that Seiko do not regulate at the factory. I didn't expect this on a 5 but on a close to £700 watch I did. I am a Seiko fan but this has put me off them a lot.

My other two watches that were magnetized were the King Seiko and my 1977 Seiko Panda.

I have owned the Panda for 14 years and never crossed my mind the idea that it could be magnetized. I have never tracked its accuracy, it could be 30 spd and I wouldn't have minded. I may start tracking it now, but the watch needs a service after all these years anyway, so I will ask for it to be regulated as best as possible.

The King Seiko, on the other hand, I suspected it was magnetized during shipment because the seller sent me a video of the watch on the timegrapher before shipping it (wise guy) and it was doing 2spd, so when I got it and it was doing +12 or more during the first days I had it, something was clearly wrong with it, so I did the test.

Here's a video of one of the YouTubers I follow, where he demonstrates the effect of magnetization in real-time and how to use a demagnetizer properly https://www.youtube.com/watch?v=TnGbbyGkYUw

·
weedge

I haven't tbh and don't have a compass or demagnetiser, another thing on the list !

We will not comment on the fact that you own an Alpinist and do not own a compass...

That's like owning an MM200 and not owning a wet suit... Oh wait! 😂

·
weedge

I have the 6r35 in an Alpinist. I did buy it as a weekend watch so didn't notice the shocking accuracy at first. At the moment its over 90 seconds plus a day. Its an expensive watch for me, I will get it regulated then its off to ebay and no more Seiko mechanics for me unless cheap.

Wait untill you spend £5.5k on an Omega like I did and it spent more time being repaired under warranty than on my wrist, I would happily take a robust movement anyday

·

Accuracy is just not so important to me as it was before I had a phone, computer, a few G-Shocks with Multiband 6 as all of these are super accurate. When I pull out a mechanical, I usually set it then and if it stays within 5 minutes a week, that is good enough as it is already back in the box and I am wearing something else. If I only had one watch, I would care more.

·
weedge

I haven't tbh and don't have a compass or demagnetiser, another thing on the list !

You can download magnetic detection apps for your phone for free. A demagnetizer is like $15 on Amazon.

Try this before spending more money on a watchmaker and then selling that same watch.

·
antoniogcar

My other two watches that were magnetized were the King Seiko and my 1977 Seiko Panda.

I have owned the Panda for 14 years and never crossed my mind the idea that it could be magnetized. I have never tracked its accuracy, it could be 30 spd and I wouldn't have minded. I may start tracking it now, but the watch needs a service after all these years anyway, so I will ask for it to be regulated as best as possible.

The King Seiko, on the other hand, I suspected it was magnetized during shipment because the seller sent me a video of the watch on the timegrapher before shipping it (wise guy) and it was doing 2spd, so when I got it and it was doing +12 or more during the first days I had it, something was clearly wrong with it, so I did the test.

Here's a video of one of the YouTubers I follow, where he demonstrates the effect of magnetization in real-time and how to use a demagnetizer properly https://www.youtube.com/watch?v=TnGbbyGkYUw

Ohh Noo, I am magnetized😂, If that is your biggest problem, you are so very blessed.

·

What’s your take on accuracy? Anyway I have a few Seiko 6R movements and they all run well within spec. I don’t have a timegraph or an app, and even if I did I don’t think it’d tell me anything. I don’t need my watches to be quartz accurate, if I did I’d get a quartz watch, but I like them to be reasonably accurate for a mechanical watch. There are so many comments disparaging Seiko’s 6R movement’s that I do check them often, because in my experience they’re pretty spot on, chronometer level accuracy or precision as the guy that posted earlier, that what we are discussing is actually precision not accuracy.

Whatever. So earlier this month I got my Seiko Presage GMT, wound it up to full power (it has a power reserve indicator, love it) set it 1 minute faster than my reference Hodinkee app watch. Checked it a couple weeks later, and it had lost 10 seconds. Checked it again tonight, I think it’s now 11 seconds slower. I think that’s pretty damn good. My SPB 187 is very similar. As far as magnetization, a myth.

Image
Image
·

Be sure to check in various positions. Most movements run at different rates in different positions -- especially seikos. By being mindful of each watch's characteristics, and resting/storing them accordingly, I'm generally able to keep them within a second or two. If you have one that runs fast or slow in all positions, it might be time to break out the tools and regulate it.

·

Totally agree with @foghorn. I have two Seiko 5 watches with the 7S26 movt. One is approx. +15s/day, the other is -15s/day. Both regulated by the watchmaker. My 1970 Seiko with the 6106C movt is -25s/day. Finally, my 4R35 dive watch is at +2s/day regulated.

I gave these watch info because:

1) As you know, all mechanical watches will always be inaccurate, and you live with its limitations. The manual that came along with the watch specifies clearly a standard deviation the movement will have (e.g +30s to -25s/day). If accuracy is a big hitter in your selection criteria (e.g 50%, etc), then look for that info before purchase. Or, just get a quartz watch. Otherwise, it's a futile endeavor to expect a mass produced mov't to perform like a limited edition or COSC certified expensive mov't.

2) I see the accuracy rate of a watch as part of its personality. If you have kids, you'll know what I mean. Each one has his/her own quirks. The +15s/ day can be the "choleric" personality, while the -15s/ day is the "phlegmatic".

My almost vintage Columbia field watch had a VJ-32 quartz mov't inside (VJ quartz I think are manufactured by Seiko too, ala the NH line for mechanical mov'ts). When I brought it to the watchmaker for repairs in Feb 2022, he installed a new VJ-32 mov't which cost me only $23 total incl. the service. To this day, it is at -1s per 90 days! Hands down this unheralded super cheap VJ quartz mov't beats most, if not all, more expensive and hyped quartz releases, including the Seiko pieces mentioned above, Tissot, Citizen, etc. It also handily beats most, if not all, quartz watches by luxury brands. Even the touted GS Spring Drive cannot compete as it only has a daily rate of +8/-1 seconds (Mechanical), and a yearly rate of ±10 seconds (Quartz).

https://www.grand-seiko.com/us-en/collections/movement/springdrive

As such, I set all my mechanical watches according to my Columbia field watch, and get a good picture of their deviation. No need for free or paid mobile apps, which are also highly inaccurate because of the magnetism a mobile phone generally gives off. Do I use an atomic clock to calibrate my VJ-32? Yes, every 90 days. ;)

3) On a practical basis, perhaps 95% of jobs, and 99% of after-work lifestyles, do NOT demand 0s/day time keeping accuracy. If it were the case, then let's all get atomic wrist watches, and demand all brands to produce such. Time is priceless, precious, and fixed, yes. But I don't pressure myself to live and work on a per second accuracy basis.

Rotate your watches. Set them when necessary. Then above all, just enjoy them on wrist.

Image
·

Which timegrapher application you been using?

·

Seems like some people get good Seiko movements and some bad. I do not expect great accuracy with a mechanical but I do expect reasonable variation. That's why my everyday is a Casio Oceanus, it beats a Grand Seiko Quartz every time, but mind you have a £30 Casio with mutiband that spot on all the time, month in and out.

·
martin1111

Which timegrapher application you been using?

This is not a timegrapher application, in this application you record the time on your watch in any moment and the application calculates the difference with the real current time obtained from the network. This one is called "Atomic clock and Watch Accuracy", for Android phones.

I have used a timegrapher application called Watch Accuracy Meter which is quite simple to use, although I do not have a timegrapher to compare with,it did show the correction on the watch rate after it was demagnetized.

·
weedge

Seems like some people get good Seiko movements and some bad. I do not expect great accuracy with a mechanical but I do expect reasonable variation. That's why my everyday is a Casio Oceanus, it beats a Grand Seiko Quartz every time, but mind you have a £30 Casio with mutiband that spot on all the time, month in and out.

I have all kinds in my collection: automatic, quartz, radio-controlled, solar and a vintage high accuracy Seiko Grand Quartz, which I am also tracking to see how it does.

Although I'd like all of them to be as accurate as possible, I know I can only expect them to behave within their stated accuracy, which for Seiko seems to be a greater margin than necessary, probably as others have mentioned to prevent guarantee claims.

In your case, 90 seconds is disproportionate, way out of the margin stated by Seiko, so I would suspect something is wrong with the watch outside not being factory regulated, that's why I asked if you checked it for magnetization. Maybe it got bumped in transportation to the dealer, who knows. That watch should have been taken under guarantee by Seiko.

I understand you have a bitter taste because that bad accuracy, I would have it fixed and see if I enjoy the watch then. It shouldn't be too expensive to have it regulated.

·
antoniogcar

I have all kinds in my collection: automatic, quartz, radio-controlled, solar and a vintage high accuracy Seiko Grand Quartz, which I am also tracking to see how it does.

Although I'd like all of them to be as accurate as possible, I know I can only expect them to behave within their stated accuracy, which for Seiko seems to be a greater margin than necessary, probably as others have mentioned to prevent guarantee claims.

In your case, 90 seconds is disproportionate, way out of the margin stated by Seiko, so I would suspect something is wrong with the watch outside not being factory regulated, that's why I asked if you checked it for magnetization. Maybe it got bumped in transportation to the dealer, who knows. That watch should have been taken under guarantee by Seiko.

I understand you have a bitter taste because that bad accuracy, I would have it fixed and see if I enjoy the watch then. It shouldn't be too expensive to have it regulated.

Image

Yes I will get it regulated. I am not wealthy so it took some time to save for so it has left me bitter, but hey ho these things happen. With negative thoughts about the watch I will take a loss and let it go. I got the watch as a "best" watch so only wore it at weekends so I did not notice at first. Worst things happen in life other than a wonky watch so all in all live and learn.

·
antoniogcar

This is not a timegrapher application, in this application you record the time on your watch in any moment and the application calculates the difference with the real current time obtained from the network. This one is called "Atomic clock and Watch Accuracy", for Android phones.

I have used a timegrapher application called Watch Accuracy Meter which is quite simple to use, although I do not have a timegrapher to compare with,it did show the correction on the watch rate after it was demagnetized.

Thanks for response. I know about Watch Accuracy Meter I have that installed on my phone but I never tried Atomic clock and Watch Accuracy app. I will try that app

·
antoniogcar

I didn't have a compass, so first I tried some mobile phone apps that detect magnetic fields using the phone sensors, but I couldn't understand the readings, so I went on to the Internet and found how to build a makeshift compass using a needle, a glass of water and the magnetic seal of the fridge to magnetize the needle, like a school science project .

It worked as a charm, I could clearly make the needle turn by moving the watch around the glass, and it only happened with three of my watches and not the rest.

So I bought the demagnetizer, compass included, for around 12 euros. Fixed the issue in two minutes with the demagnetizer.

Great compass, although not the easiest thing to carry around in the woods.